Table of Contents
7 Mrs Mac and her girls in green
10 The Special Intelligence Bureau
15 The secrets of the Coral Sea
31 Pappy Clark goes to Washington
43 The Garage: a cryptological love story
THE SECRET CODE-BREAKERS OF CENTRAL BUREAU
David Dufty is a Canberra-based writer and researcher. He completed a psychology degree with honours at the University of Newcastle, has a PhD in psychology from Macquarie University, and has worked as a statistician and social researcher at the University of Memphis, Newspoll, and the Australian Bureau of Statistics. His previous book, How to Build an Android, described modern developments in robotics and artificial intelligence.
Scribe Publications
18–20 Edward St, Brunswick, Victoria 3065, Australia
2 John St, Clerkenwell, London, WC1N 2ES, United Kingdom
First published by Scribe 2017
Copyright © David Dufty 2017
All rights reserved. Without limiting the rights under copyright reserved above, no part of this publication may be reproduced, stored in or introduced into a retrieval system, or transmitted, in any form or by any means (electronic, mechanical, photocopying, recording or otherwise) without the prior written permission of the publishers of this book.
The moral right of the author has been asserted.
9781925322187 (Australian edition)
9781911344711 (UK edition)
9781925548198 (e-book)
CiP entries for this title are available from the National Library of Australia and the British Library.
scribepublications.com.au
scribepublications.co.uk
To my parents, Don and Barb Dufty
Contents
Prologue: Names in the dirt
1939
1940
3. Code Club
4. Sailor Jack
1941
7. Mrs Mac and her girls in green
10. The Special Intelligence Bureau
11. East wind, rain
12. West wind, clear
1942
14. Beirut bookshops
15. The secrets of the Coral Sea
16. Cranleigh
18. Traffic
19. Leaks
20. Nyrambla
21. Wet boots
1943
25. Killing Yamamoto
26. Noon positions
27. Kaindi
29. Cartwheels
1944
30. The Sio box
31. Pappy Clark goes to Washington
32. Hollandia
33. The Yoshimo Maru
34. Biak
1945
38. Spies
39. Lingayan Gulf
42. Tarlac
43. The Garage: a cryptological love story
48. Radio silence
1946 and beyond
49. Secret medals
Abbreviations
ADVON Advanced Echelon. Central Bureau in Port Moresby and later Hollandia.
ANZAC Australian and New Zealand Army Corps
ASIP Australian Special Intelligence Personnel
ASWG Australian Special Wireless Group
AWAS Australian Women’s Army Service
COIC Combined Operational Intelligence Centre
FECB Far East Combined Bureau
FRUMEL Fleet Radio Unit, Melbourne
FRUPAC Fleet Radio Unit, Pacific
GCCS Government Code and Cipher School
JN-25 The main Japanese naval code system
WAAAF Women’s Auxiliary Australian Air Force
WRAN Women’s Royal Australian Navy
PROLOGUE
Names in the dirt
About 10 years or so after the end of the Second World War, in the sleepy Brisbane suburb of Chermside, a boy named Peter Hill noticed the chooks in his family’s back yard scratching at objects in the dirt. When he inspected what they were scratching at, he found several small, rusted metal disks. Some of the disks had names, as well as the ranks of military officers, on them. Others were corroded beyond recognition. He searched for more of the buried disks, found many that the chooks had not located, took them inside, cleaned them up, and arranged them in an album as a personal collection.
About 55 years later, Peter Hill donated them to the local historical society, which was baffled by the disks, what they meant, and why they had been buried at Chermside.1 The disks belonged to military personnel, but that was all they seemed to have in common. Some were from the army, some from the navy, and even those from the same service had different unit memberships. The ranks were all over the place, including privates, corporals, captains, and lieutenants. The highest-ranking disk was that of Lieutenant-Colonel Alistair Wallace Sandford. According to the identity disk, the colonel had two units, Central Bureau and a second unit that seemed to say ‘ASIPS’.
Some of the other names were Victor Lederer (Captain, Central Bureau, born 1914), Don Laidlaw (Captain, Central Bureau, born 1923), and John Lovell (private, Central Bureau, born 1923). Who were these people, and what was Central Bureau? Who buried all their identity tags, and why?
The answers are even more interesting than the questions. The people whose wartime identities were buried in the ground in Chermside had been involved in a large, secret intelligence operation against the Japanese army in the Second World War that the United States navy was later to declare had been ‘of immeasurable importance’ in the successful prosecution of the war.
When the Second World War started in 1939, Australia had little expertise in signals intelligence, the field of espionage devoted to intercepting and making use of the enemy’s own communications systems to learn their plans and movements. There were no code-breakers, traffic analysts, or radio-security experts in Australia — in part because Australia was a backwater in the international espionage game, and in part because Australia relied on Britain to do the dirty work of intelligence on its behalf.
By the end of the war, there were several signals-intelligence organisations based in Australia, all secret, and all working at full capacity against the Imperial Japanese forces. The largest and most active of these was Central Bureau, which was based in Brisbane but had tentacles across northern Australia, the Pacific Ocean, and South-East Asia, and as far as Luzon in the Philippines. Others were known as the Australian Special Wireless Group (ASWG), Fleet Radio Unit Melbourne (FRUMEL), Special D Section, and Section 22.
This book tells the story of the emergence of signals intelligence in Australia during the Second World War, and of the men and women who worked for Central Bureau and other signals organisations based in Australia. The activities of those people remained secret for decades, and were never recognised by the Australian government. During the post-war decades, they were all bound by the Official Secrets Act and forbidden from telling anyone what they had done during the war. They did not participate in annual ANZAC Day marches for 30 years. For decades, there was no official acknowledgement of their vital contribution to the war effort.
This intelligence network was made possible due to the help of the Allied powers, particularly the United States and Britain, who had both been developing their expertise in signals intelligence for decades.2 Until now, the Australians who contributed have been neglected and ignored, and their incredible stories and achievements have remained unknown.
1939
1
The fires of a distant war
The Second World War began far from Australia, on a moonlit autumn night in Poland. Before dawn on the morning of 1 September 1939, a bomb fell from the sky onto the Polish town of Wieluń, exploding on the All Saints Hospital. More bombs followed, also hitting the hospital, killing 32 people. Air-raid sirens sounded as the bombs fell like rain, destroying almost all of the town’s buildings and killing more than 1000 residents.
The bombs were being dropped by the Luftwaffe, the formidable German air force, under the command of Field Marshal Hermann Goering, as part of a coordinated series of attacks by Germany on Poland. The 32 patients and staff who died in the All Saints Hospital were the first people killed in the Second World War — a war that would ultimately cause the deaths of more than 60 million people and wreak unprecedented devastation.
There were no Polish troops in Wieluń, no heavy industries, no major transportation routes. There was nothing of military value there — only the sleeping residents, woken by the sounds of explosions and air-raid sirens. The first deaths in the bloodiest war in history were utterly pointless, even from a military strategic perspective.
Within minutes of the air raid on Wieluń, the German battleship Schleswig-Holstein, anchored in the harbour near the Polish military garrison of Westerplatte on the Baltic Sea, started shelling the Polish outpost. German naval infantry troops landed and attempted to storm the garrison. The fighting at Westerplatte continued into the days ahead until the garrison was seized.
As the sun rose on the first day of September, the German army began a full-scale invasion of Poland with one million troops. Two days after the start of the invasion, Poland activated its alliance with the nations of Britain and France, which both declared war on Germany and initiated a naval blockade of German shipping. Australia, New Zealand, and India declared war on Germany within hours of Britain doing so.
The German army reached the Polish capital, Warsaw, in two weeks. The day after the Germans encircled the city, Russia began its own invasion of Poland from the east. Russia and Germany had previously entered into a secret non-aggression pact, and as part of that pact had made a deal to carve up Poland between them.
Germany was in the grip of a wave of nationalistic fervour and a cult of personality around its charismatic leader, Adolf Hitler. Fascism was a political movement sweeping Europe that emphasised nationalism, military strength, and racism, particularly against Jews. Hitler’s Nazi party took fascism to its limits. Blaming Jews and bankers for many of the problems of ordinary German people, he promised to restore Germany to the glory it had enjoyed before the First World War. The Nazi party had its own army, a paramilitary organisation called the Schutzstaffel, usually known as the SS. Hitler also established a secret police force, the Gestapo, to locate and imprison dissidents. And now, with the invasion of Poland, Hitler had set Germany on the path to greatness once more.
Hitler was friendly with the leaders of other fascist nations in Europe and elsewhere, but there were two nations in particular that he bound tightly with Germany in a series of agreements: Italy and Japan.1 Italy had been ruled by the fascist dictator Benito Mussolini since 1929. Mussolini’s volunteer militia, the ‘Blackshirts’, persecuted political opponents, as did his secret police force. Japan’s prime minister at the outset of Germany’s invasion of Poland was Nobuyuki Abe, who had taken the position the day before. Japan’s government was engaged in a series of crises while the Imperial Japanese Army consolidated its control over the nation under General Hideki Tojo. Under Tojo’s leadership the army was engaged in military expansion of its own in East Asia.
The three nations were the core of Hitler’s network of alliances known as the Axis.
In opposition to the Axis were Britain, France, Australia, Canada, and other nations, known as the Allies. The United States would be forced into the war two years later with Japan’s attack on Pearl Harbor; but, for now, as bombs fell on Wieluń and German tanks rolled across the Polish border, America did not take sides.
Beneath all the upheaval of the beginnings of war, as troops marched by the millions over borders, as leaders of nations were discarded and new leaders arose, there was another war — a hidden war for inside intelligence and information — that had been underway for years.
A short drive out of London in the English countryside there is a large old manor named Bletchley Park, which at the time was the home of the Government Code and Cipher School (GCCS), a secret organisation dedicated to breaking the codes and ciphers of other nations and reading their most confidential messages. There, a team led by the brilliant Alan Turing were trying to crack high-level German diplomatic and military messages enciphered with a machine known as ‘Enigma’. An ingenious series of cogs and wheels enciphered each message in such a way that it could only be read with the help of another Enigma. The team had intercepted plenty of Enigma messages sent to and from German diplomats and commanders, all of which were unreadable. If Alan Turing’s code-breakers could read them, the British and their Allies would glean important secret information and gain an edge in the war.
The code-breakers at Bletchley Park also toiled to break lower-level codes of the German army, navy, and air force, as well as the codes and ciphers of Russia, Italy, and other nations.
There was a Japanese section, too, devoted to breaking and intercepting Japanese military and diplomatic messages, many of which were collected from radio transmissions in East Asia, which they called the ‘Far East’. The GCCS had a subsidiary organisation in Hong Kong, the Far East Combined Bureau, that had been intercepting Japanese military radio transmissions and breaking Japanese codes for over 10 years, thanks to the work of an Australian naval officer, Eric Nave.
But with war erupting across Europe, work on Japanese codes was naturally a low priority, and the Japanese section at Bletchley Park was small compared to the large teams attacking German, Italian, and Russian codes and ciphers. Decrypting the messages of Japanese generals was less important than trying to read Hitler’s mail.
2
Tropical sickness
Four days before the invasion of Poland, a British cruiser, the HMS Birmingham, sailed into the Seletar naval base in Singapore. The Birmingham had come from Hong Kong, Britain’s other major military and trading hub in East Asia, bringing with it almost the entire workforce of a secret British intelligence unit, the Far East Combined Bureau. The unit had evacuated from Hong Kong because the nearby war between China and Japan was too close for comfort and had become too much of a threat.
Japan was not at war with Britain — not yet — but was engaged in a protracted war with China, and was winning. The Japanese army had recently invaded Canton, the Chinese province next to Hong Kong, the Spratly Islands just to the south, and Hainan to the west, so that Hong Kong itself was now an isolated island surrounded by Japanese-occupied territory.
The Far East Combined Bureau had for some time been monitoring Japanese radio and diplomatic messages.1 Like Singapore, Hong Kong was British territory and was unlikely to be attacked by Japan, but even a small risk of the bureau being captured had been too much for the authorities to accept. Such a capture would enable the Japanese to learn exactly how much progress the British had made in signals intelligence and in breaking Japanese codes. It would send the British code-breakers back to square one.
One of the passengers who disembarked from the Birmingham was the Australian Eric Nave, now an officer in the British Royal Navy. Fifteen years earlier, Nave had established Britain’s first radio-interception and code-breaking operation in Asia, working from a cabin on a naval ship in Shanghai. He was the first person to break an Imperial Japanese military code in 1927, and had since broken numerous low-level and medium-level codes. In Hong Kong, aside from radio intelligence, he had become involved in more traditional spycraft activities, and became friends with a British intelligence agent named Ian Fleming — the man who later created the fictionalised spy James Bond.2
Nave, aged forty, was thin, quiet, and neatly dressed. He had been in the navy all his adult life; at first in the Royal Australian Navy, and subsequently the British Royal Navy. In the final year of the First World War he enlisted in the Royal Australian Navy, but was initially rejected (for reasons unknown), and was eventually accepted only on the condition that he sign up for permanent service. This meant that, unlike most young men who enlisted during the war, he wouldn’t be demobilised when it finished. Later, stuck in the navy during peacetime, he studied Japanese, spent two years in Japan attached to the British embassy studying the Japanese language, and returned as the Royal Australian Navy’s leading expert on Japan and its language. He was transferred to the British navy in 1923, and joined the GCCS. Although he was employed by the British, he was the first Australian code-breaker.
When he arrived in Singapore on the Birmingham, Nave was sick, and had been for some time. Earlier in the year in Hong Kong, he had started having bouts of nausea, diarrhoea, cramps, and vomiting. The illness recurred, each time with increasing severity, and he spent several long stints in hospital. As a result, Nave looked far older than his years. The doctors did not know what was wrong with him, and they could not cure him. In the absence of a diagnosis or a cure, and with Japanese naval activity at an all-time high, Nave went back to work.
He still did not have a diagnosis, but the hospital trips had yielded one benefit: he had fallen in love with a nurse, Helena Gray. She was not allowed to travel on the Birmingham, since she was a civilian, but she made her own way to Singapore. Nave married her within a week of his arrival.
Four days later, on the other side of the world, on the first day of September, the Second World War began. Germany, under the leadership of Adolf Hitler and his Nazi Party, invaded Poland with one-and-a-half million troops. German planes bombed Polish towns while 60 German army divisions moved eastward, no match for the relatively small and poorly armed Polish forces. Within days of the occupation, France, Britain, and nations of the British Commonwealth such as Canada, Australia, South Africa, and New Zealand declared war on Germany, and the second Great War had commenced.
As a part of the British Empire, Hong Kong and Singapore were now officially at war — but with Germany, not Japan. A war was on, but it was in Europe, on the other side of the globe. Nave and the other code-breakers and traffic analysts at the Far East Combined Bureau — or FECB, as they called themselves — focussed on Japanese naval movements.
The Japanese consulate encrypted its messages in a code known to the FECB as the ‘LA’ code, because every encrypted message began with the Japanese syllable ‘LA’. The FECB had long ago figured out how the LA code worked, and had developed techniques for breaking new versions. Nave spent most of his time decrypting consular messages in the LA code until a replacement arrived.
A British officer, Hugh Foss, Nave’s mentor and trainer from the GCCS in England, took over the work on consular messages. Like Nave, Foss was fluent in Japanese and was a highly skilled cryptanalyst. With Foss taking care of the LA code, Nave returned to what he knew best: Japanese naval radio messages.
Work on such messages was about to ramp up significantly. In early September, another British agent from the GCCS, Malcolm Burnett, arrived in Singapore. He brought with him important information about the Japanese navy’s new code. The Americans called it JN-25 (short for ‘Japanese Naval Code 25’), but apparently had not managed to break it. Back in England, at Bletchley Park, a team of analysts, including Burnett and led by John Tiltman, had not only figured out how JN-25 worked, but had devised a method for breaking it. The method was neither easy nor quick, but it meant they could chip away at the code, message by message and piece by piece.
Nave and his colleagues could not decrypt JN-25 messages. Not yet. But they understood several other Japanese naval codes, and provided the Singapore commanders with continuing, detailed briefings of Japanese navy activity.
The Far East Combined Bureau was, in theory, just one of many sources of information available to the commanders regarding foreign powers such as Japan. But British intelligence was so meagre and poorly run in East Asia that it was almost worthless. As a result, the code-breakers were in reality the main source of reliable information for the British. Unfortunately, most British military commanders in Asia did not understand how good the signals-intelligence analysts like Nave were. They did not fully appreciate the value of the intelligence they got, and did not pay sufficient heed to their warnings.
It was clear to everyone at the bureau monitoring the daily radio traffic, and as a result completely immersed in the day-to-day activities of the Imperial Japanese Navy, that the threat of Japan was real and growing. Japan’s military dominance of East Asia was almost total and was expanding southwards, ever closer to Singapore.
The bureau picked up information about a landing of Japanese troops on the coast of Cambodia that had just taken place, but were not sure of the exact location. Nave and the commanding officer of the FECB, Captain Wylie, decided that, at a minimum, this warranted reconnaissance flights around the area, to try to find out more about what the Japanese army was up to. Where exactly had the army landed? It would be good to know.
Nave and Wylie visited the air marshal. They explained what they had discovered, and suggested he send a reconnaissance plane over to Cambodia. The air marshal replied that he would like to do so, but alas, there were no planes available. He then added that, yes, there was one plane available, but unfortunately it was far too slow and ‘would probably be shot down’.
No planes were sent.
The military establishment recognised that Japan was a threat, but in an abstract, ‘we’ve seen all this before’ kind of way. The British Empire was at the height of its extent and powers. After four centuries of colonisation, expansion, and conquest, it was the most extensive empire that the world had ever seen. It was a supremely prosperous military superpower with interests across the world, including the wealthy trading ports of Singapore and Hong Kong, which it had transformed over the decades from small, undeveloped harbours to trading powerhouses. It had seen rivals come and go.
In short, the British military establishment was complacent. They knew that Japan was conquering China and nearby territories, but China’s army in 1939 was small, and poorly trained and armed. They felt that Japan would not dare take on the British Empire, and if it did, it would lose.
One officer in Malaya, the British territory to the north of Singapore, was preparing his troops for the possibility of a future war with Japan. He said to Air Chief Marshal Sir Henry Brooke-Popham, the head of British military in Singapore, ‘I do hope, Sir, we are not getting too strong in Malaya, because, if so, the Japanese may never attempt a landing.’ Heaven forbid that his soldiers not have an opportunity to face those upstarts from the north. Another regimental commander was famously heard to say of his troops, ‘Don’t you think they are worthy of some better enemy than the Japanese?’3
Singapore was multicultural, but it had a stratified caste society, with Europeans, and particularly the British, at the top. All British residents had servants, regardless of their class background back home. The British spent their days working in downtown offices, and their evenings attending the cinema, going for night-time strolls along the harbour, and socialising at exclusive venues such as the legendary Raffles nightclub. Singapore’s Chinatown, by contrast, was a slum teeming with the activity of markets, rickshaws, and alleyways.4
For Nave and his colleagues at the FECB, the mood around them jarred with the shifting geopolitical landscape to the north. These changes were barely noticed by the local population, with their light-hearted optimism about the empire that protected them. At the FECB, poring over the constant stream of radio transmissions, team members were immersed in their study of the rise of a new and very different empire. They spent their days tracking army and navy units around Asia and the Pacific Ocean. It was not any single radio signal, or the contents of any specific message, that caused them alarm, so much as the overwhelming weight of the patterns of movement, the tone of the correspondence, the arrival of new ships and new army units, and the sheer amount of activity that seemed to increase with every passing week. The conclusion was inescapable.
Nave kept working, but his illness was wrecking his body. He was thin and weak, and in pain. He was admitted to Johore Hospital, where he was subjected to a litany of tests. After questioning Nave at length, the hospital’s senior surgeon made a diagnosis: ‘No doubt about it. Sprue.’5 There was no cure for sprue, and his condition was chronic and deteriorating. Normally, someone as sick as this would be ‘invalided’ and sent home, but Nave couldn’t be replaced, so he had to stay.
The head of the bureau, Captain F. J. Wylie, compromised and gave him permission to take three months’ leave in Australia. 6 Helen and Eric Nave boarded a ship bound for Australia in January 1940, hoping that Nave’s holiday might be more permanent than Captain Wylie intended.
1940
3
Code Club
In January 1940, Professor Thomas Room, head of the mathematics department at the University of Sydney, received an invitation to meet confidentially with the army’s chief of staff at the nearby Victoria Barracks. When he arrived for the meeting, the chief, Lieutenant General Ernest Ker Squires, made a surprising and curious suggestion. Squires asked the professor if he would like to start up a code-breaking club with some of his fellow academics at the university. Squires added that it would be better to meet at the barracks than on campus. It would draw less attention, and if they made much progress the army could offer certain resources.
Squires knew nothing about code-breaking himself, and so could offer little detail about what Room should do next, other than to impress upon him that the army wanted someone local to learn code-breaking. Although Room was a mathematician, cryptanalysis — the discipline of breaking codes and ciphers — was hardly his speciality. His thing was geometry.
Thomas Gerald Room was a friendly but somewhat shy man who always spoke in a formal, correct manner, even in casual settings. He had arrived in Australia from Cambridge, England, five years earlier to take up a professorship at the University of Sydney’s mathematics department, which at the time comprised five staff, including two temporary assistant professors.1
A devout Christian, he joined the Student Christian Movement, and through that campus club met Jessica Bannerman, an Australian arts graduate. They married in 1937 and bought a house in St Ives on the city’s north shore. In 1938, he published a book, The Geometry of Determinantal Loci, a magnum opus that had taken him several years to write.
Professor Room pondered who among his colleagues he should approach to start a weekend code-breaking club. After some consideration, he invited only one other faculty member — his close friend Dicky Lyons. Since moving to Sydney, he and Richard Lyons had formed a friendship around two common interests: mathematics and sailing. They spent their summer weekends together yachting on Sydney harbour.2 It was a somewhat uneven friendship because Lyons, a less accomplished scholar, was in awe of Room’s mathematical talent and achievements.
The two academics went to Victoria Barracks, where they were allocated a private place, and started teaching themselves cryptanalysis. The university library had a copy of The Black Chamber, their most likely starting point.
The Black Chamber was written by Herbert Yardley, a code-breaker for the United States navy until 1929. Yardley’s book described in detail how his United States’ code-breaking unit — dubbed ‘The Black Chamber’ — decrypted Japanese diplomatic messages during the Washington Peace Conference of 1921 and by doing so learned Japan’s negotiating strategy at the conference. The result was that the final deal restricted Japanese naval expansion to the greatest degree possible, tethering them to a ratio of 5:5:3 against the navies of the United States and Britain. Yardley’s book provided numerous insights into the workings of codes and ciphers, as well as its role in modern-day intelligence.3
In 1929, the United States’ president, Herbert Hoover, appointed Henry L. Stimson as secretary of war. When shown the contents of decrypted diplomatic messages of other nations, Stimson was appalled. To his sensibilities, the work of Yardley’s Black Chamber was underhanded, secretive, and immoral, and not the sort of thing that civilised nations did. Stimson ordered that the Black Chamber be closed down immediately because, in his words, ‘Gentlemen do not read each others’ mail.’
Turfed out of his job, Yardley got his revenge with the publication of The Black Chamber, blowing the lid on American cryptanalysis activities against Japan in the process, and providing the ideal introductory textbook to the art of code-breaking for two mathematics professors in Sydney.
After learning the basics, the two professors were ready to practise and hone their newfound skills. The army’s chief of staff was happy to help, and they were provided with sample material to get them started. At first they were given exercises with simple, low-level codes in Japanese. Once they had sharpened their teeth on these easy problems, the army gave them some messages in the so-called LA diplomatic code.
The professors were not told where these practice messages came from, but the LA code messages weren’t just toy puzzles. They were real Japanese consular messages that had been intercepted by an army station at Park Orchards near Melbourne.
The Australian army had recently made the decision to try to build local expertise in making codes (known as cryptography) and in breaking codes (cryptanalysis). They first tried to do it in-house, establishing a small team of army personnel, but the progress of the unit was slow and disappointing. Their contacts in Singapore at the Far East Combined Bureau suggested that they follow the model of the GCCS in Britain and to try to recruit some ‘professor types’.
Room and Lyons couldn’t speak Japanese, but, since they were dealing with coded Japanese messages, they needed to learn. They took courses in Japanese from Margaret Lake, a lecturer at the teachers’ college on the university campus. Lake was an excellent teacher, and they made rapid progress.4
After their initial forays into code-breaking, and with the encouragement of the chief of staff, they sought to expand their circle, and approached the university’s two classics professors, Trendall and Treweek, to join their secret club.
Athanasius Treweek had been a member of the Sydney University Regiment for years, and was a captain in the field artillery. Treweek had long believed that war between Japan and Australia was inevitable, and had been learning Japanese since 1937 in preparation for it.
Arthur Dale Trendall was a New Zealander who had been in Australia for two years. Fluent in Greek and Italian, he was an expert in fourth-century Greek vases, and had a growing international reputation as a classical scholar.
Trendall was renowned for his uncanny intuitive judgement when inspecting a new classical item. Richard Johnson, in one of several obituaries after Trendall’s death in 1995, said of him,
Trendall had a phenomenal visual memory; he could look at a vase he had never seen before and place it to an obscure painter or school which he had last seen years previously, on the basis of the stylistic similarities. When one considers that about 20,000 vases of the period are known, and he knew every one, the magnitude of the memory may be appreciated. Nor, of course, was it only memory; he had an eye for style and for details of execution which enabled him to produce his great classificatory collections: Paestan, Campanian, Apulian, Lucanian, Sicilian.5
This ability to see patterns and connections in ancient vases seemed to translate rather well to the art of code-breaking. While the mathematicians worked on theoretical solutions, Trendall described this ability years later to the historian Des Ball, telling him, ‘You get a feeling for it. Your eye lights up on something, and … bang.’6
In fact, Trendall considered the intuitive approach to code-breaking to be far superior to the mathematical, analytical approach — in part because real messages were incomplete, or contained errors made either by the sender or the interceptor who transcribed it. These messy, error-ridden messages were called garbles in the cryptanalytical trade. Garbles were a diabolical problem for the mathematicians, but Trendall could quickly see beyond the chaos of the garble to the ordered pattern of the message below.7
They met at the barracks on weekends, where the army provided them with a steady stream of cryptanalytical puzzles, all of which were real, intercepted Japanese messages. Quite quickly, they broke the LA code, a Japanese consular code. Trendall boasted that breaking the LA code was ‘child’s play’.
The chief of staff came to them with a different kind of code-breaking task in September. He explained to them that the army censor had come across suspicious correspondence, and wanted them to investigate. A letter from a British officer stationed in China to the wife of a senior air force officer in Melbourne contained curious patterns of dots on a newspaper clipping that came with the letter. The officer and his wife had recently been living in China themselves, having returned to Australia in July, so it was possible that one or both of them had made foreign connections before returning home. And, adding to the suspicion, the letter had been sent first to another woman in Sydney, who had then forwarded it to the woman in Melbourne. The army was worried that the dots were a code being used to engage in espionage. The chief of staff handed them the correspondence in question and left them to it.
The professors studied the letters and the clipping. There were indeed patterns of dots in the margins, and it was clear that the dots were some kind of code. They dubbed it the ‘Dot Code’. Before long, they had broken it.
The secret message began: ‘Ever think of Cathay? Blissful afternoons with precious naked in my arms.’ It then got quite lurid and explicit before ending with the line, ‘Many kisses, much love.’8
One thing was clear. The secrets of the nation were safe from this pair. They weren’t engaged in espionage; they were having an affair.9
Room recommended that the army let the mail through to the lovers, but that it add a series of dots on the outside envelope in a Dot Code message that said, ‘Careful! — The Censor’.
There is no record of whether the army took the professors’ suggestion seriously.10
4
Sailor Jack
There was a sign saying ‘Artillery’ at the Sydney recruiting depot, with a bench below it. Alan Flannery sat on the bench and waited. He wanted to enlist, and had chosen Artillery because most of the other young men from Manly had enlisted in that particular corps. Flannery expected that after signing up he would join them at the nearby training grounds. It did not work out that way.
After some time, a lieutenant approached him and said, ‘I’m recruiting for Signals today. Would you like to join Signals?’ Flannery’s main concern was that he might not be enlisted into the infantry, so, with no sign of interest from the Artillery recruiters, he said, ‘Yes.’
The officer, Lieutenant Arthur G. Henry, was looking for anyone who knew something about radios or Morse code. Alan Flannery, whose job was to operate movie reels in a local cinema, had listed his profession as ‘operator’. Lieutenant Henry, it seems, had mistakenly thought he was a Morse code operator such as could be found working for the post office; his interview was simply too rushed to disabuse him of the error. After enlisting Flannery into the Third Signals Company, Lieutenant Henry assigned him to Four Special Wireless Section, and ordered him to report to the Melbourne Showgrounds two weeks later. Flannery was dismayed, as he had hoped to remain near Sydney, at least for the duration of his training. There were more surprises in store for him and the seventy-odd other new members of Four Special Wireless Section, because, as they would soon learn, the section was not an ordinary signals unit.
At the Melbourne Showgrounds in Caulfield, the newly enlisted members of the section were introduced to their new commanding officer, Captain Jack Ryan, otherwise known as ‘Sailor Jack’, ‘Uncle Jack’, or sometimes ‘The Skipper’. Ryan informed them that their work was ‘hush hush’, not to be discussed with anyone outside the section, either then or ever. That included other army signals units.
Jack Ryan was, as one person described him, ‘a chunky bulldog of a man’.1 Aged 41, he was a veteran of the First World War, having served as a wireless operator on the HMAS Sydney. Then aged 16, he was the youngest member of the crew when it sank the German raider the Emden in the Battle of Cocos.2 After the First World War, he became chief engineer for Melbourne radio station 3AW, while maintaining a membership of the Naval Reserves.3 Upon the outbreak of the Second World War, he tried to enlist once more, but was rejected as being too old. Instead, he ran training courses for the Signals Training Battalion at Mount Martha, south of Melbourne. In early 1940, an advertisement caught his eye: the army was looking for militia officers who had had experience as naval wireless officers. The advertisement said that they would consider applicants of any age.
And so it was that Ryan was appointed the officer in charge of a new section, Four Special Wireless Section. Two other navy veterans successfully applied and were appointed as his lieutenants: Arthur Geddes Henry and William Fitzmaurice Hill. Before the war, Henry had been in charge of communications for the New South Wales Railways, while Hill had worked for the Australian Broadcasting Commission. (Since television had not yet been invented, the ABC was only involved in radio at the time.)4 It was up to the three of them to locate and recruit suitable members for Four Special Wireless Section. They took men of any age who were willing to serve, and who had either knowledge of or an aptitude for the various skills needed for a signals unit, such as interest in amateur radio (known as ham radio), expertise in electric circuits, a knowledge of Morse code, or fluency in German.
They recruited men across a wide spread of ages. Ron Baines was 16, having lied about his age on the application form to meet the minimum cut-off, while Harry Ackling was over 60, having also lied about his age to meet the cut-off at the other end. John Butler, a public servant with the tax department who also happened to be a proficient Morse operator, recalled how the unit also varied in background: ‘In occupations we had farmers, public servants, businessmen, bank officers, actors, script writers, PMG [Post Master General] telegraphists and linesmen, policemen, wharf labourers and rouseabouts.’5
Stanley Robert Irving Clark was also among those standing to attention before Ryan at the showgrounds. Aged 39, Stan Clark had been selected by the army for this unit because he also worked for a radio station, Melbourne’s 3DB. But unlike Ryan and his lieutenants, he was not an engineer. He was a program manager and scriptwriter, whose passion was modern music. Clark chose the playlists for the radio station, and also wrote a children’s radio show called ‘Teenie Weenies’.6
The younger soldiers nicknamed him ‘Pappy’ because of his prematurely grey hair.7 Despite starting with no technical background, Pappy Clark was to demonstrate a natural flair for signals intelligence in the war years ahead.
They moved to the army’s signals-training camp near Seymour in central Victoria, where they were meant to be issued with standard uniforms. However, due to shortages, there weren’t enough to go around. Instead, many of them were issued with second-hand uniforms from the First World War. John Butler’s great coat even had a hole in it, which he preferred to believe was from a cigarette burn rather than a bullet.8
The mess huts, which doubled as training rooms, had hessian walls, offering scant protection against the Victorian winter. From the start, Ryan subjected them to endless marches, parades, and drills, along with signals training, gruffly telling them on numerous occasions that he intended them to be ‘soldiers first and operators second’.
There were several other units scattered over the grounds at Seymour also going through training. Unlike Sailor Jack Ryan’s section, these were regular signals units that would soon be running the day-to-day communications needs of the army. Not all signals units were involved in ‘special signals’, like Four Special Wireless Section was. Then, as now, military forces needed communication experts to transmit messages between units and around the battlefield, and for commanders to liaise and coordinate their actions. Signals specialists were the people who made this happen. The difference with Four Special Wireless Section was that instead of sending and receiving messages from their own side, they were only interested in messages sent by the enemy.
These other signals units shunned Captain Ryan’s Special Wireless Section. They knew that it was somehow different from the other sections, and were suspicious of its activities. Rumour had it that Ryan’s men were engaged in ‘snooping’ — monitoring the activity of other signals personnel, looking for infractions or misbehaviour. Neither Ryan nor his trainees corrected this misperception. They were not allowed to. After all, their work was ‘hush hush’.9
Modern radio stations can broadcast music, voices, and any other mixture of sounds, but the very first radios could not do this. Those early machines could only send a single tone, like an unchanging musical note — the only variation being the starts and stops of the tone. To send messages with radio, Morse code was used. In Morse code, all the letters of the alphabet can be sent or written as dots or dashes. For instance, the letter A is a dot and a dash, like this: • —
The most famous and well-known Morse message is ‘SOS’. The letter S is three dots, and O is three dashes, so the Morse version of SOS is • • • — — — • • •
Morse code can be sent over radio by using a series of beeps. A tone that goes for a long time before being cut off (‘beeeeeeeeeep’) is a dash; a short burst (‘beep’) is a dot. Accordingly, a string of long beeps and short beeps can spell out a message in Morse code. Radio operators used a little metal key on the table that, when tapped, closed an electric circuit and sent out a tone on the radio. In this way, pressing and releasing the key, they could ‘tap out’ messages to each other.
The warring European nations made extensive use of radio operators with Morse keys, particularly when, as occurred in many cases, other lines of communication such as telegraph lines had been cut or destroyed.
Almost as soon as radio came into widespread use, radio intelligence quickly followed: the skill of listening to the enemy’s radio signals, and using that to guide strategy against them, rapidly developed. However, the trouble with radio messages is that when you broadcast on your transmitter, anyone else — not just the intended recipient — with a radio receiver that’s tuned in to the right frequency can pick it up and listen in. Eavesdropping in this way is known as ‘intercepting’ a radio message. Unlike the interception of physical messages, the radio transmission is not disrupted (although there are ways of doing that, too). The recipient still hears the message, but so does the interceptor.
The other signals trainees learned how to send and receive Morse code, the benchmark for operational duty being 25 words per minute. But Ryan’s section did not need to practise sending in Morse, as their task was specifically to detect and transcribe Morse code messages sent by the enemy. They would not be sending messages back in response.
Some trainees dropped out, and were not replaced. In December, all the units in Seymour, which together were designated the Third Signals Battalion, were given their operational orders. They sailed by troop ship to El Amiriya in Egypt. The other signals companies were deployed in North Africa, where Allied forces were fighting the Italian army.
Four Special Wireless Section were sent to Heliopolis for additional training from British intelligence officers. There, they learned the details of German and Italian message procedures, and conducted daily drills in breaking down the components of German and Italian messages. They learned how to set up and use direction-finding equipment (often referred to as just ‘D/F’), to identify the location of an enemy radio source. Those who were designated as operators practised transcribing Morse code in a foreign language. The engineers in the section learned, and practised, the setting up and maintenance of antennas and high-frequency receivers.
From Egypt, they moved to Lebanon, and then in March were sent to Greece. The German army was threatening to advance southwards through Macedonia, and Greece was at risk of invasion. British forces were occupied in their homeland and across Africa, so those Australian and New Zealand Forces that were not already engaged in North Africa were given the job of defending Greece.
1941
5
A wireless unit in Greece
Geoffrey Ballard, aged 24 when he enlisted, was a committed Christian who had considered becoming a church minister. Unlike most Australian recruits, this was not his first trip to Europe. It was Ballard’s experiences four years earlier, travelling around Europe, and particularly Italy and Germany, that motivated him to join the army. The evils of fascism, he reasoned, outweighed the evils of killing in wartime.
In Italy, he had seen ‘strutting troops everywhere’, and a wall depicting the ‘New Roman Empire’ that showed Italian-controlled territories around the world. Ethiopia was a new addition to the wall, having recently been annexed by Italy. Ballard noted darkly that the map-makers had left lots of space on the wall for further additions.
As he travelled through Germany, he saw first-hand the fever of Nazism that was gripping the country. Shopkeepers said ‘Heil Hitler’ to their clientele. Swastikas, the Nazi flag, were commonplace. He happened to be in Berlin on Hitler’s birthday, where an elaborate parade took place. Marching soldiers, tanks, military cars, and then more troops came in an endless procession through the lightly falling snow, while the excitement of onlookers in the crowd was palpable. Men and women were calling out ‘Der Fuhrer!’ in expectation, and when Hitler himself finally appeared at the end of the parade, a forest of arms shot up all around him, giving the Nazi salute of an outstretched straight arm, with the hand flat, palm down.1
When the recruitment officer in Melbourne asked if he had any preferences for areas of specialisation, Ballard asked to be assigned to intelligence. He was hopeful of getting such an assignment, in part because he could speak German, but instead was assigned to the Australian army’s 7th Division as an orderly.
The 7th Division travelled on the Mauretania, a commandeered passenger ship, to Suez in Egypt, and from there moved by train to Palestine. After two months at a military base, Ballard was given orders to report to Corps Headquarters in Egypt with another Australian soldier, Don Inglis. They were driven to a large Australian camp about 20 miles south of Alexandria. There, they were sent to the officer’s mess to meet Colonel John D. Rogers, the director of military intelligence. Rogers took them for a walk about 100 metres across the sand, out of earshot, where he stopped and interviewed them.
Rogers asked Ballard if he could speak German. ‘Yes, Sir,’ he said. And if he heard two German pilots in conversation over the radio, would he understand what they said? ‘Yes, Sir,’ Ballard replied confidently. Rogers told him that they would undergo intelligence training with the British, and from there they would be sent to Greece.
Geoff Ballard and Don Inglis were then transported to Heliopolis, where they were given two days’ instruction in signals intelligence at the local museum.2 British officers taught Ballard the basics of making codes and breaking them. They taught Don Inglis the basics of traffic analysis, the art of understanding the enemy from the messages being sent and received, even if those messages couldn’t be read.
The science of understanding an enemy from its messages is known as signals intelligence, often just called ‘sigint’. Code-breakers, or, more correctly, cryptanalysts, try to break the enemy’s codes and ciphers in order to read its encrypted messages. But even if the codes and ciphers can’t be broken, insights into the enemy’s mind can still be gained from the radio signals themselves. Traffic analysis is the study of the ebb and flow of messages between the enemy’s units. Much can be learned about the enemy just by keeping track of who is sending messages, and who is receiving them.
Ballard’s and Inglis’s whirlwind introduction to sigint was not meant to get them up to speed on this complex art, but only to get them started. They would have plenty of on-the-job learning opportunities. With their crash courses now complete, they were made corporals, taken to the Alexandria docks, and put on a troop ship to Greece, where they joined an Australian signals unit, Four Special Wireless Section, camped in the mountains to the north.
They arrived at the encampment of the section and reported to the commanding officer, Captain Jack Ryan. After cursory introductions, Ryan explained to Ballard and Inglis that they would be part of the section’s I unit, its intelligence team, but that, sadly, the section did not yet have an I officer, and so there was nobody to run the I unit. However, that would all change soon, Ryan told them. He had been informed that ‘Mister Sandford’, an Australian intelligence officer with the required expertise, would also be joining them ‘any day now’.
In the meantime, Ryan told them, there was an experienced British signals unit nearby, 101 Section, in the hills outside Elassona, and arrangements had been made between the British and Australian armies for 101 Section to train Ryan’s I personnel.
Each day, Ballard and Inglis would travel to Elassona and work with the British unit in a disused mosque. Another member of the section, Eddy Kelson, would go, too. Kelson was a German Jew who had fled his home country to escape the Nazis. His fluency in both German and English was an invaluable asset.
The British sigint officers taught them more advanced theory and practice for each of their respective areas of expertise. Every evening they returned with new techniques and information. Although they did not know it, Ballard, Inglis, Kelson, and the handful of other individuals in the small unit of Ryan’s section were the only Australian signals-intelligence unit engaged in the war against the Axis powers.
The young sigint trainees monitored the march of German forces through Yugoslavia from the GCCS base inside the concrete mosque. The aerials picked up radio traffic from the north as well as across the Mediterranean, where they learned about Allied losses before their commanding officers did.
Eddie Kelson would report on events coming through on his radio receiver. However, with the German units moving so quickly and with the war now spreading across the entire region, the signals came from all directions, and he had no idea where those events were occurring.
‘There’s a tank battle taking place somewhere,’ he announced on one occasion. ‘There are reports of Panzers in action!’ The location of the tank battle he had identified was a mystery.
As the enemy came closer, the peaceful, rustic environment changed. Refugees came spilling southwards, escaping the advance of the Nazis.
That all changed with a hand-delivered message for Captain Ryan marked ‘Top Secret,’ informing him that the ANZAC forces planned to withdraw from Greece and that his section had to evacuate immediately. Events that evening demonstrated how urgent the situation was. Captain Ryan was taking a late-afternoon shower when artillery gunners arrived, making their way through the campsite and allocating positions to defend against the oncoming Nazi army.
They packed up their equipment and headed south to the coast. German planes strafed and bombed them, destroying their vehicles and forcing them to continue on foot. In Athens, they boarded the last departing ship for Cairo, the Else, a Greek passenger ship.
The Else, packed with as many passengers as could fit, including soldiers and civilians, pulled out from Port Pireaus and lumbered southwards into the Mediterranean Sea as Greece fell under total Nazi control.
6
A purple jacket on Crete
The Else and its passengers never made it to Cairo. Instead, the ship sailed to the island of Crete. The Else sailed into Souda Bay on the island’s north, where smoked billowed from a British tanker that had been bombed by German planes. Nearby was a British Royal Navy ship that had also been attacked. It was left crippled, unable to sail, sitting in the shallow bay like a sitting duck. But its guns were still operational, and still manned by British sailors, waiting to defend the ship against further attacks from the air. All Australian soldiers on board were ordered to disembark the Else. Once the troops had left the ship, carrying what supplies they had salvaged from Greece onto the Souda Bay wharf, the Else pulled out to sea again and disappeared.
As they walked from the wharf into the township, they were taken by surprise by an air raid. Several German fighter planes swooped down, flying with the sun directly behind them, blinding the Australians to their attackers. The planes strafed the docks as the soldiers of Four Special Wireless Section disembarked, forcing them to take cover. The attacking planes disappeared as quickly as they had come.
The soldiers’ unexpected arrival in Crete was part of a hasty mobilisation of British and ANZAC forces onto the island. The Allies controlled Crete for now, but had reason to believe that it was the next target of the advancing Axis forces from the north. Hitler had, in fact, decided to invade Crete, because he was preparing an offensive against Russia and did not want the Allies using the island as a base to attack his oil reserves in Romania.1
The section initially set up camp in an olive orchard on the outskirts of the town of Chanea, but their location must have been given away, because German planes took to bombing and strafing the orchard on a regular basis. The section moved camp closer to Chanea, to the top of a hill overlooking the town. There, they found an empty concrete water tank, which made an ideal bunker. They covered and camouflaged the water tank, and set up their radio-intercept equipment inside.
General Bernard Freyberg, a New Zealand army commander, arrived at the end of April, and was immediately placed in charge of the defence of Crete, taking over from Group Captain George Beamish of the Royal Air Force. This news cheered the troops on the island, as Freyberg’s exploits as a soldier of war were legendary. But with the haphazard way the forces had been quickly sent, Freyberg found himself in charge of disorganised and poorly equipped troops. The units had left their heavy arms and most of their equipment behind in Greece. The air defences were woefully inadequate, with no fighter protection at all. He had no support staff, and the only trained intelligence staff among the soldiers on Crete were the I section of the Australian unit, Four Special Wireless Section, on top of the hill. Freyberg recruited anyone who seemed suitable as support personnel, and established his headquarters in a quarry near Chanea, below the hill where Four Special Wireless Section was esconced.2 He named his new headquarters ‘Creforce HQ’.
Freyberg was appointed commander by General Sir Archibald Wavell, who was making a brief visit to Crete when Freyberg arrived. Before departing back to Cairo, Wavell explained to Freyberg that he would be assisted with a new kind of intelligence called ‘Ultra’, which was so secret that Freyberg could tell nobody about it. Any message that contained Ultra would merely say that the information had come from a ‘most reliable source’.
Wavell told Freyberg that Ultra information came from a high-level spy in Greece, but this was a lie.3 Ultra was coming from Bletchley Park in England, where, a few weeks earlier, Alan Turing and his team had broken the German’s Enigma cipher.
The so-called Enigma machine was an advanced cipher system, using an elaborately modified typewriter. As a message was typed, all the letters and numbers were displayed with lights above the keys, and the machine would convert each key pressed into a different one. (For example, BERLIN might be displayed in the lights as IDTMRP.) The only way to retrieve the message was to type it back into an Enigma machine with the appropriate settings to reverse the cipher. What made the Enigma machine diabolical was that it had a complicated system of cogs and wheels inside, so that every time a key was pressed, the cogs would move and the settings would change; this meant that if the key was pressed or typed a second time, a different enciphered message would result. The letters of the message changed the cipher, letter by letter, as it was being entered. The cogs could easily be swapped and replaced with different-sized ones, so there was a stupendously large number of possible settings. The operator recovering the signal would need both an Enigma machine and the current settings in order to extract the message.
Breaking the Enigma cipher was a monumental feat, and possibly one that changed the outcome of the war. The Enigma machine was so complex and secure that the Germans were sure that nobody could break it. As a result, they blithely sent messages back and forth that revealed their plans in great detail, confident that the enemy would never read those messages.
But from March 1941 onwards, the British were reading their Enigma-encoded messages. This was thanks, in part, to the brilliance of the Allied code-breakers such as Alan Turing, and to earlier work on Enigma by Polish agents. Allied generals now had access to a new kind of intelligence to help them make their plans. The British were calling this new kind of information ‘Ultra’ because it was ‘Ultra Top Secret’. Later in the war, the Americans also adopted the word ‘Ultra’ to describe intelligence from decrypted enemy signals.
Before Freyberg arrived, Ultra messages from headquarters to Crete were sent to Group Captain George Beamish, who was in command of the Royal Air Force on the island. Beamish was not trained in intelligence, but there have been some suggestions that Beamish continued to play a role as a liaison officer, assisting Freyberg with the receipt of Ultra while they waited for a specialist intelligence officer to arrive from Cairo.4
At the water tank at the top of the hill, Captain Ryan continued to assure Four Special Wireless Section, with increasing conviction, that ‘Mister Sandford’ would join them any day.
Lieutenant Alistair Wallace Sandford, their new chief intelligence officer, finally arrived on 18 May. He had trained with British intelligence in Cairo, and had been an intelligence officer since February.5 He brought with him German codebooks, as well as instructions from Middle East headquarters that the section was to discontinue its existing work and focus entirely on working for, and reporting directly to, Freyberg’s headquarters, doing ‘tactical work’.6
That evening, Captain Ryan organised a parade for the section. Lieutenant Sandford arrived, wearing a bright claret-coloured jacket. It was striking and out of place, and Ryan was not impressed. After parade, when the officers dined together, they offered the newcomer a drink of wine. Sandford inspected it, and was appalled. He said to them, ‘Crete abounds with noble wines, and you’re content with this hogwash? I’ll go into town tomorrow and buy something drinkable.’7 The next day, he did as he said, venturing down to Chanea and returning with local wine.
The jacket and the wine were only the start. He had ideas for improving the officers’ mess, such as hanging curtains on the walls and using tablecloths. When someone muttered sardonically that ‘There’s a war on,’ he responded, ‘Even so, old boy, there’s no need to live in squalor.’ He had a sense of style, and spoke, as Lieutenant Hill put it, ‘in a dandified way’. They nicknamed him ‘The Scarlet Pimpernel’.
Lieutenant Hill recalled of Sandford,
None of his equipment was uniform. He deplored regimentation. He said he didn’t see why individuality should be suppressed, and made to conform to a common, very common, pattern.
In short, he was somewhat out of place as an officer in the war-toughened Australian Signal Section. But as events were to prove, he was completely fearless and so in the end we accepted him as a brother in arms.8
Sandford’s lack of conformity was not a demonstration of a lack of social awareness, but of a complete lack of fear of the opinions of others. Similarly, he had no fear of the enemy; or at least, if he did, he never demonstrated it. A particularly striking instance of this occurred much later in Egypt during an air raid, when the section was hiding from the raid in a bunker below ground. Sandford brandished his rifle, headed outside, and strode the open streets of Cairo, firing into the air at German planes overhead.9
Within a short time, the nickname ‘The Scarlet Pimpernel’ was forgotten, and he was simply known as ‘Mic’, a name that stayed with him until the end of the war.
Given his flamboyant manner and his sense of style, some of the troops of Four Special Wireless Section wondered aloud about the possibility that he was a ‘poof’. Their suspicions were correct. Sandford was gay, and it soon became clear that he made no pretence otherwise — again, because he really did not care. The opinion of other servicemen, whether lower or higher in rank, concerned him less than bombs from German air raids. Given Sandford’s promotion later in the war to lieutenant-colonel, he was probably the first openly gay officer of that rank or higher to have served in the Australian military forces. However, given the nature of his work, his war record and everything he did during the war was kept secret for decades afterwards. And thus the fact that a high-ranking officer was a flamboyant homosexual also remained a secret for the same length of time — perhaps conveniently for the army.
Sandford’s arrival meant that the section was now operationally complete. With his expertise, and now possessing the latest German codebooks, the section was able to intercept and read coded enemy radio messages, which they passed on to headquarters at the bottom of the hill. Within a week of Sandford’s arrival, Freyberg had noticed a marked increase in efficiency of his intelligence staff, including the provision of a detailed report on enemy dispositions.10
From their vantage point high on the north coast of the island, they were picking up large numbers of radio messages from the German forces in Greece and the northern Mediterranean. One of the German units they were listening to and tracking was moving from location to location, and seemed to be involved in the preparations for the impending German attack on Crete. When, one day, they heard the German radio operator say in German a phrase that translated as ‘Goodbye, landlubber!’, that was all they needed. The radio signals they had been listening to were coming from a seaborne German assault fleet headed for Crete.
Besides being an intelligence officer for the wireless section, Sandford was also the new Ultra liaison officer that Freyberg had been promised. Sandford’s second role has been obscured subsequently, in part due to the secrecy surrounding Four Special Wireless Section, and in part due to a typographical error. In 1991, in Crete: the battle and the resistance, his influential book about Crete during the Second World War, the historian Antony Beevor named Sandford as the Ultra liaison officer,11 but in an appendix to the book mistakenly referred to him as ‘Captain Sandover’, perhaps mixing him up with another Australian officer, Major Sandover.12 Somehow the error has been propagated, and there are now numerous history books that refer to a non-existent intelligence officer on Crete, ‘Captain Sandover’.
The cryptanalysts at Bletchley Park had been decrypting messages about a planned German invasion of Crete since 6 May. These messages were summarised into reports, relayed to Cairo, then sent in encrypted form to Crete. On 13 May, Freyberg received an Ultra message that provided him with a comprehensive description of the German invasion plans. Here is the first part of the message:
The following summarises intentions against Crete from operations orders issued.
Para 1. The island of Crete will be captured by the 11th Air Corps and the 7th Air Division and the operation will be under the control of the 11th Air Corps.
Para 2. All preparations, including the assembly of transport aircraft, fighter aircraft, and dive bomber aircraft, as well as of troops to be carried by both by air and sea transport, will be completed on 17th May.
Para 3. Transport of seaborne troops will be in coordination with admiral south-east, who will ensure the protection of German and Italian transport vessels by Italian light naval forces. These troops will come under the orders of the 11th Air Corps immediately on landing at Crete.
Para 4. A sharp attack by bomber and heavy fighter units to deal with the Allied air forces on the ground as well as with their anti-aircraft defences and military camps will precede the operation.13
It continued from there, describing plans for fuel supplies, and an instruction not to plant mines in Souda Bay. This was a wealth of information, describing both air and sea components of the imminent invasion, but Freyberg misunderstood it. He thought that the primary invasion force would arrive by sea, and planned his defences accordingly.
Freyberg was wrong. The paratrooper attack was not a prelude to the main invasion; it was going to be conducted by paratroopers, and the naval landing was secondary.
Freyberg’s forces on Crete had little hope of defending themselves against the German paratrooper onslaught, even with the assistance of insights from Ultra. When the attack came, it was overwhelming.
Four Special Wireless Section worked through the days and nights tracking the ever-increasing volume of German radio messages, even as German planes flew overhead, as bombs dropped, and as German fighters strafed the hillsides around them.
They tracked the movements of the German navy fleet that they had identified, and reported those movements on a regular basis. There were 15 ships. When the fleet finally arrived in the waters to the north of the island, the soldiers of the wireless unit watched from their water tank on the hill as British planes bombed the German ships. It was one of the few times that sigint operatives saw first-hand the final outcome of their work.
On 20 May, in the midst of a ferocious air attack, ten German gliders landed, or crashed, onto the terrain nearby. Lieutenant Hill formed a small group to investigate the crash and find out if there were any survivors. A second group of four men from the wireless unit went in a different direction, looking for German paratroopers. Hill’s posse found the wreck but no survivors, and returned to the base. The second group discovered the bodies of several German soldiers, but when they moved forward to investigate more closely, German soldiers hiding nearby fired on them with automatic weapons.
Three of the group — Jim Wood, Bob Leopold, and Frank Kelton — took cover near each other. The only one hit by the initial burst of gunfire was Jim Wood, who was shot in the thigh. The fourth man, Billy Wells, dived for safety in a different direction, but, realising that he had been separated from his mates, stood up to make a run for it and join them. There was another burst of gunfire, and Billy Wells was killed.
The other three were now hidden, and thus were no longer an easy target, so the German soldiers fired mortar shells at them. Bob Leopold was hit in the back as he lay in hiding. Kelton and Wood were able to wriggle to safety and make their way back to the section, but Leopold still lay there, alone and wounded.
The Germans found Leopold and took him prisoner. A German officer spoke to him in impeccable English, and offered him Australian chocolate. The officer explained that they had found stores of this chocolate when they invaded Greece, left behind by the Australian troops. It became apparent that the two German soldiers who had captured him had recently parachuted in, and since most of the other paratroopers in the same drop were dead, they were on their own. The Germans did not tell him they were releasing him, but that’s effectively what they did: they left him alone, giving him ample opportunity to escape. He did so, and made his way back to the section. Ryan sent him to the military hospital in Chanea.
The air attacks continued. Many German paratroopers were killed as they dropped to the ground or soon after, or died when their gliders crashed into the rocky hills of Crete. But enough survived to re-group and take control of strategic points.
The airfield of Maleme, just to the west of Chanea, was one of the first Allied bases to fall. Freyberg’s Ultra message of 20 May warned him that Maleme was a target of the Germans, but he did not want to defend it too heavily for fear that the Germans might twig that their messages had been intercepted and decrypted. This was a strategic error.
An airfield was a natural and obvious site to defend even without enemy intercepts informing the commander that it was a specific target. Freyberg was correct in his belief that protecting the Ultra secret was vital, but defending Maleme more heavily would not have given the secret away.
The same day as the landing took place near the wireless section, German paratroopers landed in large numbers near Chanea and nearby Maleme airfield. They suffered heavy casualties from the Australian, New Zealand, and Greek troops defending Crete — of 600 paratroopers who managed to form one battalion on the ground, 400 were killed within 24 hours of their landing. The local Cretan population had also taken up arms, and resisted the attackers with whatever they had. One elderly Cretan found a German paratrooper tangled in parachute ropes and beat him to death with a walking stick.
But, overnight, the New Zealand battalion defending the airfield withdrew by mistake to the west. The Germans cut their communication lines, leading their commanding officer, based to the east, to wrongly believe that the battalion had been destroyed. This mistake and a series of other errors, made due to poor communications and misjudgements, allowed the Germans to quickly move in and capture the airfield.14 Its seizure was the turning point. With the airfields in their control, the Germans were able to move in more planes and reinforcements. Chanea came under further aerial bombardment, and more ground was lost to the Germans. With the Allied forces in disarray, Freyberg issued the order to evacuate.
Ryan gave the section an order to move out immediately, but informed them that there were no vehicles available to transport them. They would have to walk the 40 miles of mountainous terrain to Sphakia on the south coast of Crete.
It was this long march across the Crete interior that taught Mic Sandford a lesson in the value of military regimentation. Lieutenant Hill discovered that Sandford did not possess a standard-issue army water flask. All he had was a pretty little silver flask that he had purchased himself. It carried about a quarter the amount of a regulation flask, and Sandford quickly ran out of water. The section would, from time to time, refill their flasks at rivers and other drinking sources, but even if Sandford filled his to the brim, it never lasted long. Hill and Lieutenant Henry, taking pity and not wanting him to die of thirst, shared their water bottles with Sandford for the entirety of the trek.15
Southern Crete was still in Allied hands, but the German air force dominated the skies, strafing and bombing the evacuating troops wherever they spotted them, and causing the retreating wireless section to frequently hide.
On arrival at the evacuation point, too, they had to remain hidden. With German planes patrolling during daylight hours, transport ships could not approach by day, and would only come close to shore between midnight and 3.00 a.m. Only during this three-hour window could anyone leave Crete.
They boarded a transport on the beach that ferried them across the black night sea to a waiting transport ship. From there, they sailed south, again to Cairo, but this time it was for real. In Egypt, the sick and wounded were hospitalised, and the dead and missing members of the unit were replaced with newcomers. After their short respite, the section was sent back into the field, this time to Palestine, as a new battlefront was opening in nearby Syria.
With Crete firmly in their control, the Germans rounded up the remaining ANZAC and British troops, taking them prisoner. A special security group, the Sonderkommando von Kuhnsberg, ransacked the now-abandoned Allied quarters and places of operation for anything that might reveal useful intelligence.16 When they searched Freyberg’s ‘Creforce HQ’, they found a sheet of paper with a handwritten message dated 24 May, which gave precise information about the location of German troops the previous day, as well as an accurate and precise description of General Julius Ringel’s plan of attack.
Here is the message the German agents found:
Personal for General Freyberg. According to most reliable source, by Midday May 23 German troops had reached coast near AG, Marinas, and cut off British units east of Malemes. Units of coastal group then to attack Suda Bay. It was proposed a … more companies …17
They sent the scrap of paper back to Berlin, where it was studied.
In fact, the note was the first page of a several-pages-long Ultra message. Not only did the message use the key phrase ‘a most reliable source’, but it described enemy positions and even enemy plans. The information had been obtained from Enigma decrypts; more importantly, it could only have come from Enigma decrypts. As a result, the German agents had in their hands clear evidence that the Allies were reading Enigma messages; but when intelligence officials studied the note in Berlin, they overlooked its significance.
Despite Freyberg’s poor defence of Maleme to protect the Ultra secret, the German army suspected that he had advance knowledge of their invasion plans. An enquiry concluded that Freyberg’s intelligence came from spies in Greece. Luckily for the Allies, they did not suspect that Enigma had been compromised.
7
Mrs Mac and her girls in green
Florence Violet McKenzie stood at the front of the woolshed at 10 Clarence Street on a mild Sydney summer day, waiting for the visitor.1 She was wearing the crisp, green uniform of the Women’s Emergency Signalling Corps, a uniform that she had designed. The pillars of Sydney Harbour Bridge loomed over nearby Observatory Hill and the run-down alleyways of The Rocks.
Commander Jack Newman of the Royal Australian Navy arrived. Aged about 41, he was about ten years younger than her. She introduced herself, and as she smiled her eyes crinkled into a maze behind her thick glasses.
She hoped that Jack Newman’s visit would finally open the minds of the military establishment to the opportunity she was giving them: a fully trained corps of female signals experts. The air force and army had already knocked her back, and various government ministers had sent polite refusals in response to her letters. But if Jack Newman was in town to visit her, that suggested the navy, at least, might be prepared to take her seriously.
With the armed forces stretched to the limit, there were severe shortages in specialised skills, particularly signals. In late 1940, the navy had placed an advertisement calling for anyone proficient in Morse and other areas of signals expertise to apply to the navy. They had received a reply in the mail from Florence Violet McKenzie on 27 December, who informed them that she had established an organisation called the Women’s Emergency Signalling Corps, and had trained 1,200 women in Morse.2 She was pleased to have the opportunity to offer the services of these women to the navy.3
The Naval Board was interested but, but given the political sensitivities of employing women, was cautious, so it sent the director of naval communications, Jack Bolton Newman, to investigate further. Newman was acutely aware of the shortage of skilled signals experts in the Royal Australian Navy.
With her characteristic bright smile, McKenzie led Newman into the woolshed and upstairs to her signals school for women.
Even as a girl, Florence Violet Wallace was interested in electricity. She recalled her childhood, saying, ‘I used to play about with bells and buzzers and things around the house. My mother would sometimes say, “Oh, come and help me find something, it’s so dark in this cupboard” — she didn’t have very good eyesight … So I’d get a battery and I’d hook a switch, and when she opened that cupboard door a light would come on… I started sort of playing with those things.’4
Born in 1890, Florence decided early in life that she wanted to be an electrical engineer, despite it being an exclusively male occupation, ‘I practised at home,’ she recalled in an interview late in life, ‘and learned from books. You can learn anything from books.’
After two years of science at Sydney University, she applied for entry to the Sydney Technical College’s engineering department. A condition of enrolment was having a relevant apprenticeship, but because she was a woman no business would give her one. She took the problem to the head of the department, who reaffirmed the policy.
‘You can’t come here and do engineering unless you’re working at it. You’ve got to be working at it,’ the head told her.
She asked, ‘Well, now, supposing I had an electrical-engineering business and I was working at it. Would that be all right?’
The head reluctantly agreed. ‘Yes, if you produce proof.’5
So Florence Wallace opened an electrical store in the Sydney Royal Arcade, and took her first contracted work, wiring a house in Marrickville. With newly printed business cards and a contract for work, she enrolled in the course, and upon graduation became Australia’s first female electrical engineer.
For people in the emerging subculture of do-it-yourself electrical gadgetry, Florence Wallace’s shop was known as the go-to place. The most exciting new technology in the electrical world was radio, and Florence Wallace was at the leading edge of it. In 1922, she became the first woman to gain an amateur radio licence. The same year, she co-founded the magazine Wireless Weekly with William John MacLardy, the founder of radio station 2SB. Wireless Weekly was an instant and enduring success, years later becoming Electronics Australia. Florence Wallace was also a radio presenter at 2SB (which would later become 2GB), including a weekly show about tropical fish.
Another electrical engineer, Cecil McKenzie, ordered through her shop an expensive radio valve from America, but when he took the valve home, it rolled off the table and smashed. When Cecil returned to the shop and told her what had happened to his valve, he looked so miserable that she burst into tears in sympathy. Cecil’s broken radio valve and Florence’s tears over it ignited a romance. A year later, they were married. She became Florence Violet McKenzie.
Electrical appliances were revolutionising Australian homes and freeing women from a lot of the drudgery of housework, but Mrs Mac was dismayed by the reluctance of many Australian women to embrace these new technologies.
In 1938, she and several other women founded the Australian Women’s Flying Corps at Sydney’s Feminist Club. She became treasurer, and started teaching classes in Morse code for the women in the corps.6 She became aware of the acute demand for courses to teach Morse to Australian women, and so the following year she and her husband, Cecil, founded the Women’s Emergency Signalling Corps, an organisation whose sole aim was to provide signals training to women. This included Morse code with a Morse key, as well as semaphore (flag signalling) and sending Morse with lamps.
The woolshed in Clarence Street offered a central location and cheap rent. She covered the greasy old floors with sheets of linoleum, and the wooden walls with hessian. Friends and relatives donated tables and chairs. McKenzie rigged telegraphic wiring between the two upper floors of the building so that women could practise signalling in Morse to each other between floors. She wanted this to be viewed as a genuine corps, a militia of technical experts, so she designed a uniform for its members. The uniform consisted of a bottle-green knee-length frock, a brown belt, and a black, tailored blazer.7
Women would come to Mrs McKenzie’s signals school — ‘Sigs’, as it became known — in their lunchtimes, after work, or on Saturday mornings. They called Florence ‘Mrs Mac’. The whole thing was funded by donations, and Mrs Mac never took any money.
There was little socialising at Sigs, at least during practice, when a visitor would hear nothing but an incessant clacking of Morse keys. Mrs Mac invented various teaching aids to help the women learn Morse, many of which included short bursts of song.
For example, the letter Q in Morse is dash-dash-dot-dash — or, as a Morse operator would say, ‘Dah-dah-dit-dah.’ Mrs Mac would send the letter Q to a student and then sing ‘Here Comes the Bride.’
She used catchy phrases. The letter L is dot-dash-dot-dot, or, in Morse vernacular, ‘dit-dah-dit-dit.’ Mrs Mac would send the letter L and then call out, ‘To L with it!’8
Learning Morse was slow going at first, but with persistence the breakthrough would come. Sue Timbry, one of Mrs Mac’s students, explained years later: ‘All of a sudden a wonderful magical moment came when you forgot about the ‘dits’ and ‘dahs’ and you just actually heard the musical sound of each letter. And then you were really on the way.’9
Mrs Mac took Jack Newman upstairs to the training rooms, where he was introduced to her class for the day. She showed him the Morse-sending apparatus, and explained how he could test the women by sending and receiving from the floor above.
Newman tested several of the women in Morse sending and receiving. He was blown away. He was impressed by their skill, by the training facilities Mrs Mac had established, and especially by Mrs Mac herself. That meeting on a summer day in early 1941 was the first of many. The two developed mutual respect and admiration. Newman later told her that, through her school, she had provided him with ‘the cream of Australian womanhood’.10
Jack Newman reported to the Naval Board that the members of the Women’s Emergency Signalling Corps were ‘almost as good as the men’. This remark was less sexist than it sounds; the ‘men’ that Newman was comparing them to were experienced naval personnel, fully trained and operational on Australian warships and in shore stations. That Mrs Mac’s women, with no formal training or operational experience, were almost as good as them was itself astonishing. Indeed, Jack Newman’s shocking endorsement of Mrs Mac’s students was a pivotal moment for Australian signals and for the role of women in the Royal Australian Navy.
While Mrs Mac was pleased with Jack Newman’s support, his praise of her ‘girls in green’ was a bit weak for her liking. She maintained to everyone who would listen that, rather than being ‘almost as good as the men’, they were ‘as good as the men, if not better’. Over time, Newman came to agree with her that this was indeed the case.
Admiral Ragnar Colvin, the First Naval Member, was not about to act rashly.11 Colvin had joined the Royal Navy as a cadet 45 years earlier in 1896. With failing health and at the end of his career, he was not about to change longstanding tradition on a whim. Nonetheless he was aware that in Britain, his homeland, the navy had recently established a special women’s-only service, the British Women’s Royal Naval Service (known as the WRENS).
Under Colvin’s leadership, the Naval Board decided, after considering Jack Newman’s report, to investigate the matter further. For one thing, there was the issue of accommodation. If women were to be based at shore stations, where would they sleep? Would existing naval accommodation even be appropriate?12 Colvin told Newman to prepare a more detailed report that addressed all of the board’s concerns.13
A month later, Colvin’s health took a turn for the worse. He stepped down as First Naval Member, and his second in command, Commodore John Walter Durnford, temporarily stepped into the role as head of the navy until a replacement was appointed.14 Durnford had been keen on the idea of recruiting the women from the beginning.
Newman submitted his homework as Colvin had insisted. He also brought Mrs Mac down to Melbourne to appear before the Naval Board in person and tell it about her Women’s Emergency Signalling Corps. The board, chaired now by Durnford, promptly decided to employ a small number of the women on a trial basis. If it all worked out well, they could bring in more women, and the board thought it might even get requests from the British to send its female telegraphists to Singapore.15
But first they needed permission from the government. Durnford wrote to the minister of the navy, Billy Hughes, asking for permission to do so. Hughes, a former prime minister, came along to the next board meeting. He gave permission to employ 14 women at Harman, a signals base near Canberra, on two conditions. First, they had to keep looking around for suitable men who could do those jobs; and, second, they had to keep it quiet.16
Since Newman had been so enthusiastic about bringing women into the navy, the board decided that he should be the one to take them on a trial basis at Harman, the naval communications station that he oversaw.
Mrs Mac and 14 women from her Women’s Emergency Signalling Corps arrived at Queanbeyan railway station on 28 April 1941, where they were met by Jack Newman and the new officer in charge of Harman, Lieutenant Commander Archibald D. McLachlan. Newman had personally established Harman two years earlier and had moved from Melbourne to live at the base. He had created the station’s name, ‘Harman’, from a blend of his own name and his predecessor, Neville Harvey.17 A week before the women arrived, Newman had transferred to the Navy Office in Melbourne’s Victoria Barracks, handing control of Harman to McLachlan.
Outside the railway station, Newman had several cars waiting to transport them to Harman. When they arrived, the naval base appeared to be deserted, but the women soon noticed faces peering out of windows. Newman, concerned about how the men might behave when Mrs Mac’s women arrived, had ordered them all to stay indoors and out of sight, but the curious young men took any chance they could to glimpse the new arrivals.
The women were allocated four cottages, three for sleeping and one as a mess. After showing them around, Newman took them to the recreation hall and gave them a pep talk, concluding with a variation on his standard admonition to new recruits. ‘I’m told there are two kinds of women doing war work — ones that sit around and drink tea, and ones that work. I hope you are the latter.’
Having staked his career on the success of Mrs Mac’s Signals Corps, he was sincere in his expressed hope that they were ‘the latter’. He need not have worried, though. The women were excellent.
The navy established a new service, the Women’s Royal Naval Service, but they still considered the Harman intake to be very much an experiment. None of the women were officially enlisted into the newly created service, which they nicknamed the ‘WRANS’, until October. In the meantime, they were just civilians who happened to be living on site at a signals base and helping out with ‘telegraphy’ duties. For that matter, the newly created Harman was not yet officially a naval shore station, and wasn’t established as one until 1943. So, for the time being, the women at Harman weren’t officially in the navy.
Despite their unofficial status, the women at Harman considered themselves to be very much ‘in the navy’, and made a point of dressing in uniform at all times. Since they didn’t have naval uniforms, they wore the green-and-brown WESC [Women’s Emergency Signals Corps] uniforms made for them by Mrs Mac. When the navy finally got around to issuing uniforms, there weren’t enough.
Shirley Huie, in her history of the WRANS, wrote that in the early days at Harman, ‘It was not unusual to see three WRANS out shopping in Canberra, one in a green WESC uniform, one in the WRANS naval rig and a third in a Girl Guides outfit.’
The commander of Harman had to deal with new kinds of problems — from how to get around the standard health-check procedure of pissing in a bottle, to dealing with the fact that there were no navy regulations allowing someone to be discharged because of pregnancy. Although he had relocated to Melbourne, Jack Newman’s wife remained at Harman for the time being, where she took on a motherly, mentoring role to the women.18
In March, the air force established the Women’s Australian Air Force, and began Morse training with a small number of women.19 Many who enlisted were members of Mrs Mac’s corps, having arrived with full proficiency in Morse code. The Australian army soon followed suit, founding the Australian Women’s Army Service in August.20 By the time the navy officially established the WRANS in October, they were the last to form a women’s service, despite having employed women since April.
Mrs Mac accompanied another eight women from Sydney to Melbourne’s Victoria Barracks at the start of June. As with the previous intake, Jack Newman tested them, gave them additional training, and sent six of them to Harman. Newman retained Mrs Mac’s two most talented Sydney students, Joan Cowie and Jo Miller, in Melbourne for additional training. Cowie and Miller became the nucleus of the navy’s special signals workforce.
With the services now enlisting women, demand for training at the Women’s Emergency Signals Corps skyrocketed. Before long, the army and air force started sending men to the Clarence Street school, where they were taught by women who had previously been trained by Mrs Mac.
Mrs Mac herself was never given any rank nor paid a salary by the Australian government or any of the armed services. As an acknowledgement of her contribution, the air force bestowed on her the title of Honorary Flight Officer, but this was not a military rank — it was merely a ceremonial gesture of goodwill. She continued to work for free throughout the war, and for years afterwards when she trained Australian pilots, until the owner of the Clarence Street woolshed evicted her.
Later in life, she corresponded with Albert Einstein, at one stage sending him a didgeridoo and a boomerang, and she wrote several books. In 1982, on her deathbed, her last recorded words were, ‘It is finished, and I have proved to them all that women can be as good as, or better than men.’21
8
Fabian’s tunnel
In March 1941, a British Royal Air Force bomber landed in the Philippines on the island of Corregidor. On board the plane was a British agent, Malcolm Burnett, who had travelled from Singapore to visit a United States navy sigint team based on the island.
Corregidor is a small, rocky outcrop near the mouth of Manila Bay. Standing as it does in the way of ships entering the bay on their way to the Philippines capital, Manila, it had for centuries been a Spanish fort. The Spanish lost control of the Philippines in 1898 when they lost their war against America. But America, having gained the territory, did not at first know what to do with it. With no history of European-style colonisation, taking control of another nation and running it as a colony was politically impossible; the American public would have neither understood nor accepted such a policy. So, as with other territories won from the war, the American government invented the notion of a ‘protectorate’ as a political solution. The Philippines people had civil autonomy to run their own affairs, but would be ‘protected’ by a US military presence. Given its key strategic location at the mouth of the bay, the island morphed from a Spanish fort to a base for the United States military.
The American sigint team on Corregidor that Burnett was visiting were called Station CAST. Just like the FECB team in Singapore, they were also monitoring Japanese naval radio signals. But, unlike the British, they had so far been unable to crack the new naval code known as JN-25. The unit was originally based in Shanghai, but had pulled back to the Philippines when the war in China heated up. They spent their days working in a tunnel below ‘Monkey Point’ at the island’s northern end. Here, they had made great progress with traffic analysis, but they had not made any inroads into the code system itself.
Burnett was one of Tiltman’s team of four, who, back at Bletchley Park in 1939, had found the solution to JN-25 and devised a method for breaking it and any new version of the code. Burnett was the man who flew from England to Singapore to teach the FECB code-breakers such as Nave and others how to break JN-25. And now Burnett was on Corregidor Island to teach the Americans how to do it.
Station CAST was run by Lieutenant Rudolph J. ‘Rudy’ Fabian. Fabian was a short, barrel-chested man from Montana. He was loud and confident, and leadership came naturally to him. The navy had recently sent another officer, Lieutenant John Leitwiler, to take command of Station CAST, but Fabian had stayed on as an advisor and effectively continued his position as commander, despite the fact that Leitwiler outranked him.
In addition to Leitwiler, there were two other officers in the team who had higher ranks than Fabian, but both were happy to leave the running of the team to him. They were Lieutenant Commander Gil Richardson and Lieutenant Commander Swede Carlson. Both of them were fluent Japanese linguists, and were intellectuals at heart who enjoyed the challenges of code-breaking and translation, and were relieved not to have to bother with the administrative work of commanding a navy unit.1
This was Fabian’s first meeting with Burnett, but was not the first meeting between Burnett and members of the Station CAST team. In February, Fabian had sent Jeff Denis to Singapore to visit the FECB and discuss the latest version of the JN-25 code with Burnett and others. He had come back with key information for Station CAST to work with.
In an interview years later, Fabian recounted what happened:
Denis went down and on his return he brought from the British a solution to this five number problem ... He brought back the solution, how to recover the keys, the daily keys. How the code was made up and a lot of code values. And since it was a very heavy-volume system, it was the heaviest-volume system on the air. I talked with my people and I went back to the CNO and requested permission to drop everything else and go to work on this five-number system. And we did pretty well on it. We couldn’t do any solid reading, no, but we could pick up stop phrases like ‘enemy’, ‘enemy submarine’, or ‘enemy aircraft’. But we were coming along pretty fast.2
Now that Denis had taken the secrets of JN-25 to Corregidor, and the signals unit there was ready to ramp up work on it, the British agent Malcolm Burnett was there on a reciprocal visit, to cement the partnership between the FECB in Singapore and Station CAST in the Philippines. There was a lot of work ahead for both teams. Even knowing how JN-25 worked, building up the codebook — the code dictionary that was being used — was a mammoth task, and if the two teams could exchange information they would make faster progress. However, the most practical way of doing this — broadcasting their discoveries to each other with long-range radio transmissions — was unthinkable, because it would run the risk of providing evidence to the Japanese that the British and Americans were piecing together their codebooks. Instead, they agreed to swap their discoveries on JN-25 by sending them back and forth between Singapore and the Philippines on a plane. Just in case the plane ever got shot down and the Japanese got hold of the documents, all the documents on board would themselves be encoded.
Burnett’s visit was made possible because of a new partnership between American and British cryptographers that had been forged four months earlier in London. In January, a delegation of American cryptanalysts had visited Britain, led by one of America’s brightest minds in code-breaking, Dr Abe Sinkov. The Americans came to refer to it as the ‘Sinkov Mission’. They visited Bletchley Park, and met with British code-breakers, including Turing. By the end of their visit, they had negotiated an agreement between the two nations to collaborate on cryptological work. This agreement had ramifications across British territories, including the FECB in Singapore, and for Fabian’s team in Manila Bay, because it formally allowed them to exchange information and collaborate on Japanese codes. From that point onwards, Allied code-breaking in the Pacific was a team effort.
This was true for the efforts in 1941 to unearth the JN-25 codebook, and it remained true throughout the war.
The Japanese naval code, JN-25, had a fatal weakness. Unlike Enigma or Purple, the code made no use of machinery. The whole thing worked, as did all Japanese code systems, with codebooks. But the use of codebooks was not the weakness. In fact, it was a strength.
Machine ciphers are expensive to create and to run, and are not practical on a large scale in a shifting battlefield. All armies make use of low-tech solutions, specifically codebooks. If they are constructed well, codebooks can be very hard to break.
The JN-25, for the most part, was constructed very well. There were three layers of protection for each message. Everyone using the JN-25 was issued with two books. The first was a codebook, for translating standard words into code words. The second was an ‘additive book’, for making additional changes to the message.
In the first step of encipherment, the person coding the message translated each word in the message to a new word using the codebook. The codebook worked pretty much like a dictionary; you looked up the word and wrote down the entry. All the ‘code words’ were just random strings of five digits. If you were a Japanese radio operator, and you had to translate the word ‘attack’, you would look up ‘attack’ in the codebook, and might find that the entry was ‘05337’.
Already the message was unreadable. Someone peering over your shoulder, or secretly listening in to your radio broadcast might see or hear the number 05337, but would have no idea what it meant. Only a friend listening in with a radio receiver who also had your codebook could use the book to discover that the word you were sending was actually ‘Attack’.
But now there was a second step. Just in case the enemy had your codebook, or had some other way of figuring out what the number string meant, that number could itself be encrypted. The additive book had no words — just pages and pages of random numbers. If you and your friend agreed to start at the top of page three of that second book, you could add a digit to each of the five numbers 05337 to get a completely new number.
(By the way, if you added the two digits and the result was 10 or more, you just removed the 1 and kept the ‘singles’ part, in what was called ‘false addition’ or modulo addition.)
Now, the code word, or really ‘code number’, to be more accurate, had itself been encrypted, and the only way to get back to the original was if your friend (or your enemy) had the codebook, the additive book, and knew what page and line of the additive book you started the second step with.
If that sounds diabolically difficult to crack, that’s because it was.
The weakness that Tiltman found and exploited was that all the numbers in the codebook — all the ‘code words’ — were divisible by three. The Japanese navy did this on purpose. If an operator received a message and was concerned whether he or she had decrypted it accurately, or even whether the sender had possibly made a mistake somewhere, it there was a simple check available to see whether the message was right: the numbers simply needed to be divided by three.
On the face of it, there was nothing wrong with making the numbers divisible by three. After all, they were not really numbers anyway. They were just code words, written in meaningless strings of digits.
But any pattern in a code or cipher system is enough to weaken it. This seemingly innocuous trait gave the code-breakers a foothold into the cliff face of the naval code. If they could figure out which messages were using the same additives — and for various reasons, this could be done — it was just a matter of subtracting all the meaningless strings of five digits from all the other meaningless strings, until they stumbled across a result that was divisible by three across the entire message. It was long and laborious work, especially when done by hand, as it was at the time. But it could be done.
Every victory was a small victory. If a message encrypted with a cipher machine such as Enigma was broken, all the words in the message were decrypted at once. But breaking a codebook, even when knowing how the code worked, was a tedious and back-breaking endeavour. Breaking a message might yield a single new code word, for a book that contained about 300,000 entries. But the teams in Singapore and Corregidor were relentlessly chipping away at it, and by sharing their progress, they began to make inroads into JN-25.
9
Stone frigates
Joan Duff was never quite sure how she came to work for Jack Newman at the Navy Office. In August 1941, aged 20, Joan heard that the armed forces were starting to enlist women. Since three of her four brothers were in the army and serving in combat areas, she decided to join up. Her mother was vehemently opposed, but she had made up her mind.
The Melbourne city recruitment station was a temporary shack next to the town hall. A recruiter inside the shack asked what service she wanted to join. She told him ‘army’, and provided him with her school reports. The recruiter read them slowly, then opened a notepad, wrote a brief letter, and placed the letter into an envelope, which he then sealed and addressed to ‘Commander J.B. Newman’ on the outside. ‘Take this to Victoria Barracks and deliver it to Commander Newman,’ he told her, handing her the envelope.
Joan did as he instructed. She caught a southbound tram to Victoria Barracks, where she showed the letter to the guards at the gate. The letter seemed to act as an automatic pass into the barracks. The guards directed her to go the third floor of a building called ‘Block E’. She spoke to nobody else until she was inside the old brick building, had gone up a flight of old wooden stairs, and found the office of Jack Newman.
Newman was middle-aged, stern-looking, and smoked a pipe. He didn’t wear glasses, but had a mannerism that gave the impression he did. He looked sternly at Joan with his chin pointed down and his eyes bulging up, as if staring over the top of invisible bifocals.
Newman opened the envelope, read the letter, then folded it, looked up at her, and said, ‘Any special qualifications?’
Joan said, ‘No, Sir.’ Many women her age who were entering the workforce could type, but she, as a trainee journalist, could only do two-finger typing.
Newman said, ‘Except intelligence.’ It was a statement, rather than a question.
Joan said, ‘Well, I hope so, Sir.’
Newman asked her, ‘When could you start?’
She said, ‘When would you want me to?’
Newman said, ‘Would you like to start now?’
And from that moment, Joan Duff joined Newman’s staff. He paid her in cash in an envelope at the end of the week. As Joan was not part of the swelling ranks of Morse telegraphists, and instead reported directly to Newman and worked with ‘the men’, she missed out on the bulk enlistment of women telegraphists into the WRANS, and remained a civilian employee of the navy, a ‘temporary clerk’, throughout the war. In that capacity, she worked for Jack Newman’s directorate, and in later years of the war for other Melbourne-based intelligence units.
On one occasion, Jack Newman told her, ‘Since you’re a civilian here, if the Japanese invade us, you’ll be shot.’
‘What, as a spy?’ Joan said, jokingly.
Newman gazed at her and replied, ‘We are engaged in espionage.’
The second batch of Mrs Mac’s female telegraphers left Melbourne at the end of July to begin work at Harman. Joan Cowie and Jo Miller remained in Melbourne, where Newman gave them additional training in methods related to signals intelligence, then brought them in to work at Block E in the same room that he and Joan Duff occupied. The three women — Joan Cowie, Jo Miller, and Joan Duff — studied the call signs of Japanese ships, sorting lists and looking through patterns of communication for particular ships.
As Joan Duff explained it in an interview:
If someone found a copy of my telephone bill, all the numbers are there that I have rung in the last six months. Now if, for some reason, they didn’t know whose bill it was, they could find out by looking at those numbers that I frequently call. If you know someone’s calling pattern, you can tell it’s them by who they call. If Frank calls Fred, Josie, and Eleanor all the time, if you find a phone bill belonging to someone who called Fred, Josie, and Eleanor, it’s not going to be someone else now who’s calling, it’s Frank. Now, it might not be Frank, but you can more or less presume it is until you find Frank in another signal. And that’s what we were doing. We were identifying ships by monitoring who they called and how often.
They located and tracked Japanese ships across the Pacific Ocean, and, as they did so, marked them with coloured pins on a map of the Pacific Ocean on the wall. The map itself had a metal blind that was rolled down and locked to conceal it at the end of each day. One evening, Joan Duff, who was working late, returned to the office to discover that they had forgotten to lock away the map. A cleaner was standing in front of it, leaning on her mop, studying the map.
‘Those little dots,’ the woman asked, pointing at the map. ‘Would they be ships?’
‘I really don’t think we should be discussing this,’ said Joan.
‘Oh, it’s all right, luvvy,’ the cleaner said. ‘I’m Mrs Gathercole, and I’m Navy Intelligence.’
Joan reported the incident to Jack Newman the next day. Mrs Gathercole was reassigned, and Block E got a new cleaner.1
Newman’s ‘office’ was just a desk in the far corner with a tall partition in front of it. A frequent visitor was Commander Rupert B. M. Long, known as ‘Cocky’ Long, who was Australia’s director of naval intelligence. Long, sometimes dressed in naval uniform and sometimes in expensive suits, would saunter wordlessly past Joan’s desk beside the door and disappear behind Newman’s partition, where the two men would have long conversations in low, murmuring voices. As they talked, smoke from their pipes would waft from behind the partition to the ceiling.
Jack Newman and Cocky Long had known each other since 1913, when they were both 13 years old. That was the year they joined the navy, as part of the Royal Australian Naval College’s first intake of cadets. Their parents signed them over to the navy for five years of schooling plus a guarantee of 12 years of service after graduation — effectively locking them into 17 years of service. The benefit was that the boys would be career officers, and would join the inner circles of naval command. They lived, studied, and trained together at Osborne House in Geelong until 1915, when they relocated to the newly constructed college facilities at Jarvis Bay.2
In 1917, with the First World War in full swing, the boys graduated and were posted to ships: Cocky Long and three other cadets were sent aboard HMAS Australia, while Jack Newman served with the British on HMS Canada and HMS Columbine.3
After the end of the First World War, Jack Newman, considered by his superiors to be intellectually gifted, was posted to London for several years to undertake a specialist course with the Royal Navy in signals and communication.4 There, he developed an interest in signals intelligence.
As the new director of naval communications in March 1939, Newman travelled to Singapore for a conference on ‘wireless intelligence’, and to inspect the direction-finding apparatus in use there.5 On his return, he planned the construction of three direction-finding stations in Australia: at Canberra (Harman), Perth (Jandakot), and Darwin (Coonawarra), to be linked with the British radio-intelligence operations based in Singapore and Hong Kong.6 In August of that year, the navy created a new position, director of naval intelligence, and appointed Cocky Long to it. Newman wasted no time in convincing him of the merits of signals intelligence.
Like Newman, Cocky Long had also studied in London after the First World War, but while Newman was learning science and cutting-edge technologies such as radio, radar, and codes, Long trained as an old-school spymaster. Since 1934 he had effectively been running naval intelligence in Australia, so his promotion was more a matter of formalising what he was already doing.
Cocky Long was a charming and popular socialite. Barbara Winter, in her biography of Long, described him as ‘an extraordinarily skilled manipulator of people’:
He was friendly, constantly and professionally cheerful, likeable, helpful, a sympathetic listener, generous with his praise, he made people want to do what he intended them to do. A pair of elderly ladies touring a restricted area of Japanese territory would never have thought of themselves as spies; they were just doing a favour for that nice young naval officer.7
Long’s wife was the daughter of a wealthy Sydney businessman. Through her, he had entered the upper echelons of Sydney high society, and was well known and well connected. His connections included Captain Francis Edward de Groot, the man who in 1932 charged the opening ceremony of the Sydney Harbour Bridge on a horse and slashed the ribbon with his sword before the premier, Jack Lang, could cut it.8 Through de Groot, he was introduced to The Association, a discreet and informal ultra-right-wing organisation. Long cultivated a large personal spy network across the south-west Pacific, in large part through his father-in-law’s shipping company. He personally ran over 150 agents, whose files he later burned at the end of the war.9
Long was also in charge of the Coastwatchers, a network of individuals across Australian coasts and the south-west Pacific who would monitor and report on shipping movements. Some were locals, and some were Australian agents placed in the field. In the coming years, their reports on Japanese warship movements would be invaluable.
Long understood the value of sigint, but Newman was the expert on signals, so Long was happy to support his colleague’s drive to get it going. They lobbied the Australian government, but the prime minister, Robert Menzies, knocked them back, telling them by letter that there was no point in having an Australian cryptanalysis capability, since the British were already taking care of that in Singapore.10
Eric Nave’s arrival in Melbourne in May 1940 was the turning point. Long summoned Nave to Victoria Barracks for a meeting with himself and Newman to talk about establishing a cryptographic organisation. Nave was interested. Still sick and with a heavily pregnant wife, he had no desire to return to Singapore. He was assigned to work for Newman.11
Later in 1940, Newman invited Nave’s old boss, Captain Wylie of the Far East Combined Bureau, to a secret conference in Melbourne, at which Newman and Nave outlined detailed plans for an Australian cryptanalysis organisation.
And in March 1941, just before relinquishing his command of Harman and moving to Melbourne, Newman flew to Singapore again to attend an intelligence conference between Britain, Australia, New Zealand, and Holland, which had a signals-intelligence unit on Java called ‘Kamer 14’.
On his return, Newman prepared a secret report for the Australian government. He described how the Singapore operation had two sections, a ‘W’ and ‘Y’ section. The W section (which stood for ‘wireless intelligence’) was doing traffic analysis and direction-finding, while the Y section was doing cryptanalysis — actual code-breaking. He also reported that the other nations, including Britain, supported the proposal that Australia should have its own cryptographic organisation. Getting the British on side was the finishing touch in convincing Menzies, who finally relented.
The plan had come together by May. A new unit was formed, the Special Intelligence Bureau, with Nave in charge but reporting to Newman. The army offered the services of its intercept station at Park Orchards, and informed Newman about its secret little network of code-breaking professors in Sydney. The air force offered to contribute, too. It was willing to provide a signals team, so long as Jack Newman trained them. It also sent an officer, Henry Roy Booth, to Singapore for training and experience in sigint. And, of course, Newman had discovered an untapped well of signals talent and expertise through Mrs Mac’s Women’s Emergency Signalling Corps.
The air force kept their part of the bargain to establish a signals unit … barely. They sent a mere seven men to Jack Newman for training at Victoria Barracks. Their ranks were bolstered by two men sent from the army.12
Newman sent six of them to Darwin in September, where they began intercepting Japanese radio signals, sending the results back to Newman’s office in Melbourne. None were officers. When they got to Darwin, they reported to the local commanding officer, but operationally they reported directly to Newman in Melbourne.13 The Darwin RAAF command was told not to interfere or harass the small unit. They set up two Kingsley AR7 radio receivers on the top floor of the ‘Camera Obscura’ building at Darwin aerodrome, where they began monitoring Japanese naval radio signals across the western Pacific.14
The downside of Jack Newman’s stellar career was that the Naval Board would no longer let him serve at sea. He was stuck working on shore stations, known in the navy as ‘stone frigates’. Ship’s protocols applied on stone frigates, and they all had names as if they were ships; for example, the signals station he had established in Darwin was HMAS Coonawarra; the nearby Melbourne naval base was HMAS Lonsdale; and Harman, when the official designation came through, would be HMAS Harman. But these weren’t real ships. Calling them frigates didn’t change that.
Jo Miller said of Newman years later,
He took the trouble to visit our families in Sydney and assure them we were doing a good job. He met our Service friends, brothers and cousins who spent various periods in Melbourne training, in transit or on leave, and shared our concern when some were listed missing, wounded or killed. We knew his heartbreak at being rejected for sea-going duties time and time again. It was no consolation that he was found to be indispensable where he was. When my young brother was killed I remember him saying, ‘I feel so guilty, J.J., he was only a kid. I’ve spent my whole life being trained for war and have ended up with a safe shore job.’ No doubt this was why he worked so hard, at times working 19–20 hours a day.15
The mild and charming Newman differed markedly in personality from the introverted, anxious Nave, but the two men had several points in common: they were the two foremost experts in signals intelligence in Australia at the time, both were workaholics, and both were career naval officers stuck on land.
10
The Special Intelligence Bureau
Eric Nave made a trip to Sydney in the late autumn of 1941 dressed in plain civilian clothes. He met with the four academics who, led by Professor Room, had formed a code-breaking club that met at Sydney’s Victoria Barracks on weekends. Nave told them that he intended to set up a cryptanalysis organisation, but had no cryptanalysts. He asked them whether they would come and work with him in Melbourne.1
The professors were interested, so on his return Nave organised for them to travel down to discuss the idea further. Room and Treweek arrived in Melbourne in May, where they met with Eric Nave, Jack Newman, Cocky Long, and other high-ranking intelligence officers. It was noted in the minutes that, ‘Appreciation was expressed of the progress made by the unofficial Sydney group, which started without any assistance.’ But this was not true; the army had been actively assisting and guiding them for a year and a half.
Curiously, the minutes systematically spelt Room’s name as ‘Roon’ and Treweek as ‘Treweeke’. Perhaps this was just an oversight, or a muddying of the waters to protect the identity of the professors. At any rate, the reason is unknown.2 The professors accepted the request and agreed to move to Melbourne.
The navy contacted Sydney University and arranged for each of the four to be placed on indefinite leave from their academic postings. Treweek, being in the militia, was subject to the rules of call-up: he went to the army in Melbourne by arrangement, where he was immediately seconded to the navy. The other academics stayed as civilians. Room, Lyons, and Treweek were all in Melbourne by mid-year, while Trendall continued teaching until the end of the year, relocating to Melbourne in early 1942.
Captain Wylie, head of the Far East Combined Bureau, had by this time given up on the hope that Eric Nave would return to Singapore. Australia and Britain had engaged in a protracted paper war over the code-breaker’s role, and Australia had won.
The admiral commanding Malaya — Nave’s superior officer — insisted that he return to duty, but Admiral Colvin, with encouragement from Cocky Long and Jack Newman, replied that Nave’s doctor had ruled him unfit.3 The dispute went back and forth for months.
The Japanese naval flag officer’s code changed, rendering it unreadable for the Singapore cryptanalysts. Nave was the expert on that code, having first broken it, or rather an earlier version of it, years before. The British needed him to return and break the new flag officer’s code.
Admiral Colvin declined on Nave’s behalf, but suggested that the code-breaker could work on it from Melbourne, if the British sent some messages in this new code for him to study. A paltry bundle of enciphered Japanese flag officer messages subsequently arrived from Singapore, but too few for Nave to study and break.4
The British couldn’t argue with Nave’s sprue diagnosis. But surely, they suggested, he could return to England and resume his position there? That wasn’t possible, reported the Australian navy’s medical experts, because to get to England he would have to sail through tropical regions, again endangering his health. In that case, countered the British, they would fly him to England. The navy doctors ruled that out on medical grounds, too, since he would have to fly across a tropical region.5
Eventually, the British conceded defeat and, in a face-saving gesture, offered to ‘loan’ Nave back to the Royal Australian Navy on a permanent basis.
Nave filled two critical gaps for Australia in sigint: he provided expertise in breaking codes, and in the Japanese language. With the arrival in Melbourne of the Sydney code-breaking club, the cryptographic strength at Victoria Barracks was immediately boosted, but there was still the problem of finding the capacity to translate the Japanese language. Both Nave and Newman had been hunting around for local Japanese experts, but without much success. Nave himself was fluent, but if he was to run the show he could not be spending his time translating.
Nave found and recruited a Japanese linguist, Arthur Barclay Jamieson, known as Jim, who had returned to Melbourne in April after living in Japan for some time.
Meanwhile, Treweek already knew Japanese, and the other three Sydney professors were picking it up quickly. On Christmas Eve, Room wrote a letter to Walter Selle, the University of Sydney registrar, in which he praised Margaret Lake for the private lessons in Japanese that she had given him: ‘The two terms I have spent under Miss Lake have proved as useful in my present job as the 20 years’ mathematics!’6
The naval intercept stations and the army’s site at Park Orchards provided Nave’s unit with a steady trickle of material. There were already large, coordinated teams working on JN-25, the main Japanese naval code, in Singapore and at the American base in the Philippines. So, instead, Nave’s bureau focussed on Japanese commercial shipping, which the receivers in Australia could monitor more easily than those in Singapore.
The Australian stations could also intercept coded diplomatic messages between distant international locations, and the British encouraged Nave’s bureau to monitor Russian as well as Australian radio traffic.7
Nave’s bureau was also getting daily intercepts of messages from the local Japanese consulate in Melbourne. It is not clear how they got these, as the consulate had neither radio transmitters nor telegraphists, communicating instead by cable. But somehow all diplomatic cables in and out of the consulate were intercepted and sent to Nave’s team at Block E.8
When Jack Newman visited Singapore in March, Nave told him to ask them for the latest solution to the Japanese consular codes. Newman did so, but Captain Wylie refused.
Newman reported Wylie’s rejection on his return: ‘The head of the ‘Y’ section stated, however, that the Consular and Diplomatic Codes were now so complicated that a large staff of experts is required to obtain results, and that anything of interest read from this or other codes or ciphers would be forward to Naval Board.’
Newman then added, ‘This did not appear to me to be an ideal arrangement, and it is suggested that Paymaster Commander Nave makes further request for codes and cyphers which are regularly intercepted and which he considers may be of value to Australia.’9 As a final sarcastic dig, Wylie suggested that perhaps Nave could instead work on breaking the ‘45 sign code’, meaning the latest flag officer’s code.10
But by June it was all but out of Wylie’s hands. The Australian government approved the establishment of the Special Intelligence Bureau, and the British government approved it as well. The code-breaking professors from Sydney moved to Melbourne, and Nave’s new team was up and running. In June, a package arrived at Block E from Singapore, containing the complete solution to the latest Japanese diplomatic and consular codes.
The FECB also sent the solution to a non-military code, the Japanese merchant vessels code.11 Since Japanese merchant vessels moved around Australian coastlines and visited Australian ports, there would be plenty more material to work with that the two radio receivers could easily pick up.
Quite suddenly, the small team at Block E in Victoria Barracks had plenty to do. Along with diplomatic and merchant vessel traffic, they also investigated Japanese air force codes and JN-4, used by submarines.
In September, Jim Jamieson and Professor Room travelled to Batavia to visit the Dutch sigint unit, Kamer 14, and from there went to Singapore to study cryptanalysis under Wylie at the Far East Combined Bureau.12
Now that the four professors were engaged in actual code-breaking, some differences in approach began to emerge between the two mathematicians, Room and Lyons, and the classicists, Treweek and Trendall. The classicists felt that they had a pragmatic, holistic approach to the task, viewing the codes as a language problem; by contrast, they viewed the mathematicians as too technically focussed, looking for purist theoretical answers in a morass of imprecision, uncertainty, and human failings.
Treweek explained it this way: ‘The mathematicians wanted 100 per cent accuracy … But often you would get a garble in the text. And it was little things like garbles that gave the show away.’13
Long asked Eric Nave if, as a break from code-breaking, he could help with a code-making task. Long’s network of coastal spies, the Coastwatchers, needed a secure means of transmitting information back to Australia. Given their isolation and the fact that many of the watchers would be working on their own, the code needed to be simple and easy to use, but hard to break. Nave was keen to help, and designed the Coastwatchers’ Code with help from Professor Room.
Everyone acknowledged that Nave was brilliant, but he was not exactly a people person. Maybe it was due to 20 years of living in boat cabins in far-flung ports. Maybe it was the disruption of rotating between posts in Britain and Asia, never staying long enough to put down roots. Maybe it was two decades of immersion in the world of intelligence, code-breaking, and, at times, general espionage. Or maybe it was just that he was chronically ill. Whatever the reason, he came across as cold and aloof.
To make matters worse, despite having spent a lifetime in the navy, he had no instincts for military politics. In particular, he was oblivious to the rivalry between Jack Newman and Cocky Long.
While Newman expanded naval signals and code-breaking, Long had recently become the director of a new unit of his own, the Combined Operations Intelligence Centre (COIC). It was an intelligence one-stop shop for military command — gathering all the intelligence output of the army, navy, and air force, analysing it all, and passing it on to the relevant authorities. It was, in short, supposed to be the supreme intelligence unit in Australia.
Cocky Long had established COIC, with 13 personnel, in a special new facility called N Block, in nearby Coventry Street. The security at N Block was extreme. The building only had one door in or out, and that door had a hatch for messages to be passed through, in order to minimise the number of individuals that needed access.14
The products of the Special Intelligence Bureau should have gone to COIC via Jack Newman. But, instead, Long persuaded Nave to cut Newman out of the loop. Everything produced by the Special Intelligence Bureau went straight to Cocky Long; Newman was given a trickle of information on a need-to-know basis. Nave and Long did not tell Newman about their arrangement.
The issue came out in the open in the weeks before the outbreak of war in the Pacific.
In late November, Nave’s team discovered from intercepted merchant-shipping messages that Japanese ships in Australian waters had all stopped moving and were standing off Australian ports along the coast, awaiting instructions from Tokyo. He didn’t know why, but it seemed significant. Why were they not entering the ports? What instructions? Maybe war was around the corner, and they didn’t want their merchant ships stuck in Australia when it happened. Nave contacted Commander Long and told him what he had learned about the ships.
Somehow Newman discovered that Nave had passed this critical information on to Long, but not him, and used this as an opportunity to raise his objections with Nave. According to Nave’s memoirs, Jack Newman told him, ‘We keep the DNI out of these matters and handle them ourselves.’15 (The DNI was the director of naval intelligence, Cocky Long.)
Nave refused. He believed that Newman was telling him to cut Cocky Long out of the loop. He told Newman that wasn’t how things were done in Singapore and Hong Kong, where they did not play such political and bureaucratic games. It was, in his experience, ‘contrary to all practices both in London and the Far East, where we worked for the intelligence branch and they were trained to handle such with proper security’.
There was nothing Newman could do about this. The relationship between the two men was never quite as congenial from that point onwards.
11
East wind, rain
At the Special Intelligence Bureau in Block E, a daily stream of diplomatic messages arrived from the Japanese consulate. The code-breakers decrypted them and then passed them to the translators, Nave and Jameson. Immersed in conversations between the diplomats of a foreign nation, they gained a sense of the political mood in Japan. There were increasingly clear signs that war with Japan was imminent.
In November, for example, they read a message instructing the local consulate to find a consulate or embassy other than the Japanese consulate to represent Italy henceforth. The significance of this was that when the war started in Europe, Australia was immediately at war with Italy and severed all diplomatic ties with it. An arrangement was made whereby the Japanese consulate could, where necessary, make representations to Australia on behalf of Italy, acting as an intermediary.
The message from Tokyo instructed the local consulate to end this arrangement and pass Italy’s intermediary role to some other country. The reason for this seemed obvious to Nave. Clearly, Japan would no longer be able to represent Italy because if it declared war, it too would no longer have a diplomatic presence here. The inescapable conclusion, as it appeared to Nave, was that Japan expected to be at war with Australia soon.
Alarmed, Nave tried to bring this to the attention of head of the navy, but Nave did not have direct access to the admiral — only to his secretary, Captain Foley. He provided Foley with a copy of the message and explained what it meant, urging him to bring it to the attention of the highest levels of command. Foley was not convinced of Nave’s reasoning; to him, the message seemed like an arcane piece of bureaucratic trivia. Nave explained the implications again, and left it in Foley’s hands.
Foley ignored Nave’s advice. The admiral was never told of the message.
The relationship between Japan and the Allies, including Australia, was frosty — a far cry from the warm relationship between Japan and Australia during and immediately after the First World War. The friendly visit of a Japanese warship to Australia and New Zealand, when the young Eric Nave had been the official escort and interpreter, was a dim memory. By 1941 the public mood in Australia, and indeed in Japan, was typified by suspicion and hostility, driven largely by Japan’s militarisation and a series of wars in China.
In 1933, the Japanese army had invaded Manchuria, a coastal region to the north of China near Japan. They replaced the Manchurian government with their own puppet government, and stationed large numbers of troops there in a military occupation. There was international outrage. The League of Nations, the precursor to the United Nations, condemned Japan’s unprovoked expansion into mainland China. Rebuked but defiant, Japan withdrew from the league.
Manchuria was just the beginning. The Imperial Japanese Army began a low-grade campaign against China, using saboteurs and other agents to engineer a series of battles and skirmishes before a full-blown invasion in 1937. They first invaded Nanking, easily overcoming the defenders. The Imperial Japanese Army’s troops ran wild through the city, committing atrocities that would later be referred to as the Nanking Massacre, or the ‘Rape of Nanking’. It is unclear how many Nanking residents died at the hands of the Japanese army, but historians agree the civilian toll was somewhere between 40,000 and 300,000 dead. The details of some of the crimes committed on the residents of Nanking are horrific.
As details emerged, world opinion turned sharply against Japan. In America, there was a widespread belief that the government should ‘do something’. In response to this domestic pressure, the United States government started a trade embargo against Japan. In reality, it was a weak and toothless embargo: it was limited to direct sales of aircraft and military equipment by American companies to Japan, and in any case was non-binding. It was a symbolic gesture more than anything.
The embargo had no effect on the Imperial Japanese Army’s enthusiasm for expansion. The war escalated as Japan scored victory after victory. In 1938, Japan attacked and captured the major Chinese cities of Wuhan and Canton, took control of most coastal regions, and mounted massive air raids against civilian Chinese populations in several cities.
The world held the Japanese government responsible, but the Imperial Japanese Army was so powerful that the government may not have had the power to stop the war even if it wanted to. Observers had doubts about how much control the Japanese army’s commanders in Tokyo had over commanders in the Chinese battlefields. For example, the Nanking Massacre was neither policy nor strategy, but rather a symptom of an out-of-control military. Domestically in Japan, pro-military political forces were ascendant. The army made several attempts to seize outright control of the government.
In 1939, America cancelled its trade agreement with Japan. Since America produced half the world’s oil, this move had the potential to restrict Japan’s oil supply. Japan had anticipated this development, and had been stockpiling oil for some time. As its oil reserves started running down, the embargo effectively started the clock: Japan had a limited amount of time to resolve the issue.
In early 1940, Japan signed a new agreement with Germany and Italy, the Tripartite Pact. This new alliance bound the three more tightly, codifying a mutual goal of military conquest. The pact also stated that if one of them was attacked by a nation they were not currently at war with, the others would assist. This meant that the nations would not join forces immediately, but any new declarations of war from other nations would apply to all three.
As Japan’s territorial gains grew, so did the rift with the superpower on the other side of the Pacific Ocean. America banned Japanese ships from using the Panama Canal. To resolve the escalating trade war and situation in China, negotiations started between the two countries.
The talks between Japan and the United States continued through most of 1941. In late November, the Japanese ambassador to the United States, Admiral Nomura, told American journalists that, despite rejecting the latest American proposal insisting on his country’s complete withdrawal from China, he was ‘very hopeful’ that the negotiations would succeed. But, by this time, Americans were cynical. Senator Popper, not involved with the negotiations but an outspoken politician of the day, told journalists, ‘If they don’t like it, they can lump it. We have watched them murder, rape and ravage over a large part of the earth. If they want peace now, let them stop aggression and get out of China.’1
In late November, Nave’s team decrypted a message that left them in no doubt that war with Japan was imminent. The transmission would become known as the ‘Winds Message’.
Here is Nave’s translation:
19 November 1941
Owing to the pressure of the international situation we must be faced with a generally bad situation. In that event communication between Japan and the countries opposing her would be severed immediately. Therefore, should we be on the verge of an international crisis we will broadcast twice — during the Japanese news broadcast to overseas and at the end of it, the following in the form of a weather report:-
1. Japanese-American crisis: ‘East wind, rain’.
2. Japanese-Russian crisis: ‘North wind, cloudy’.
3. Japanese-British crisis, including the invasion of Thailand or an attack on Malaya or the Netherlands East Indies: ‘West wind, clear’.
Action should be taken as regards (? codes) and (? documents) in accordance with the above.2
The message made it clear that Japanese leaders believed they would soon be at war. To learn whom they would be at war with, the Japanese diplomats would have to listen to the news broadcasts from Tokyo and check to see if one of the three hidden ‘winds’ signals was buried in a weather broadcast.
If Japan went to war against America, the weather report would contain the words, ‘East wind, rain.’ If it was war with the Soviets, the weather report would contain the words, ‘North wind, cloudy.’ And if Britain, the report would say ‘West wind, clear.’ This kind of code — embedding a secret message inside a publicly available message — is called an ‘open code’.
Nave’s team later learned that American and British code-breakers had intercepted and read the message at around the same time. There were minor differences between Nave’s translation and those produced for the American and British governments. Nave liked to do translations that captured turns of phrase and subtleties of expression, while translations produced for the other major Allied powers were often more terse, to-the-point renderings of the main informational content.
The leaders of the Allied nations were informed of the Winds message. Allied intercept stations in Asia-Pacific made the monitoring of all Japanese news broadcasts their highest priority.
But in their haste to get all their signals people monitoring the Japanese news, coordination and organisation was lacking. The American cryptologist Lawrence Safford, the man who hurriedly issued the orders, did not take into account that some news broadcasts were in Japanese Morse code while some were in plain language, or the fact that news transmissions did not always occur in Tokyo time. Many interceptors were not even told what they were supposed to be listening for. They had no idea that phrases such as ‘East wind, rain’ or ‘West wind, clear’ might be important. They were merely told to make sure they listened to the news. It was not done systematically, but every available intercept station was monitoring Japanese news as effectively as they could.
On 1 December, all major Japanese naval ships ceased radio transmission. With the warships in radio silence, it was impossible for Allied sigint units to track and monitor their movements. The Japanese navy had done this before, in 1939, prior to the invasion of Hainan. If Hainan was any precedent, this did not bode well.
Curiously, Allied intercept stations were picking up just as many messages as they had in November, but none of the radio signals were coming from any naval ship of any importance. To hide the fact that its warships had gone silent, the Japanese navy was filling the airwaves with fake messages.
The biggest concern for Allied commanders was not knowing the whereabouts of Japan’s formidable fleet of aircraft carriers. Instead of guns, a carrier’s weapons are planes. They can engage in battle, attack cities, and destroy enemy targets, while safely hidden far out of sight beyond the blue line of the sea’s horizon.
The United States had seven aircraft carriers, and it believed that Japan also had seven. In fact, Japan had 10, and nobody knew where they were.
On 2 December, Park Orchards intercepted another important message from Tokyo to the local consulate. Nave’s team decrypted it to find that it was an order that the consulate destroy its codebooks and burn other important documents, in preparation ‘for the worst eventuality’.
Here is Nave’s translation of that message:
To: Japanese Consul Melbourne
From: foreign minister Tokio Circular 2445 2nd December, ’41
Subject: Orders for Destruction of Documents
URGENT MOST SECRET
Taking care that outsiders are unaware of it please see that the following measures are carried out.
1. Keeping one copy each of the ‘O’ code and TU code, burn all codes including the three services special code and the code for communication with the Army.
2. When the above has been completed cable the one word ‘HARUNA.’
3. Destroy by fire all files of inwards and outwards telegrams.
4. Take the same action with all other documents, being careful to avoid the suspicion of outsiders.
Further, as this is a preparation for the worst eventuality, please see that your staff proceed with their duties calmly and circumspectly and refrain from unlawful action.3
This message had not been sent just to the Australian consulate; it was a circular to all Japanese embassies and consulates. The single code word, Haruna — which was a mountain in Japan, as well as a Japanese girl’s name — was to be transmitted to Tokyo once the consulate had destroyed its documents as instructed.
It wasn’t long before the station at Park Orchards intercepted a message from the Japanese consulate back to Tokyo. The message was simply one word: ‘Haruna’. As soon as it was received, the officer in charge at Park Orchards, Captain Ian Lloyd, phoned Nave to inform him that the Haruna message had been sent.
On 4 December, all Japanese naval traffic sent in the latest version of the JN-25 naval code became unintelligible. It was immediately obvious that the Japanese had made a change to the code, either by issuing new codebooks, or new additive books, or both. In a practical sense, this made almost no difference, because the British and American teams in Singapore and the Philippines had not yet made enough inroads into the new version of the code, which they had named JN-25B, to read anything particularly important.
On the other hand, the fact that the Japanese navy had decided to do this now was significant. There had been changes to JN-25B every few months, normally at the start of a new month. This most recent change had been made three days after the warships went silent and only 24 hours after the Haruna message was sent.
The opinion was now universal across the intelligence and military organisations of the Allied countries in the Pacific that Japan was about to strike. The questions now were, who would they strike, and where? Given Japan’s radio silence, traffic analysis had become impossible, so it came down to guesswork. The American signals team in Hawaii, Station HYPO, had a theory that at least two of the Japanese carriers were near the US-mandated territories in the South Pacific.
Amid all this uncertainty, there was one thing that Allied commanders knew for sure: a large Japanese fleet was currently sailing south down the east coast of Asia. With nothing better to go on, the consensus was that Japan was most likely planning a military attack somewhere in the South Pacific.4 They were correct that an attack was imminent, but wrong about where it would happen.
Six Japanese aircraft carriers arrived on the morning of 7 December near the Hawaiian island of Oahu, home of the United States naval base Pearl Harbor. The first wave of the attack began at 7.48 am. About 180 planes launched from the carriers, bombing and strafing the ships in the harbour. Torpedoes dropped from the Nakajima B5N2 bombers, nicknamed ‘Kate’ by the Americans, and zipped through the water to their targets.
There had been eight United States battleships in the harbour, but seven were now destroyed. Of 400 aircraft based at Pearl Harbor, the vast majority were immediately either destroyed or damaged.5
The planes wheeled back to their carrier bases, giving a brief respite before the second wave of the attack began. The first wave had caused so much damage that the sky above the harbour was filled with billowing clouds of smoke from burning ships, planes, and buildings, and the attackers had to dive-bomb in order to find their targets.
By the end of the second wave, more than 2000 Americans were dead, and the fleet of battleships in the harbour was crippled. The one fateful mercy was that the US’s aircraft carriers were on a training exercise, and were not in the harbour at the time. Having destroyed as much as they could, and with their aircraft returned to the decks, the Japanese fleet departed.
The United States president, Franklin D. Roosevelt, gave a speech the next day: ‘Yesterday, December 7th, 1941 — a date which will live in infamy — the United States of America was suddenly and deliberately attacked by naval and air forces of the Empire of Japan.’
The Winds code has been a cause of controversy from the Second World War up until the present day. No less than seven congressional hearings were held into the matter. After thousands of hours of investigation, and countless books, articles, and hearings, it turns out that none of the American stations received the Winds code before Japan attacked. One man, the American cryptologist Lawrence Safford, was the original source of a claim that someone in the American intercept network had received the message and that there had been a cover-up by the authorities, but in the course of investigations he turned out to be a surprisingly unreliable witness. As time went on, despite him claiming to have seen confirming documents, it became apparent that Safford’s memory — apparently faulty — was the only source for his claim.
In the hours immediately following the Pearl Harbor attack, the Winds code was broadcast multiple times. For instance, later that same day — after the attack — a Japanese news broadcast described the Pearl Harbor battle in great detail. After breathlessly recounting the day’s events, the newsreader then abruptly said, ‘Allow me to specially make a weather forecast at this time: West Wind, Clear!’
An American translator who heard and recorded the broadcast made a note that ‘this may be some sort of code’. Apparently, he had not been given any instructions to watch out for special messages.6
It was a retrospective fantasy to hope that an intercepted Winds code might have given a clue to what the Japanese were planning in the days before Pearl Harbor. The original message stated that its intention was to inform the Japanese ambassador that war had broken out, or, to use the language of the message itself, that ‘communications have been cut’. It was never meant as a pre-emptive warning; it was explicitly an alternative communication channel to be used after the fact, to let diplomats know that they were now operating in enemy territory. And even if such a broadcast had contained the coded words prior to the war, it would have been of no practical use anyway, because it said nothing about Pearl Harbor: the Allies still would not have known when or where Japan was going to attack.
There was, in hindsight, no need to scan the airwaves for newsreaders saying, ‘East wind, rain’ or ‘West wind, clear’. The winds message itself gave the Allies all the information they needed about Japan’s intentions, regardless of whether the code was ever sent. In short, Japan was ready for war with the Allies.
Forty years after the end of the war, Nave, at the age of almost 90, was approached by James Rusbridger, a writer, who suggested that they collaborate on a book about Second World War signals. Rusbridger became a regular visitor to the retired officer’s home, holding long conversations about the war, signals, and Nave’s role. Rusbridger’s visits became so frequent that Nave complained to his friend, fellow veteran Jack Bleakley, saying, ‘Can you talk to him?’
Rusbridger’s true motive was to use his book, and Nave’s reputation, to promote a conspiracy theory about Pearl Harbor. He titled it, Betrayal at Pearl Harbor: how Churchill lured Roosevelt into war. For Rusbridger had come to the conclusion that the wartime prime minister of Britain, Winston Churchill, had been given advance information about the attack on Pearl Harbor, but had withheld it from the Americans because he wanted them to be drawn into the war.
Rusbridger’s theory did not even make sense on its own terms. If the Japanese navy was about to attack Pearl Harbor, and if Churchill knew, what would be the point of keeping that information from the Americans? The attack was going to draw the United States into the war anyway, no matter what happened. Telling them it was coming would not have prevented that.
Rusbridger was right about one thing: Churchill knew that some kind of attack was coming. That’s because the Americans had shared the Winds message with him. But, like them, he did not know who it would be against, or where it would take place.
Rusbridger’s key mistake was that he misunderstood the main Japanese naval code, JN-25. Rusbridger noted that it had been solved prior to Pearl Harbor. If it had already been solved, Rusbridger reasoned, then surely the British were reading the crucial naval messages in the preparations for the Pearl Harbor attack, and therefore knew it was coming.
What he did not grasp was that solving a code and breaking a code are two different things. The Allies had solved it, meaning that they knew how it worked: it used five numbers, an additive book that was applied in a certain way, and an indicator system.
But they had not broken the code. Breaking a code involves learning enough code words and indicators so that any given message can be read. At any rate, even if it had been broken, the Japanese had changed their codebooks and additive books ten days earlier, so it would have been unreadable regardless. Rusbridger was wrong, because he did not understand code-breaking. The bottom line was that he had no idea how hard it was or how much work was involved.
Nave was appalled by the claims in the book. It tarnished his reputation, marring his legacy as a pioneer in Australian sigint and obscuring his wartime accomplishments, by association with Rusbridger’s fevered conspiracy theory. Nave publicly disowned the claims about Churchill in the book, even appearing on Japanese television to do so.
Nave died in 1993, two years after the publication of Betrayal of Pearl Harbor, at the age of 94. Rusbridger committed suicide the following year.
12
West wind, clear
As Japanese aircraft carriers laid waste to the United States navy in Pearl Harbor, the 18th Division of the Imperial Japanese Army landed at Khota Baru in Malaya. Within hours, Khota Baru was in total Japanese control. Troops also landed in Thailand. To avert war, the Thai government quickly cut a deal with Japan, signing a treaty allowing the Japanese free passage. Later that day — still the same day as the Pearl Harbor attack — Japanese bombers flew over Singapore, dropping bombs on the island city. There were also air raids against Guam, a small Pacific Island occupied by the United States. The following day, Japanese troops landed on the beaches of Guam. The governor surrendered quickly, but not before the signals station there had destroyed all its equipment. They did such a thorough job that the occupying Japanese forces never realised that the installation had existed.
Immediately following the Pearl Harbor attack and the various other military operations across Asia and the Pacific, Australia, Britain, and the United States declared war on Japan. In turn, Japan activated the Tripartite Pact, and by the following weekend Germany and Italy had declared war on the United States. In hindsight, this was a mistake, as it meant they were now fighting on too many fronts. But most of Europe was now under Axis control, and the German army had advanced into Russia, and was only ten miles from Moscow. For the Axis nations, victory seemed within reach.
The British command in Singapore received reports of a possible impending invasion. To stop the invasion force before it reached the Malayan coast, they sent out a fleet of five ships, known as Force Z, from Singapore to intercept the Japanese. It has been sometimes claimed that the commander of Force Z, Admiral Sir Tom Phillips, made the rather foolish decision to proceed in radio silence, rather than call for air support.1 The reality is that low cloud-cover and unfavourable weather made flying difficult, and support from the local air force bases was not supplied. Two days later, when it seemed that Phillips was going to get his air force support, he was informed that the air base at Khota Baru had been abandoned, and that the northern airfields were quickly slipping out of British control.2
The fleet was located by the approaching Japanese forces. Japanese bombers moved in, dropping torpedo bombs and following them with surface bombing. They sank two of the ships, the Prince of Wales and the Repulse. Eight hundred British sailors on the two ships died, most of them by drowning.
Prior to Pearl Harbor, Singapore command had suffered from a torpor of complacency. Now, with a powerful enemy force approaching faster than anyone had thought possible, complacency was replaced by panic.
The Japanese advance toward the Malayan peninsula was rapid, in part due to high-quality espionage before the start of the war. They knew where all the British airstrips were, and they knew the roads, the towns, the locations of British troops, and the details of Singapore’s defences. British military planners believed that the Japanese could not possibly invade Singapore by land, in part because the jungles and the poor state of the roads meant that they would not be able to drive there. The Japanese army arrived with bicycles.
Within ten hours of the Pearl Harbor attack, air raids had begun on United States military bases in the Philippines. Two weeks later, the invasion of the Philippines began in earnest, with the Japanese army landing in the north of the island of Luzon and pushing south toward the capital, Manila.
Hong Kong surrendered on Christmas Day. On 31 December, the Japanese army seized Manila. Singapore, the vaunted bastion of the British Empire, surrendered on 15 February 1942.
Australia’s new prime minister, John Curtin, made a request to the British prime minister, Winston Churchill, for Britain to come to Australia’s aid in accordance with agreements made years earlier. But those plans had been drawn up in peacetime, and Britain was currently embroiled in a full-scale war across Europe, Africa, and the Middle East with the Axis nations.
From 1940 to the middle of 1941, the Luftwaffe — the German air force — had engaged in a prolonged series of air raids on Britain known as ‘The Blitz’, (named after the German word for the operation, Blitzkrieg, meaning lightning). The German Blitz had killed over 40,000 civilians, turning swathes of London and other British cites into wastelands of rubble. British troops were spread across Africa, while German U-boats were patrolling the Atlantic, sinking British naval and merchant ships. And the nations of Finland, Hungary, and Romania were now all under the umbrella of the Axis, and Britain was at war with them as well.
Churchill refused Curtin’s request.
With no aid from Britain, Curtin decided to recall Australian troops fighting in Africa and the Middle East back to Australia. Curtin’s relationship with Churchill was strained as the two argued in fractious telegram exchanges. Churchill was particularly angered by an article by Curtin published in the Melbourne Herald, in which he said, ‘Australia looks to America, free of any pangs as to our traditional links or kinship with the United Kingdom.’3
Curtin recalled Australia’s 6th and 7th Divisions to Australia. As they returned in troop ships across the Atlantic, Churchill tried unsuccessfully to have the 7th Division diverted to Burma to assist British forces there. As it later turned out, they would have been lost if he had succeeded.
Moving the Far East Combined Bureau out of Singapore was a priority for the British. It was imperative that the Japanese not learn how far advanced the Allied code-breaking efforts had gone, particularly on JN-25.
They sent an urgent request to Australia asking Jack Newman, the head of naval signals intelligence now working with Eric Nave, if he would be able to house the FECB in Melbourne, and if they would be able to commandeer all of Australia’s intercept stations for their own use. Newman had no intention of letting the British arrive and run the show, and wrote back telling them that, regretfully, there was no room for them.
Newman did not tell Eric Nave about the request or his response, but Nave discovered a copy of the exchange in paperwork that was left about the office. Emotionally connected to Singapore and oblivious to the politics, he was mystified as to why Newman would refuse to allow his former workmates to come and join them.
Rebuffed by the Australians, the Far East Combined Bureau FECB relocated to Colombo, boarding a ship at Selatar naval base in such a hurry that important equipment — including a replica ‘Magic’ cipher machine — were left on the docks.
In the chaos of evacuation, many people were left stranded and without assistance. Anyone not directly on the FECB payroll was left behind and had to find their own way out of Singapore. As a result, several British sigint experts followed Nave’s footsteps to Melbourne.
1942
13
Escape from Manila Bay
On the island of Corregidor in Manila Bay, the daily duties of the sigint agents had morphed from an abstract puzzle of codes and call signs to something much more personal. They were monitoring Japanese naval radio activity all around them relating to the invasion of the Philippines.
Several miles across the bay, a United States army sigint unit in the city of Manila was monitoring Japanese army and air activity. Under the command of Colonel Joe Sherr, the unit known as Station 6 was frantically monitoring incoming air raids. But in the early days of the war the importance of signals intelligence was poorly understood in the wider military, and they had trouble getting the army to take them seriously.
Howard Brown was one of the staff at Station 6, monitoring the radio traffic of the Japanese naval air force, which was launching air raids from Formosa (now known as Taiwan). The incoming planes often gave their targets away by transmitting bearings back and forth to each other and their bases. One morning, he called his colleagues at the Air Warning Centre to inform them that a hundred ‘heavies’ — the nickname for Japanese bombers — were on their way. The response on the other end of the line was dismissive and sarcastic: ‘That is interesting.’1
Undeterred, Brown called again later in the morning. He now had specific and solid information that there was an incoming air raid bound for Cavite, the American naval base on the south shore of Manila Bay, and that it would arrive in 15 minutes. Brown told the officer to sound the air-raid warning. But the air-warning officer replied that there was no need to do so because none of their own observers had seen any planes.2
As Brown had said, an air raid was imminent, and within minutes the naval base at Cavite was under attack from Japanese bombers and zero fighters. With the enemy planes now in sight, the air-raid alarms finally sounded. The bombers sank the submarine USS Sealion, destroyed the minesweeper USS Bittern by fire,3 and damaged several other ships.4 Cavite also had a large munitions store, which was destroyed.
Less than a week later, Brown called again, with detailed information about the flight paths of incoming enemy planes. Again, no action was taken, and the bombs dropped without an air-raid siren having sounded.
Brown complained to the Station 6 commander, Colonel Joe Sherr, who explained the problem to his superior, General Akin. The Air Warning Centre quickly became more cooperative, sending a delegation of three officers to coordinate and liaise. But it was all too late. The Japanese invasion of Luzon was now so far advanced that the sigint team at Station CAST could no longer be of much help.5 The enemy were now so close that by the time Station 6 detected an incoming air raid, the raid was almost upon them. Joe Sherr ordered his unit to evacuate to Manila, taking refuge on the island of Corregidor, where Fabian’s Station CAST was located.
The United States army made its final stand against the invading Japanese army on Bataan, the peninsula on the north of the bay near Corregidor. The fighting across that coastal outcrop of Luzon continued from January to April. By March, Bataan and nearby Corregidor Island were the lone holdouts in northern Luzon against the Japanese advance. Corregidor was subjected to daily air raids. The signals teams were hidden in the underground tunnel at Monkey Point, where they were protected from the bombs.
Joe Sherr was given a seat on a flight to Brisbane, but he could not bring the rest of his team on the plane. Instead the rest of the personnel of Station 6 evacuated Corregidor in four P-40 fighters, and travelled in a series of short flights to Del Monte airfield, on the southern Philippines island of Mindanao.6 They carried with them high-level written orders instructing that they were to be given priority evacuation to Australia. But at Del Monte airfield, they discovered that all seats on south-bound planes had been claimed by high-ranking officers already at the airfield — their written orders carried no weight. Although Howard Brown and four others somehow managed to board a flight to Australia, the remaining five personnel were stranded in Mindanao. They kept trying to find transport to Australia, eventually leaving in a small boat. After becoming lost at sea, they made landfall in New Guinea, only to be taken prisoner by the Imperial Japanese Army. Three of them survived until the end of the war; the other two died in a Japanese prisoner-of-war camp.7
Fabian told his naval code-breakers that their work on JN-25 was top secret and that he could therefore not allow any of them to be captured by the enemy. He told them that he would shoot them himself if it became clear they were to be taken prisoner.
It must have therefore been a relief when Fabian informed members of Station CAST in March that they were to be evacuated by submarine. They had to leave by night; with Japanese warplanes patrolling overhead, daylight was too dangerous. There were more than 50 evacuees, necessitating three submarines, which came on three different nights. On each occasion, the submarine entered Manila bay and then surfaced after dark, while the sigint team on the island travelled to the submarine by boat.
Beyond the bay, one of the submarines moved under the nearby Japanese fleet. The captain did not know the location of these ships, but the code-breakers did. An argument broke out amongst them about whether to tell the captain, because they were facing an excruciating problem: their lives would be in danger if the captain decided to surface in the submarine’s current position. On the other hand, their work, and particularly the breaking of JN-25, was a classified and well-kept secret. They were not authorised to tell the captain what they did or how much they knew, or how they could tell where the enemy ships were. Fabian — who was not on the submarine, but whom they would no doubt see again if they survived the journey — would be displeased and would probably discipline them if they tipped their hand. In the end, nobody broke protocol, and they kept quiet. The submarine surfaced, was located by the Japanese fleet on the surface, and had to immediately dive in order to escape.
They duly arrived in Perth, travelled by train across the Nullarbor desert to Adelaide, and went from there to Melbourne. There were no commandeered trains or carriages at their disposal, so they had to buy whatever tickets were available, and therefore had to go on different trains and on different days, in small groups or as sole travellers.
Jack Newman found them working space near his naval signals unit, whereupon the quiet offices and corridors of Block E became crowded and noisy. They told the locals, such as Joan Duff, harrowing stories of the Philippines, of hiding from air raids every night, of sheltering in their underground tunnel while the Japanese aircraft ruled the skies, and of their escape by submarine.
In contrast to those in Melbourne, for whom the hostilities felt distant and abstract, the war was very close for the American code-breakers, and they seemed almost in a panic about it. Joan Duff observed that, ‘The way they talked, you’d think the Japanese were on our doorstep, and we would be invaded any day.’
Another soldier from Corregidor had arrived in Melbourne before them. General MacArthur had been the US military commander in the Philippines at the time of the Japanese invasion. Feted as America’s greatest general, MacArthur was now humiliated by defeat. When it became apparent that the Philippines were lost, president Roosevelt instructed him to evacuate with his family to Australia.
This was easier said than done, because a fleet of Japanese warships outside the bay had set up a blockade. MacArthur could not leave by ship, and leaving by plane was even riskier. It seemed that he was stuck on Corregidor. The president offered the services of a submarine to help him escape, but MacArthur declined.
He, his family, and close friends and colleagues left the island in a small boat at night, headed out to sea, and snuck past the Japanese ships. They made their way to Mindanao, the southernmost island in the Philippines archipelago, and from there were flown in two B-17 Flying Fortress bombers to Darwin. On the day they flew in, the city was under attack from a Japanese air raid, so they avoided Darwin, detouring to Batchelor airfield, a military airstrip about 100 kilometres south of the city.
En route to Darwin, MacArthur received a briefing about the state of Australian readiness, and it wasn’t good news. His plan had been to go to Australia, marshal the mighty ANZAC forces assembled there, unite them with US forces, and recapture the Philippines. The president had led him to believe that there was a large Allied force in Australia ready for action. Unfortunately, that force did not exist; instead, the bulk of Australia’s military forces were engaged in Europe, the Middle East, and Africa.
This changed things somewhat.
At Batchelor, they boarded another plane and were soon airborne again, flying south to the desert town of Alice Springs in the geographical heart of Australia. From Alice Springs they travelled by train to Adelaide, and then to Melbourne.
They had to change trains in the small outback South Australian town of Terowie, as the rail gauge south of Terowie was a different size from the rail gauge to the north. Apparently, MacArthur’s ‘secret’ arrival in Australia was not as secret as he thought, because a group of locals and passengers on a nearby train cheered when they sighted him. Someone called out, ‘Welcome to Australia.’
MacArthur never let an opportunity for publicity pass him by. He stepped out where he could be better seen, and when he realised a local reporter was there, gave an impromptu speech.
The report of MacArthur’s Terowie visit the next day in the Adelaide Advertiser was gushing: ‘This remarkable soldier looked an impressive figure as he walked easily out of the train.’8 The newspaper said that ‘his flashing smile was a tonic for those around’, as he rejoined his wife and son. MacArthur apparently then asked if his words in Terowie would be reported back to the United States. When learning that they would be, he gave a speech. This is how the Advertiser recorded it:
The President of the United States ordered me to break through the Japanese lines and proceed from Corregidor to Australia for the purpose, as I understand it, of organising an American offensive against Japan, the primary purpose of which is the relief of the Philippines. I came through and I shall return.
That last line was a winner, proving singularly effective at promoting his presence in Australia and shoring up public morale both in Australia and in America, where his words were relayed. The phrase ‘I shall return’ was an effective wartime slogan, getting such traction that he repeated it in several subsequent speeches, always delivering the line with panache and confidence.
Australia’s new prime minister, John Curtin, was star-struck by the famous American general, and organised to meet him as soon as possible. When they met in person, MacArthur told him, ‘Mr Prime Minister, you and I will see this thing through together.’ But he would be in charge of the war itself, adding, ‘You take care of the rear and I will take care of the front.’9
14
Beirut bookshops
‘Uncle’ Jack Ryan’s signals team, Four Special Wireless Section, were camped in the mountains east of Beirut when Pearl Harbor was attacked.
After gaining some respite in Cairo, the unit had spent the intervening time since the Battle of Crete in Palestine and Lebanon, initially for the campaign in Syria, and, once that wound down, gathering intelligence on Russian movements in Turkey.
With the war now opening in the Pacific, they shifted their attention to what Geoff Ballard called ‘Thinking Japanese’. One important thing they started to consider was the Japanese version of Morse code, which was quite different from the version they were used to. The signals unit knew all about Morse code and were experts in it, because Morse was the standard method of transmitting radio messages.
Morse code, invented in America and widely used in Europe, was poorly suited to the Japanese language. There are three writing systems in Japanese. Kanji uses Chinese pictographs, each word having its own unique symbol. Since words expressed this way can’t be broken down into a small number of elements, kanji was unsuitable for Morse. The other two systems, Hiragana and Katakana, resemble a European-style alphabet, but instead of having symbols for sounds (such as ‘k’), they use symbols for syllables (with separate symbols for each of the sounds ka, ki, ko, ke, ku).
As a result, Japan had developed its own Morse-style system, called the Japanese Morse telegraphic code (or sometimes the Wabun code), with 75 symbols for all the orthographic elements of Katakana. The Allied signals units called it the Kana code.
In the Kana code, the syllable ka was represented like this: • — • •
The symbol for ki, on the other hand, was represented like this: — • — • •
All the other syllables had their own combination of dots and dashes, just like standard international Morse. But for the signals unit this was a completely new system, which demanded the knowledge of a much larger number of patterns. Lieutenant Hill obtained details of the Kana code from British agents at the base in Sarafand al-Amar, and they all began lessons with him.
The intelligence staff drove down to Beirut to scour the local bookshops for anything on the Japanese language. Only a few months earlier they had made a similar errand to Beirut looking for Russian-language textbooks. By the end of the day, they had acquired three books on the Japanese language.
In mid-January, ‘Uncle’ Jack Ryan asked them a very leading question.
‘How long would it take us to pack up if we had to move out of here in a hurry?’1
Two days after Ryan’s hint, they were given orders to move out. They moved to Palestine, where Mic Sandford rejoined them.
After Crete, Sandford had travelled with the unit to Palestine and Beirut, but once the Syrian campaign was won by the Allies, he had departed to North Africa on unspecified intelligence assignments.
On his return, it was apparent he had lost none of his exuberance, as he regaled them with stories, most notably of a hair-raising truck journey he had made near Tobruk with a senior British officer. Ballard documented Sandford’s anecdote:
Sighting the dust of a transport column overhead, they scarcely knew what to expect but as the convoy drew nearer, Mic thought he recognised the insignia on the lead truck.
‘Looks like Ariete (one of the Italian divisions), Sir,’ said Mic.
However, on close inspection, they both recognised the sign of Rommel’s Afrika Korps!
‘There was nothing else to do,’ said Mic, ‘but to drive along in the swirling dust, saluting each truck as it went past — and I’m sure we put a spurt on when we got to the end of the column!’ 2
Four Special Wireless Section waited in a holding camp for several days before taking the night train to Suez, and from Suez boarded a ship that took them back across the Indian Ocean to Adelaide.
The voyage took five weeks, during which time they learned Japanese under the instruction of two new additions to the unit, William E ‘Nobby’ Clarke and Bill Kalbfell, making use of the language textbooks they had bought in Beirut. Vic Lederer, who had grown up in Germany but had fled because of his Jewish heritage, was the section’s most proficient German translator, and was frustrated that he was ‘starting from scratch’. But, like the others, he applied himself to the new task, and before long was mastering Katakana and learning some of the rudiments of Japanese.
On arriving in Adelaide, they were billeted in local homes — Mic Sandford, being an Adelaide local, was fortunate enough to be able to stay with his family — for three weeks before moving down to Melbourne.
General MacArthur was already in town, having arrived in Melbourne a few weeks before Uncle Jack Ryan’s wireless unit, and had established a new headquarters at 41 Collins Street, courtesy of the Australian government. MacArthur was appointed as commander of the south-west Pacific area, and Prime Minister Curtin enthusiastically offered Australia’s army and air force for him to command. Within two days of MacArthur’s arrival, Curtin recalled General Blamey, who was on duty in the Middle East, to Australia. When Blamey arrived, Curtin appointed him as the new commander-in-chief of Australia’s armed forces. Blamey’s promotion to supreme military commander was short-lived, however, because Curtin immediately instructed him to henceforth report to MacArthur, who would be his commanding officer.
Blamey, raised on a farm near Wagga Wagga in New South Wales, had served in the First World War at Gallipoli and the Western Front. In the period between the wars he became the police commissioner for Victoria. He was a short, round man with a neat moustache. He was bright, tough, and known as someone who had no patience for fools. He immediately began a wholesale clean-out of career soldiers in the upper ranks of the Australian military, believing many of the long-serving senior officers to be over-promoted and inept, and replacing them with fresh talent.
MacArthur met Blamey for the first time at the Menzies Hotel on 29 March. According to journalists Shane Maloney and Chris Grosz, ‘The American supremo swept into the room as if preceded by a flourish of trumpets. Lean and clean-shaven, sporting his trademark corncob pipe, ostentatiously braided cap and Hollywood-style aviator sunglasses, he found himself face-to-face with a short, rotund figure wearing a grey moustache and Bombay Bloomers. At 57, seven years younger than MacArthur, Blamey was the very image of Colonel Blimp.’3
Blamey was able to work with MacArthur, and in fact over the following years was the only non-American of the five generals who comprised MacArthur’s inner circle of command. The other four were Charles Willoughby, also from MacArthur’s Philippines base at Bataan, now assigned to intelligence; Lieutenant General Stephen J. Chamberlin, the operations chief; Major General Richard K. Sutherland, MacArthur’s chief of staff, and Spencer B. Akin, who had escaped with MacArthur from Corregidor, and who had been in charge of signals at the United States army base in Manila.
In the Philippines, Akin ran the ‘Voice of Freedom’, a propaganda radio program broadcast three times a day.4 From a temporary, makeshift office in a commandeered school in Melbourne, MacRobertson Girls’ High School, Akin began investigating what was around. He understood the value of signals intelligence, and started working on the idea of building his own organisation in Australia that he could use entirely for MacArthur’s benefit. He learned about Newman and about the wireless unit that had recently returned from Palestine.
Mic Sandford was notified that Akin wanted to meet with him at his office at the school.5 There is no record of Sandford’s meeting with Akin or what they discussed. But in the days following that meeting, things moved quickly. The next day, Blamey appointed a new chief signals officer, Colin Simpson, who also promptly visited Akin. The two of them decided to persuade MacArthur and Blamey of the need to build a signals-intelligence capability in Australia.6
Three days later, the head of Australia’s army intelligence circulated a high-level memo suggesting that Sandford’s intelligence team could form the nucleus of a new signals bureau, referred to in the memo as a ‘Central Bureau’.7
And the day after that, Simpson held a high-level meeting at Victoria Barracks, inviting the key individuals in Melbourne who could play a role in signals in Australia. Jack Newman and Eric Nave were there, naturally. ‘Uncle Jack’ Ryan and Mic Sandford were in attendance. Joe Sherr, the head of Station 6 in Manila was there, and so was Rudy Fabian, commander of Station CAST from Corregidor Island. Norman Webb of the British army, who had fled Singapore, was also present, as was H. Roy Booth, an Australian pilot and intelligence agent who had also escaped Singapore in the weeks before its capture two months earlier.
The unpredictable currents of wartime had carried a diverse assortment of people with interest and expertise in sigint to Melbourne: American army traffic analysts from Manila; US navy code-breakers from Corregidor; British army code-breakers from Singapore; Mic Sandford’s team from Palestine; and indeed Eric Nave, who was a hybrid creation of the Royal Australian Navy, the British Royal Navy, and Bletchley Park.
Nave’s unit had collected three British diplomats who had fled Singapore before its capture by the Japanese. They were Henry Archer, Hubert Graves, and Arthur Cooper. Cooper was a particularly eccentric character, and was well known for his habit of carrying a pet gibbon ape everywhere he went. The gibbon’s name was Tertius. It had been his pet in Singapore, and he had smuggled it with him when he had fled to Australia. They were now working with Nave on diplomatic traffic and the Japanese Merchant Shipping code, JN-40.8
None of them had expected to find themselves in Melbourne, but since they were, they had an opportunity to create something new.
At the conference at Victoria Barracks, Simpson proposed that they create two organisations: one for interception, and the other for cryptography and research. The Australian Special Wireless Group would take control of existing intercept stations and equipment, and deploy new intercept units around the south-west Pacific. The other component, Central Bureau, would be a research-and-control centre for traffic analysis, code-breaking, and intelligence reporting that would liaise with the sigint services of America and Britain. These two organisations were closely intertwined, and had a symbiotic relationship from that moment forward.
For the intelligence component, they wanted a bland, nondescript name. If someone learned about the existence of the organisation or accidentally overheard it mentioned in conversation, that person’s curiosity should not be piqued. ‘Central Bureau’ was perfect, because it gave nothing away at all. It was, quite literally, the most boring name they could think of.9
Central Bureau would be directed by Spencer B. Akin, with three assistant directors: the Australians, Mic Sandford and H. Roy Booth, and Joe Sherr of the US army. Their assignment was to break the Japanese army codes, which were not only completely unbroken, but had not been seriously investigated by any Allied nation.
Rudy Fabian had no desire to get involved in what he saw as ‘secret inks and all that crap’.10 As far as he was concerned, the naval code JN-25 was the main game, and that code was his single-minded focus. It had the potential to turn things around in the Pacific War immediately. As a concession to the multi-national spirit of things, he conceded he was willing to cooperate with the Royal Australian Navy. Working with army and air force personnel, whether US, British, or Australian, was out of the question.
Fabian got his wish, and a separate organisation was set up, a joint operation between the United States navy and the Australian navy, to tackle Japanese naval codes. The organisation was initially named FRUMEL, ‘Fleet Radio Unit, Melbourne,’ to parallel the American signals base in Hawaii, which was renamed from HYPO to FRUPAC (Fleet Radio Unit, Pacific). Jack Newman was head of the Australian contingent at FRUMEL. Eric Nave, although he was still on the payroll of the British, remained seconded to the Royal Australian Navy, and also joined FRUMEL. Nave’s diplomatic section came along, too, with Trendall in command.
With FRUMEL unravelling the Japanese naval codes, and Central Bureau assigned to tackle Japanese army codes, the question remained of who would take on the Japanese army and navy air force codes.
Imperial Japan was in the unusual situation of having two air forces, with the army controlling one and the navy the other. Each of these two air forces had their own distinct code systems. The Allied code-breakers referred to these code systems as the ‘Army Air to Ground’ and ‘Naval Air to Ground’ codes. Fabian had no interest in either of these. The responsibility of breaking all the air-to-ground codes, including the navy air-to-ground codes, was given to Central Bureau. 11
15
The secrets of the Coral Sea
On an April morning at the Naval Signals Office in Block E, Newman announced that they were all to move from Victoria Barracks. The Royal Australian Navy commandeered a nearby apartment block, Monterey Flats, just a short walk away overlooking Port Phillip. Holes were knocked in the walls between the myriad one-bedroom apartments to create larger work spaces.
With the civilian tenants evicted, Monterey Flats filled up with the various units working for Newman and Fabian: the Americans from Corregidor; Nave and his team; Newman’s own traffic-analysis team; the professors; some former members of the FECB who had escaped from Singapore before its capture; and numerous new Australian recruits.
Monterey was focussed on Japanese navy traffic, particularly JN-25. They had the advantage that the code had already been broken: even if they did not know all the code words and additives, they understood its overall structure. They also benefitted from the fact that navy messages, broadcast as they were from ship-to-ship, or ship-to-shore, across the vast distances of the ocean, were easy to pick up from radio-intercept stations in Australia.
The Royal Australian Navy had already started constructing a new intercept station south of Melbourne on a farm at Moorabbin. However, there was a chronic lack of manpower due to the war, and work was progressing slowly. The Americans from Station CAST drove down to Moorabbin and offered their assistance.
They installed poles for the antennae, converted a shack on the farm into an office, and built a barracks for the male troops who would be stationed there. It was decided that a pre-existing farmhouse on site would house the women. With the entire Station CAST personnel on the job, the Moorabbin site was soon operational. As American navy radioman James B. Capron put it, ‘In short order, we converted a farm into a first-rate intercept station.’1
Newman had an office on the ground floor at the northern end, with Joan Duff and two senior WRANS, and nearby, closer to the main entrance, was a roomful of newly recruited WRANS performing the tedious but critical tasks of sorting messages, collating call signs for traffic analysts, transcribing Kana, and various other tasks.
Fabian installed himself upstairs, collaborating with his colleague Joe Rochefort in Hawaii, head of the naval code team FRUPAC, as well as still collaborating with the FECB, which had fled Singapore and was now in Ceylon. The three centres coordinated their teams for a massive three-pronged attack on the JN-25 code. Treweek, one of the four Sydney professors, also moved in and, unlike Nave, established a good working relationship with Fabian.
Fabian, a master of military politics, consolidated his control at Monterey. MacArthur wanted access to the fruits of Monterey’s labours, but since FRUMEL was part of the US navy, it was out of reach of MacArthur’s command. Fabian met with Vice Admiral Herbert Leary of the US navy and discussed how to handle this. They did not want to anger MacArthur with an outright refusal, but did not want to allow him into their command structure either. The solution they came up with was that Fabian would personally deliver all key intelligence to MacArthur — which was what he wanted — but he couldn’t then distribute it to his own organisations. Having talked it over with Fabian, Admiral Leary signed an agreement with MacArthur, but baked some caveats into the agreement:
(1) Fabian or one of his unit’s representatives will report to MacArthur’s headquarters each day at 1400 hours. The FRUMEL representative will never be kept waiting in MacArthur’s outer office.
(2) No one will be authorised to make copies of any material provided by FRUMEL.
(3) During the briefing of FRUMEL material, only MacArthur and his chief of staff, General Sutherland, will be present. Everyone else, including MacArthur’s chief of intelligence, General Charles Willoughby, will be excluded from these briefings.2
The upshot was that, yes, Fabian reported to MacArthur, but MacArthur had no control over Fabian or his FRUMEL unit, and although MacArthur would be given intelligence briefings from Fabian, he had to keep them to himself.
Fabian was the winner from this. The arrangement entrenched his control at Monterey, and gave him complete control over the distribution of intelligence produced by FRUMEL.
Fabian enforced the guidelines, personally reporting to MacArthur each day at his headquarters in Collins Street with the latest information on Japanese navy signals. While information out of FRUMEL was tightly controlled, he maintained and strengthened communication lines with the sister naval sigint operation in Hawaii, FRUPAC (previously known as Station HYPO), and reported upwards to Vice Admiral Leary. In keeping with Leary’s directive, he did not report to Central Bureau or exchange any information with them at all.
Joan Duff, Newman’s clerk, found the Americans curiously informal compared to the protocols she was used to in the Royal Australian Navy. Nobody had to wear jackets, for instance, if they didn’t want to (although they still had to wear ties). Even more striking was that their commander, Lieutenant Rudy Fabian, was known to everyone under his command as ‘Rudy’. It was unheard of in the Australian navy for ratings and junior officers to refer to their superiors by their first name. It didn’t seem to lessen his authority, however. Fabian made a great show of being companionable with everyone, but his jokes often had an edge, and he could switch from being ‘one of the boys’ to an authoritarian in an instant. In return, his subordinates joked around with him, but grumbled behind his back about what a taskmaster he was.3
As for Jack Newman, he knew what he had to do: get along with Fabian. The American was irascible and difficult, but Fabian was brilliant in his role as commander of a code-breaking operation. Fabian was fixated on a single idea: breaking JN-25 would win the war.
The Japanese naval code they had to unravel, JN-25, was an old-school system using codebooks. Their task was rather different from the problem confronting the British code-breakers in England, led by Alan Turing, who were cracking the German’s Enigma machine. The Enigma was a machine cipher that, with its ingenious array of wheels and wires, encrypted each letter separately, one by one. The Enigma saw only letters, never words.
That is the fundamental difference between a code and a cipher. A cipher machine changes each letter, or each symbol, on its own. A code changes the word as a word, using a codebook.
When you crack a machine cipher like Enigma, if you know all the settings, you can unlock the entire message, all at once.4 But for a code, unless you stumble across a copy of the codebook, there’s no short cut. You have to try to figure out every word in the codebook, one by one.
Abe Sinkov explained it like this, years later: ‘Solving a code message or code system is different from solving a cipher. Normally when you solve a cipher, you can read the entire communication without any problem. In solving a code it’s a process of little-by-little reconstruction of the book. It means that messages are readable only in part in the early stages and there will be gaps.’
The point is that the word ‘solve’ can be misleading. ‘Solving’ a code system just means that you know how the thing works. It doesn’t mean that you can read any of the messages.
John Tiltman, with some help from Alan Turing, had ‘solved’ JN-25 by 1941 — or, to put it another way, the code-breakers at Bletchley Park knew how JN-25 worked. For example, they knew that it was a five-digit code, that it used a codebook to translate words into five digits, and that there was a second, additive book that the message sender would use to then add to the original numbers. But knowing all of this didn’t help them read a single message. 5
Breaking the latest JN-25 codebooks that had come into use before Pearl Harbor was slow work: although the code system was now well understood, reading any of it relied entirely on mistakes being made by Japanese operators. As Duane L. Whitlock, one of Fabian’s cryptanalysts from Corregidor, said, ‘No matter how much talent was thrown at the problem on either side of the Atlantic, entry into the Japanese navy operating code in either case moved forward initially with only such speed as the Japanese errors would allow.’6
At this early stage of the war, mistakes were still plentiful, but would soon dry up. For instance, the Japanese language at the time was very formal, laden with introductions, greetings, and polite preambles. Even in completely encrypted messages, these were easily recognised by experienced code-breakers, and provided small pieces of the codebook. Weather reports were also useful. The Imperial Japanese Navy had a habit of sending these at the same time each day in the same format, allowing an alert code-breaker to pick up some more easy wins. Preambles and weather reports were not going to give up the whole book, and were unlikely to provide clues to important book entries to do with military movements and strategy, but they were a crack in the JN-25 wall that, over time, could be levered into a gaping hole. When the books changed again, as they did every six months, they would be faced with a brand-new wall to break through, but that was a problem for the future.
At the time they were still newly installed in Monterey, the JN-25 code was only about 20 per cent readable — which meant that, for any given message that arrived encoded in JN-25, the code-breakers could read about one in five words.7 Traffic analysis was proving far more useful, however, so that, when combined with patches of decrypted messages, the Monterey code-breakers discovered some important and urgent information.
On 9 April, a message was received from the Japanese commander-in-chief, enquiring how repairs were going on a damaged Japanese aircraft carrier, the Kaga. The Kaga had already been deployed against the United States and Australia with destructive effect. In December of the previous year, it was one of the carriers involved in the Pearl Harbor attack, launching its planes on deadly missions against the American port from beyond the horizon. Three months later, it attacked Darwin, again out of sight from land, launching fighters against the northern Australian city. Returning from Darwin, it had hit a reef, taken damage to its hull, and limped to the safety of a Japanese-controlled port for repair.
The interesting thing about this message was that the reason for the commander-in-chief’s curiosity was that he wanted Kaga to participate in a new campaign against the Allies, the ‘RZP campaign’.8 The traffic analysts at Monterey knew that RZP was a Japanese naval placename code for Port Moresby, the capital of Papua New Guinea, not far north of Australia. The message did not reveal the nature of the campaign was. Perhaps it would be a major air raid, in the same vein as Pearl Harbor and Darwin. However, other information in their possession was leading them to believe that the RZP campaign would be an invasion, not an air raid. If the Japanese managed to control Port Moresby, they would control all shipping around the east coast of Australia, and Australia would be completely isolated from the United States. It would be a major strategic win for Japan, and a disaster for the Allies.
Fabian told MacArthur of this development at his daily briefings with the general, and showed him the translation of an intercepted message in which a Japanese commander had stated the campaign’s objective:
FIRST TO RESTRICT THE ENEMY FLEET MOVEMENTS AND WILL BE ACCOMPLISHED BY MEANS OF ATTACKS ON THE NORTH COAST OF AUSTRALIA.
MacArthur was incredulous, but Fabian explained the code-breaking process, and how signals intelligence worked, to persuade MacArthur that it was true.
Fabian later said of a meeting with the general:
When I told MacArthur about the messages that we read about Japanese plans to invade Port Moresby, New Guinea, he said, ‘That can’t be right.’ My idea of their strategy was that the Japanese would go to New Caledonia and then proceed to close off north-west Australia. I explained the whole COMINT process including how the information was derived from code groups. I convinced MacArthur to change his plans. A transport was scheduled to leave Townsville for New Caledonia the next day. Instead, they were sent to Port Moresby.9
The other US naval centres, including the team at Hawaii led by Joe Rochefort, were notified. More nuggets about the Port Moresby plan soon came in. The intercept station just south of Melbourne at Moorabbin intercepted a message that, when decrypted at Monterey Flats, said, ‘The Fifth Carrier Division is on its way to Truk.’ They knew that the Japanese ‘Fifth Carrier Division’ was a fleet of just two ships, the Shokaku and the Zuikaku. Truk — also known as Chuuk — is a small island in the Pacific almost 2000 kilometres north of New Guinea, and at the time was the headquarters for Admiral Yamamoto, commander-in-chief of the Japanese combined fleet. Moving to Truk meant that the Fifth Carrier Division was moving within striking distance of Port Moresby. Another intercepted message said that they were going to be joined at Truk by a second naval fleet, the Fifth Cruiser Division, and by yet another unit that FRUMEL could not identify.
On 23 April, Fabian provided this intelligence briefing to MacArthur, revealing that the Japanese navy was amassing a large force at Truk:
Forces at or approaching Truk now include 5th carrier division, SHOHO [another aircraft carrier], 5th Cruiser Division less Nachi, 29th Destroyer Division, Commander Submarine Force, and half of the 8th Submarine Squadron.10
Their fears were confirmed the very next day from traffic analysis of unread messages. The commander-in-chief of the Japanese Fourth Fleet sent radio messages to a unit or units called the ‘RZP Occupation Force,’ and ‘RXB occupation force’. The analysts at Monterey did not have to read the messages to understand what this meant. The Imperial Japanese Navy was planning to invade and occupy both Port Moresby (RZP), and Tulagi, an island in the Solomon Islands to the north-east of Australia. If the Japanese navy controlled either, or both, of those locations, Australia would really be in trouble.
In late April, Central Bureau, analysing all the available signals intelligence, advised General Headquarters to prepare for likely carrier attacks on Cape York, Townsville, and possibly Brisbane.11
Through late April and into May, from messages passed across desks at Monterey Flats, they watched the build-up continue on an island thousands of miles to the north and far from the eyes of Allied aircraft. New additions to the Port Moresby invasion force included a squadron of destroyers, a submarine fleet, and nine Zero fighters that would be loaned to one of the aircraft carriers from Rabaul, closer to New Guinea. Then, characteristic of all major Japanese naval operations, there was a lull in radio activity. Of the messages that did make it to the intercept stations and then to Monterey, the call signs for all ships involved in the assault had been changed. There was an increase in ‘high priority’ radio messages, which the analysts at Monterey decided had to be an attempt to throw them — the interceptors, or any others who were listening — with spurious, ‘dummy’ radio traffic, making it hard to know which messages really were high priority.12
There was no doubt about the target, and there was no question that the Allies had to stop the attack from succeeding.
At Monterey Flats, everyone was working to capacity as the code-breakers tracked the approaching fleet. Joan Duff, still working on traffic analysis for Newman, recalled that one of the American code-breakers, Ensign Ralph Cook, rushed in and asked her if she knew the call sign of a particular ship.
‘I don’t know,’ said Joan. ‘I’d have to look it up. They’ve all changed recently.’
‘Oh never mind,’ said Cook. ‘They’ll be here soon enough and I can ask them myself.’ He then pretended to tap an imaginary person on the shoulder. ‘Excuse me, Mr Moto,’ he said, ‘Could you please tell me your call sign?’13
MacArthur contacted Nimitz, the US general commanding the war in the central Pacific, and commander of the US fleet. They hurriedly organised a fleet of ships to stop the Port Moresby invasion force, but did not want to alert the Japanese that they knew of the attack or that they were preparing a response, so they arranged for fake orders to be sent for the ships involved to return to Hawaii and then they put them into radio silence.
There were two invasions planned: Tulagi Invasion Force was to arrive on 4 May, and then the Port Moresby Invasion Force would sail for Port Moresby.
The Tulagi invasion force landed at Tulagi, and took control of the island on 3 May, completing the first stage of the operation. The Allies learned of the Japanese landing from one of the Coastwatchers, the network of spies across the south-west Pacific established by the Royal Australian Navy. The coastwatcher was Donald Kennedy, a New Zealander who was the district officer of the nearby island of Santa Isabel to the north. Kennedy sent his report of the landings by radio, then fled Santa Isabel for the neighbouring island of New Georgia, where he set up a new, hidden base and continued to operate as a coastwatcher throughout the war.14
The aircraft carrier USS Yorktown sailed north towards the Solomons, and its aircraft were in striking distance of Tulagi by 6.30 the following morning. The Yorktown then launched a series of raids against the Japanese naval ships still berthed at Tulagi, sinking the destroyer Kikazuki as well as five landing barges and five floatplanes. In return, the Japanese at Tulagi shot down three of the Yorktown’s aircraft. Late in the day, the Yorktown returned south to rendezvous with the rest of the Allied fleet and make a refuelling stop with the oil tanker USS Neosho.15 The Yorktown’s attack had reduced Japanese air power; but, on the other hand, it had alerted them that there was at least one carrier nearby.
Tulagi was just the prelude; the main game was Port Moresby. In addition, the code-breakers had established that a Japanese strike force consisting of two aircraft carriers and several other ships were silently circling down to the east of the Solomons, toward the Coral Sea. This strike force was planning to ambush the Allies as they moved to defend Tulagi and Port Moresby.
The Allied commanders were ready to attack the attackers, but were still determined not to give them any reason to suspect that their codes had been broken, and that the Allies were reading their messages. The invasion force was coming; the Allies knew the day, and they knew the path, so MacArthur ordered a reconnaissance plane to fly out so that it could just happen to find the Port Moresby invasion force. That way, the Japanese could put their discovery down to bad luck.
And that’s how it went down. On 5 May, an Australian reconnaissance plane flew north-east across the Coral Sea in the direction of the advancing enemy strike force, found the Japanese fleet, and reported its location.
The Allied forces only had one day of surprise up their sleeves, because the very next day a Japanese reconnaissance plane found the Allied fleet. But with the vagaries of pilot’s reports and no precise coordinates available, both sides failed to locate each other’s main strike force.
The Japanese did locate the Allied support ships further south in the Coral Sea, but the pilot mistook it for a carrier fleet. They launched a strike at the ships, sinking the oil tanker, the Neosho, and another ship.
Meanwhile the Allied fleet had split in two. While the carrier the USS Lexington hunted for the strike force, the USS Yorktown, along with other ships, including the Australian warships HMAS Australia and HMAS Hobart, sailed north to intercept the Port Moresby invasion force as it rounded the eastern tip of New Guinea. The invasion force had conveniently been spotted by Allied reconnaissance planes that had gone looking for it that morning.
The Yorktown’s fleet suffered several waves of attacks from planes operating out of Rabaul, as well as from Japanese carriers, but did manage to sink the small Japanese carrier the Shohu. With the Yorktown blocking the way and with the loss of its most powerful warship, the invasion force turned around and sailed back to Rabaul in the north.
The Allied ships then re-formed into a single fleet, trying to find the Japanese strike force.
Both sides located each other in the early morning of 8 May, with the Lexington’s radar indicating that the Japanese fleet was only 70 miles away and closing.
The two sides launched swarms of aircraft against each other’s ships. The enemy planes passed by each other, missing each other due to heavy cloud cover over the sea.
The inexperienced American pilots dropped their torpedoes too early and too far from the Japanese ships, so that not a single torpedo found its mark. The Japanese pilots were more accurate, hitting the Lexington with two torpedoes and following this up with devastating dive-bomb attacks.
The Lexington, weighing 36,000 tonnes and over 880 feet in length, was the largest ship in the battle. The huge carrier started to tilt, with the petrol tanks for the planes leaking. A spark ignited an explosion, causing several fires. By 5.00 pm the ship was abandoned; once everyone had left, the order was given to sink it with a torpedo.16
The Yorktown was still in action, but had taken damage and was slowed to a maximum speed of 26 knots.
The Japanese fleet was unable to take advantage of the crippled American fleet, as the Japanese carrier Shokaku had also sustained damage, and there were still several Allied warships in action. The Japanese strike force sailed north, leaving the Coral Sea and ending the battle.
The Battle of the Coral Sea was the first naval battle in history in which no ship ever came within sight of an enemy ship. Every attack was conducted by warplanes launched from the two sides’ aircraft carriers.
The interceptors at Monterey ‘watched’ the entire battle unfold from Japanese naval radio messages they intercepted and decrypted. By Saturday, 9 May, the battle was over, and FRUMEL decrypted the following message from the commander-in-chief of the Japanese third fleet, a message that Fabian passed on to Hawaii, Vice Admiral Herbert Leary, and, of course, to MacArthur:
(a) occupation of Moresby postponed.
(b) ‘RZP’ Moresby striking force is to proceed to NE of Solomons in accordance with previous orders and to standby to accompany the RYC (Nauru) occupation force.
(c) Moresby occupation force is to proceed to Rabaul.
(d) ‘RYC’ (Nauru) operations will be carried out in accordance with general directive.
Japan’s occupation of Port Moresby was postponed, and there was never another attempt to take it by sea. Following the battle, there was a large amount of JN-25B radio traffic.17 The traffic analysts at Monterey tracked the progress of the various Japanese naval ships as they left the area, some of them limping back to far-distant ports for repair. A message from an unknown source at Rabaul revealed the official Japanese stocktake of their losses:
Shoho hit by 7 torpedoes and 13 bombs and sunk. 22 of her aircraft made forced landings resulting in 16 seriously injured and 64 injured. Number of ship’s company drowned not yet known.
Shokaku hit by 3 torpedoes and 8 bombs and damaged in engine room, petrol bunkers etc. 4 officers and 90 ratings killed and 96 seriously wounded.
On the Japanese side, the toll was significant. The Shoho, a small carrier, had been sunk, and Shokaku, a larger carrier, was damaged. Another aircraft carrier, the Zuikaku, was still sailing and undamaged, but most of its planes had been shot down into the sea, along with their crews. Four other Japanese ships had also been lost.
The loss in tonnage for the Allied fleet was even greater. The carrier Lexington had been sunk and the carrier Yorktown had been damaged. The destroyer Sims was also sunk, and the oiler Neosho damaged beyond repair.18
Fearing further losses from the Allied fleet, the Port Moresby invasion force retreated north, back to the Japanese stronghold of Rabaul.
Nevertheless, undaunted by his losses in the Battle for the Coral Sea, the Japanese navy’s commander-in-chief, Admiral Yamamoto, quickly prepared a new offensive in the North Pacific. The plans for this new battle were discovered at Monterey by luck.
Every day, piles of intercepts would arrive at Monterey by motorcycle from nearby Moorabbin. Some messages were nice and clean, while others were garbles: partial messages, fragments, or riddled with transmission errors. The team prioritised clean messages, as these were easier and gave the biggest bang for the buck. Garbles were put into a box. They would fish out some messages from the garble box at the end of the day if they had time, and see if they could make sense of them.
Late one evening in May, William Trembly, an American at Monterey, plucked out a message from the garble box to have a shot at stripping the additives from it. He was not a translator, but he had been working in code-breaking for long enough to recognise the JN-25 code word for ‘Attack’. Trembly alerted the watch officer.
The intercept that Trembly had discovered was not a complete message, but one of several parts of a much longer message, not all of which had been picked up by the Australian intercept stations. Sleeping staff were woken, and Monterey was soon alive with activity as other parts of the message were located, the additives stripped, and the message decoded as far as possible. They discovered that the target of the attack was a placename referred to as AF. They knew that this was a code word for Midway, a Pacific island occupied by US forces.
The most senior translator, Gil Richardson, translated what they had. They notified FRUPAC, the code-breaking centre at Hawaii run by the brilliant workaholic Joe Rochefort, whose team searched through their collection of recent intercepts and located other parts of the message. On 20 May, Rochefort presented Admiral Chester Nimitz, commander-in-chief of the United States Pacific fleet, with a detailed, 12-part message about the forthcoming battle, including Yamamoto’s battle plan and the composition of the Japanese fleet.
At least, that’s how Rochefort, Richardson, and others at the Monterey and Hawaii units remembered it after the war. But that’s not exactly what happened. Nobody has ever found a record of this 12-part message. Edward Layton — Nimitz’s intelligence officer, who was there when Rochefort presented his astounding discovery to Nimitz — claimed in 1970 that ‘there was no such message’. He added that, ‘There was never a complete battle order as it is reported in some of the books.’
Rochefort’s briefing for Nimitz and Layton was assembled from bits and pieces of intelligence across multiple stations — from traffic analysis, fragments of JN-25 messages, and translators in Melbourne and Hawaii translating what they could, and synthesising all of this into a coherent picture. They lived their lives in an ocean of symbols, codes, translations, and message counts, and were forbidden to talk about their work for years afterwards. It would have been easy for the code-breakers to be confused, over time, about which documents were inferences, and which were intercepts.
The historian Craig L. Symonds obtained Layton’s personal diary from the war years. Layton kept meticulous records of the important intercepted messages. Symonds writes,
There is no record in that journal of a unified 12-part message on May 20, but it does indicate a dramatic upswing in the volume of message traffic that day. Some of the messages concerned a planned Japanese ‘fleet exercise,’ but a dozen others obviously referred to a forthcoming operation. One revealed the presence of ‘occupation forces’ for both Midway and Alaska. Another mentioned that Japanese forces would approach the target from the northwest. As the Navy analysts worked on these messages, the results would have been collated and compared so that Rochefort could present the collected findings to Nimitz. Very likely, therefore, the men who achieved this intelligence coup recalled their effort as having focused on a single message rather than a group of shorter messages.
Symonds concludes,
None of this detracts from the crucial contributions of the code-breakers, but it does remind us that the subsequent decisions made by the commanders on the scene were more complex and open-ended than might otherwise be assumed. The Battle of Midway was not won by the code-breakers alone but by the analyst, the decision makers who trusted them, and finally by the men who drove the ships, manned the guns, and flew the planes at the point of contact. Certainly there is enough glory for all of them.19
The Japanese plan, as Rochefort explained to Nimitz and Layton, was this: Yamamoto’s fleet was going to pretend to attack the Aleutian Islands near Alaska, but at the last minute would divert south to the true target, the Midway Atoll. Yamamoto believed that occupying Midway would give the Japanese forces an impregnable command of the western half of the Pacific Ocean. Once the surprise occupation was complete, Yamamoto was sure the Americans would respond by quickly trying to recapture it. In anticipation of this, he developed the third and final part of the plan: a large fleet of Japanese battleships and aircraft carriers would wait silently to the west of the Midway Atoll, below the horizon, and then, when the American fleet arrived to stop the invasion, the Japanese fleet would move in, taking them by surprise and smashing them. This complex, three-part plan was in keeping with a deep-seated Japanese principle of warfare: the value of a decisive battle. Almost every available ship in the Japanese navy — 200 in all — were involved.
The commander of the US navy, Admiral Nimitz, now informed of the ruse, prepared a counter-ambush.
Fabian entered the code-breaking room after a briefing on the preparations in the North Pacific and said, ‘You boys better be damn sure you know what you’re writing about, because they’re moving carriers on the basis of what you’re saying.’20
At 6.30 in the morning of 4 June, American installations in the Midway Islands came under heavy aerial attack from Aichi D3A carrier bombers (known as ‘Val’ bombers), Nakajima B5N (‘Kate’) torpedo planes, and numerous fighters. The Americans knew they were coming. There were 26 Marine Corps fighter planes, a mix of Brewster F2A (‘Buffalo’) and Grumman F4F (‘Wildcat’) fighters, in the skies above Midway ready to meet them. They were overwhelmed by the Japanese onslaught — 17 of the 26 American planes were shot down — but they prevented the wholesale destruction of the bases on the ground.
At 9.30 in the morning, American torpedo bombers departed westward, searching for the Japanese fleet. They found it, but almost all of them were shot down by Japanese carrier-based fighters. However, they did manage to draw away and engage enough Japanese fighters to make way for the next American wave of attack: dive bombers. These had more success, mortally damaging the Japanese aircraft carriers Kaga, Akagi, and Soryu.
Three hours later, the one remaining Japanese carrier, the Hiryu, located the American carrier Yorktown. Its torpedo bombers attacked the Yorktown, damaging it so badly that it had to be abandoned. Meanwhile, the United States carrier Enterprise had moved in to assist the Yorktown, and it launched bombing raids that seriously damaged the Hiryu.
With the loss of all four aircraft carriers involved in the battle, Yamamoto abandoned his plan to occupy Midway, and ordered that the remainder of the fleet retreat to the west.21
Professor Treweek, working under Nave, saw the events unfold through his work at FRUMEL. After the code-breakers had compiled their estimate of Japanese losses, he went home to his wife and told her, ‘Japan has just lost the war. I can’t tell you more. Japan has just lost the war.’22
Fifty years later — when the nature of his work was no longer classified — he explained to Sydney journalist David Jenkins the reason for his optimism: ‘It was when they got rid of those aircraft carriers. We got the last message, Akagi mo chinbotsu. “Akagi also sank.” All those aircraft carriers, you see, then they’d had it. They could then be picked off, atoll by atoll.’23
Treweek might have thought the war was already won, but neither the Japanese navy nor army saw it that way. In fact, despite the setbacks in the Coral Sea and the retreat of their Port Moresby invasion force, they had not given up on occupying Port Moresby. They devised a new strategy — an overland invasion across the New Guinea mountains. There was already a stream of intelligence revealing the existence of this plan. On 19 May, FRUMEL intercepted the following message:
WE CONSIDER IT NECESSARY TO BEGIN THE CONSTRUCTION OF A LAND ROUTE TO MORESBY FROM LAE.
Further messages showed that they were indeed planning such a land route.
On 23 May, Monterey received a message from Japan’s 8th Base Force in Rabaul which said that, after investigations that included talking to the natives in New Guinea, they had learned of the existence of an old mining road that passed very high through the Owen Stanley ranges, ‘over a range of mountains about 2,300 metres in height’, went through a town called Kokoda, and then went back down to Port Moresby in the south.24
The message concluded, ‘There is no other road running across the island. Although this would be a difficult route for a campaign, we are studying the possibility.’
Fabian passed on the message to MacArthur.
More messages came in revealing the new plan, specifically that the Japanese army based at Rabaul was planning to land at Buna on New Guinea’s north coast, near the northern end of the high mountain road. MacArthur’s generals, including Blamey, knew of this intelligence — but nothing happened.
Everyone at Monterey was frustrated by the lack of action on this new Japanese plan. An Australian at Monterey — possibly Nave — anonymously told a journalist after the war of everyone’s frustration. ‘We had got the message that they were going to land at Buna … We told MacArthur and Blamey. They said, “Nobody in their right senses would land there.” We told them this was going to happen. Blamey couldn’t have cared less about it.’25
One reason was that MacArthur’s head of intelligence, Willoughby, was convinced that the Japanese would never be so foolish as to try to take Port Moresby overland. He simply didn’t believe the Japanese would attempt to come over the mountains. Another factor was that the intelligence about the landing at Buna was only one of many pieces of information about Japanese strategy, and nobody really knew what they were up to. MacArthur hedged his bets, and warned Blamey that there might be an overland invasion by ‘minor forces’.
Blinded by this cloud of uncertainty, Blamey’s eventual action was paltry. He sent the 39th Battalion up the Kokoda Trail to stop the new line of enemy attack.
The Australian battalion arrived in Port Moresby, then headed north, with the objective of stopping the advance of Japanese troops south, on a road through the mountains known as the Kokoda Track. It was a militia battalion, completely untested in battle.
With so many troops still scattered across the globe, Australia had only a minimal fighting force at its disposal, but there had been other options, including calling on troops stationed in Townsville. However, Blamey was concerned about the need for flexibility and the possibility of other battle fronts opening up, and didn’t want to commit the core of his forces to a single remote jungle operation, where his troops, once in, would be hard to retrieve in a hurry. This turned out to be a miscalculation. The Kokoda Track was the new main battlefront. When they met their adversary on the mountain track, the troops of Australia’s 39th Battalion suffered heavily.
16
Cranleigh
Central Bureau moved into a large, old three-storey house named Cranleigh at 225 Domain Road, a leafy suburban street on the other side of the Botanical Gardens, just a short walk from Victoria Barracks.1 They started with a total staff of six, all from Mic Sandford’s intelligence team, soon to be joined by the miscellaneous assortment of sigint experts from across the south-west Pacific that had been blown by the storms of the Second World War to Melbourne.
Sandford was promoted to major, and, as assistant director under Akin, was effectively in charge. The first two sigint unit members to arrive in Greece two years earlier, Geoff Ballard and Don Inglis, were now promoted to lieutenants. The code-breakers who had initially been recruited to the unit on the strength of their German-language skills, including Ballard as well as Vic Lederer, Don Inglis, Eddie Kelson, and Stan Winn, were there not because of their language ability but their knowledge of sigint.
Lieutenant Norman Webb was the British army’s contribution to the fledgling organisation. Webb had been running a special wireless group on Singapore that had operated during the Battle for Singapore, departing the city on a ship on 11 February, only four days before Singapore surrendered. Webb brought with him a collection of Japanese army radio messages that his unit had intercepted before they fled.2
The personnel of Station 6 from the Philippines were attached to Central Bureau, and they had also brought with them a civilian translator, Clarence Yamagata, who had now been assigned to Central Bureau. Yamagata was a nisei — an American born to Japanese immigrants. He had studied at the University of California at Los Angeles before moving to the Philippines to establish his own practice. When war broke out, he worked for General Akin on creating Japanese-language propaganda, before being evacuated to Australia.3
They were joined in time by more Americans, starting with cryptographers Larry Clark and Charles E. Girhard, and 12 enlisted men deemed suitable for the work. The next wave included Captain Hugh Erskine and another ten enlistees.
They spent their working days in the airy rooms of Cranleigh, and in the late afternoon would stroll down for drinks at one of the nearby pubs and bars, which would be crowded with American soldiers and sailors.
Their task seemed insurmountable. No Allied nation had broken the Japanese army codes, or even made much headway against them. And although Sandford’s team were competent traffic analysts, they did not have enough incoming traffic to work with.
By contrast, the navy code-breakers at Monterey had an abundance of material, because navy radio messages were broadcast across the oceans from high-powered transmitters; they could be detected even at the receiving stations near Melbourne at Moorabbin and Park Orchards. But army radio communication was usually done with low-powered transmitters over short distances. For this reason, they needed, but did not yet have, intercept units close to the enemy.
They also had the responsibility of solving the Japanese naval air-to-ground codes, which the US and Australian navies had no interest in working on. Monterey was preoccupied with JN-25 and diplomatic traffic, and had passed the buck for these comparatively minor codes to Central Bureau.
Eric Nave arrived on temporary secondment to teach the staff of Central Bureau — which initially consisted mostly of the intelligence staff of Ryan’s wireless section — the science and art of code-breaking. He held classrooms each day.
They learned that Nave had solved the Japanese navy air-to-ground code. He taught his eager students — Sandford, Ballard, Inglis, Kelson, and the rest of the wireless unit intelligence team — how it worked.4
Ballard said later,
It was the solution of the Japanese Naval Air codes which brought the most rewarding results at that time. One of my indelible memories of those early days in Central Bureau was of the scene in Captain Nave’s upstairs room in Cranleigh, with the winter sun slanting through the lead-paned windows. There, in a class-like atmosphere, Captain Nave taught a small group of us to unravel the Japanese Naval Air codes so effectively that when we were posted to field sections, we were able to read them continuously. Thanks to his instruction also, when the codes changed, we were able to reconstruct them with a minimum of delay. The atmosphere of excitement and discovery that marked our first encounters with Japanese codes has never left me.5
The code involved a method for scrambling and unscrambling the order of the letters in the message. Nave had experience in breaking scrambled-order codes from his earlier years with British intelligence, particularly when stationed with the Far East Combined Bureau in Hong Kong.
The basic idea was simple. The sender of a message in this code would have a grid somewhat like a crossword, which Nave called a ‘cage’. If you were working at a Japanese naval air base in the Second World War, to send a message in this code you wrote your message in the squares of the cage in Kana, going from left to right, as if you were filling out a crossword. Then to get the enciphered version of the message, you copied down all the Kana symbols in the cage, but you copied them from the cage in columns, starting at the left column, and working across to the right.6
At the other end, the receiver of the message would get the scrambled Kana letters. But since they would also have a copy of the crossword-style grid — the cage — they could write them into the columns and then read them off as rows, and would be able to read the original.
The nice thing about this code system was that it was easy to create variations. The operators would be issued with the latest cage, with different letters blacked out, or they would have a system of starting all messages at column two on Mondays. And there were other ways, indeed many ways, of making slight changes to how the cages were used in order to thwart would-be code-breakers.
The first messages in this code were received in June 1942, from the intercept base at Townsville. The encoded messages were peppered with frequent occurrences of the Kana symbol, yo. The code-breakers called it the ‘YO code’. You can see two examples of a cage used for the YO code in the diagram on the next page.
A partially burned YO code cage found in northern Australia and sent to Central Bureau for examination
A reconstruction of a complete YO code cage by code-breakers at Central Bureau (this is a different cage from the captured one shown on the left)
The first is a copy of a cage, partially burned, found by Australian forces and sent to Central Bureau for study. The second picture, on the right, is a complete cage that the cryptanalysts at Central Bureau pieced together from studying numerous YO code messages.7
From time to time, an air-to-ground substitution table, lost or left behind by the enemy, would be discovered by Allied forces. One such table, partially burned but partially readable, found by Allied forces and sent to Central Bureau, is shown in the picture on the next page, taken from the bureau’s technical records.8
A YO code substitution table found by Allied forces and sent to Central Bureau for examination.
The effect of the cage was to jumble the different parts of the message into a seemingly random order. That was known in code-breaking circles as a ‘hatted’ code, because it was as if someone had put all the pieces of the message into a hat and then pulled them out at random.
Eric Nave had been studying hatted codes like this one for years. In Hong Kong, in the years leading up to the outbreak of war in the Pacific, his main responsibility was breaking and translating messages in the main Japanese naval code. At that time, JN-25 had not arrived, and the Japanese navy was using a hatted code as its main system. Nave broke it, and supplied a steady stream of intelligence on Japanese naval activity from the Far East Combined Bureau’s offices in Hong Kong.
Most of Fabian’s team was occupied with JN-25, and he had no interest in Nave’s assistance. Excluded from the inner circles of Fabian’s empire in Monterey, Nave focussed on the navy air-to-ground code, and before long it was broken. The downside was that the Japanese navy changed the books for this code very quickly, with new books seeming to come into effect every couple of weeks.
It turned out that yo meant nothing at all. It was simply inserted into messages as a dummy, filler symbol. Some of the squares in the cage were pre-filled with the Kana symbol yo (which looks like a backwards E in the cages shown earlier), so that yo would occur throughout the messages in a pre-determined way, depending on the layout of the cage.
When a coded YO message was written back into the cage, all the yos would match up with the yo squares in the cage. This was a useful check to ensure that the message had been accurately received and transcribed. But redundancy is a crack in the surface of any code system. Although the scattering of yo throughout the encrypted messages looked random at first glance, their placement was driven entirely by the structure of the cage. Central Bureau knew this, and they also knew that different columns of the cage were used on different days. Knowing those two facts was enough to reconstruct the cage for any set of messages in the YO code, as long as they had a collection of messages for each day.
A new naval air-to-ground code came into use in September. It worked the same way as the YO code, with some minor changes. Nave named it the NE code because, instead of yo, the messages were peppered with the syllable ne. There were some differences in how it worked, but it was broadly similar to YO, so the NE code was quickly broken by Nave. Once it was broken, most messages in NE were deciphered and translated on the same day they were received.
Doctor Abraham ‘Abe’ Sinkov, a mathematician and US army cryptanalyst, had arrived in Melbourne from Washington in July. Sinkov was one of America’s top code and cipher experts. In the previous year, 1941, he had led a contingent of American code-breakers to England’s Government Code and Cipher School at Bletchley Park, where he had long conversations with British code-breakers such as Alan Turing and John Tiltman. That visit, which was later dubbed ‘the Sinkov Mission’, established a working relationship between the two major Allied powers on cryptanalysis, and gave the Americans their first insights into Britain’s cracking of the German Enigma machine. Sinkov arrived in England with a gift for the British cryptanalysts: a replica of the Japanese ‘Purple’ machine.
Germany’s Enigma was a complex cipher machine with settings that changed with each keystroke. The Germans believed it was unbreakable, but Turing had found a way to decipher Enigma messages. Japan had developed its own variant, nicknamed ‘Purple’ by the Allies. Large and cumbersome, Purple machines were given limited use, mostly in diplomatic embassies, and were not widely used by the Japanese armed forces. While Turing was cracking Enigma, an American team led by William Friedman cracked Purple. Turing and Friedman solved their puzzles independently of each other. Sinkov’s gift to the British, a replica of a Purple cipher machine which had been constructed by Sinkov’s colleagues, allowed the British to instantly read Japanese diplomatic communications, just as the Americans were already doing.
The Sinkov Mission, and the closer relationship it forged between the British and American sigint operations, was the reason Malcolm Burnett flew from Singapore to Corregidor in 1941 and provided Fabian’s team with the structure of the Japanese naval code, JN-25. The mission laid the foundation for creating Central Bureau and FRUMEL, by clearing away many of the obstacles encountered when sharing top-secret information between agencies from different countries. In large part because of the mission, the Allies were now cooperating on sigint. The joint Allied operations that had begun in Melbourne built on the partnership that Sinkov had helped develop.
The Japanese army codes, like their navy codes, came in two basic varieties: Kana number systems and Kana systems.9
There were two kinds of code systems used by the Imperial Japanese Army: those that used digits and those that used Kana. Important messages between major bases and senior commanders were sent in a code that was known at Central Bureau as the ‘Mainline Army Code’, in which all messages were strings of four digits. The Mainline code did not use Kana; instead, the numbers were transmitted in universal Morse code. Since it was a high-level system, they knew it would be tough to break.
But not everything was sent by the Mainline code. Smaller operational units needed to send encrypted messages, too. Besides the high-level codes for important communications, every nation had developed simple code systems for use in the field. These were called ‘low echelon codes’ or ‘tactical codes’. As they were being sent with strings of three-figure code symbols, they became known as ‘three figure codes’.
Every nation’s tactical code systems had already been broken by their enemies. Britain, the United States, Russia, Germany, Italy … everyone had broken into everyone else’s tactical systems years earlier. This didn’t matter to any of the combatants, because if they changed their codebook often enough, there would still be a delay before the enemy could build the new codebook and decrypt the tactical messages. A message about an incoming air raid or troop movement was useless if a code-breaker only managed to hand it to a local commander two weeks later. Tactical codes were not supposed to be as secure as high-level codes; they were usually just there to buy time.
Kana systems were based on the orthographic Katakana alphabet described earlier, which translates mostly as ‘letters’, or, to be more precise, syllables such as ra and mi. The Kana systems were used for lower-level units and tactical communication. The air-to-ground codes were also in Kana.
Sinkov figured that Mainline Army Code would be too hard to break at first. There weren’t enough field units, enough code-breaking personnel, and enough data on the mainline code. He decided to go for some easy wins, and to tackle the tactical three-figure code systems.
Norman Webb had brought a bag of Japanese army three figure Kana messages from Singapore. Sinkov decided to use that bag of messages as the starting point until some field units were deployed and more radio traffic started coming in.
Sinkov’s optimism was misplaced. The solution to the tactical codes was frustratingly elusive.
17
Morse and Kana code
The Australian army now had the task of deploying intercept field units to the north to feed the new cryptographic centre, Central Bureau. After all, without a steady diet of enemy radio messages, the sigint experts would have nothing to study. There was already a unit in Darwin, established by Newman, sending a trickle of radio traffic back to Melbourne, but more was needed.
The plan they came up with at the April conference at Victoria Barracks was to build a group within the army using personnel from Ryan’s Four Special Wireless Section. The intelligence staff from that section, including Sandford, had moved out to establish Central Bureau, but the rest of Ryan’s personnel, with their knowledge of signals and radio interception, could be the foundation to build special signals units in Australia.
The consensus was that Ryan would be promoted to become head of this new signals group, which would be called the Australian Special Wireless Group (ASWG). But Simpson, Australia’s recently appointed head of signals, had other ideas. On 19 May, he informed Captain Ryan by phone that, unfortunately, his promotion had been cancelled, and that furthermore he and Lieutenant Henry would not be needed in the area of special signals. He explained that he had interviewed both of them, and was not convinced that they were right for the job. Instead he had appointed a Lieutenant Colonel Moulds as the commander-in-chief of the ASWG. When Ryan and Henry asked what duties they would be transferred to, he told them that that issue was not a high priority for him, as he had ‘several officers whom he had to place somewhere’. Central Bureau were notified in the afternoon.
In late May, Four Special Wireless Section was transformed into the Australian Special Wireless Group commanded by Lieutenant Colonel H.L. Moulds. Simpson’s hand-picked head of the ASWG may well have been an excellent choice, but the bait-and-switch did not sit well with Ryan’s former unit members at Central Bureau, who made moves to keep Ryan in the game.
Three weeks after Ryan was frozen out, the Central Bureau intelligence committee decided to formally retain Captain Ryan as a permanent technical advisor — a slap in the face for both Moulds and Simpson. Ryan was also invited on an ad hoc basis to many of the intelligence committee meetings, sometimes even chairing them. Reporting directly to Akin and ultimately to General MacArthur, who saw it as his baby, Central Bureau had friends in high places. By the end of August, Ryan had replaced Moulds as commanding officer of the ASWG, and he remained in that position for the rest of the war. 1
One of the originals from Greece and Crete, Sergeant Ralph Thompson, was promoted to captain and given command of the first section created by the special wireless group. Moulds named it ‘51 Wireless Section’, and departed Melbourne in May for the Northern Territory.2 There, the section set up their campsite and interception equipment at Coomalie Creek, a new airfield about 60 miles south of Darwin.
Assembled in a hurry, Ralph Thompson’s section was at less than half-strength; he had a total of 24 personnel. Eight were Kana operators, and the rest were involved in the diverse range of roles needed to run the intercept site: mechanics, drivers, cooks, and so on.3
The ASWG began at Park Orchards, the nearby army intercept site on Melbourne’s north-east. Within days, however, plans were made to relocate 200 miles north to Bonegilla, a sprawling military camp outside Wodonga near the New South Wales border.4
In training camps across the country, large numbers of recruits were being trained in signals to meet the day-to-day communication needs of wartime. These trainees knew nothing about special signals — intercepting enemy signals — but they were learning how to set up aerials; how to maintain and operate army radio sets; and, most importantly, how to send and receive messages in Morse code. Officers from the ASWG visited these training grounds, selected those who showed the best aptitude for Morse code, and ordered them to be transferred to the special-signals training camp at Bonegilla.
Bonegilla was established in 1940 on a rural property to accelerate the training and recruitment of Australian soldiers, with the army constructing 600 huts on the site in a three-month period. When war broke out in the Pacific, an additional 248 huts were built.5 These huts, built in haste, were rudimentary and shoddy. The walls were made of corrugated iron, so that the men were freezing cold in winter and roasted in summer. The recruits slept on the wooden floors.
All the training camps at Wodonga were self-contained and separated from each other by considerable distances. This isolation was ideal for the special wireless group. The trainees could be kept completely separate from the regular military camps nearby, reducing the problem of prying eyes from outside.
The army traditionally recruited men, but the war had created a shortage of them, and the special wireless group was hungry for more interceptors, so in August, it began training with a female cohort from the Australian Women’s Army Service (AWAS). Just as with the men, these women were selected for their Morse code proficiency, and were given all the same training.
Mic Sandford, now a lieutenant colonel, made occasional visits to Bonegilla. When the first AWAS contingent arrived, Jack Ryan assembled them and introduced them to Mic Sandford. One of those women was Jean Hillier, who, in her memoir about life as a Kana operator, described Sandford as a ‘mercurial man about whom, outside sigint circles, far too little is known’.6
He then asked them if they had any questions. One asked, ‘Why can we not go straight to work?’ Some there had already been operationally deployed, after all, and did not see the need for more training.
With a smile, Sandford asked the trainee if she knew all the characters of the Japanese Morse code. There was a stunned silence. Nobody had yet told them about the Kana code. Hillier described the moment as being ‘as though a train had quietly jumped the tracks and headed off into the fourth dimension. In a way that is what happened because, from that moment and for many years afterwards, part of our lives remained invisible.’
Having stunned them into confusion, Sandford explained the deal:
Mic, using all his charm and eloquence, then set about enfolding us into our destiny and duty. We were to intercept enemy wireless traffic and police our own and Allied traffic. He told us something of the history of the Unit, the work already accomplished and the job that must be done now.
He told us we were all needed urgently and must train as quickly and thoroughly as possible and he introduced us to the word ‘Kana’. Knowledge of Kana, the word for the symbols of the Japanese Morse code, was the keystone of the job.
It was a word we were forbidden to use outside our lines and then inside only with care. I never had any trouble obeying that rule and yet it was to give me the most fascinating, enthralling years of my life.
The Kana code, being a Morse version of Japanese orthographic Katakana, comprised 75 separate characters that had to be learned. (Three versions of the code, as learned by interceptors, survive, each with a different number of symbols — perhaps due to modifications in the code or to operational requirements.)7 It had to be learned in a hurry, so learning the actual Katakana symbols was out of the question for Australian recruits who mostly could speak only English.
Instead, they learned to transcribe Kana code using a modified version of the English, or, more accurately the Latin, alphabet. Extra letters were created with bars, dashes, and circles. For example, the trainees learned to write the Katakana syllable ba, sent in Kana code as dash-dot-dot-dot-dot-dot, as Bl, while a regular B was used for the Katakana syllable ha. They did not learn Japanese. They simply learned to recognise 72 different combinations of dots and dashes in a radio signal, and to write those dots and dashes down in various ways as letters with bars, dots, and dashes. That’s all there was to being a Kana operator. The fact that they had no idea what they were transcribing or why made the job difficult and confusing.
In Melbourne, Central Bureau was starved for cryptographic material. Pappy Clark suggested that the trainee Kana operators at Bonegilla and elsewhere should practise on real intercepts. That way, at least some material would start coming in through the door, even if it was being done by novices.8 As a result, the ASWG trainees at Bonegilla honed their interception skills and knowledge of the Kana code with live enemy transmissions.
In June, a second unit, ‘52 Wireless Section’, was formed and ready to go, with Lieutenant Litchfield in command. In contrast to Ralph Thompson’s 51 Wireless Section, now camped in remote scrubland of the Northern Territory, this section was deployed much closer to home, at Ferny Creek, only 20 miles east of Melbourne. The reason was that 52 Wireless Section was dedicated to intercepting diplomatic messages. In August, the section relocated to Bonegilla, not for training purposes, but as a base of operation, and later in the war moved to a location on Victoria’s Mornington Peninsula.
By October, four more sections were formed and deployed. 53 Wireless Section and 55 Wireless Section were sent to New Guinea, while 54 Wireless Section was set up in Darwin and was later based at Kalinga near Brisbane.9
The AWAS women, including Jean Hillier, became 56 Wireless Section, and departed Bonegilla in October by train for Perth. They established an intercept site in Perth that monitored Japanese radio traffic to and from Singapore until mid-1944, when they also relocated to Kalinga in Queensland.
While the Australian army was forming and deploying sigint field units, so too was the Royal Australian Air Force, under the command of Flight Lieutenant H. Roy Booth.
Henry Roy Booth had been a lawyer before the war with a successful business practice in North Sydney.10 Booth enlisted with the air force in 1940, was selected for intelligence work, and was posted to the Far East Combined Bureau in Singapore to study British sigint techniques. With Singapore under threat from the advancing Imperial Japanese Army, he returned to Australia, was assigned to Central Bureau as the head of its air force component, and became the third assistant director, along with Sandford and (initially) Sherr.
Jack Bleakley, a kana operator who served under Booth, recalled his unassuming nature:
He was a quiet man, a behind-the-scenes man. He was overshadowed by Sandford. Sandford was a very strong personality, and whenever the Army and Air Force met together, Sandford won the arguments. Booth did his work well, but without too much publicity.11
While the army’s new deployments at Bonegilla were called ‘Wireless Sections’, the air force called its intercept operations ‘Wireless Units’, and started the numbering of them at 1. Booth had already started the process of establishing wireless units earlier in the year; he now continued that process, but under the auspices of Central Bureau rather than the air force directly. The Ballarat Gunnery School in rural Victoria was an obvious place to start.
Allan Norton was one of Booth’s early recruits. He was 16, having lied about his age in order to enlist, and came from a potato farm in Western Australia, about 30 miles north of Albany. Fascinated by radio, Allan wanted a role as a radio technician in the army, so he learned Morse code at home. He would awaken at dawn each day and start working the crops with his father. The only night-time lighting in the homestead was a single hurricane lamp, making any kind of evening study impossible. But in the middle of the day, in between his early-morning and late-afternoon work, he had time to himself. He sat at the dinner table with a book of Morse code dots and dashes, and practised tapping out Morse code messages with his finger.
Every day Allan practised, tapping the wooden tabletop. His father, not wanting to lose his son to war, registered him as an employee in a critical industry — farming — so that he was exempt from the draft. But Allan had other ideas, and enlisted at Albury for the Australian army.
For those who wished to qualify for signals work, there was an optional Morse code proficiency test. To pass, applicants had to be able to transmit more than 20 words per minute. Allan passed, and was sent to the gunnery school. There, he was selected to join a new special wireless unit. The catch was that, despite having spent hours practising Morse on the farmhouse table, he would no longer be required to send or receive Morse code as such, but a Japanese variant that they called the Kana code.
Allan Norton was sent to Melbourne, and was billeted with the other selected recruits, all as bemused as each other. They were taken to stalls in Melbourne showgrounds in the inner-north suburb of Ascot Vale, and there they were given a course in the Kana code.12 They were not told about the existence of Central Bureau or FRUMEL, or about the activities being undertaken by those organisations at Cranleigh and Monterey. All they knew was that they were learning the Kana code and that they would soon be transcribing Japanese radio messages in this code. The course lasted for several weeks. At the end of it, when their instructors were satisfied that the recruits could listen to and transcribe Kana at an acceptable speed, they were told that they had successfully graduated and were now part of the newly formed ‘1 Wireless Unit’, and were sent to Townsville in Queensland. Howard Brown and three other US army signal personnel from Station 6 in the Philippines went with them to lend their expertise as the unit became operational.13
Townsville was effectively a frontline garrison town at the time, teeming with US and Australian troops. The town endured sporadic air raids, although these failed to have the impact of the raids on Darwin and Port Moresby further north.
The members of 1 Wireless Unit did not live in regular army bases with regular army troops, since they were kept isolated as a matter of policy. Instead they had their own quarters in the suburb of Pimlico. Their workplace was a rural property some way out of town in a locality known as Stuart Creek, not far from a local prison. A driver took them by army lorry to the location.
The building, as they approached it, looked like a standard Australian farmhouse, with wooden window-frames and a large verandah with a white wooden railing. Up close, though, it turned out not to be a farmhouse at all, but a concrete bunker. There were no wooden-framed windows, nor a white wooden verandah. These had been painted onto the outside walls to disguise the bunker as an ordinary house.
Inside was a cavernous room lined with desks, and on each desk was an AR7 radio receiver set, with headphones attached. This was their new workroom. They were now, in the new lingo of Australian signals intelligence, ‘Kana operators’.
The operators were organised into three shifts, eight hours per shift, to work around the clock. Shifts were called ‘tricks’, and the officer in charge of a shift was the ‘trick chief’. The enemy transmitted radio messages at all times of the night or day, and therefore the bunker at Stuart Creek was always operational, with 12 Kana operators and a trick chief on every trick.
They wore headphones, and moved their AR7 frequency dial to the Japanese military frequencies they would be listening to on that shift. Once they were tuned in, they would listen for a message to start — a series of dots and dashes in Kana code — and they would write it down on paper as fast as they could, using the Kana shorthand they had learned at the Melbourne showground.
In trying to intercept messages, they immediately found that the real thing was nothing like their training for it. Kana code was new to them, but it was not new to the Japanese radio operators. It was their language, their code, and their Morse system. Unlike the fresh recruits of 1 Wireless Unit, they were seasoned — some having been doing it for years. So much for the 20-words-per-minute pass mark at the showgrounds. The wireless unit in the bunker quickly learned that Japanese operators routinely sent and received messages at speeds above 70 words per minute. The strings of dots and dashes in Kana code streamed into the recruits’ headphones as they struggled get them down on paper.
It was sometimes hard to stay locked on to a message and to transcribe it from beginning to end, especially for high-traffic frequencies. The Kana operator would be straining to hear a faint but important message and get it all down on paper just as another broadcaster, much closer by, would start to send a message on the same channel, blaring out of the headphones with a loud womp-womp, drowning out the fainter message the operator was trying to catch.14
Over time, they became more proficient. They learned how to recognise urgent and important messages, depending on which radio frequency was being used, who was sending it, and other things. They learned to recognise certain operators by the rhythm and speed of their key taps. The recognisable cadences of the way a particular Morse sender tapped out the keystrokes was called a ‘fist’. Every Morse operator had a fist, and when the team in 1 Wireless Unit listened to enough of his or her messages they got to know it well. Recognising a fist was a useful skill when the enemy changed their call signs — those code words at the start of the message to identify the sender or receiver. A unit that tried to disguise its identity might give the game away when a Kana operator recognised a fist he had heard before.
Howard Brown from Station 6, a seasoned career army officer and veteran of the Philippines invasion, was a cause of fascination and alarm for the young Australian recruits.
Jack Bleakley, a kana operator at Townsville, recalled, ‘The only immediate danger to 1WU personnel at Townsville came from one of its attached Americans, Captain H.W. Brown. A ‘gung-ho’ type, he was known to draw his service pistol and shoot the occasional bullet through the ceiling of his quarters in French Street — once actually shooting out his light globe before going to bed for the night! On some occasions, to pass the time away, he would take pot shots through the open window at pigeons sitting on telegraph wires outside.’
When he woke up one morning to find that his army command car had been stolen overnight, he was furious, and ordered several off-duty officers to track it down. The vehicle was soon located, abandoned on a road just out of town.
Bleakley recalled, ‘In true Hollywood style, he deployed his men in a stake out across the street from the car, to await the return of the thieves and gave his orders to “shoot to kill”. There was no doubt whatsoever that he would have carried out his threat, but by a stroke of luck the culprits had flown and an international incident was averted.’15
They focussed on aircraft in the early days. Depending on the code system, many of the messages were encoded and made no sense at all; but sometimes, particularly from small units like aircraft crew, the messages were in plain language, or P.L., as the Kana operators called it. Sure, it was still Kana code, and it was in Japanese, and the Kana operators had not been trained in the Japanese language at all — they just knew how to transcribe Kana code. Even so, they learned the meanings of commonly used words and phrases, such as ‘danger’ and ‘repeat that’.
Surfing the radio frequencies in search of traffic, they soon learned the busy frequencies. The busiest two on the dial were 8915 Hertz and 6050 Hertz.16 These frequencies carried traffic from Rabaul on the island of New Britain, north-east of New Guinea, and Truk, an island about 1000 miles north of New Guinea. Rabaul and Truk, with their large harbours, had been transformed into major Japanese military bases.
The busy frequencies at peak times were, in Allan Norton’s words, ‘a cacophony of different sounds, like a hen house’, with Kana messages at different volumes starting and stopping over the top of each other. Some, transmitted from a long distance or on a low-power radio, were faint, whereas others — important messages from large bases or from units close by — blared out of the headphones, hurting the eardrums. Each Kana operator monitored his own frequency, and each cacophony was private, hidden from the outside world by the headphones over their ears.
For the trick chief looking at the rows of operators, the only sound was the scratching of pencils on paper.
At the end of each shift, the army lorry, having delivered the Kana operators for the next shift, would be waiting to take them back to Pimlico, where they would talk, listen to the radio, and sleep. The operators of each shift got to know each other well, but barely knew those on other shifts, beyond their names. When shift one was at work, shift two was idle, and shift three was asleep. Their paths rarely crossed.
With the war now extended across New Guinea and the south-west Pacific, and suffering from uncertainty about enemy plans and movements, MacArthur wanted sigint up and running as soon as possible. The Australian army and air force both began the production of intercept units, but this was a slow process.
To meet the pressing need for more special signals units, the United States army also formed a unit in Australia in 1942 that performed interception for Central Bureau. Once he had finished helping 1 Wireless Unit get established in Townsville, Howard Brown (from Station 6 in Manila), trained a US army platoon that had recently arrived in Melbourne in the various aspects of special signals. The unit was named the 126th Signals Radio Intelligence Company. Although the platoon had arrived knowing little about signal interception, they had been supplied with excellent radio-interception equipment — far better than anything the Australians had — that they had brought from the United States. After some initial training in Melbourne and a short stint in northern Victoria, Brown relocated them to Townsville, where they shared equipment with 1 Wireless Unit, and where the American operators learned Kana code. The unit later moved to Stafford, north of Brisbane.17
Meanwhile, Roy Booth moved the signals-intelligence training from the makeshift facilities at the Ascot Vale showgrounds to Point Cook, the air force signals school near Werribee, south-west of Melbourne. Colin Brackley was one of the recruits who trained there. He recalled that there was a motto on the wall in Latin: ‘Nilus panicus’, which translates as ‘Don’t panic’.18
With the army now deploying wireless sections in the field, Sandford was alarmed that they had no intelligence units of their own on location. Captain Ryan’s section in the Middle East had had an ‘I’ unit headed by Sandford, and the assumption was that these newly formed sections would follow the same template. They didn’t, and the army seemed disinclined to remedy the situation.
Sandford took matters into his own hands, and started raising I sections in Melbourne himself. At the end of June, he wrote to the director of military intelligence, John Rogers (whom he knew from the Middle East campaign), informing him that no intelligence personnel were currently operating at the army’s intercept sites, but that Central Bureau was in the process of raising two such sections itself, and planned two more. Sandford wanted these sections to be included in the ‘order of battle’ — the official list of operational units. Rogers agreed, and a new division of Central Bureau was created, the Australian Special Intelligence Personnel, whose members were always referred to as ‘ASIPS.’ These sections, it was now officially decided, would be attached to wireless sections. In August, the first ASIPS unit departed Melbourne for the long journey to Coomalie Creek in the Northern Territory; first by train to Brisbane, then Mount Isa, and then by a series of plane trips to the airfield at Coomalie Creek.
The officer in charge of the ASIPS at Coomalie Creek was Captain William E. Clarke, known as ‘Nobby’. Nobby Clarke had done a tour of duty in the Middle East. He was fluent in both Japanese and Chinese, so with the outbreak of war in the Pacific he had been transferred to Ryan’s unit, joining it in Egypt before the journey back to Australia.19
The ASIPS units were responsible for liaising with Central Bureau and supplying intercepts taken by the Kana operators, and also for decrypting messages. With the naval air-to-ground code solved, incoming messages in that code could be decrypted, read, and translated on the spot without having to be sent to Central Bureau. In the handling of air-to-ground intercepts, speed was crucial. These were often messages about forthcoming air raids or other tactical operations. By the time an intercept made its way back to Central Bureau for processing, the air raid had already come and gone, or the Allied ship had already been sunk, or the reconnaissance plane had long since passed overhead and returned to its base.
The units decrypted a steady stream of material relating to aircraft movements around New Guinea and the sprawling islands of the Netherlands East Indies. They warned local bases of aircraft landings and take-offs at nearby enemy bases, incoming air raids, the location of reconnaissance aircraft, enemy sightings of Allied aircraft and ships, and more.
Once a new version of the naval air-to-ground code was solved at Central Bureau, the codebooks, cages, and substitution tables were sent to the ASIPS at the intercept sites in northern Australia and later New Guinea. The ASIPS were in direct communication with the local air force commanders. In all cases, even for those messages decrypted on site, copies of the intercepts were sent to Central Bureau for further analysis and storage.
Local commanders were never given direct access to intelligence from the so-called ‘high grade’ codes and ciphers (such as JN-25, Purple, and Enigma) that were used by the highest levels of enemy command. However, the naval air-to-ground codes (YO and NE) were more tactical in nature, used by Japanese flight crews and airfield-base personnel for rapid communication. Allied commanders believed that allowing field commanders to make use of decrypted tactical codes was unlikely to betray the extent of the Allies’ code-breaking success against the high-grade systems.
The air-to-ground decrypts were vital in combating the constant air raids against Darwin. It soon became clear from traffic analysis of air-to-ground messages that standard procedure for the Imperial Japanese Army was to keep their bombers far from the Allied front line, at rear bases safe from Allied air attack. When called upon to conduct an air raid — for example, on Darwin or Port Moresby — the bombers would move to an advance base, fly out on their mission from that base, and then retire back to a rear base. This reduced the likelihood of their bombers being destroyed on their airfield by air raids from the Allies’ bombers.
This procedure provided an opportunity for the cryptanalysts to deduce when air raids were likely. When the intercept units decrypted a message about bombers moving to a nearby base to the north, such as Timor, they notified the local command. Other indicators of impending air raids included messages about bomber take-offs from nearby, or weather reports for Darwin.
On several occasions, based on information from Coomalie Creek, the Allied air command in Darwin sent bombers and fighters out to pre-emptively attack the Japanese planes, either destroying the planes as they sat on the airfield readying for take-off or intercepting them in-flight on their way south. One particularly large air raid against Darwin was thinned out to the extent that by the time the enemy planes arrived, it was a mere ‘token attack’, being a fraction of the intended force.
18
Traffic
The Japanese, now occupying the Solomon Islands to Australia’s north-east, had almost complete dominance of the south-west Pacific beyond Australia. The long-awaited first Allied counter-attack came in August, with landings on two islands in the Solomons: the large central island of Guadalcanal, and the smaller island of Tulagi to its north.
On the morning of 7 August, an American fleet sailed north under the command of Rear Admiral Richmond Kelly Turner, with the First Marine Division under Major General Alexander Archer on board. John Prados described the fleet’s approach in his history of the Solomons Campaign, Islands of Destiny:
The weather closed in as they approached. Low cloud ceilings had kept Japanese search planes away. They might achieve surprise in this, the first major Allied offensive of the war.
With dawn a carefully choreographed sequence began with the order — the first of many times it would be issued — ‘Land the landing force.’ Soon afterward, aircraft from American carriers one hundred miles away swept in to strafe the invasion beaches.1
The Japanese defenders had no idea that the Allies were coming, and were caught completely off-guard. The United States marines landing at Tulagi met no resistance at all, despite having to wade one hundred yards to get to the beach, while the landing force at Guadalcanal met a token resistance before the Japanese defenders retreated to other parts of the island. Tulagi was quickly secured as an Allied base, but the battle for Guadalcanal was only just beginning. It would last for months, with both sides trying to sneak reinforcements to the island without getting their ships sunk by the other side’s planes and submarines.
General Alexander A. Vandergrift and Admiral Turner were faking when they looked at the clouds with crossed fingers. The morning could have brought an unbroken blue sky from horizon to horizon, and yet no reconnaissance planes would have sighted them. They knew this because Pappy Clark, a traffic analyst at Central Bureau, had been charting the Japanese reconnaissance planes over the Solomon Islands, and he knew their daily patterns.2
Since taking part in that first parade at the Caulfield showground two years before as an original member of Jack Ryan’s new special signals unit, Pappy Clark had become an expert traffic analyst. His talent extended beyond applying techniques he learned from others: his creative, music-obsessed mind allowed him to invent new methods and techniques, and to find interesting ways of using the data they had.
Using signals from the air force’s network of direction-finding units in Queensland, Clark plotted the flights of the Kawanishi H8K Flying Boats based much further north at Gasmata, on the island of New Britain. The Kawanishi, nicknamed ‘Emily’ by the Americans, were long-range Japanese aircraft that scoured the Coral Sea for any sign of Allied shipping movements.3
Pappy Clark advised the invasion fleet to wait below the horizon until a certain time, when the Emilys had done their flyover and were heading back to their base. Clark told an ABC interviewer in 1976 that this was the reason he was later given a Member of the British Empire (MBE) award; because, as he put it, the Americans were ‘able to land marines at Tulagi and catch the Japanese at breakfast’.4
The success of the Solomon Island landings did not translate into a quick, bloodless victory. Although the Japanese on Tulagi had been caught by surprise, they mounted a brief but fierce resistance when the marines advanced across the island’s interior.
On Guadalcanal, the easy landings led to jubilation and overconfidence among the green, untested marines, leading them to fall for a simple ruse. A lone, drunk Japanese soldier wandered into the camp, and, when interrogated, told them that the Japanese on the island were starving, weak, and ready to surrender. Vandergrift’s intelligence officer, Lieutenant Colonel Frank B. Goettge, was in charge of the interrogation, and immediately asked Vandergrift to let him go on a mission to find these Japanese soldiers, telling his commander that ‘hundreds of Japanese troops were roaming the jungle west of us ready to surrender’. There was no shortage of marines willing to volunteer. Vandergrift assented, but sent the expedition by boat to a landing point around the coast that was away from known enemy strongholds.
Goettge’s expedition left at night, lost their way, and made landfall at the mouth of a river close to a Japanese encampment. They were surrounded and picked off, one by one, in the darkness. Three of the marines escaped, separately and alone, swimming back to the American beachhead. The last marine to leave the beach was Sergeant Frank L. Few, an American Indian from Buckeye, Arizona. As he swam out in the dark beyond the view of Japanese snipers, his last vision of the beach was the flashes of descending swords as the Japanese soldiers overran the marine’s position.5
In the following days, the main beachhead attracted sustained enemy attack. Meanwhile, the Japanese forces there readied themselves for combat in the overgrown interior. The support fleet came under such fierce aerial bombardment that Admiral Turner moved it away from Guadalcanal’s shores two days later, leaving them stranded without naval support.
This was the beginning of the long campaign for Guadalcanal, a battle that was to drag out for another six months.
Signals were arriving at Central Bureau from Townsville, where 1 Wireless Unit had a team of newly trained Kana operators, as well as a series of direction finders in place across Northern Queensland. The Americans were investigating Japanese army codes, but the volume of messages was still too low for them to make much headway. However, the Australians at Central Bureau, led by Mic Sandford, were putting the output from Townsville to good use in traffic analysis, as Pappy Clark showed with the August landings in the Solomons.
Traffic analysis (known these days as ‘Signals and Network Development’6) was not, as the name might imply, a study of vehicles. It was the investigation of radio traffic: the ebb and flow of the enemy’s radio messages between its military units. Now, as then, a traffic analyst isn’t interested in breaking an encrypted message, or learning what the message says. The analyst is simply interested in the fact that a message has been sent. What can that tell you? With a bit of ingenuity, quite a lot, and it can tell you even more if you look at the patterns across large numbers of messages.
A spike in radio traffic coming from a certain base could mean that large numbers of units were gathering there, or that an operation was about to begin. If an enemy stronghold increased its air-to-ground traffic, that might suggest a build-up of aircraft; the use of tactical codes, by contrast, could suggest coordinated activity with battlefield units.
In 1942, there was a goldmine of information embedded in the preambles to coded Japanese messages.
If a signals officer had a message to send, there was no point in just turning on the transmitter and tapping out the message in Morse or Kana, even if encrypted — the radio broadcast would be picked up by any receiver in range. In the Second World War, the airwaves were full of radio signals from both Allied units and Japanese Axis units, from navies, armies, and air forces, as well as from merchant ships and other units. Nobody would know who you were sending them to, and that included your intended recipient, without sending and receiving identifiers being tagged on. This wasn’t true just for Second World War radio messages — it’s true for all messaging systems. Packages need a mailing address to reach their intended destinations; phone calls can only be made between two unique numbers.
Therefore, every single radio message broadcast by either side contained a preamble — what we today call ‘metadata’. The most important information to include in a preamble was the identity of the sender and recipient. This was done with simple codes known as call signs.
For example, in the naval air-to-ground code, at one point the call sign for the air base Rabaul, to the north of New Guinea, was ‘290’, while the Japanese air base at Dili on Timor was ‘336’. A Japanese signal operator who had to send a radio message from Rabaul to Dili would include the call sign for both bases in the preamble before transmitting the message itself. Thus, the operators at Dili, when their radio receiver picked up the message, would know that it was a message for them, and that it had come from Rabaul.
Depending on what code system was being used, other information in the preamble included the code that the message was encrypted in, the time the message was sent, whether it was a multi-part message, and, in the case of the mainline codes, the unit or individual the message was specifically addressed to.7
Traffic analysis is the forgotten child of signals intelligence, overshadowed by code-breaking, but code-breaking is almost useless without it. For example, there is no point in breaking a code and reading an enemy message that says, ‘We are sending you the 21st Regiment’ if you don’t know who the message was sent from, or who it was sent to. On the other hand, if a traffic analyst tells you that the message was sent from the Japanese commander at Rabaul to the port of Wewak, the message is suddenly quite useful.8
Sinkov, being a mathematician and a leading expert in cryptology, was naturally suited to the code-breaking side of things. He and his expanding cohort of recruits arriving from Washington had cryptanalytical expertise, but no expertise in traffic analysis. Prior to the Second World War, the US army had not engaged in traffic analysis at all. (The US navy had been doing so, however, and did have such expertise, both at Monterey and Hawaii.) The Station 6 officer from the Philippines, Howard Brown, now attached to Central Bureau but currently in Townsville, was the first United States army officer to engage in traffic analysis when he started monitoring air raids against Manila.9 Sinkov’s colleagues at Arlington Hall in Virginia had sent some sigint personnel to England for training in traffic analysis; these personnel had begun some traffic-analysis work at Arlington Hall in April, but it was still a fledgling operation.10
For the Australians, the reverse was true. All of Australia’s cryptanalysis experts — Nave, Room, Trendall, Treweek, Lyons, and a handful of others whom Nave had been training — had been absorbed by the naval operation in Monterey Flats, leaving their compatriots over at Central Bureau with barely any code-breaking expertise. But Mic Sandford’s Australian army contingent at Central Bureau did have expertise in traffic analysis. In the course of travelling across Greece, Crete, Africa, and the Middle East on various campaigns, they had acquired knowledge from the British, and had become proficient in setting up their operation from scratch.
To help them get started with traffic analysis, Newman had supplied Central Bureau with a list of known Japanese army radio frequencies that the Australian navy had collected.
The United States navy also helped. Fabian provided them with a list of known Japanese army call signs, as well as a Japanese army address code that had been broken by his colleagues at Station HYPO in Hawaii.
The American naval sigint analysts were monitoring Japanese naval messages, not army messages. Even so, they had picked up some knowledge of Japanese army call signs, because these call signs would be used in Japanese naval messages during joint operations.
It turned out that, in 1941, the Japanese army had run out of call signs because of its rapid expansion across Asia and the Pacific. It was operating in locations in far-flung parts of the world from Burma, to Singapore, to Timor, and elsewhere — remote places that were not listed in the army’s call sign index. It got around the problem with a quick fix: if there was no call sign for a particular place, operators would simply spell it out using Kana. These were usually phonetic adaptations of the English names of these places. Rabaul, for example, was transmitted as RA BA RU, Sydney was SHI DO NI, and Timor TI MO RO.
This quick fix, with its mixture of proper call signs and on-the-fly spelled-out placenames, had the potential to cause confusion for message recipients. Therefore, if a Japanese army signals officer sent a placename in this way, he would first send two Kana letters, WE RA, and end it with MU WE. Thus, Timor would be sent as the following Kana string: WE RA TI MO RO MU WE.
These were easy to spot in a radio broadcast, because the placenames sent in this way were always bracketed at the start and finish with the Kana letter we. This code was broken in 1941 by Edward Layton, an American naval cryptanalyst. Layton called it the WE-WE code.11
As a token effort to disguise what they were doing, the Japanese army issued a Kana substitution table to be used with these placenames (in which RA became KA, for example). But once Layton realised what they were doing, this was quite easy to unravel.
Although it was only in use for a short time, this address code turned out to be of great value for Sandford’s traffic analysts. The WE-WE code enabled them to identify the location of various units moving into the south-west Pacific. Japanese signals headquarters in Tokyo soon released a new, expanded address-code index that included call signs for all the new locations, so that the WE-WE code was no longer used, but the traffic analysts had used the interval to learn the Imperial Japanese Army’s structure and movements.
Over the second half of 1942, as Japanese and Allied forces clashed in a series of battles in the Solomon Islands and the eastern end of New Guinea, MacArthur’s two main sources of Ultra came from the JN-25, courtesy of the United States navy, and from Sandford’s team of traffic analysts at Central Bureau.
Every transmitted message was of interest to the team. They recorded the preambles, particularly the call signs, before passing the air-to-ground messages to the Australian team and the other army messages to Sinkov.
In June, Booth’s newly operational 1 Wireless Unit in Townsville intercepted a message that used a WE-WE encoded placename, GA DA RU KA NA RU, as the location of an airfield under construction. It did not take long for the traffic analysts to conclude that this probably referred to the island of Guadalcanal in the Solomon Islands. The American forces were already planning an operation in the Solomons, so they expanded it to include a landing at Guadalcanal to stop the airstrip from being finished. Eleven thousand United States marines landed on the island on 7 August. The ensuing battle for Guadalcanal continued for several months until February the following year, when the Japanese forces withdrew.
In July, Sandford reported to Allied commanders that there had been an increase in overall radio traffic from Japanese bases to the North, which he thought strongly indicated that a major operation was imminent, probably within 30 days.12 Based on traffic analysis alone, he was able to advise Willoughby, MacArthur’s chief of intelligence, that the likely target was either Buna, on the New Guinea north coast, or Milne Bay, which was right at the far-eastern tip of New Guinea. Sandford also advised Willoughby that patterns of air-to-ground activity were similar to those observed in the Philippines during the Japanese invasion in early 1942 — an insight almost certainly given to him by Howard Brown from his hectic days at Station 6.
With the recent change in JN-25, many naval messages were now unreadable, and much of Fabian’s advice to MacArthur was also based on traffic analysis. Fabian advised MacArthur that there would be a major landing in New Guinea or the Solomons, and that, based on the movements of troop transports, the initial landings would be consolidated by the development of bases.
The Japanese army duly landed at Buna in July. However, even though the landing had been anticipated, heavy storms prevented the Allied aircraft from harassing them.13
More evidence came in that Milne Bay was the next target. Sandford’s team detected the movement of the Japanese 17th army into the area in readiness for an operation, and a partially decrypted intercept involving them referred to a unit as an ‘occupation force’. MacArthur’s new air force chief, General Kenney, sent reconnaissance planes across to Rabaul and confirmed a build-up. Japanese documents captured by Allied forces also referred to Milne Bay, dovetailing with the traffic analysis.
MacArthur conferred with Blamey about the need to reinforce Milne Bay, where the Australian army had recently established a base. Blamey sent the 18th Australian Infantry Brigade, bringing the total number of Australian soldiers in the bay to about 10,000 troops.
The Japanese army made landfall at Milne Bay on 25 August. Allied commanders waited anxiously for updates from the Australian troops there, but radio messages were infrequent and provided little information. The generals had no idea what was happening.
It was the wet season, and in the hasty, makeshift buildings of the Australian encampment, the army radio sets had got wet, and were out of operation. That was why there were so few update reports.
The Japanese army had landed in the middle of a torrential, lashing storm. Visibility was poor, and neither side was sure what was going on. The battle on the beach and into the sodden coastal jungle was chaotic, with shells ripping through the drenching rain, and soldiers on both sides stumbling through the foliage, hunting each other. The sun broke through and air support arrived, but with soldiers scrambling around in the dense undergrowth below, the pilots could not tell who was on which side.14
In the midst of the communication blackout, MacArthur finally got updates on the progress of the battle, but from naval decrypts supplied by Rudy Fabian. The navy was intercepting Japanese naval messages about the battle, and Fabian was able to inform him that the Australians were winning. One Japanese commander had been killed in the fighting, and another one wounded; only one-third of the Japanese landing force was still operational.
Despite this, the Japanese army established a beachhead, and would not let it go. Sandford discovered reinforcement shipping nearby. Soon afterwards, based on intercepted naval messages, Fabian advised that the shipping in question was on its way to Milne Bay, but that it was in fact an evacuation convoy. Bad weather again prevented planes from attacking the convoy, and the Japanese forces on the bay were evacuated to Rabaul in the north.
The Australian army thwarted the Japanese army’s attempt to capture Milne Bay, in part due to information gleaned by Sandford’s traffic-analysis team. This in no way diminishes the actions of the Australian troops who fought in the Battle of Milne Bay, some of whom were killed in action. The intelligence provided by Ultra was the reason they were there, but Ultra did not fire the guns. The battle still had to be fought on the ground. It was the first time in the Pacific War that a Japanese land-based operation had been successfully repelled.
19
Leaks
The whole inter-Allied and inter-service cooperation thing didn’t sit well with Rudy Fabian, who believed that too much swapping and sharing of information with other agencies only increased the risk of Ultra falling into the wrong hands. As the United States naval commander at Monterey, he exerted tight control over the intelligence generated there.
Some Central Bureau staff complained that Monterey, although it was a short walk across the park, might as well have been on the opposite side of the world, such was the lack of cooperation.
Two young American cryptanalysts, Hugh Erskine and Charles E. Girhard, newly arrived in Melbourne and working for Sinkov, visited Monterey Flats. After they returned to Cranleigh they discovered that General Akin already knew about their visit; consequently, he personally instructed them to stay away from Monterey in future.1
Central Bureau personnel were not alone in finding FRUMEL difficult to deal with. When an inspector-general from the US Seventh Fleet arrived at Monterey and tried to enter, the guard at the front door stopped him.
‘I cannot let you go in,’ the guard told the inspector-general. ‘I’ll call the commanding officer.’
The inspector-general turned and left, furious. He complained about his treatment to the chief of staff, who advised him to leave FRUMEL alone.2
Fabian had an agreement with MacArthur to deliver intelligence briefings in person, which he did at MacArthur’s headquarters. He had no respect for General MacArthur’s intelligence chief, Charles Willoughby, and refused to brief MacArthur if Willoughby was present. On one occasion, Fabian arrived with an important intercept, showed it to MacArthur, and then, as he passed Willoughby’s desk on the way out of the building, he stopped, lifted the top-secret document, and set it alight, dropping the ashes into Willoughby’s waste-paper bin before returning to Monterey.
Fabian became increasingly concerned that his operation was not as watertight as he hoped. He suspected that Eric Nave, in particular, was sharing information with other agencies behind his back. He was right about that. Nave was providing briefings to Australian military intelligence officials at nearby Victoria Barracks, as well as to Australian naval intelligence.3 Nave had the view that high-ranking intelligence personnel could be trusted, regardless of which service or department they belonged to — an attitude that had previously got him into trouble with Jack Newman.
Newman’s fears were not allayed by a security breach in the United States. On 7 June, the Chicago Tribune newspaper published a front-page article with the headline, ‘US Navy Knew in Advance All About Jap Fleet’.4 The article then gave details about how the United States navy knew about the Japanese plans for Midway well in advance, including their tactic of pretending that the target was the Aleutian Islands. Alarmingly, the article specifically mentioned code-breaking as the source.
Japanese intelligence agents never saw that edition of the Chicago Tribune, so the secret that JN-25 had been broken remained safe. Nonetheless, the response from the United States government was swift. The journalist who’d written the story, Stanley Johnson, was brought before a grand jury for treason, although in the end the jury decided not to indict.5 It could not have escaped Fabian’s attention that the journalist who almost destroyed all his achievements, and who jeopardised the entire code-breaking operation, was Australian.
And earlier in the year, Australia’s prime minister, John Curtin, himself made a similar slip-up, briefing journalists in April that the Allies knew about a build-up of Japanese warships in the Marshall Islands. This, too, was information that had been obtained by the painstaking, secret work at Monterey Flats in Melbourne and by their naval colleagues in Hawaii.6
Neither of those leaks came from Eric Nave, but Fabian became increasingly suspicious over time, and openly accused Nave of breaching security. He wanted Nave out of Monterey. An opportunity to do so was at hand, as the Allied nations negotiated a new intelligence arrangement.
In October 1942, Britain and America entered into a new secret agreement for cooperation in intelligence, the Holden agreement. This allowed for the increased sharing of information on Ultra, as well as a clearer division of responsibilities. Previously, both major Allied powers had been working on Japanese naval codes; now, under the Holden agreement, the United States took overall responsibility for naval codes, although with some continued British participation.7
FRUMEL was placed under the direct control of the United States navy, whereupon Fabian insisted that civilians no longer be employed. The only civilian on the ground floor was Newman’s personal assistant, Joan Duff, and she was transferred to a new intelligence unit dealing with radar and counter-radar called Section 22.
It has been noted by historian Ian Pfennigwerth that Rudy Fabian must have had friends in high places, because Eric Nave was personally named in the Holden agreement; it specifically stated that he could not work at FRUMEL. From a cryptanalytic perspective, this was a loss to FRUMEL, because Nave was actively engaged in the work on JN-25 until the day he was expelled from Monterey; however, security was an even higher priority for Fabian than code-breaking success.8 Professor Room also left FRUMEL, but for different reasons: he had been focussing on diplomatic cipher work, which was no longer a US navy responsibility, and he was also a civilian.
Sandford was quick to seize the opportunity to gain these two star recruits. He offered them both a job at Central Bureau, which had vacated Cranleigh and moved north to Brisbane a month earlier.
The diplomatic section at Monterey was also transferred to the Australian army, and moved from Monterey Flats in St Kilda to a site at Bonegilla. The Sydney professors, Arthur Dale Trendall and Dicky Lyons, stayed with the diplomatic section, which was renamed ‘Special D Section’. Professor Treweek, who had a good relationship with both Rudy Fabian and Jack Newman, remained with FRUMEL at Monterey Flats.
Nave was not the only individual at Monterey who was more inclined than Fabian to cooperate with other agencies.
Newman berated Sandford early in May about Central Bureau’s lack of success and for encroaching on signals intelligence that was the navy’s responsibility. ‘To be brutally candid’, Newman thundered in an official minute to Sandford, ‘a review of the Central Bureau reports to date reveals no real W/T intelligence, but shows that effort has been expended on matters that are incidental to Naval Y intelligence and are not within the province of Central Bureau or within is capabilities which are necessarily limited by shortness of experience.’9
It was a withering assessment, but Newman’s rudeness was largely theatrical, designed to curry Fabian’s trust and approval. He sent Sandford a very different minute in July:
Last time I wrote you a minute it was a very rude one. I would ask you to look back on that with a grain of salt, since I did it deliberately.
Now I would like to say (for what it is worth to you) that I think your outfit has got on to the right lines and is likely to make a good job of it.
Don’t forget, however, that the foundation on which your house is built is thoroughly trained and thoroughly reliable personnel — both analytical staff and ‘Y’ operators — and both officers and other ranks. In this regard it may be of interest that the recent change of major call signs at 1500 G.M.T. on 24th has been completely and accurately flogged out by a well-trained gun’s crew of four telegraphists W.R.A.N.S. within 23 hours.
Please do not look on this patronising epistle because that is the last impression I want to give. If however I can be of any help towards the general war effort I should be delighted to receive any private enquiries from you.
Best of good luck, and don’t put much trust in R.A.A.F. D/F bearings!
Yours Sincerely,
Jack B Newman10
This letter to Sandford, suggesting that they exchange information privately, was an offer to establish clandestine communications behind Fabian’s back. A sort of ‘black market’ in sigint material from FRUMEL to Central Bureau was established. There was no need to be secretive about the other direction because Central Bureau, in keeping with the terms of Simpson’s conference, freely provided anything to FRUMEL that was likely to be of use to them.
Newman’s public hostility toward Sandford must have worked, because Fabian never suspected that Newman was responsible for what he called ‘security breaches’. Eric Nave was the focus of his distrust.
20
Nyrambla
By the second half of 1942, the Japanese controlled an arc of islands to the north of Australia, including Timor, Bougainville, New Britain, and much of the Solomon Islands, as well as the north coast of New Guinea. From airfields in these locations, they launched continual air raids against stretched and depleted Allied forces in northern Australia and southern New Guinea.
Despite this, the Australian mainland was no longer considered to be under direct threat of invasion. The Japanese plans to capture Port Moresby had been thwarted at the Battle of the Coral Sea, and their forces were now re-grouping at Rabaul. Melbourne now seemed too far from the action for MacArthur’s liking, so he moved his headquarters to Brisbane in July. The Australian government provided him with use of the AMP office tower in the heart of the city — a building chosen in part because of its solid, reinforced-concrete structure, which meant that it would withstand an air raid better than most. General Akin went with him, and, with Akin, so did Central Bureau.
Central Bureau were assigned a large, two-storey house about ten minutes’ drive north-east of the city centre, at 21 Henry Street, Ascot. It was conveniently located, close to Akin’s office in the AMP building in one direction and a new airfield that the US army air force had recently built at Eagle Farm in the opposite direction. This airfield became the delivery point for incoming material from intercept field units.1
The house was a huge, old double-brick homestead with high verandahs and woodworked awnings, built in the 1880s on an acreage on the outskirts of the city. The original owner was a wealthy banker named Henry P. Abbott, who named the house ‘Nyrambla’, an Aboriginal word whose meaning was not recorded. When Abbott retired, he sold Nyrambla and moved to Sydney. The property passed through a succession of owners. As the city grew, the surrounding farmland gave way to suburbs, and the acreage itself was subdivided and sold. The street in front of the building was named Henry, and the street behind named Abbott.
The Queenslander magazine ran a feature article on the property in 1932, noting the six bedrooms, upstairs sitting room, and a two-storey rear wing for maid’s quarters, kitchen, and breakfast room. The magazine described the cavernous ground floor:
There are a number of fireplaces in the house with tiled hearths and marble mantelpieces. The hall is wide and high-ceiled. It is divided by two arches. The one nearest the front door is occupied by a doorway of cedar. The arch itself is filled in with glass, divided by bars of wood, the whole resembling an open fan. The door below and its side panels are of red glass. The panels on each side of the big front door also are of red glass, each with a woman’s figure marked on it in white, and the name ‘Providencia’ beneath one and ‘Hospitalita’ below the other.
A visitor approaching Nyrambla from Henry Street was met with the showy opulence of the front entrance:
Two bay windows open on to the front veranda on each side of the porchway and on to the balcony above. This porchway is arched above the set of front steps, with the name ‘Nyrambla’ marked thereon in black letters. The paving of the porch is of black and white tiles in large squares. Columned arches form the sides of the porch. Beyond the second arch, towards the rear end of the hall, is the staircase. It is of cedar, like all the rest of the fittings in this well-built house, but its railing is of iron.2
The small American contingent of Central Bureau, all of whom had been in Melbourne for only a few months or less, were the first to move up to Brisbane. After inspecting Nyrambla, they found it to their liking and moved in, taking up residence on the upstairs balconies. When Sandford arrived, they were evicted and sent to find billeted accommodation in surrounding suburban homes. The organisation was about to expand, and the whole building, as large as it was, would be needed for operations.
Army renovators had fitted it out in preparation: the ballroom was divided with makeshift walls into a rabbit warren of small office spaces, while other areas had similarly been sectioned and divided in ways that the army acquisitions people imagined might be useful.3 Various rooms were allotted to either traffic analysis or cryptanalysis, or for processing bags of intercepts arriving from Eagle Farm airstrip.
New personnel arrived in dribs and drabs. Most of the Australians sent to Central Bureau had been hand-picked as suitable for intelligence work from military training camps across the country. Those that came from interstate arrived at Brisbane Central railway station, where they would be met by an army intelligence officer and escorted directly to Nyrambla, or sometimes first taken to MacArthur’s headquarters in the AMP building.
They were billeted in local apartments and hotels, but, in keeping with Sandford’s policy to isolate intelligence staff from the rest of the military, never stayed at regular barracks, and were discouraged from mixing with the thousands of Australian and American troops now stationed in Brisbane.
General Akin made it a policy not to visit Central Bureau, because he did not want to draw attention to its activities. Central Bureau was not his only responsibility; as MacArthur’s chief signal officer, he was responsible for the entire United States army’s signals operations in the south-west Pacific area. His arms-length approach gave the impression to many there that he had little involvement with Central Bureau, but in fact he was in regular contact with its senior officers, and over the course of the war made visits to Central Bureau’s intercept units and other field operations.
Brigadier John Rogers, the director of military intelligence, established a second office in Brisbane, where he spent most of his time from that point forward. As a result, the Military Intelligence Headquarters in Melbourne effectively fell under the command of his second in charge, the assistant director of military intelligence (ADMI), Lieutenant Colonel Robert Little. Like Rogers, Little had served in the First World War, and had been decorated for his role as field artillery battery commander in northern France. According to the citation, ‘by his personal bravery and initiative kept guns in action under adverse conditions’. Rogers increasingly focussed on operational matters and on liaising with the Americans, while Little took care of the back-room politics and liaising with Australian politicians.4
Rogers did not share Akin’s qualms about visiting from time to time, and neither did the head of the Australian military forces, General Blamey. Mic Sandford would give them tours of the Nyrambla operation, providing a breezy commentary as he escorted them from place to place.
On one occasion, Sandford was showing Blamey around when the pair walked into an office where a man sat at a desk, hunched down with his hands over his head, groaning. The staffer was Harry Waters, who suffered from chronic migraines. Sandford, not being familiar with Waters’ condition, was momentarily taken aback but quickly recovered, and with a wave of his hand remarked to Blamey, ‘It would drive you nuts, doing this stuff all day.’ Blamey nodded. With that, Sandford guided Blamey back out of the room and on to other parts of the building, his breezy commentary continuing with barely a break.5
Sandford was too busy for the hands-on work of traffic analysis. Pappy Clark had emerged as a natural talent in the area, so in September he was promoted from lieutenant to captain (and later to major), and was put in charge of all traffic analysis.
He had been, from the start, the joker of the section. During the group’s return voyage from Egypt to Australia aboard the Mendoza, the mood was bleak in the sweltering, inescapable tropic heat. The ship was overcrowded, and they were cramped and sore. Fist fights would sometimes break out. On a particularly hot day, Pappy Clark appeared on deck dressed in coat, gloves, and balaclava, shivering and swinging his arms as if to warm himself up. The sheer absurdity of Clark’s antics broke everyone up into laughter.6
Captain Clark’s priority now was to compile a list of Japanese army call signs and figure out what each one stood for. Other sigint centres, particularly the British Wireless Experimental Centre in Delhi, had been collecting call signs and had provided what they knew. From these, and from the daily stream of intercepts arriving in bags from the Eagle Farm airfield, Clark built the call sign catalogue.
He learned from Sinkov about the new traffic-analysis team at Arlington Hall who were keen to contribute. Clark would assign them analytical tasks, sending material to Washington for them to work on via the direct communication line between the two centres.
The Japanese army was not making his life easy. They were using a very complex system of call signs, which had at first baffled everyone. All the signs were encrypted in their own special code, completely separate for the code systems used for the message itself. This was not like the WE-WE code, which used a simple substitution method. This was a full-blown code of its own, complete with codebooks and additives. Clark called it the ‘place name code’. It was eventually broken at Central Bureau by Pappy Clark’s traffic-analysis section, with some help from Sinkov and Sinkov’s colleagues at Arlington Hall.7 The naval cryptanalysts also contributed with the provision of army call signs that had been sent on navy channels along with the naval equivalent of the same call sign.
Sinkov was still stymied by the three-figure codes. It started to become clear that the problem was so hard because the Japanese army was not using a single tactical code; it was using dozens of codes. This was made possible by the unusual policies of the Imperial Japanese Army. Low-level units were not allowed to communicate with each other, only with their parent units. Regiments didn’t talk to each other, only to their division command. Division commanders, in turn, couldn’t communicate with other division commanders, but always had to deal directly with their army headquarters, and so on, up and down the chain. Although it might sound very inefficient, this meant that each communication line could use its own unique code system. The traffic analysts contributed to this astonishing discovery as they pieced together the structure of the enemy’s army.8
Japanese tactical-code systems proliferated across the south-west Pacific. They emerged from mysterious units in the uncharted wilderness of New Guinea, from islands to the east and west, and from further north. They came and went like fireflies in the dark. There was never enough traffic in a three-figure code; just as a solution began to emerge, the code would vanish, and the intercept stations would hear it no more.
Solving the various puzzles involving codes, traffic, ciphers, and call signs needed lots of data — and that meant more intercepts. The Japanese army was not making it easy. Central Bureau was stymied by certain Japanese army practices that thwarted their attempts to collect large numbers of messages for each system. The army units were talking to each other on low-power transmissions with limited range, making the messages hard for the intercept stations to pick up clearly unless they were close by. A unit would send a brief radio burst at high power, enough to make contact, and then turn down the power so that the transmission was barely strong enough for the recipient to receive it, and no more.
An Australian intercept site at Coomalie Creek or Townsville was too far away to pick up many of these quiet, low-power transmissions. More intercept units were needed, and they were needed as close to the enemy as possible.
21
Wet boots
In late September 1942, a signals unit arrived in Port Moresby on a troop ship from Townsville. The unit was the 55 Australian Special Wireless Group, with John ‘Jack’ Vasey as their commanding officer.
Port Moresby was the main Allied base in Papua New Guinea and the staging point for forward operations, such as the Diggers engaged in jungle warfare inland along the Kokoda Track.
The town had suffered numerous air raids, sometimes daily, from the Japanese army air force. The harbour was littered with the wrecks of bombed ships that jutted from the water like rusty islands. The beach was long, wide, and shallow — too shallow for swimming — but bomb craters filled with seawater made for good swimming holes. Most of the buildings in town were Port Moresby thatched huts, clustered around the central market, where the local currency was cowrie shells. Papua New Guinean tribespeople would visit the town, the men carrying spears, and the women carrying children slung in woven slings on their naked backs.
Arriving in Port Moresby in the dry season, it seemed to Central Bureau’s Doug Pyle that ‘everything was yellow’ — the grass, the earth, buildings, the occasional forlorn palm tree, and the skin of the soldiers taking Atebrin tablets to ward off malaria.1
By 21 September 1942, Imperial Japanese Army troops had reached the south side of the Owen Stanley Range, and were so close to Port Moresby that they could see its lights in the distance.
Australian troops had suffered heavy casualties fighting them along the Kokoda Track, in an extended battle to stop the Japanese overland advance to Port Moresby. Australian forces, including troops of the newly arrived 25th Brigade, had dug in on Imita Ridge. The troops of the Imperial Japanese Army, so close to their objective, were in a buoyant mood, and shelled the Australian defenders with heavy mortar bombardments.
Twenty-five miles away down on the coastline, Port Moresby had changed from a quiet colonial British outpost to a large Allied military base in an urgent and grim mood. Jeeps and trucks moved through the dirt streets.
The beach was contoured with bomb craters. The surrounding countryside was strewn with troop encampments, and landing fields were lined with the remnants of the Allied air force in the south-west Pacific.
General George C. Kenney, a new US army air force commander, had recently arrived from America to serve under MacArthur. Kenney learned that Allied air power in the south-west Pacific area — the combined number of planes of the Australian and US army air forces — totalled a paltry 80 aircraft. This wasn’t the fleet on the airstrips at Port Moresby, but the sum total for the entire of North Australia, New Guinea, and the south Pacific islands. Because of the focus on the war in Europe, much of what airpower the Allied nations had was deployed in Europe and Africa. Most of the rest had been shot out of the sky by Japanese Zero fighters or bombed by Japanese ‘Betty’ bombers as they sat on the grassy airstrips of northern Australia and southern New Guinea. Reinforcements were, of course, on their way from the United States.
Even for the operational planes in Port Moresby, air operations were a hit-and-miss matter, due to the extreme tropical weather of New Guinea. To the north, the towering Owen Stanley Range was an unbroken labyrinth of mountain ridges and jungle valleys, frequently covered with thick clouds and mist. Torrential rains frequently came over the far-northern side of the range, and would soon be coming to Port Moresby itself when the wet season began.
The signals unit did not stay in Port Moresby for long. They were taken by barge to their designated site on Fairfax Harbour to the west, at a place called Seven Mile. Their intercept site had already been set up, with antennas erected and radio receivers installed, thanks to two men who had arrived before them from 51 Wireless Group in Darwin.
The two operators from Darwin had been assigned to join Vasey’s unit to add much-needed experience and to bolster the unit’s numbers, which were currently at half strength.
Jack Butler, one of the two from Darwin, explained later how they had made the most of their early arrival:
When Gordon (my mate) and I were unloaded in Port Moresby nobody wanted to know us. The Japanese were advancing down the Kokoda Track and things were decidedly unpleasant for two unattached personnel. We made ourselves known to the area CSO and asked for a receiving set, aerial wire, stationery and a work tent so we could begin mapping out the Japanese groups in New Guinea. We did, of course, have an ulterior motive. We reckoned that, if there was to be another evacuation, we were going to be the CSO’s bag carriers.2
Jack and his mate, Gordon, operated the site on their own for four weeks before the arrival of Vasey’s 55 group. They had two kinds of radio receivers — the stock-standard American-made HROs, just like the ones in Townsville and Darwin, and an Australian model, the Kingsley 101, made in Melbourne. The HROs were normally reliable, but suffered in the damp, humid New Guinea weather. The Kingsleys were encased in rubber, making them more resilient in the jungle conditions and less likely to suffer damage in transportation.3
A direction-finding station was set up about two miles away. Enemy radio traffic was high, and the unit worked hard to keep up. There was a large amount of Japanese air activity; air raids against Port Moresby and other Allied bases were constant. The unit monitored army air-to-ground and naval air-to-ground traffic, and even in cases where they were not able to decrypt the message, alerted the local commanders whenever an air raid was imminent.4
The unit had two tents: one for interception, and one for intelligence. This was the structure for all the special signals units, including the 51 Section at Coomalie Creek, and 1 Wireless at Townsville (although they worked in a concrete bunker rather than tents). Wires connected the intercept aerials to the receivers in the intercept tent where the Kana operators worked, listening to Japanese Kana codes and transcribing them throughout the day and night. The transcriptions were then taken to the second tent for the intelligence staff to work on. The intelligence staff included personnel trained in decoding air-to-ground messages, a translator, and the commanding officer, who would make operational decisions about what to do with the intel gained from the intercepts.
Those in the intelligence tent were designated by Central Bureau as the ‘Y’ staff. The work of Y intelligence people was mysterious to the regular soldiers. They had been trained in intelligence, apparently, and probably thought they were better than everyone else. The regular signals personnel, in keeping with the Australian tendency to cut ‘tall poppies’ down to size, renamed them the ‘Shit Staff’.5 The name stuck.
A lot of air-to-ground messages that arrived on the desks of the Y staff at the camp at Fairfax Harbour were either partly encoded or not encoded at all. For example, sometimes the sender had composed the message using the codebook, but for whatever reason had done no further encryption, such as applying additives or scrambling the syllables using a cage. These secondary, more complex, techniques were called ‘encipherment’ (whereas merely using a codebook was called ‘encoding’). So, at Central Bureau, these partially encrypted messages were called unreciphered messages. And since the codebooks were often solved or mostly solved, such messages could be read on the spot by an experienced cryptanalyst.
On the other side of the mountains, at Wanigela on the north coast of Papua New Guinea, an advance party of four men from 55 Section had set up a direction-finding station. Just getting there had been an odyssey. When Jack Ryan had given them their orders back at the Bonegilla training camp, he had made four points:
1. You are going into enemy territory;
2. Never let your equipment be captured;
3. If attacked (on the way to location), retreat and go another way;
4. Best of luck.6
They went by ship to Port Moresby, from where they were flown to Milne Bay, and then travelled by ship 100 miles along the north coast to Collingwood Bay. There, they went ashore in a longboat, through the swamps to an Australian army camp at a small airstrip in a field of kunai grass.
They arrived with a single large tent, a Bren gun and ammunition, grenades, and two tons of direction-finding apparatus. They installed the direction-finding equipment, and, to protect it from the rain, constructed a hut out of natural material they found near the campsite. But the hut was no match for the tropical climate, and rapidly deteriorated. The jungle weather took a toll on the equipment, too, which had to be dried out in the sun periodically to kill the mould.
The basic principle in radio direction-finding — or ‘D/F’, as it was called — was to get two readings of the same radio signal from different locations, and use them to estimate the location of the source of the signal. When the D/F unit at Wanigela picked up the direction of an enemy radio transmission, and the D/F unit at the main base also got a fix on the direction from their location, the section could put the two together and estimate where the transmission came from.
Each day, Jack Vasey called the headquarters of the US army’s Fifth Air Force on a makeshift landline that ran from Seven Mile around Fairfax Bay to Port Moresby, giving them information on enemy aircraft movements, and particularly on incoming air raids. Sometimes the messages came in plain language that a local translator at 55 Section could read, and on those occasions Vasey could even give the commanding officer the intended target of the raid.
All messages and direction-finding readings were flown in mailbags back to Central Bureau, where they were sorted and processed by Pappy Clark’s traffic-analysis section. Here, insights were gleaned about the movements and plans of the enemy, particularly its troop movements up and down the northern New Guinea coast from places such as Palau and Rabaul. These insights were provided to Akin in a constant stream of updated intelligence, who in turn provided them to MacArthur and to MacArthur’s head of military intelligence, General Willoughby, whose disposition toward sigint was growing progressively warmer.
By the end of 1942, the Japanese army had retreated back across the Owen Stanley Range, and the data coming in from Wanigela was accordingly less useful with every passing week. It was getting to the point where most of the locations of the enemy transmissions were simply described as ‘a long way west’.
Vasey sent the order to Wanigela to pack up the D/F equipment and move to a new location. They moved out in early January 1943; but on the way out, the boat transporting them became beached in the mud, and was stranded there for two days. As they sat trapped in the shallow water of the bay, a Japanese reconnaissance plane flew overhead. They were now sure they were doomed, and anticipated a bomber would follow after the plane had spotted them. But apparently they were too small a target and too far away from the Japanese airstrips to bother with, because no bomber came.
After heading back around to Milne Bay and then to Port Moresby, they moved west to a place called Kerema, on the southern coast of New Guinea — just as remote as Wanigela had been — where they re-installed their D/F receivers and resumed their daily routine of mosquitoes, tinned food, and radio interception.
1943
22
Sausages and sandshoes
Hitler’s invasion of Russia was not going to plan. His forces had advanced rapidly in late 1941, but by the start of 1943 were far from home with dangerously extended supply lines. The German army was engaged in a prolonged battle with the Soviet army at the city of Stalingrad, where the freezing cold of the Russian winter and the entrenched Soviet defenders were taking a huge toll. The battle of Stalingrad ended when General Paulus surrendered to the Soviets. His Sixth Army had been reduced to 91,000 troops, having lost over 150,000 during the campaign.1
To Australia’s north, the Allies were locked in a back-and-forth air war against the Japanese army across the Owen Stanleys, the spiny mountain range across New Guinea. The Japanese controlled Singapore; Irian Jaya (then called Dutch New Guinea); Indonesia (then the Dutch East Indies); most of the north coast of New Guinea; and Rabaul, further to the north. Meanwhile, the Allies controlled Australia and the south coast of Papua New Guinea. Both sides were locked in close fighting across the Solomon Islands.
The war in the south-west Pacific area was being fought on a shoestring. The Japanese supply lines, like those of the German army in Russia, had become severely stretched — in their case, across Asia and the Pacific — and the Japanese army had suffered heavy losses on the Kokoda Track and Guadalcanal. On the Allied side, the leaders in Washington and London were focussed on defeating Hitler in Europe and Africa. The south-west Pacific was a lower priority, drawing less of their attention; it was a backwater of the war. MacArthur and Kenney had to make do with whatever they could get from Washington, while Australian forces, led by General Blamey, provided what support they could.2
Signals intelligence was critical, given the limitations on troops and equipment. The Allied generals had to get as much value as they could from every battle, every bombing raid, every movement. And signals intelligence was not letting them down; it was providing a stream of first-class information about the enemy.
Every day, the wireless units at Port Moresby, Coomalie Creek, and Townsville sent bags of hand-written radio transcripts to the nearest airstrip, where they were loaded onto planes and flown south to Brisbane. There, the planes landed at Eagle Park airfield to the north-east of the city. The sealed bags were offloaded and delivered by army lorry to Central Bureau, where their contents were spilled onto sorting tables.
In a ground-floor room at the back of Nyrambla, Nev Wintin and Alan Flannery invented a word game. They were compiling Japanese messages based on the places of origin, all of which had been abbreviated to a four-letter phonetic shorthand. Thus, Singapore was ‘Sing’; Narimasu was ‘Narm’; Sasebo was ‘Sase’; and Hiroshima was ‘Hirs’.
Wintin and Flannery became bored with these abbreviations, and twisted them for fun. ‘Sing’, the code for Singapore, rhymed with ‘Bing’ and so became ‘Crosby’, in homage to the singer Bing Crosby. ‘Narm’ reminded them of the Sydney shoe manufacturer North Australian Rubber Mills, so it became ‘Sandshoes’. ‘Sase’ became ‘Sausages’. The code for Hiroshima, ‘Hirs’, sounded like ‘Hearse’, and so their word-game version of Hiroshima was ‘Funeral’.
They coded the Japanese bases according to the whims of their wordplay, compiling daily traffic for Sausages, Crosby, Sandshoes, Funeral, and other quirky names. When ‘Flan’ wrote the daily report, Wintin would call out the made up names, and Flan would have to translate the places back to their true names. The system worked until one day a report was accidentally sent to Washington littered with Flannery and Wintin’s names, meaningless to anyone but them.3
The two men were working for Stan Clark in his traffic-analysis section, having been sent on temporary loan by Jack Ryan’s Special Wireless Group to help fill a shortage in staff. While Sinkov and his American code-breakers toiled upstairs on the impenetrable Japanese army’s Mainline codes, Clark’s section was having more luck. This was because Clark’s traffic analysts did not need to read a message in order to learn something useful. Knowing that the message had been sent was enough.
Pappy Clark was now closing in on a complete solution to the thorny problem of Japanese army call signs.
Upstairs, Sinkov’s cryptanalysts struggled to make sense of the Japanese Mainline army code, the code for the highest levels of Japanese army communication. Meanwhile, Clark downstairs had found that even doing traffic analysis on the Mainline messages was tricky because the Japanese army went to great lengths to conceal its call signs.
Nev Wintin and Alan Flannery in the traffic-analysis section were having fun with word games, but getting to the point where they could even play those word games with Mainline messages was far from trivial. Clark didn’t care about the contents of the messages; air-to-ground messages were Eric Nave’s problem, and Mainline messages were a headache for Sinkov to deal with. Clark was only interested in the preamble. Who was the message from, who was it to, when was it sent? Most of the code systems, particularly early on, contained fairly straightforward preambles, but the preambles for the Mainline army code were complex and mysterious. For one thing, they varied in length. For another, the call signs seemed to move and change position. And most frustratingly of all, the call signs themselves were encrypted with their own special code, completely separate from the code that was used for the rest of the message. He needed to untangle the preambles to give his traffic analysts something to work with.
If the preambles couldn’t be penetrated, there would be no more jokes about sausages and sandshoes coming from the downstairs traffic team. Clark would not be able to track the movement of enemy forces, and the Allied generals would have to make their decisions across the south-west Pacific area without knowing what they were up against.
An early insight, the first crack of light in the wall, had come a few months earlier in October 1942 from his team’s work on radio nets.
The Imperial Japanese Army’s communication structure was very hierarchical, even using different tactical codes for various divisions. This was a headache for the code-breakers, but it made life easy for the traffic analysts, because they were able to quickly classify radio messages according to the command structures they fell into. These radio networks were called nets. At the start, they tracked three nets: the army’s, the army air force’s, and the navy air force’s. Before long, they were finding nets for different regions. There was a net for Japan, for Korea, the Philippines, Rabaul, Singapore, and so on. They tracked the geographical coverage of each net, and this gave them the zone of control for various army headquarters. As the networks became more detailed, the traffic analysts started naming them after animals. Doing this made it easier to keep track of them all and to reduce the chance of getting lost in a sea of frequencies and call signs.
‘Kangaroo’ was the net used by the Japanese 8th Area Army, with its headquarters in Rabaul and operating in New Guinea and the Solomons. ‘Buffalo’ was a net based in the Philippines. ‘Cuckoo’ was a flying-army circuit based in Singapore.4
With all these different circuits and their vertical structures, it was clear why the preambles were such a tangled mess. Clark realised that the Japanese army was using a routing system. Messages would pass through many nets from origin to destination, going first from one signals office to another, and then to another. All the routing information had to be included in the preamble.
If military chiefs in Tokyo wanted to give orders to a regiment on the frontlines in the Solomon Islands, they would not — and could not — broadcast a message directly to the local commander. Instead, the Tokyo signals office would radio the message to Manila. The Manila headquarters would pick it up and re-broadcast the message to Mindanao, then to Rabaul, and then to whatever frontline unit was supposed to get it. Similarly, if an army commander in Singapore wanted to contact Rabaul, the message would skip from base to base across South-East Asia, passing through Manila, before arriving at its destination.
This meant that the preamble included not only the sender and receiver. It also included the point of origin and the ultimate destination. It sometimes included multiple recipients, and sometimes the specific addressee, such as ‘Chief of Staff, 14th Army’.
The preambles were complex, but contained a lot of useful information once they were unravelled. It was all food for traffic analysis; but from Pappy Clark’s point of view, the real gold lay in the routing codes for the point of origin and the destination.
Kana operators at intercept sites were sometimes instructed to focus on what they called the ‘DD’ codes. These were the codes within the preamble that identified the ultimate sender and recipient. The sender was identified by DD in Kana code (representing the Japanese syllable HO, repeated), while the recipient was designated in Kana code as P- (representing the Japanese symbol for a long vowel).
Thus a message from Manila to Rabaul would contain the following string, somewhere in the preamble:
DD MANILA P- RABAUL
If the recipient had relocated by the time the message arrived — something that can easily happen in wartime — the message was forwarded with additional routing information. The old, incorrect recipient location would be appended with the word MUTO. For example, if the recipient in the example just given had moved to Palau, Rabaul would send it on with the following string somewhere in the preamble:
DD MANILA P- PALAO MUTO RABAUL5
Of course, no Japanese army message ever came through to the interceptors with the plain text, ‘DD Manila’. That would have been too simple to solve. Japanese signals units had their own specialised code system for placenames in the preamble. Instead of sending the word ‘Manila’, a Japanese signals operator would convert Manila to a string of digits, using a placename codebook. The operator would then apply additives to those numbers, using specialised additive books. This created an entirely new layer of encryption — not only were the messages encrypted, but the address system for sending the message was, too, as a stand-alone cipher separate from the one used to encrypt the message.
Central Bureau had been studying the address codes since 1942 with help from the code-breaking centres at New Delhi and Arlington Hall. As each message bounced its way across Asia and the Pacific, it would pass through multiple command radios that would encrypt the routing codes in different ways. Singapore, controlled by the Japanese 7th Area Army, would encrypt the placenames with its own local system. When the message arrived in Tokyo it would be decrypted and re-encrypted into a separate system before being broadcast elsewhere, such as to Rabaul.
But sometimes the Japanese operators would forget to change the encryption before re-sending the message. As a result, the next station down the line would not be able to read the preamble, and would ask for it to be re-sent. This helped Central Bureau, because then they would have a ‘crib’: an example of the same preamble encrypted two different ways. This could help them break into an unknown code system, if the other example was in a code that had already been broken.
The value of all this was that Clark could now track high-level messages across the entire theatre of war. He could also make guesses about likely movements. Distant military units often made contact with each other before one of them relocated to the same place as the other. If a Japanese army division in China passed a series of messages to Rabaul, Clark would start to think that they were going to be deployed in Rabaul. When messages from the same division arrived, but now with a DD code of Manila, they were probably on their way and had broken their journey in the Philippines port, maybe waiting for a troop transport for the next leg of the trip.
When the routing code system was completely solved by Central Bureau in March 1943, in partnership with the American cryptanalysts at Arlington Hall, Japanese army movements became an open book to Pappy Clark.
Clark’s traffic analysis had earned a reputation as reliable and trustworthy, and was the bread and butter of the daily intelligence summaries produced at MacArthur’s headquarters by COIC (the Combined Operational Intelligence Centre, still staffed by Australians).
An American commander who was particularly impressed by Clark’s consistently useful product was Admiral William Frederick Halsey Jr., who had taken command of the South Pacific fleet in October 1942. Since then his forces had battled their way along the Solomon archipelago, and in February 1943, after months of fighting, had finally gained control of the central Solomon island of Guadalcanal. Admiral Halsey learned how Ultra advice had led to the successful landings on Guadalcanal and Tulagi prior to him taking command, and that on Tulagi, in particular, the marines had landed on the beach without a shot even having to be fired. Halsey’s opinion was that someone should get a medal for this.
Australians at Central Bureau didn’t get medals, but a decision was made to put Pappy Clark up for a prestigious award.6
In September, at the urging of General Akin, Mic Sandford nominated Clark to be awarded an MBE — a ‘Member of the British Empire’ — with Abe Sinkov, Roy Booth, and Brigadier John Rogers, director of intelligence, as co-signatories on the nomination. Sandford wrote,
Lieutenant (Temporary Captain) Clark served as an NCO and later as an officer with No. 4 W/T Section in Greece, Crete, Syria and Egypt, and with Central Bureau in Australia.
His work throughout has been of an outstanding character, particular in the initiative he has shown and his ability in devising ways and means of picking up special intelligence, of breaking down and of turning that information to good account.
A report was recently prepared by a senior officer in the RAAF in which he cites many instances in which Intelligence emanating from Central Bureau has been of major operational value even to the extent of guiding the course of our Naval Task Forces against the enemy.
The success of this work has been largely due to the exceptional ability of Lieutenant (Temporary Captain) Clark, who richly deserves the honour for which is recommended herewith.7
Clark was given the award on 1 January of the following year. A small notice appeared in the Sydney Morning Herald awards listings, stating, ‘Captain Clark served with an army wireless telegraphy section in Greece, Crete, Syria, Egypt, and in Australia. His work throughout has been of an outstanding character and of major operational value.’
It was the only acknowledgement that Pappy Clark would ever receive from the Australian government in his lifetime. Honours, such as promotions and medals, were almost unheard-of for Australians in wartime signals intelligence. Indeed, Mic Sandford, who had successfully nominated Pappy Clark for the honour, never received any Australian award or medal himself.8
23
Planes in daylight
Mic Sandford boasted to military intelligence that his gamble in acquiring Nave had paid off. In February 1943, he wrote to Lieutenant Colonel Robert Little, the assistant director of military intelligence based in Melbourne: ‘Domestic progress has been tremendous. Nave has been quite invaluable. We are now reading air operational traffic of the utmost importance.’1
And thanks to Nave, with so much useful data coming in, Sandford was making plans to send important decrypts directly to the highest levels of command:
Hitherto, however, while a certain amount of it has been used in the field, a good deal reaches us too late to be of any but historic value. Ryan (who is now a Lt. -Col) has arranged to institute two watches on this traffic in the Brisbane area and we are having a teletype line installed. This should enable our air headquarters to receive the information simultaneously with the Japanese.
On the whole, things look much brighter than they did a week ago.2
Eric Nave and Professor T.G. Room, having arrived at Central Bureau after being evicted from FRUMEL in October, were both enjoying success at breaking into army codes. Nave had found the solution to naval air-to-ground codes the previous year, and this had become virtually a production-line process.
But the wireless sections and units to the north were now picking up a different kind of air-to-ground traffic. As the Japanese army increasingly took control of the south-west Pacific operations, the Japanese army air force was more active.
Nave solved this new code with Room’s assistance.3 They had a head start because Nave had a copy of an old Japanese army air-to-ground codebook, discovered by Allied forces in Singapore before the city was captured. The codebook had made its way to Australia (possibly with Norman Webb). It was completely out of date, and none of the current collection of messages was using it, but nonetheless it gave them insights into how the code was structured.
The system used Kana, which was converted into digits and then had additives applied. There were variations from area to area, and the code changed frequently. The call signs were in Kana (or sometimes a combination of Kana and digits), and changed every ten days. Additive books were issued every month.4 With such rapid code changes, reading messages at a timely pace was going to be challenging. Fortunately, it turned out to be a lot easier in practice than the theory suggested.
Nave discovered that, most of the time, the Japanese army air signals officers didn’t bother with additives. If they had a message to send, they would use the codebook to convert it into the standard code, and then just transmit it, completely unenciphered. This was understandable. The officers using this code were mostly pilots and airstrip ground crew. They probably didn’t want to bother with additive encipherment in the middle of a bombing run, or when fighters were providing air cover for a ground battle. Messages needed to be sent quickly.
But because the codebooks themselves didn’t change very often, the message was only hard to crack if the extra work had been put in with the additive encipherment. The result was that Japanese army air-to-ground messages were usually read in full, as soon as they were intercepted, by the local wireless unit or ASIPS team.
For the Allied intercept sites, a message about enemy aircraft movements could reveal the location of those aircraft, almost as clearly as if viewing them overhead in daylight.
Nave’s encyclopaedic knowledge both of code-breaking and of Japanese communication systems gave Sandford another idea. He was planning to establish an intelligence school in Brisbane, capitalising on the concentration of expertise at Central Bureau in Nave and others. Sandford told Little that he hoped to have an ‘extensive training program’ up and running by the end of the year.
A lot of the army air-to-ground messages picked up by the wireless sections turned out to be weather reports. They tended to have a distinct pattern, making them instantly recognisable to experienced Kana operators and ASIPS personnel. Weather reports were of no interest to MacArthur’s intelligence analysts, but Eric Nave had a hunch about how they could be put to use.
These sporadic meteorological reports were almost certainly being sent by Japanese reconnaissance planes. He compiled a collection of them, got hold of data for the timing of recent air raids against Port Moresby and other Allied bases, and matched up each weather report with an air raid. As he suspected, the Japanese reports and the air raids matched up neatly: a few hours before every air raid, a weather report was broadcast. The lag time between the two was the flight time from a Japanese airfield to the target of the raid.
Once Nave made this connection, weather reports on air-to-ground channels took on a whole new meaning. They were effectively air-raid warnings.
Sinkov set Professor Room the task of breaking a code system named JN-16 that Japanese bases were using to send synoptic weather reports to each other. Room managed to turn this seemingly dull and unimportant backwater of the signals landscape into a valuable source of intelligence.
At the time, the only way to get reliable meteorological information was to either use a reconnaissance plane or a weather balloon. The resources of the air force were already stretched, and they could not invest much effort into weather reporting, even though such information was critical for the timing of tactical decisions. With Room’s solution, the air force was able to tap into the Japanese system, accessing the enemy’s own daily weather reports about conditions over enemy territory for free.5 Professor Room acquired a promising young code-breaker, Judy Roe, in early 1943 through AWAS. Judy’s father had told her that she ‘must do something to help the war effort’, then had her assigned to Central Bureau through his friendship with Mic Sandford. Roe, who was later commissioned as a captain, was the first female Australian code-breaker, and the only woman engaged in actual code-breaking — specifically, the study and attempted solution of enemy code and cipher systems. After the war, she became a professor of mathematics at Sydney University.6
The army intercept unit at Coomalie Creek, 51 Wireless Section, tracked the movement of 20 Japanese army bombers eastwards towards Timor over a three-day period. Geoffrey Ballard, who had served in signal intelligence in Greece, Crete, and Palestine, was the intelligence officer on location at the section’s Northern Territory base. On 28 February, he notified the north-west area headquarters of the air force that the bombers had reached Koepang, an airfield on the island of Timor, in preparation for an air raid on Darwin the next day. The air force immediately launched a bombing raid of their own on Koepang.7
Air Commodore F.T. Bredin, responsible for the air raid, commended the wireless section’s work, telling his commander, ‘The successful attack on Koepang aerodrome this morning when twelve aircraft were destroyed on the ground and nine others severely damaged (figured subject to further conformation) was made possible by the information supplied by No. 511 Radio Station over the past few days.’8
Radio signals across the Timor Sea came in clearly from the islands across the water. This was not the case for enemy bases far to the north across New Guinea’s rugged interior, where the towering Owen Stanley mountain range dampened the radio waves, like ripples on a lake hitting a sandbank.
On 29 January, the Japanese army launched an attack on the remote mountain town of Wau, held by Australian diggers — the A Company of the 2/6th Infantry Battalion under Captain W.H. Sherlock. The Australians quickly ran out of ammunition, and Captain Sherlock himself was killed on the first day of fighting as he attempted to charge through enemy lines to safety with only a bayonet for a weapon. Reinforcements and ammunition arrived for the Australians by plane at the Wau airstrip later that day.
A week later, with the battle still raging, eight United States P-39 fighters came in toward Wau to discover Japanese ‘Zero’ fighters and ‘Sally’ bombers strafing and bombing the airstrip and attacking five transport planes coming in to land. The P-39 squadron dived in and engaged with the Japanese attackers, shooting several down while the others departed, but not before the Japanese bombers had destroyed two Dakotas, damaged another, and blown up the Australian signals hut, killing everyone inside. (The ‘signals unit’ at Wau had no intelligence or interception capability; their job was simply to keep A Company in contact with the Australian army command at Port Moresby.)
There had been no warning of the Japanese air attack on Wau; the Owen Stanley mountains stood like silent sentinels, shielding distant Japanese bases from the prying eyes of the intercept units.
If they were too far away from the action, the units could not provide tactical intelligence. The only way to solve this was to raise more units and deploy them further north, closer to the enemy. In late February, a sigint planning group chaired by Mic Sandford recommended the creation of three new wireless units.9
24
Computing machines
Two United States army officers arrived at the Sydney waterfront in March 1943, asking the wharfies questions about shipments from California, and searching around through stacks of crates. They found what they were looking for — several boxes that had been sitting on the dock for several weeks — loaded them into a truck, and drove away. The two Americans were Major Larry Clark and Lieutenant Richard C. Frazier, and the boxes they had located contained the parts for electric calculating machines built by an American company called IBM. The machines were needed in Brisbane at a secret organisation called Central Bureau.
They transported the boxes to Brisbane and offloaded them into the garage behind the main house at 21 Henry Street, Ascot. There, the boxes were opened and their contents examined. There was some weather damage and some pieces were corroded, but nothing that looked too serious.1 The bigger problem was that the IBMs needed to be assembled, and there were no assembly instructions.
Inside the boxes were parts, neatly stored in envelopes with numbers written on them, but there was no parts list to tell them what the numbers meant. Luckily, Frazier had worked with IBMs in Washington. He recognised many of the parts, and could explain to the others what they did and where they fitted into the machines.
Another clue was a photo of an individual machine, as it would look completely assembled, on the outside of one of the boxes. With the photo on the box, Frazer’s recollections, and using their combined expertise, they were able to assemble the IBMs. Then they discovered that the Australian power grid runs at 240 volts, while the IBMs were designed for the American grid, at 120 volts. They needed step-up transformers, which they ordered to be custom-built in Melbourne and shipped to Brisbane, from where they were unloaded into the garage behind the building.2
An older IBM, a tabulating machine, was already operating in a commercial business in Brisbane. Upon Central Bureau learning of its existence, it had been commandeered, loaded onto a lorry, and installed at Nyrambla. To house it, the roughly built garage walls were insulated with two layers of caneite (compressed boards of sugar-cane fibre), and a large air-conditioning unit was installed.
There were now 12 new machines installed. When they were configured correctly for a code that had been broken and for which the codebook was known, the punch-card operators could set up cards with the numbers from an incoming message, and the machine would read the card, strip the additive, decode the groups, and print out a Katakana version of the text.3
A punch card, a thin piece of cardboard with holes punched in it, was the primary way of providing input. There were no electronic keyboards or touchpads; the computer could only respond to the patterns of holes in a punch card. As it was fed in, brushes would pass over the card, triggering switches if they dropped into a hole and thereby relaying the contents of the card. Levers triggered by the switches would flick the cards into piles according to the configuration of holes. The machines made a lot of noise.
The air conditioning was necessary, not for the machines themselves, but for the cards. If the cards got damp, the machines would not read them correctly. The cards had to stay cool and dry.
WAAAFS — personnel from the Women’s Auxiliary Australian Air Force — were put to work on it. Eve Scott, who had enlisted in the WAAAF, started out as a store clerk before being abruptly moved to Central Bureau and a secret world of ciphers and machines.
‘We were sent to 21 Henry Street, Ascot, Headquarters at Brisbane,’ she said. ‘We were only there a short time, and all of a sudden we were moved from there to Oriel Park, Ascot in the fire station. And when I got there I was so scared of these machines. I didn’t want to learn.’4
An American officer patiently helped her overcome her trepidation, and soon it was ‘head down, tail up’, and she was part of a workforce that processed up to one thousand Japanese radio messages a day.
Before the computers arrived, everything was done with paper and pencil.
Interceptors at the field stations transcribed radio messages by hand; the messages were transcribed into Kana by hand; and by hand they were sorted, decrypted, and translated. Abe Sinkov had seen a better way of doing this in Washington. There, his colleague and mentor William Friedman had pioneered the use of machines that could churn through computations many times faster than humans.
At the outset of the war Sinkov and Friedman, the founder of cryptanalysis in America, and the man responsible for breaking the Purple machine, had approached the founder president of IBM, Thomas J. Watson Sr., for help.5 Watson’s company, IBM (which stood for ‘International Business Machines’) manufactured business equipment such as typewriters, and collating machines of various kinds, and had a reputation for innovation and the commercial application of technological advances. It had grown from around 6000 employees and $19 million in revenue ten years earlier to a company with more than 18,000 employees and $86 million in revenue in 1942. This was due to innovations such as the first commercially successful electrical typewriter, the invention of an automated carriage return on typewriters, and the first automatic marking system using cards, known today as ‘fill in the bubble’ answer sheets.
Watson put all the resources available in the company at the army’s disposal, and chose Stephen W. Dunwall, a product developer at the IBM factory at Endicott, New York, to work with Friedman on developing and running IBM machines in the code-breaking effort.6 They brought in IBM punch-card machines, and adapted them to new uses. The machines could strip additives off code rapidly and in bulk, and look for repeating patterns in the resulting layer of code beneath the additive. Repetition might mean that the same message had been encoded twice with different additives. In this way, they could try thousands of combinations until they hit one that looked promising.
Now in Australia, Sinkov knew what he needed: he was up to speed on the use of such machines by the American cryptographers; Central Bureau could benefit from the power and efficiency of these machines; and IBM was producing the best model for the job, the NC4. It was a no-brainer. Sinkov ordered a shipment of IBMs to be transported to Brisbane from America.
In the meantime, Sinkov’s colleagues at Special Intelligence Service (SIS) were willing to run calculation jobs for Central Bureau on their IBMs in Washington. SIS was expanding rapidly, and had recently relocated to a building called Arlington Hall in a girl’s school that the US army had acquired for the war effort.
Sinkov sent them a batch of 2000 messages in a code known as the 666 code, which was a three-figure Kana code, with a larger sample of messages than most of the other tactical codes. It was being used by an unknown Japanese unit in New Guinea. Arlington Hall’s machine technicians tried what they called a ‘brute force’ attack. This basically involved getting the IBMs to churn through every possible encoding, looking for a meaningful message underneath. It did not work, because the messages did not have enough overlapping indicators.
This did not dampen Sinkov’s enthusiasm to acquire some IBMs for his own use at Central Bureau. After several months, for some reason, they still had not arrived. He sent an enquiry to Washington asking when they would be shipped, and was informed that the IBMs had been sent months before.
Doug Pyle, previously a Kana operator, was for a time an IBM mechanic after a dreary stint working as a clerk for Eric Nave. Having returned from the field due to ill health, he was stationed at a desk in the hallway at Nyrambla outside Nave’s office, poring over pages of code under a dim light. He complained to Eddie Kelson that it was ruining his eyesight, and added that he would like to work with the IBMs, as he was ‘good with machines’. Pyle was immediately appointed as IBM mechanic, under the tutelage of Van, an affable American with an endless store of ribald stories.7
The IBMs needed to be kept cool, and so large industrial-cooling systems were installed, but they were so noisy that the neighbours complained. The machines were relocated to nearby Ascot Fire Station.
The machines were used for simple tasks at first, like sorting words. A basic trick in code-breaking, if you can do it, is to find the high-frequency code words. This is a useful first step in finding out what those words mean. Sorting words meant creating punch cards for every word in every message — a laborious task. By the time the IBMs arrived at Nyrambla, Sinkov’s colleagues at Arlington Hall, under the direction of Solomon Kullbach, had devised more efficient methods that involved much less labour.
The tabulators could do other useful tasks. They could sort by call sign, for instance, so that all messages coming from a particular enemy unit could be grouped together. Or, alternatively, they could sort a large pile of messages by indicator values, such as all messages using the same set of additives to encipher the code words.
The IBMs, marketed and sold as sorting machines, were not computers in the modern sense. But the code-breakers, over the course of the war, found ways of making them do more and more complex tasks. Army engineers built ‘add-ons’, plug boards that could be installed onto the machine as a kind of program.8
25
Killing Yamamoto
On 13 April 1943, a Japanese naval message was picked up by the US navy intercept sites, and was received by FRUMEL at Monterey Flats as well as FRUPAC in Hawaii. The message, which had been sent to several recipients in the south-west Pacific area, was encrypted in the latest version of the Imperial Japanese Navy’s high-level code JN-25. This version had been classified as JN-25E14.1 The cryptanalysts had not yet broken this code, as the Japanese navy had only started using it in February. Even so, they recovered enough words to realise it was important.
This was the initial fragment, recorded in a US navy memo on 14 April:
C-IN-C COMBINED FLEET ON APRIL 18 WILL AS FOLLOWS X BALLALE ISLAND (RXZ) BLANKS
An analyst at FRUPAC wrote a comment underneath the translated fragment: ‘Probably a schedule of inspection by C-in-C Combined Fleet. Lacks additives but we continue work.’2
That was enough to get everyone’s attention. The commander-in-chief of the Japanese navy, Admiral Isoroku Yamamoto, considered to be a master military strategist, had been the architect of both the Pearl Harbor attack and the Battle of Midway. According to this message fragment, he was about to make a tour of bases in the south-west Pacific. They had not unravelled the new book of additives that had been used to encrypt the message. Nonetheless, they would ‘continue work’ on it.
US navy memo about the Yamamoto message, 1943. The memo states that the naval cryptanalysts knew it was important but could not break it.
But the Japanese army signals office at Rabaul had made a mistake. One of the addressees of the message was the garrison commander on Ballale, a tiny island to the east of New Guinea. As the commander of a small frontline unit, his staff did not have the latest JN-25 code and additive books, so the signallers had helpfully re-transmitted the message to him in a Japanese army air-to-ground code.3 This code had been broken by Eric Nave at Central Bureau earlier in the year, so that messages in it were completely readable by the Allies.
The re-transmitted message from Rabaul to Ballale was picked up by an Australian Kana operator, Keith Falconer, at six minutes past midnight at 1 Wireless Unit’s forward detachment in Port Moresby. Since it was encoded, Falconer paid no more attention to it than any other message he transcribed that night; it meant nothing to him, so it was bundled along with the rest of the messages from that shift and delivered to the headquarters of 1 Wireless Unit in its bunker near Townsville. The message was processed by Alan Marsland under the supervision of the Townsville ASIPS unit, who realised its significance.4
The message in full said,
From CINC SOUTHEAST AREA FLEET.
CINC COMBINED FLEET will visit (?) RXZ (Ballale), R__, and RXP (Buin Air Base) on 18 April as follows:
Will depart Rabaul at 0600 in medium attack plane [6 fighters escort] and arrive Ballale at 0800. He will arrive R__ at 0840 in a subchaser [Comdr. #1 Base Force provide]. Depart R__ at 09.45 in subchaser and arrive Ballale at 1030 (|?). [ _____ ] at 1100 depart Ballale in a medium attack plane and arrive RXP (Buin) at 1110. ___6 at 1400 departs RXP in a medium bomber and arrive Rabaul 1540 ___. In case of bad weather delay one day.
During this trip the CINC will look into existing conditions and make visits to the sick wards. [some blanks in this paragraph].
In case bad weather should interfere with this schedule it will be postponed one day.
The US naval code-breakers had to be told of this at once. The itinerary outlined in the message was set for four days’ time, on 18 April.
The officers in charge at 1 Wireless Unit, dealing as they did with tactical intelligence, had direct access to local commanders.5 The message was immediately sent to FRUMEL, the naval sigint unit at Monterey Flats in Melbourne, where Lieutenant Gil Richardson translated it and passed it to Hawaii. Rear Admiral Gil Richardson recalled the event in the NSA publication about signals in the Pacific, The Quiet Heroes of the Southwest Pacific Theater:
I also remember working on Yamamoto’s schedule, which led to the shootdown of his plane. The message that I worked on was in a Japanese Army code system. Although we were unfamiliar with Japanese Army codes, we got this message out in a hurry because it was a substitution system.6
Fabian, too, claimed in an interview with the NSA after the war that the Yamamoto message came in on army air-to-ground, and that United States’ navy code-breakers ‘had nothing to do with it’.
Luckily, the message had been intercepted in Port Moresby by Australia’s top interceptor, Keith ‘Zero’ Falconer. He was known as Zero because, back in the Kana code training course in Melbourne, Falconer would score zero — zero errors on the test — every single day. He had become exceptionally good at Kana code, and was considered to be the best Kana operator in 1 Wireless Unit.
The Yamamoto message was replete with specific details, so merely getting the gist of it would not have been enough. To gain the maximum benefit from it, it needed to be intercepted and transcribed by the very best — and, luckily, it was. Falconer was so good that when he was later needed at Darwin to take command of 3 Wireless Unit in a hurry, he was flown there despite Central Bureau being aware that he had malaria, and that sending him to Darwin was therefore a direct contravention of the law. He was later flown to Leyte in a hurry, too, bypassing numerous high-ranking officials stuck in a bottleneck on the island of Biak.
Falconer had taken down the message in unenciphered army air-to-ground code, which was easily translated. This meant that, in Hawaii, the cryptanalysts could fill in the many blanks of the message where the words were locked with unsolved additives. This allowed them to inform Nimitz of the JN-25 message and that they had completely decrypted and translated it. The clear implication was that they had decrypted the message themselves, when in fact there was an essential extra step involved: an Australian Kana operator in New Guinea had given them critical help.
The US navy quickly planned an operation, which they called ‘Operation Vengeance’.
According to the message, the first leg of Yamamoto’s trip would be from Rabaul to the island of Ballale. This part of the trip was within striking distance of the US airbase at Henderson Field, on the island of Guadalcanal.7 The message said that Yamamoto would be travelling in a ‘medium attack plane’ escorted by six fighters.
But the message said twice that the trip might be postponed by a day, depending on the weather. The implication to the forward Japanese bases was that they should wait for confirmation. This meant that the US navy had to wait, too. Were its planes to arrive a day early, when Yamamoto’s trip had been delayed, they would run the risk of being sighted by Japanese ships or reconnaissance planes, giving the secret away and wrecking the operation.
Just as the commander on Ballale waited for confirmation about suitable flying conditions, so did the Allies. They got it the next day. A follow-up message was sent in army air-to-ground, completely unenciphered. It was picked up by a Kana operator at 51 Wireless Section at Coomalie Creek, near Darwin, and sent immediately to the naval air-to-ground solution centre in Brisbane, where Major Nobby Clarke, head of the centre, forwarded it to FRUMEL in Melbourne.8
The Allied commanders now had confirmation that Yamamoto would be making the scheduled journey. Admiral Nimitz, the head of the Pacific Fleet, gave the order to put Operation Vengeance into motion.
In the early morning of 18 April, a group of 18 pilots at Henderson airfield on Guadalcanal in the Solomon Islands were briefed on their mission. They were told that Admiral Yamamoto would be on board a flight to the Pacific island of Ballale, but they would intercept his plane near Bougainville, an island off the east coast of New Guinea.
Ultra was still a closely guarded secret known by very few people. The pilots could not be told how the information had been obtained, so they were told that one of Australia’s Coastwatchers based in Rabaul had seen Yamamoto boarding a Betty bomber that morning, and had reported this important news via the Australians.
Having had their cursory briefing about the mission, the pilots took off. The squadron of American Lockheed P-38 Lightning twin-engine fighters flew north-west across the Pacific Ocean toward Bougainville. They flew low, close to the water, to avoid detection.
The planes had been fitted with extra-long-range fuel tanks to allow them to make the 400-mile round trip. Four of the planes were designated as the ‘Killer’ section, whose pilots were tasked with shooting Yamamoto’s bomber. The others were given the job of engaging the Zero fighters escorting the bomber, and were the ‘Cover’ section.
With the green, mountainous island now visible to their left, one of the pilots spotted their target, and announced over the radio, ‘Bogeys, eleven o clock high!’ They climbed rapidly as the enemy aircraft came into view, and saw not one but two Betty bombers being escorted by Zeros. One of the bombers was transporting Admiral Yamamoto, while the other was carrying another high-ranking naval commander, Admiral Ukagi. Despite this surprise development, they stuck to the plan. The squadron split into two teams, the Cover section engaging the Zeros while the Killer section pursued the bombers.
An American pilot hit Yamamoto’s Betty with machine-gun fire. The Betty spiralled down to the island, streaming black smoke behind it, and crashed into the jungle. Yamamoto was already dead from the bullets fired by one of the P-38s. Ukagi’s plane was also hit and crashed into the ocean, but he and two crew members survived.
Rex Mitchell, the Killer section leader, radioed, ‘Everybody, get your ass home.’ Low on fuel, the P-38s wheeled back for the return flight to Guadalcanal.
After their return to Henderson airfield, one of the pilots, Thomas Lanphier, claimed credit for the kill, exclaiming, ‘I got Yamamoto!’ Lanphier was lauded as a hero until several of the Killer section pilots voiced their belief that a different pilot, Rex Barber, was the one who had shot Yamamoto. The controversy about who actually shot down Yamamoto was to continue for decades. The US navy at one point tried to please everyone by giving both pilots credit, but this didn’t work because the two accounts of what had happened were so different that both could not have been true.9 The weight of evidence now suggests that it was Barber, not Lanphier, who shot down Yamamoto’s plane.
Yamamoto’s body was recovered by the Japanese army, but his death was kept secret until 21 May when his ashes arrived in Tokyo. The emperor declared a national day of mourning for Japan to honour its fallen hero. It was reported that Yamamoto had been found clutching his ceremonial sword, but this seems implausible given the machine-gun damage and the plane’s plummet to earth. In all likelihood, this was an embellishment for propaganda purposes, or his body had been rearranged by a dying companion.
Although numerous books and articles have quoted the United States president, Franklin D. Roosevelt, as having given the two-word command, ‘Get Yamamoto’, he never uttered those words, and was not involved in the decision. Roosevelt did not intervene much with operational army and navy decisions, and in any case was travelling and was not in Washington at the time. The navy secretary, Frank Knox, was informed, but left the decision to Admiral Nimitz, who made the call after discussing it with his intelligence officer, Ed Layton.10
The United States navy gave credit to the wrong decision-maker and the wrong pilot, although the record has been set straight on both of these. There has been another oversight in awarding credit, as we have seen, for the person responsible for the interception of the message that betrayed Yamamoto’s location in the first place.
After Yamamoto’s death, the Japanese navy investigated the incident, which they called the Navy kō incident.11 The investigation concluded that the Allies could not have learned of Yamamoto’s whereabouts from the message, because it had been broadcast in the most up-to-date, secure version of the JN-25 code. Rather, they concluded that the Allies had learned of Yamamoto’s whereabouts when the message was re-broadcast on local frequencies in low-level army codes.12
The struggle to decrypt the content of the original message was understandable, since new versions of JN-25 were multiplying and were getting successively more secure. According to the Australian historians Peter Donovan and John Mack, by this time the Japanese codes were more complex and were harder for the code-breakers to crack. They wrote, ‘Later in the JN-25 saga, the indicator systems became much more difficult. This was not an intractable problem in 1939–1943. However, the trend is clear enough.’13
Historical accounts of the Yamamoto shootdown have provided various explanations of how the JN-25 message was decrypted. Earlier accounts claimed that the message had been sent in an old JN-25 system, ‘JN-25D’. Or that the additive book, although new, had been discovered by United States marines, allowing FRUPAC to unlock the message. Or that ‘whiz bang computers’ were used. All these and other explanations of the US version of events can be found in well-known accounts in many books and articles — and all are refuted by FRUPAC’s own original archival documents. The simple explanation is that, by having been re-broadcast, the message gave the code-breakers a ‘crib’, a second version of the message in a code that had already been broken.
The translator in Hawaii who worked on the message was Alva B. Lasswell. When asked how the message had transformed overnight from an unreadable message encrypted with a poorly understood code system to a perfectly decrypted and translated message, he simply said, ‘I personally did the whole thing overnight.’14
For decades after the war, the intercepts at Port Moresby and Darwin were not mentioned by the United States navy.
There are numerous well-documented examples of credit-taking and political turf wars in the United States navy during the Second World War. The misdirection about the Yamamoto message was merely one of many instances. Station HYPO, situated so close to the Pacific Fleet Headquarters at Pearl Harbor, had easier access to the highest levels of US command than Fabian’s team in distant Melbourne.
One of Fabian’s code-breakers, Duane Whitlock, claimed after the war that FRUPAC (previously named Station HYPO or ‘H’), under the leadership of Joe Rochefort, had claimed an unjust amount of credit for code-breaking achievements, often at FRUMEL’s expense.
In an interview with the NSA as part of their oral history program, he said of Rochefort, ‘I can’t believe that he, single-handedly, with his little pill-pushing crew over at Station ‘H’, were the ones that pulled the Coral Sea and the Battle of Midway and the rest of the World War II chestnuts out of the fire. That isn’t the way it works.’15
The Australian role in Operation Vengeance, and Keith Falconer’s role, in particular, was obscured for decades. Wireless Unit veteran Jack Brown posed the following question in his memoir, Katakana Man: ‘If recognition can be awarded to these two P-38 pilots, why has no recognition been awarded to Keith R. Falconer, whose interception gave the United States the message regarding Yamamoto’s flight?’16
The US navy certainly deserved much of the credit for the operation. Sadly, they took all of it.
26
Noon positions
The young cryptanalysts arriving in dribs and drabs from America were clever, but they had never seen action. Despite being enlisted in the United States army, a rather un-military attitude was starting to colour their behaviour, influenced by the Australians they were working with. That undisciplined attitude might have been what prompted the gun drill.
In December 1942, Lieutenant Chester Ray of the United States army decided to hold gun-drill classes with the cryptanalysts. He organised an instructional lesson for some of the young American code-breakers on breaking down and reassembling the soldiers’ 45-calibre automatic pistols but, foolishly, held the lesson indoors, in an upstairs room at Nyrambla.
Lieutenant Ray’s class was interrupted by the entrance of John Bartlett, an American code-breaker who had come in to ask for leave. One of the guns discharged accidentally, and Bartlett was shot in the stomach. Lieutenant Ray and others carried Bartlett downstairs and called an ambulance, but the ambulance took too long to arrive, and, while they waited, Bartlett kept bleeding. When the paramedics eventually arrived, they drove him to Brisbane General Hospital, but it was too late. Bartlett died at the hospital.1
The gun that had discharged belonged to Joe Richard, one of Sinkov’s most promising code-breakers, and a friend of Bartlett’s. Richard, usually a warm and upbeat person, became consumed with grief and guilt. He decided to work extra hard to make up for Bartlett’s loss, and so he asked Sinkov for additional tasks. At that time, everyone was still working on the three-figure codes, but he was frustrated with them and sick of working on what seemed a hopeless cause.2
Richard had noticed that the traffic analysts working in Pappy Clark’s team had finished with a stack of Mainline army messages. Richard suggested to Sinkov that he could work on them as well, sorting them by code group, and then by message within each code group, in order to find the duplicates.
Sinkov was happy for him to do this, and sent him to Webb, who was in charge of all intercepted files. Richard collected a large stack of unread messages, and spent the following days and evenings sorting them.
Because the Japanese navy and army had so many codes, they used a number in the message to indicate what code the message was in. The number that identified the code was repeated in the message, so it was called the repeater. The cryptanalysts called a group of messages with the same repeater a repeater group.
These messages were not in groups of five numbers as in JN-25; they were in groups of four numbers. Pappy Clark’s team called them ‘four figure’ messages. In the hot summer Brisbane evenings, Richard worked away, sorting messages at a desk in an upstairs room, under a single hanging light and with a blackout curtain on the window. He sorted the messages according to the digit strings at the beginning, and, since the messages were all date-stamped, he picked a date, 19 December, and started working forward through them.
Richard starting sorting messages for hours each evening. One night, he was sorting a set of messages that were designated by the number ‘2468’, which identified the code that the message was encrypted in. This particular code system was new, and its workings were, so far, a mystery. The first message starting with 2468 had arrived from the intercept units only a couple of months before, but since then it had increased in frequency. It was originating from Japanese army bases, and so it was definitely an army Mainline code. The traffic analysts believed that 2468 stood for ‘water transport’ — in other words, for troopships that were being used to move the Imperial Japanese Army around the Pacific.
Somewhere early in the message, after the call signs, the number 2468 would appear, followed by a string of incomprehensible digits, all in groups of four. As Richard sorted them, he started to notice little things about the messages — things that seemed like patterns. For messages in the Water Transport Code, he sorted according to the third block of digits, because this block was always repeated at the end of the message — suggesting that it was important in some way. Richard recounted the discovery many years later:
By about the third night of sorting, I noticed that that first digit was nonrandom. In other words, when I sorted on that digit every day I got similar height piles. The zero one four and five piles would have several messages and the two seven eight and nine piles would only have one or two. I told Colonel Sinkov that the 2468 indicator was not behaving randomly. Later, some time in February 1943, Col. Sinkov gave me a three-page typed letter that I think was from WEC India. This letter pointed out — as I had discovered also — that the first digit in the repeated group in the 2468 code was nonrandom and it showed this behavior graphically.3
The code-breakers at the Wireless Experimental Centre (WEC) in New Delhi had come to a similar conclusion — namely, that the first digit was not random. Because any lack of randomness is a sign of a design feature, thereby a crack in the otherwise impermeable surface of a code, Sinkov knew that this discovery might break the shell. Deciding that the time was ripe for a concerted attack on the code, Sinkov circulated Richard’s findings to New Delhi and Arlington Hall, and suggested that all three centres assign more analysts to the code.
One night, soon after his conversation with Sinkov, Richard was sorting new incoming messages when he realised that the pattern had changed. Maybe there had been some change in a codebook or additive book. He couldn’t tell, but it was curious. Two weeks later, a message arrived from the US code-breakers in Washington about the code. They had investigated Richard’s hunch by processing hundreds of messages through their IBM computing machines to try to find a pattern; but, sadly, there was no pattern. The Arlington Hall code-breakers concluded that the digit was, in fact, random after all. Richard’s theory was wrong, they reported.
Richard was confused and disappointed. Later that day, he noticed another cryptanalyst, Harry Shiffman, putting red-pencil marks on stacks of 2468 messages. He asked Shiffman what he was doing. Shiffman replied that he was getting the messages ready for machine coding because Sinkov wanted to find out if there was a pattern in the supposedly random digit.
That was good. Even though Arlington Hall had found nothing, Sinkov was going to double-check Richard’s hunch himself, using Central Bureau’s own IBMs. Richard thumbed through the stack of 2468 messages that Shiffman was working through. There was a wide range of date stamps on these messages. Then he remembered the night when the pattern seemed to change.
If this mixture of messages from before or after the change was fed to the machine, there was no way the machine would find any pattern. They needed to be separated into two groups: the messages sent before the change, and the messages sent after it. Richard wondered if Arlington Hall had made the same mistake. He asked Zach Halpin, who was in charge of the machine section, to keep the messages in separate batches, depending on where he thought the changes were occurring. It was a bit of extra work for Halpin, but he agreed to help.
The date of that change turned out to be important.
That night, Richard went back over the messages around the time he thought he had noticed the change, trying to determine exactly when it had happened. He found the change, at the time that he had remembered. But he also noticed something else.
After the repeater 2468, there seemed to be more relationships between the first few number groups — which everyone knew were words disguised as numbers. In particular, there was a pattern between the first digit of the second group and the first digit of the group after that. Whatever it was, there were only ever three possibilities for the first digit of the next group along. For example, if the first group after 2468 started with a 1, the second group could only start with a 3, 4, or 9; if it started with a 2, the second group might could start with a 2, 4, or 6; and so on. This made no apparent sense, but it had to mean something.
He decided to make tables for the different time periods, showing the changing relationship between the first digit and other parts of the message, but got very tired while doing so. By the time he had finished writing out a table for the first time period, before the change, he needed to sleep.
The next day, Sinkov looked at the table of numbers that Richard had made the night before. It seemed to be a step forward to cracking the 2468 code. He held a meeting with other Central Bureau cryptanalysts to talk about what do with the table.
Abe Sinkov was an accomplished mathematician, and he had a pretty good idea about what these patterns meant. Whoever had designed the Japanese army’s 2468 code had made a critical mistake: they had created a redundancy of some kind. If one supposedly random number in the message was somehow linked to another number, Sinkov was hopeful that it would only be a matter of time before he figured out why. And when that happened, the whole code would break apart. He immediately notified the other code-breaking centres at Bletchley Park, Arlington Hall, and in India and Canada about the table that Richard had discovered, and what it meant.
Work ramped up on the 2468 code, at 21 Henry Street as well as at Arlington Hall, where a team was now giving it full attention. Messages were sorted for a new IBM run. More people were on the case now; it was no longer Richard’s private project. But, even so, he was still deeply engaged. Once the others had gone out to lunch, he went over to the ‘IBM sort’ stack of messages and thumbed through them.
Joe Richard had found a pattern in the numbers, but he did not have the advanced theoretical understanding of ciphers needed to make use of it. Sinkov told Richard to keep working on it and to write out more tables of numbers like the one he had shown them.
The Australian mathematician and code historian Peter Donovan believes that the weakness in the 2468 code was so subtle that it required advanced mathematical expertise such as Sinkov’s. Donovan also notes that the weakness in the indicators bore a similarity to the weakness in the Enigma machine’s indicators, which gave the British their first inroads into that cipher. Maybe Sinkov had recognised the situation from his conversations with the British code-breakers, helping him find a way into the Japanese Water Transport code.4
Sinkov alerted his colleagues at Arlington Hall to the discovery, where there was immediate interest. They realised that the digits were yielding the coordinates for additives in an additive book. They completed the tables, tested them on some messages, and found that they worked.
The code had been cleverly designed, but it had two weaknesses. The first was the fact that each additive book issued to Japanese army signals units comprised fewer than three hundred pages. This mean that, for any given message, the starting point in the additive book could have been on page 3, or maybe page 15, or perhaps page 250, but whatever the page number, it could never contain more than three digits. However, the Water Transport code used groups of four digits, not three. Japanese cryptographers had therefore needed to invent a rule for their army signals units to be able to send the starting-page number in a way that converted it to a four-digit string. One option might have been to just use trailing zeroes (for example page 0003, or page 0015, or page 0250), but that would have been a disastrous mistake in code design. Putting a zero at the same place in every single message would have been a gift to someone like Eric Nave or Abe Sinkov.
The system the Japanese army had decided to use was that the first three digits were the page number (for example, 250) and then the last one would be a filler digit, derived by adding those three digits. (For example, 250 would have the filler digit ‘7’, so the indicator group would be 2507.).
Unlike the rest of the message, the starting-page number couldn’t be encrypted using the additive book, because the recipient needed the page number to know where in the additive book to start. To get around this paradox, the page number (and the dummy fourth digit) were encrypted with their own special 10-by-10 table.
The group of ‘indicator’ digits was then disguised with additives, but even this wasn’t enough to hide the dependency between the last digit and the others from an expert code-breaker.
A well-designed code system will disguise this information because, even without having access to the book, a code-breaker could collect all the messages that started on the same page, and use that as an entry point. To guard against this, the indicator group itself was encrypted, by using a 10-by-10 conversion square. But again, this only required two digits — one for across, and one for down, so that, for example, ‘3, 2’ would mean third column, second row — when four were needed. The Japanese army’s solution was to turn the coordinates into a group of four by repeating each one (3, 2 becoming 3322, for example), and then encrypting them by using the first part of the message itself. This doubling turned out to be a critical flaw. With enough messages in his possession, Sinkov could discover the conversion square, and from that he could move to retrieving the indicators, and then to cracking the Water Transport codebook itself.5
Sinkov fired off a message about the table to all the other Allied code-breaking centres working on the problem. Arlington Hall sent a reply saying that they had made the same discovery at the same time.
Doug Pyle, a traffic analyst working on the front verandah, recalls that Richard burst onto the verandah, shouting, ‘We’ve done it! We’ve broken the water transport code!’6
Richard said many years after the war, ‘Historians say that Central Bureau and Arlington Hall broke the 2468 code in parallel. I think that more of the credit belongs to Central Bureau, because we alerted Arlington Hall that 2468 was a unique, separate system and that it should be sorted by time period.’7
That breakthrough came in April 1943. By June, Central Bureau had already translated some water transport messages into plain Japanese text.8 But even with the indicator system available to them, they still hit troubles in breaking the code. The Japanese operators would scramble the messages and re-order them randomly, with the code ‘6666’ to indicate the true start of the message. This starting point was not obvious, because the code word 6666 would be disguised with an additive.
But with the indicator system broken, these were relatively minor challenges, and within six months the code had been completely broken to the point that Central Bureau could decrypt and read any water transport message that arrived.9 It seemed as if they went from a few hundred messages a month to about 20,000 messages a month ‘almost overnight’. 10
Over the following year or so, three-quarters of a million messages sent in the water transport code were intercepted. About 5 per cent of these were convoy-related. Analysts scoured the translations for anything that might betray the location of an enemy convoy or other important information, such as the units the convoy was transporting, their destinations, or their points of departure. The most useful item in any message was the planned ‘noon position’, a standard for shipping movements.
Translations containing important intelligence were sent directly from Central Bureau to Akin, who in turn passed them directly to either MacArthur or General Sutherland. Orders were then issued to submarine commanders, with a warning prominently given: ‘For Your Eyes Only.’ Once the exact coordinates of the noon position had been given away, a submarine commander would simply make his way to that location and wait. The loss of Japanese lives and ships due to the reading of noon positions in water transport messages was immense.11
It was now much harder for the Imperial Japanese Army to relocate its divisions, send reinforcements, and reprovision existing units. The endless stream of noon positions relayed from Central Bureau to the American submarines took a huge toll on the significant strategic advantage previously enjoyed by the Imperial Japanese Army in the south-west Pacific.
27
Kaindi
In July 1943, a detachment from 55 Special Wireless Section was flown to Kaindi, an old gold mine in the Wau region of New Guinea.1 Vic Lederer recounted what it was like at the abandoned mine:
Kaindi, high up in the mountains, had a cool climate and it rained almost every day. The trees were covered with greenish-grey lichen and a smell of rotting vegetation was ever-present. When you went for a walk, or a run to get some exercise, you had to be careful not to fall into one of the disused mine shafts, concealed by grass, which honeycombed the area. A lot of gold had been found here and in the Wao-Bulolo area, and there was plenty of evidence of mining, such as abandoned dredges and rusty machinery.2
There was a track known as the Bulldog Track nearby. The Imperial Japanese Army, having no Japanese names for local geographic features, had apparently adopted the English name for the road, because the detachment’s operators would frequently hear radio messages referring to ‘BU RO DU’ — a Kana phonetic rendition of ‘bulldog’. But although they were close by, the enemy never found their campsite at the abandoned mountain mine.
On the first day of operations, they discovered three new enemy radio networks, all sending tactical three-figure Kana traffic, linking various Japanese army units and bases on the north coast of New Guinea. They soon found two more: one that linked Ambon to units in the Rama Valley, and another that was being used by barges between Rabaul and New Guinea.3
Security seemed to be lax on these circuits, and the Kaindi detachment frequently picked up unencoded, plain-language messages on them. In one case, they picked up the movement of a regiment from Wewak to Madang — an important piece of intelligence, as the regiment was later involved in the Lae-Salamua battle. The Kaindi intercept gave the Allies forewarning of the regiment’s location there.
At Wanigela, Fairfax Harbour, Port Moresby, and other intercept sites, the Kana operators worked long, gruelling hours without being told what their efforts had achieved. Geoff Ballard’s weekly briefing at Coomalie Creek was the exception. For most sections at most intercept sites, unless you were in the intelligence loop, you didn’t know.
There were no newspaper reports of their accomplishments, and no commendations from high-ranking commanders. There was only the 24-hour roster: three shifts of eight hours, around the clock, every day. It never changed and never stopped, week after week, or month after month.
They listened to the whistles and beeps of enemy wireless traffic, wrote them down in a never-ending stream of symbols in pencil and on paper, and gave those bits of paper to the intelligence staff.
What happened to those scrawls after they were carried in bundles into the intelligence tent was a mystery that, for the Kana operators, was never solved.
The intelligence staff worked in a similar information void, albeit one step along. They received transcripts, processed them, and discussed them with the local intelligence commanding officer. The transcripts would be sent to Central Bureau, after which they sometimes heard that their CO had passed actionable local intelligence along to local commanders; but, mostly, once the transcript passed from their hands, their job was over, and they did not learn of the fruits of their labour.
At Central Bureau — just as they did at FRUMEL — teams of clerks sorted messages by sender or by call sign, or transcribed Morse to Kana, or stripped additives, or converted numbers to Japanese words. They were all cogs in a huge machine.
Their contact with the outside world was through mail delivery, but these were infrequent, and the letters from family would have words and sections cut out of them by military censors.
The isolation, the tedium, and the heat all played on the mind.
Ronald Courtney, a Kana operator for 1 Wireless Unit in Port Moresby, became progressively sluggish and incoherent. Allan Norton reported this to the commanding officer, but was told in response that the man was simply being lazy.
Those in the unit who had worked with Courtney for months knew that there was something wrong with him. He was previously a neat and clean person, but now, along with his vague and dopey speech, his hygiene was deteriorating. However, despite their concerns, they could not identify anything that was sufficiently indicative of illness to convince the commanders to take action. Eventually, the man’s mind deteriorated to the point where he could no longer perform his duties, and he was pronounced unwell and ordered to a military hospital in Melbourne. He was flown to Brisbane and sent the rest of the way by train. He had to change trains in Sydney, where he wandered away, missed his connecting train to Melbourne, and was listed as AWOL for two days. When he was relocated, he was taken the rest of the way to Melbourne military hospital. He died a short time later.
There were occasionally other tragic incidents at various locations. A Kana operator at Townsville flew into a rage during a shift to such an extent that three of the other operators on duty had to hold him down until he could be removed, never to return. Another Kana operator shot himself.
The air force sent a British physician, Dr Edgar A. Pask, to investigate the health conditions of air force personnel in New Guinea and other places. Pask’s investigation included a visit to Nadzab. He reported that the conditions were harsh, and conducive to mental-health and physical-health problems, and poor performance.4
Jack Bleakley, a Kana operator at Nadzab, recalled the doctor’s visit:
He interviewed all of us, and his report to the RAAF and CB was that if wireless interceptors weren’t given leave almost immediately he wouldn’t be responsible for their mental health.
They started to give us leave then for those that wanted it.
And those that applied for longer leave — they were granted it but the most amazing thing was that the interceptors realised how important their job was, but I would say 90 percent of us decided to stay on.5
28
Joe Sherr’s final flight
By mid-year, the intelligence school that Sandford had proposed in February was up and running for wireless-unit personnel. Eric Nave oversaw an extensive training course in the interception and decryption of Japanese air-to-ground traffic. Other components of the course provided an introduction to the range of signals-intelligence activities undertaken by Central Bureau and the wireless units. There were also guest lectures from heads of the various sections at Central Bureau. The ‘I’ course took seven weeks to complete.1
Joe Sherr, the United States army’s assistant director in Central Bureau, was in close contact with the American special wireless units. The original company that had been trained by Howard Brown in Victoria had recently relocated from Townsville to their own camp and operations base at Stafford, four miles north of Brisbane. They established 15 grounded V-antennas for direction finding at their camp east of Trout Road, on a hill that the locals still call ‘Radar Hill’.2
The American signals units, expanding with a steady flow of arrivals from the States, were keen to benefit from the course, too.
When Pappy Clark came to give guest lectures on traffic analysis, Nave introduced him as Captain S.R.I. Clark (since his full name was ‘Stanley Robert Irving Clark’). This caused a misunderstanding with some of the Americans, who thought these were the initials of his role; one suggestion was that maybe his title was ‘Special Radio Intelligence Clark’.3
Earlier in the year, representatives of all the Allied cryptographic centres had met in Washington and formed an important new agreement, the BRUSA agreement, named after the two major powers at its heart: Britain and the USA. Despite the name, other nations were signatories as well, including Australia, Canada, New Zealand, and India.
India was, at that time, a British colony, and the operations there were integrated with the British intelligence network.
Like Central Bureau, the Wireless Experimental Centre outside Delhi was also in the business of intercepting Japanese radio communications and breaking Japanese codes. While Central Bureau provided intelligence on the south-west Pacific area, the Wireless Experimental Centre provided signals intelligence on East Asia, where Allied and Japanese forces were also engaged in war across China, Burma, and India.
The head of the centre, Colonel Peter Marr-Johnson, was keen to build a closer relationship with other Allied intelligence centres, so he organised a conference for September.
Mic Sandford had been the Australian army representative at the BRUSA conference, but he had other commitments, so he nominated Pappy Clark as his obvious replacement. Roy Booth agreed, and the two of them put their recommendation to General Akin.
Akin responded a few days later, sadly explaining that MacArthur would never let that happen, because Pappy Clark was merely a major and this was not a sufficiently high rank. Akin told them that Colonel Joe Sherr, on the other hand, had a high-enough rank, and he should go instead. It appeared that Sherr had gone behind their back.
In September, Joe Sherr flew to India bound for Delhi, in a journey that involved several flights. He landed at Calcutta, where he boarded a Dakota DC-3 to take him the final leg of the journey to Delhi. Tragically, the plane crashed on take-off, killing Sherr and the other three people aboard.4
Howard Brown (now Captain) was particularly affected by Sherr’s death. After having helped set up 1 Wireless Unit in Townsville, he had rejoined 126 Signal Radio Intelligence Company at Radio Hill outside Brisbane. His friendship with Sherr had stretched back for years, to when they were both living in the Philippines. Brown was working for a radio station in Manila in 1936 when Sherr personally encouraged him to apply for a commission. In 1941, the two of them had kept their signals unit, Station 6, operational even as Japanese air raids wreaked devastation around them across Luzon. They had both fled to Corregidor when it became clear that the situation was hopeless, and had both separately found their way to Australia in 1942. Sherr, who was seen as indispensable, was transported with MacArthur; Brown had a longer, and far more harrowing, journey.
When the two men reunited in Melbourne, Sherr gave him steady reassurance. Brown said later, ‘[Sherr] let me cry in his beer, and he told me that things were bad, but that everyone was doing the best they could.’
Joe Sherr, as assistant director of Central Bureau, had been the United States army’s representative. General Akin needed to replace him, so he promoted Abe Sinkov to the rank of colonel and appointed him as assistant director.
The contrast was stark. Colonel Sherr had been a seasoned career army officer who inspired confidence. Sinkov, a short, bespectacled mathematician, was not a career soldier, and promoting him to colonel did not make him any more soldierly.
For Howard Brown, in particular, having Abe Sinkov as a commanding officer did not sit well.
The United States’s Signal Radio Intelligence companies lobbied to change their reporting lines in Australia. Instead of reporting directly to Central Bureau, as they had when Joe Sherr was alive, the suggested new plan was for the units to report directly to the army Signal Corps. This would not affect intelligence gathering. They had deployed to New Guinea in August, and would keep doing wireless interception for Central Bureau, but administratively would report through regular army channels.
Sinkov didn’t see anything wrong with this. The units were moved out from under Sinkov’s command, and changed their name to Radio Squadrons Mobile. The impact of this change would not be felt for another year. The Radio Squadrons Mobile, over time, became less and less integrated with Central Bureau, until by the end of the war they were operating independently.
This really was an internal issue within the United States army, but the Australians still felt slighted: they had trained many of the Radio Squadrons Mobile, only to see them leave and form their own organisation. Pappy Clark, in particular, complained to his friend and American colleague Joe Richard about it again some decades after the war was over. Richard reported, ‘He was still indignant about it 40 years later!’5
However, considering the valuable training that Clark, Nave, Sandford, Inglis, Ballard, and others got from the British under similar circumstances, and with a similar outcome, it was probably reasonable to view it as fair play.
Some time after this change was instigated, Howard Brown was promoted to lieutenant colonel, and was placed in command of all United States army signals units in the south-west Pacific area until the end of the war.6
Rudy Fabian’s replacement, Captain E.S.L. ‘Sid’ Goodwin, arrived to relieve him of command of FRUMEL in September 1943. Captain Goodwin was a well-connected career cryptanalyst from Washington D.C. who had been involved in Japanese signals intelligence for years.7 In fact, Goodwin was the person who in 1936 had set up the first navy signals intelligence unit in the Philippines, the precursor to Station CAST and indeed to FRUMEL itself.
Fabian had caused plenty of headaches, particularly due to his refusal to work with the United States army, but since October 1942, when he had purged Nave and all the civilians, things had settled down into a workable arrangement for everyone. Fabian got on well with the Australian director of signals, Jack Newman, and Newman got on well with the Australians at Central Bureau. There were still behind-the-scenes communications being conducted between Fabian’s unit in Monterey Flats and Central Bureau in Brisbane, but he didn’t seem to get as upset about it.
Most importantly, both FRUMEL and Central Bureau were churning out top-quality intelligence, tag-teaming the Japanese naval and transport fleets as they tried to move their forces around the south-west Pacific.
Fabian’s problems started with the promotion in March of Captain Arthur McCollum as the intelligence officer for the United States navy’s Seventh Fleet. McCollum was an extrovert and a networker, and was hard at work building a power base for himself at Brisbane headquarters. He had already tried to curry favour with Central Bureau; learning of their frustration with FRUMEL, he approached Mic Sandford with a quiet offer to freely share naval intelligence with the Australian army component of Central Bureau, just as long as they did not share it with the Americans there.8
FRUMEL was outside McCollum’s command, sending its output directly to MacArthur’s head of intelligence, General Charles Willoughby, via COIC (the Combined Operations Intelligence Centre). This did not sit right with McCollum and he wanted to change it, but he discovered that the walls protecting FRUMEL had been built before he arrived and that Fabian had no intention of dismantling them.
Relieved by Goodwin, Fabian was transferred to Brisbane to work directly for McCollum, helping him build a naval intelligence centre. It quickly grew from 30 personnel to over 300.
Fabian couldn’t stand working for McCollum, whose approach was, in Fabian’s words, ‘I want this, and I want this, and I want this.’
He made use of his connections, going above McCollum’s head to Admiral James Fife. Fabian explained, ‘I went in to the Admiral and I said, “Look, this fellow’s got too many wants to suit my blood, and this is the way I propose to do it.” ’
Fife’s response was, ‘That’s all right, you do it your way.’
Fabian recalled, ‘McCollum hated my guts. We did arrive at a fairly equitable solution and we had a format for stuff that we put in a bag to send it up to Brisbane every day. And things went fairly smoothly despite his thoughts about me.’9
Things might have been going smoothly, but it couldn’t, and didn’t, last. Fabian soon got another transfer — this one, out of Australia. His new assignment was a sideways move to a far less exciting job. He was sent to Colombo to be the American liaison officer with the resident British code-breaking operation, the Far East Combined Bureau. Fabian’s job was to deliver American intelligence in a mailbag each day. He later described it as ‘boring as hell’.
‘It was run typically British. I can’t say anything more than that, really. It was lackadaisical. Everything stopped for tea, and all that sort of thing.’
‘And, furthermore, the senior liaison officer to the Eastern Fleet was sent home without relief, so I took his job, too. So I’d go to the staff meetings; it was really to have something to do.’10
And that was how the colourful, controversial, and unarguably brilliant code-breaker Rudy Fabian spent the rest of the war. He sat it out in a boring office job in Colombo.
Fabian, with a hide of leather, treated McCollum the way he treated everyone else: he did things how he wanted. Goodwin, a more easy-going person, had more trouble. The day he arrived in Melbourne he found a message waiting for him from Captain McCollum, telling him to report to Brisbane immediately. Goodwin rushed up to Brisbane, only to get a lecture that he should have reported to McCollum when he first arrived in Australia.11
In contrast to the tension with the naval intelligence office in Brisbane, Goodwin found an ally in his new colleague, Commander Jack Newman of the Royal Australian Navy. He said after the war, ‘All of us had excellent relations with Jack Newman. Working with the Australians was a great experience.’12
Around this time, the owner of the Monterey flats contacted the Royal Australian Navy, complaining that he was losing money, and asked for the building to be returned. No doubt, Fabian would have brushed aside the issue if he was still in charge, since the outcome of the war was more important than one person’s livelihood, but Goodwin was not so hard-nosed.
After a flurry of exchanges between Melbourne and Washington, the joint naval unit agreed to vacate the premises as soon as a suitable alternative was available.13 Construction began on a new facility at nearby Albert Park, and when it was completed in 1944 the entire operation moved there.
29
Cartwheels
With the Allies on the south coast of Papua New Guinea, and the Japanese on the north coast, both sides engaged in a protracted air war across the interior, attacking each other in tit-for-tat bombing raids, back and forth over the Owen Stanley mountains. United States army fighters and bombers that had been damaged in combat above the rugged interior or over the Bismarck Sea to the north would sometimes make it back to the Port Moresby airfields, only to crash on their return as they tried to land.
The steady toll of these accidents had an impact on the interceptors at Fordet — the 1 Wireless Unit ‘Forward Detachment’ near Port Moresby. Keith ‘Zero’ Falconer, an artist, drew pictures of the wrecks. Vic Lederer described a particularly devastating event when an aircraft failed to land on the runway, crashing into an encampment and killing numerous servicemen.
The wireless unit couldn’t stop the death of Allied air crews, but with their access to the Japanese air-to-ground radio traffic they could help in a significant way: the unit was efficient at detecting incoming air raids. The enemy aircraft would typically fly in radio silence on their way to a bombing raid, but even in this situation increased activity at their base would often give the raid away.
Air-raid alerts were sent to the local Port Moresby headquarters, but Taff Davis, an officer at Fordet, found an even faster way of alerting the nearby troops and air bases. Whenever one of his interceptors learned of an incoming raid, he walked outside with his rifle and fired three shots into the air. After the raid was over or the threat of a raid had passed, he walked out and fired a single shot.
All the nearby bases learned what Taff Davis’s three shots meant. They must have wondered how he knew — but nobody ever came and asked.
There were two notable times when the wireless unit thwarted large air raids. The first was on 12 March, when a large incoming attack on Port Moresby was detected. The local commanders were notified, and MacArthur’s air chief, General Kenney, sent fighters out to meet them, destroying one-third of the bombers.
In mid-August, the wireless unit picked up a series of radio messages about Japanese aircraft movements at Wewak on the north coast across the mountains. According to these naval air-to-ground intercepts, numerous bombers had arrived at Wewak airstrip and were preparing for major raids on Allied air bases. The information was passed to Kenney, who wasted no time in responding. He ordered an immediate air raid on Wewak. All available aircraft were organised for a bombing run. They took off in the middle of the night, arriving in the skies over Wewak at dawn the next day.
The Allied bombers arrived in the nick of time. The Japanese bombers were sitting on the airfields, wing to wing, about to depart on a massive bombing run: hundreds of aircraft were ready for take-off, their pilots and crew aboard, their engines running.1 Kenney’s planes plastered the airfields of Wewak with bombs before the Japanese bombers could get into the air. The Allied bombers also inflicted damage on the runways, and struck nearby fuel depots, sending plumes of black smoke into the sky.
They did as much damage as possible before circling back to the Allied airfields at Port Moresby and Buna. Then, learning that his surprise reverse air raid had been a success, Kenney decided to keep going. Realising that the airfields at Wewak were now in such disarray that they would likely to be vulnerable to more bombing runs, he sent waves of bombers across the mountains to Wewak and back for three days. By the time they were finished, 150 Japanese aircraft were destroyed, mostly as they sat defenceless on the ground.
Questioned about the raids after the war, Major-General Tanikawa of the Japanese Fourth Air Army at Wewak, said,
At the time of the air attacks on Wewak on 17th and 18th August our defences were not alert. We lost 100 planes including light bombers, fighters and reconnaissance planes. It was a decisive Allied victory. We were planning to regain the balance of air power and were making plans to bomb Port Moresby and other areas. A few days before our projected plan was to materialise, we were bombed at Wewak and our air power was severely crippled. Consequently our air power was rapidly diminishing and was unable to aid our ground forces effectively which, in the end, constituted one of our chief reasons for losing the war.2
The Wewak raid had been unplanned. It was a gift that had been handed to MacArthur and Kenney by Central Bureau’s 1 Wireless Unit. In fact, prior to Central Bureau informing them of the build-up, the Allies had no idea that Wewak was becoming a major enemy base. MacArthur, impressed by this unexpected windfall, ordered the formation of five new wireless units. Given the rigors of recruitment and Kana training for wireless units, there would be considerable time-lag before they were operational, but the process of signals-intelligence expansion had begun.
Before MacArthur’s order came in, two new units were already being formed in Townsville — named, naturally enough, ‘2 Wireless Unit’ and ‘3 Wireless Unit’. The more experienced operators of 1 Wireless Unit, which now had over 200 personnel, moved to Port Moresby and joined its forward detachment — Fordet — in September.
The Wewak raids happened only weeks after the US 6th Army division captured the Woodlark and Kiriwana islands to the east of New Guinea, with air cover from US and Australian air forces. Once the islands were captured they were turned into air bases to allow Kenney to launch air raids on Rabaul. This was the first step in MacArthur’s new strategy, Operation Cartwheel, a series of battles across the north coast of New Guinea and nearby islands with the goal of capturing Rabaul.
But as Cartwheel got going, and with increasingly useful intelligence coming in from both Central Bureau and FRUMEL, MacArthur changed his objective. An invasion of Rabaul would involve a series of large-scale battles with heavy casualties on both sides. Instead of capturing the Japanese stronghold, would it not be better to bypass it?
This suited MacArthur’s personal objective of recapturing the Philippines, where America’s great general had suffered a resounding defeat, and from where he had had to flee to Australia. And having been reported in newspapers across the globe as promising ‘I shall return’, MacArthur was resolved to do so.
The next move in Operation Cartwheel was to capture Lae, a town on the New Guinea north coast about 200 miles directly north of Port Moresby. On 3 September, paratroopers from the US army 103rd Infantry Regiment and the Australian army 2/4th Field Regiment dropped into Nadzab, a village inland from Lae. Meanwhile, the Australian 9th Division landed on beaches near Lae. After putting up a brief resistance, the Japanese withdrew from Lae. The town itself was captured five days later.
Soon after Lae was captured, a forward detachment from 1 Wireless Unit was moved there, with more of the unit following later in the year. Joining them was the 53 Wireless Section from the Australian Special Wireless Group, who flew by bomber from Port Moresby to Nadzab on Christmas Day, 1943.
Jack Brown, who wrote about his experiences as a Kana operator in his book Katakana Man, described the journey there:
Normally aircraft today would fly over the mountains, but the aircraft moving us were not capable of this and the only way to get there was to fly through the valleys and gaps. Going there was rather hairy because of the twisting and turning. Sometimes it looked as though the aircraft’s wingtips would hit the cliffs on the side. If the clouds had dropped there was no way you could get out; you would have to crash.3
Nearby, another coastal village named Finschhafen had also been captured and occupied by Allied forces. Fifty-three Wireless Section was relocated there, along with an ASIPS unit. According to Doug Pyle, one of the ASIPS personnel, they were initially camped with the US army air force, who were well provisioned and ‘provided us with everything we needed’.
The airstrip was made of punched metal laid over bleached coral. The US Kittyhawks made a zinging sound when the wheels made contact with the metal as they came into land. If it was a bad landing, instead of a zing there would be a woomp as the plane bounced up off the metal before coming back down to make contact again. Sometimes they would return with visibly damaged wings, or parts missing. Sometimes they would not come back at all.
Reception along the coast was too poor for the wireless section, so they moved to the top of a nearby escarpment, where they could pick up messages from far across the sea. They were now an all-Australian unit. They still had their American tents, cots, and other equipment, but no longer had the luxury of US army air force food — only bland and monotonous Australian army-issue meals. Their only connection to the outside world was a dirt road back down the hill to the US army air force base.
The units at Finschhafen and Nadzab were getting brilliant reception from enemy units. Nadzab, in particular, was a terrific location for interception. The reception of enemy radio signals was much clearer without the towering Owen Stanley Range in the way. Here, the interceptors could listen to signals on a range of frequencies, including low-power transmissions that had until now frustrated them so much. And Nadzab itself, up the valley from Lae, was higher above sea level, providing an unencumbered radio view of Rabaul.
While radio reception was of a high quality, the living conditions unfortunately did not match. The region hosted a range of tropical diseases, including typhoid, malaria, and a sometimes-fatal disease transmitted by grasshopper mites known as ‘scrub typhus’:
At Nadzab we had a good supply of running water, which we thought was great until we discovered that the stream flowed over two dead Japanese bodies just up the hill.4
Jack Bleakley, another Kana operator in Nadzab, who later wrote The Eavesdroppers, a book about Second World War signals operations, recalled that the camp was plagued by rats and bull ants. He recounted how the wireless units kept the bull ant problem at bay with an improvised solution. They split open bullets from machine-gun cartridges, planted them in the ants’ nests near the camp, and then ignited them.5
Rats and mice were a problem, too, according to Jack Brown:
The rats were big. They were possibly eating dead bodies and were everywhere. To catch them we used kerosene tins. We would fill the tins with water and sometimes with 100 octane petrol. We put a bottle over the top of the tin and then put the bait in the end of the bottle. Where the neck was leading into the middle of the tin we would put tropical spread to make it a bit slippery so the rats and mice could run along the top of the bottle and when they got to the neck to get the bait they would fall off into the liquid. We used to catch a lot this way, and some chaps produced traps from 44 gallon (200 litre) drums.6
When he was not in the signals tent taking Kana code, Brown would go exploring the nearby wilderness — a habit of his at every intercept site. The jungle was thick, hot, and wet from the constant rain, and littered with the bodies of dead soldiers. There were also exotic, beautiful butterflies. Brown recalled, ‘One in particular I remember was the Blue Emperor. It was a magnificent butterfly; its iridescent blue-black wings were so beautiful. I did manage one day to catch one, but I let it go after I had had a close look at it.’7
As was the practice at other intercept sites, intelligence officers from Central Bureau attached to the wireless units decrypted air-to-ground messages on the spot. One day, an interceptor at 53 Wireless Unit intercepted a message from Wewak about a high-level Japanese official who was travelling to Wewak and would arrive soon. They passed the information back to Port Moresby, where the decision was taken not to act on it, as they would not be able to organise an interception in time.
A US army air force contingent operating at the airstrip at Nadzab was camped nearby. Apparently, the secrecy provisions of the signals unit were more relaxed in remote jungle bases, because the interceptors told the local American pilots, two of whom took it upon themselves to take out the high-ranking officer.
The American pilots, Tom Lynch and Dick Bong, were in a competition with each other to see who could score the highest number of kills. Lynch arrived first, but found that in his haste he had forgotten to install a gunsight. It was up to Bong to take the shot, and he did, destroying the aircraft and strafing it with gunfire. They learned afterwards, again from Ultra decrypts, that they had killed not just one high-ranking official, but several more, as well as many staffers.
Lynch died in action the following week. Bong continued to fly throughout the war, and was eventually promoted to major.8
1944
30
The Sio box
Central Bureau got a lucky break in January, thanks to a chance discovery by an Australian soldier in northern New Guinea.
The coastal towns of Lae and Finschhafen had been captured by the Allies, and signals units were operating in both locations. From here, the objective was to take the entire peninsula — the Huon peninsula, a craggy network of ridges, valleys, and cliffs enveloped in jungle. The Australian 9th Division fought its way along the coast towards Sio on the northern side of the peninsula. Dengue fever was common — in the month of December, 291 soldiers were evacuated sick.1
They were worn down and weakened by a series of obstacles: enemy outposts in pill-boxes; minefields; snipers; the hostile terrain of cliffs and mountain rivers; and mosquitoes. But it was a war of attrition, and the Japanese defenders had also been weakened by the Australian advance.
Fifty miles to the west, the Americans landed at Saidor, in order to cut off the Japanese army’s line of retreat westwards. Heavy rain filled the mountain rivers, making them impassable and stopping the Americans’ advance. Consequently, Japanese scouts discovered that their retreat had not been cut off. When the Australians approached Sio, they met little resistance, as most of the Japanese defenders had fled into the mountains, taking only the bare essentials for the long hike to Wewak. Only a few rear units stayed behind to slow the Australians’ advance. The Australians climbed a nearby cliff with rope ladders and stood on a precipice overlooking the town below, which was almost deserted. Two days later, on 13 January, they entered Nambariwa unopposed, and secured Sio two days after that. The neighbouring towns had served as a north-eastern base for the Japanese army but were now abandoned, the dead left unburied.
The Australian historian David Dexter described the scene:
The Sio-Nambariwa area contained large dumps of all kinds which the enemy had made no systematic attempt to destroy. It was found that Nambariwa was the principal enemy supply base for the Finschhaven area. Both banks of the river had been used to provide barge off-loading points, barge hideouts, or dump areas. A huge fuel dump, a barge workshop, and an engineer stores dump were found. In the Goaling’s upper reaches there were large bivouac and hospital areas. The various arms of the river had been used for hiding and off-loading barges and there were large dumps of various kinds. All principal dumps had suffered from aerial bombardment and six sunken barges were visible in one deep river...
Mopping up continued for the next six days. Patrols from Sio Mission killed eight Japanese and counted 16 bodies on the 16th. Nine Japanese were killed on the 17th and a sergeant-major was taken prisoner. Patrols up the Goaling found further huge dumps of equipment and weapons. In the afternoon Captain Pursehouse, the devoted Angau officer2 whose local knowledge had been most valuable, was killed by a lone sniper south of Sio lagoon after interrogating about 350 natives collected by the 2/17th Battalion.3
Before the Australian 9th arrived, the Japanese defenders at Sio, the Japanese 20th Division, had a problem that needed solving: they were in possession of a set of Mainline army codebooks.
There was a chance the defenders would be either killed or captured by the Australians. Whatever happened, the codebooks could not be allowed to fall into enemy hands. For this reason, a unit from the 20th Division was told to burn their codebooks. As standard procedure, they were also ordered to cut off the books’ red covers and hand them in later to prove that the books had been destroyed.
But it was the monsoon season. There was a strong wind blowing, and it was raining. The codebooks themselves were wet, and in the monsoon weather would not burn. Even if it was not raining, the Japanese unit was reluctant to burn the books, because a fire would create a column of smoke and might give away their position to the advancing Australian army.
They stuffed the codebooks into a metal strongbox, then dropped the box into a deep, watery pit in a nearby bog before retreating into the jungle.
When the Australian 9th Division entered Sio, they proceeded carefully, wary of Japanese stragglers and snipers, and concerned that the Japanese might have left land mines before they departed. Sappers — Australian combat engineers — scanned the area with minesweepers brought in from Finschhafen.
Two Australian Field Security soldiers, Richard Henry ‘Dick’ Smith and John Burke, got word from a patrol of the remnants of a nearby enemy headquarters, and organised a patrol to investigate, where they found various papers as well as the box in the nearby pit.4 After the army established that the object was not a land mine, they pulled it out of the pit, opened it, and discovered the books.5 The books were taken back down to Finschhafen and from there flown to Brisbane, where they were delivered to the Allied Translator and Interpreter Section, known as ATIS.
ATIS was basically the Australian military’s team of Japanese translators. They were located at a prisoner-of-war camp on the outskirts of Brisbane, in a large villa named ‘Tighnabruaich’ in the suburb of Indooroopilly.6 They were located at the camp because one of their duties was to interrogate prisoners of war.
As both organisations dealt with classified military information, ATIS and Central Bureau had a close working relationship. The few Japanese translators in Australia were in high demand, and since Central Bureau had its own in-house translation unit, the two organisations helped each other and even exchanged translators, depending on workload. So when the still-sodden codebooks arrived, ATIS knew exactly who to call.
Two Central Bureau translators, Charles E. Girhard and Hugh Erskine, came over to inspect the red-bound books. They immediately realised what the books were, and excitedly took them back to Nyrambla.
Girhard said of it, ‘What a mess! The books had been buried near a stream bed in a steel trunk that looked like a footlocker. There was so much mildew on the material that each page had to be dried in the large commercial cooking ovens in our kitchen.’7
At Nyrambla, the books were carefully pulled apart and each page dried — some in the kitchen ovens, others in the open air. The Sio box, it turned out, contained the Mainline army code that had so eluded Central Bureau. One of the books was ‘Rikugun Angosho Number Four’, the latest version of the Mainline army codebook. But it was even better than that. In addition to the main codebook, there were two additive books, a book with instructions on how to use all the other books, and descriptions of the message forms for a range of Mainline army codes, specifically codes with the following traffic identifiers: 7870, 2345, 6666, 5555, and 7777.8 Any Japanese army message sent with one of those numbers at the start was now completely readable to Central Bureau.
Until recently, the code-breakers had thought that each of those numbers represented a different code system. But Arlington Hall, liaising closely with Central Bureau through Sinkov, had studied them carefully and had concluded a few months earlier that they were all variants of the same code system, each one using a different indicator book and message form. (The form refers to how the words and other message items are arranged and ordered.) The contents of Sio box were proof that Arlington Hall was right about the Mainline army codes. It was just one code with variations, and now Central Bureau had the key. They communicated everything they learned to the US cryptanalysts at Arlington Hall, and to the British at New Delhi and Colombo.
The Japanese army was making heavy use of the Mainline army codes across the south-west Pacific, including for transmitting high-level orders and memos. Now Central Bureau changed its work priorities to focus on solving and translating as many Mainline army messages as it could.
One of the first Mainline messages that was decrypted at Nyrambla was about the Sio codebooks themselves. It was a report from the lieutenant whose job it had been to destroy the books, reporting to his superiors that he had done so. Here is his message:
No. 1 SEMPAKUDAN Adjutant Message No. 501, Part 1. As per No. 1 SEMPADUKAN Restricted Message No. 1, we returned only the covers of Rikugun Angosho No. 4, Additive Book OTSU No. 4 and GO No. 2 Special Conversion Square NO. 11, and burned their contents. The personnel of the code section took along the codebooks that were left, and during the change in direction of the overland route from SHIO to HANSA… (a)
(a) To be continued.9
The lieutenant’s lie would keep the discovery of the books in the Sio box a secret. The lieutenant had reported to Tokyo that he had taken those books and ‘burned their contents’. No, he hadn’t. He and his unit had disobeyed a direct order, in order to save their own lives. Central Bureau knew it was a lie, but Tokyo did not. This meant that not only did the bureau have the complete set of army Mainline codes, but Tokyo did not suspect it, and was therefore unlikely to issue an emergency code change.
With the Imperial Japanese Army active across northern New Guinea, there was a massive flow of army Mainline messages coming in from wireless units and sections across northern Australia and New Guinea. The only limit to the number of such messages that Central Bureau could solve each day was how fast they could work.
Sinkov’s IBM machines now proved their worth. Installed a few blocks away at Ascot Fire Station, the machines were operated by Central Bureau’s ‘Machine Section’, the head of which was US army Major Zach Halpin. The beauty of the machines was that they could do the work of an entire team of clerks, and could do it faster. In this case, the entire codebook, Rikugan Angosho Number Four, was entered into the machine. A message was entered onto a punch card; the relevant additives for that message were found in the additive book, and they were entered on a separate punch card. Both cards were then fed into the computer.
The machines would print two decoded versions of the message, one with the codebook entry numbers, and one with a romaji translation — a version in the English alphabet. With the IBMs running around the clock, this was industrial-level decryption.
The entire process was streamlined.10 Incoming messages arriving at the nearby airfields were first delivered to Pappy Clark’s traffic-analysis section. There, the analysts recorded what they needed — such as the identities of the senders and receivers, and other information in the preambles — before sending the messages to Sinkov’s cryptanalysis section. Since the Mainline code was completely broken and solved, the only thing the cryptanalysts had to do was find the additives for each message in the additive books. They would write out the additives and attach them to each message before sending them to the machine section, where the messages and additives would be punched into punch cards, laid in stacks, and fed into the computers.11 Before long, they were processing about 2000 Mainline messages per day.12
With the vast collection of mainline messages that could now be decrypted, the translators could not process them all. ATIS translators moved across to Nyrambla to help, as part of which Fabian sent two of his translators in Melbourne, Forrest Biard and Thomas R. Mackie, to Central Bureau. Biard and Mackie had much-needed expertise in translating and understanding more complex military messages, because FRUMEL had been dealing with the large, versatile JN-25 code for some time.
The codes that had preoccupied Central Bureau until now — the air-to-ground codes, weather codes, and the Water Transport code — were all based on relatively limited vocabularies and smaller books. As valuable as the Water Transport code was from an intelligence perspective, it was mostly about shipping movements, which was a relatively narrow topic. The newly broken Mainline army code, on the other hand, covered a wide range of topics, and its messages contained a huge amount of Japanese military jargon. Biard and Mackie’s knowledge of this jargon was invaluable.
The Mainline solution also provided a rare opportunity for the code-breakers to assess their own techniques. They retrieved and looked back over some of the work they had done in the past on the Mainline codes, using the Sio books to find out where they had made mistakes and where they had gotten it right.13 It was not the case that they had made no headway on the Mainline codes — there had been some progress — but it was slow, and it would have been some time before they solved it. Going back over that work, they discovered that there were intermittent mistakes throughout the process, from interception through to the code-breaker’s desk. Interceptors sometimes misheard; sometimes they wrote down the wrong Kana symbol; sometimes they wrote it down correctly, but their handwriting led to a mistake later. There were errors in transcribing, in filing, in stripping additives … mistakes that compounded themselves and slowed the solution process.
To the north-west of Rabaul is a small group of islands known as the Admiralty Islands, the largest of which is Manus Island. MacArthur knew that these islands would make an excellent air base, and that occupying them would help isolate the Japanese stronghold at Rabaul. Japanese forces had occupied the Admiralties earlier in the war, but recent evidence suggested that they had abandoned them. Allied reconnaissance planes flew overhead without trouble, and without seeing any signs of activity. But Central Bureau learned otherwise.
Fifty-three Wireless Section at Nadzab, which was monitoring the Admiralty Islands radio traffic closely, intercepted messages that revealed the presence of 4000 Japanese troops hiding on the islands, patiently waiting in ambush for the Allies to attempt a landing. With MacArthur intending to personally accompany the landing force, such an ambush could have been disastrous.
Central Bureau discovered the existence of the hidden units through decrypted Mainline army messages sent between Manus Island and Rabaul. These messages revealed that the Japanese army was expecting the Allies to try to capture the Admiralties in the near future, and had correctly guessed that the initial landing would occur on Manus Island. The decrypted messages further revealed that the army had stationed most of its troops on Manus Island to launch a surprise attack once the Allies made landfall. Pappy Clark estimated, with traffic analysis, that the Japanese army’s strength in the Admiralties was between about 3000 and 5000 troops.
Knowing of the trap and how it had been set, MacArthur instead landed on the smaller neighbouring island of Los Negros on 29 February. He came ashore from one of the frigates, inspected the airstrip at the small township of Momote, announced that the invasion had been successful, and then departed.
Before long, the wireless units at Finschhafen and Nadzab intercepted Mainline army messages from the Japanese commander on Manus Island, reporting that the Allies had landed, and declaring his intention to fight to the death.
All-out close combat began. The battle for the Admiralty Islands was bloody, with a heavy loss of life on both sides. Even with the advantage of pre-battle intelligence from Central Bureau and with continuing updates about radio signals from the Japanese army, the Americans found the going tough. At one point, the Japanese defenders broke through the American line and came within 15 feet of the makeshift command headquarters that had been established near the landing.14
Eventually, the Americans strengthened their hold on the Admiralty Islands, and did not hurry to defeat the enemy on Manus Island. Over the coming days and weeks, using their new codebooks for the army Mainline code, the wireless units at Nadzab and Finschhafen were able to read messages about the diminishing forces on Manus. They were able, therefore, to monitor the action from afar — not based on reports from their own command or from the Americans, but from enemy radio signals. As was the case in battle after battle in the Second World War, the Japanese forces refused to surrender. Instead, they followed the instructions of their commander on Manus Island, Colonel Ezaki, who sent a radio message at the start of the battle, ‘Be resolute to sacrifice your life for the emperor … and commit suicide in case capture is imminent.’15
Five days later, the wireless units intercepted a message that originated in plain Japanese from the Admiralty Islands, not encoded at all, which said, ‘The time of the last hour is drawing nearer … we are striving for the fatherland.’
The next day, 6 March, they read a message from Colonel Ezaki that said, ‘The troops opposing the landing are reduced to 800, with ordnance reduced by 2/3rds and ammunition reduced by 7/8ths.’
A week later (on 15 March), they read of ‘about 120 men left on Los Negros, SW of Momote strip.’
A week after that (23 March), they learned that ‘entire admiralty garrison Manus and Los Negros, reduced to 750 … prepared to die to the last man … please drop by air more hand grenades and mg ammunition if situation of the enemy permits.’
There were now few enemy messages from the direction of the Admiralty Islands. On 30 March, the wireless units detected a radio message from Tokyo announcing, ‘We regretfully report our forces on the Admiralties have been overcome while fighting to the utmost.’16
There were soon no more enemy radio signals coming from the Admiralty Islands.
31
Pappy Clark goes to Washington
In March 1944, a conference was convened at Arlington Hall, the United States army’s cryptanalysis centre, for all Allied code-breaking agencies. This was a follow-up to a successful similar event the previous year that Mic Sandford had attended on behalf of Central Bureau, at which the landmark intelligence-sharing agreement, BRUSA (for Britain-USA) was crafted.1 This next meeting, named the ‘Second Joint Allied Conference’, was to be convened by Colonel W. Preston Corderman of the United States army. Abe Sinkov flew back to America to be part of the American delegation. Pappy Clark was chosen to represent Australia as a member of the Commonwealth delegation.2
Rather than going directly to the conference, Clark was instructed to first go to Britain, so that he could travel to America with the British, Canadian, and Indian representatives as a group. Clark’s trip to London consisted of a series of short flights: first to New Caledonia; then Hawaii; then San Francisco; Salt Lake City; Omaha; Washington; New York; Ottawa; and finally Prissic in Scotland.
In typical Pappy form, Clark’s trip was peppered with colourful incidents. In San Francisco, he was entertained by a lieutenant-colonel and his wife, and was served martinis in a bath; on the plane to New York, he gave Australian coins to a passenger as a present for the man’s children; and from Ottawa to Prissic, he flew in a Liberator at 30,000 feet in an arc across the Northern Atlantic Ocean. Since Liberators were not designed for such flying conditions, everyone on board wore oxygen masks. On a train in England he was mistaken for a doctor, because he was carrying all his top-secret papers in a doctor’s bag that was a gift from a friend, Dr Norman Glover, an osteopath. A woman was giving birth in another carriage on the train, and he was called to assist.
The leader of the Commonwealth delegation was Commander Edward Travis, the head of Bletchley Park. Before they left for America, the delegates went over the agenda for the forthcoming conference and agreed on a negotiating strategy.
The delegation caught the train to Glasgow, then boarded a ship at nearby Greenock for New York. In order to avoid U-boats, the German submarines, the ship sailed north as far as Greenland before returning south.
The conference took place on 13 March 1944. In attendance were luminaries of Second World War code-breaking, including the man who solved JN-25, John Tiltman; the ‘father’ of American code-breaking and Sinkov’s mentor, William Friedman; Solomon Kullback, one of the two men who solved ‘Red’, an early Japanese cipher machine; and others. The participants agreed on new intelligence-sharing arrangements, including a clear division of labour between the various centres. Results obtained by one centre were to be radioed to the others, and each centre would provide regular progress reports to the others.3
Building on the BRUSA agreement of 1943, the conference strengthened and streamlined intelligence-sharing between the Allied nations, and laid the groundwork for the later UKUSA intelligence agreement that continues to this day.
After the conference finished, the Commonwealth representatives left for New York, where they stayed overnight at the Rockefeller Centre. There, Jumbo Travis asked Pappy Clark if they could have a private chat. Travis told Clark, ‘You know, the Canadians have got a whole Special Wireless Group, highly trained operators. They’re sitting on their backsides in Canada. They’re too far from the action to intercept. Now I know that Colonel Marr Johnson is going to ask for these people to be sent to help in Burma. Why don’t you slip up to Ottawa and beat him to it?’4
Clark made a trip up to Ottawa the next day and called in on Lieutenant-Colonel Edward Drake, head of signals intelligence in Canada, who had also been at the conference. He suggested to Drake that perhaps his Canadian special wireless operators could come and work in Australia. Drake was keen on the idea. He had a social engagement the next day at Laurier House, the residence of the Canadian prime minister, and invited Clark to attend to discuss it further over lunch. And so the following day Pappy Clark found himself at lunch with, among others, the prime minster of Canada, Mackenzie King, and Colonel Drake. They made the decision on the spot to send their wireless unit to Australia.
Flushed with the success of his visit, and at having mixed with such salubrious company, Clark hurried back to New York to telegram Sandford with his news about the Canadian signals unit he had scored.
Sandford’s reply was brief: ‘Great news. I trust you kept General Lavarack in the picture.’
General John D. Lavarack was head of the Australian military mission in America, several ranks above Clark, and he should have been consulted — indeed asked — before Clark charged off to Canada and arranged for the deployment of their troops on Australian soil. There was no way around it: Clark had to tell General Lavarack what he had done. He visited the general, ‘with knees trembling and heart thumping’, told him the truth, and apologised.
General Lavarack sat in silence for a long time, letting Clark squirm. Then he said, ‘I had some experience with signals intelligence in Africa.’ This was an oblique reference that the General knew about Clark’s exploits — or at least of the achievements of Four Special Wireless Section — when they were on active duty in the Middle East in 1941. The general then said, ‘I would have approved of what you have done. Would you mind taking back some nylon stockings for Mrs Lavarack?’5
And so, in smuggling two pairs of stockings back to Australia, Clark made amends for his egregious breach of protocol.
32
Hollandia
Pappy Clark had some bad news for the Allied generals. With sections now deployed across Northern Australia and eastern New Guinea, Clark was able to piece together a detailed picture of Japanese activity across the south-west Pacific. And it did not look promising for Operation Cartwheel, the grand plan that had been cooked up by the Americans to take out the string of Japanese bases to the north.
Central Bureau produced daily and weekly reports for headquarters and, as part of the new intelligence-sharing agreement, to other Allied cryptographic centres. The daily ‘UBJ’ Ultra report contained estimates of enemy troop movements gained from traffic analysis as well as from important intercepted messages. These had an exclusive readership, including John Rogers, Australia’s director of military intelligence, and the main overseas intelligence centres. They also went to the Combined Operations Intelligence Centre (COIC), still staffed by Australians, but now basically support staff for MacArthur’s intelligence chief, General Willoughby.
Pappy Clark made it clear, day after day, that the Japanese army was reinforcing Wewak. It had obviously guessed — correctly — that Wewak was the next target, but wasn’t abandoning the New Guinea coastal town. The army was digging in. Clark gave a stream of reports about reinforcements flooding into Wewak. He estimated that there were now about 50,000 troops in or near the town.
Mainline army messages from Sinkov’s section backed him up, and added to the alarm. These messages revealed detailed and clever defensive preparations, including carefully choreographed ambushes for any troops that tried to invade the town. In short, the Japanese army was setting a trap.
An assault on Rabaul was also out of the question: Clark estimated that about 90,000 troops, including some tanks, were stationed there. MacArthur had already got the message that Rabaul would be a suicide mission, which was why he was getting ready to attack nearby Japanese-held towns on the New Guinea north coast: Wewak and Hansa Bay, in particular.
The reports coming out of Central Bureau, in particular from Captain Clark, made Willoughby feel worried. If Clark was right — and he was usually right — MacArthur’s plan was doomed. Even the best-case scenario was probably a long, hard jungle campaign like Guadalcanal.
After fretting over the situation, Willoughby went to MacArthur with a crazy idea. What about just forgetting those towns along the coast, leapfrogging right past them, and setting up a base further north and further west?
Willoughby even had a new target in mind: Hollandia.
Hollandia had been a Dutch colony before the Japanese invaded it in 1942. It was several hundred miles to the west, on the northernmost region of Irian Jaya. It had a good port for ships, Humboldt Bay, and would be a powerful base from which to conduct further operations in the north, toward the Philippines.
According to Clark’s team at Central Bureau, Hollandia was lightly defended, and was really just a staging post for planes heading further south and east. Aitape, a coastal town near Hollandia, was also lightly garrisoned, according to Central Bureau’s intelligence.
The main catch, and it was a big one, was that sending an invasion fleet that far, right past a series of enemy strongholds, could end in disaster. If the fleet was discovered, Japanese submarines and bombers would destroy it.
And there was another risk. If the Japanese figured out that Hollandia was the next target, they’d defend and reinforce it just as they were doing right then at Wewak. If that happened, there really would be no point in exposing the fleet to such danger, and the Allies might as well go into Wewak and slug it out there.
The other generals didn’t think much of Willoughby, who they saw as an introspective brooder, and one who frequently changed his mind. They laughed at his plan. General Sutherland, in particular, was scathing.
MacArthur didn’t laugh. He quite liked the idea, and, after some consideration, changed the entire strategy for Operation Cartwheel. Forget about Wewak; forget about those other nearby bases. The Americans would invade Hollandia, and the Australians would capture Aitape.
To keep the Japanese believing that Wewak was still the target, the Allied forces did everything they would normally have done before embarking on a large-scale attack. General Kenney sent reconnaissance planes over Wewak and the surrounding areas, and launched bombing raids against nearby defensive positions. Pamphlets were even dropped on Wewak, urging the troops there to surrender.
The invasion of Hollandia was dubbed ‘Operation Reckless’, which it was. The First Corps of the US 6th Army was selected for the operation, totalling about 50,000 soldiers and 30,000 support staff. A smaller force of United States and Australian troops, named ‘Operation Persecution’, was also organised to capture Aitape at the same time. A total of 217 Allied ships were involved to take the forces to their destinations.
The main invasion force departed from the Admiralty Islands on the afternoon of 21 April with MacArthur on board one of the vessels. Other ships involved also departed from Finschhafen and Goodenough Island on the same night. The fleet sailed north for 200 miles until sunset, then in the darkness turned westward.
Also on board the fleet, alongside 50,000 troops, tanks, a large fleet of ships, and aircraft cover, were journalists and news film-makers with movie cameras. MacArthur, a publicity hound, wanted to capture his victory on camera so that it would be displayed on newsreels back in America.
If they were discovered by enemy reconnaissance planes, they would be sunk. The signals units at Nadzab and Finschhafen were given the job of monitoring enemy communications, particularly air-to-ground channels, for that possibility. They provided a sort of ‘signals cover’.
Jack Bleakley, operating at Nadzab, recalled in his book The Eavesdroppers that he detected a Japanese reconnaissance plane scouting around near the Admiralties and finding the fleet. The Kana operators then tracked the three-seater Dinah aircraft as it tried to get back to its base to deliver a report on its discovery. The little plane made an incredible journey that they monitored through its radio transmissions: the pilot struck bad weather and had to make a detour; he was shot at by Allied ground forces; and he almost ran out of fuel. The plane stopped at Hollandia to refuel, and then flew the final leg to its home base, Noemfoor in Irian Jaya, where the pilot reported that he had sighted a large convoy of Allied ships.
Japanese command now knew about the fleet, and Central Bureau’s wireless units and sections immediately picked up communication about it. It quickly became clear, though, that the ruse was still working. The commanders believed the Allied convoy was heading to Hansa Bay, a Japanese stronghold near Wewak.
The operators at Nadzab picked up subsequent messages about the pilot of the reconnaissance plane, learning that he had received a promotion for his efforts to ‘First Class Private’. Jack Bleakley and his mates at the Nadzab base were outraged on his behalf. They knew from the day they monitored that pilot’s flight what an incredible, daring journey he had made, and they knew the importance of the message he had delivered, even if his commanders had misunderstood its meaning. Surely he deserved to be promoted to a higher rank than First Class Private? They couldn’t believe it. They asked each other, ‘What on earth does the poor fellow have to do to make corporal?’1
At dawn on 22 April, Australian and United States troops landed at Aitape. The defenders, who numbered less than 1000, saw that they were outnumbered, and fled into the jungle.
Simultaneously, the much larger force, Operation Reckless, arrived off the shore of Hollandia, shelling the town from a distance before making landfall. As soon as the shelling began, the airwaves at Port Moresby, Nadzab, and Finschhafen were filled with frantic messages.
The main landing point was at Humboldt Bay, where the United States army quickly established a beachhead and pressed forward. The Japanese defenders, taken completely by surprise, fled. However, the army met with resistance as it moved forward, and became quickly engaged in a ground battle.
With the beach secured, MacArthur made landfall with a camera crew to document his arrival. The subsequent newsreels shown in Allied cinemas captured MacArthur in profile travelling to the shore with a stirring, exciting musical backing track. ‘The general lands with his troops!’ the voiceover declared. ‘Bypassing some four Japanese divisions around Madang to the south-east, the united nations are in complete control of this vital base.’2
The Australians were bemused by MacArthur’s self-promotional use of journalists. Ballard, stationed in Port Moresby at the time, recalled, ‘The Americans brought a new technique to their landing operations, something that seemed revolutionary and bizarre. We used to say that the movie cameras went first, then the refrigerators, and finally, the combat troops!’3
The battle continued for four days as the Americans pressed toward the most important landmark in Hollandia, the airfield. After four days of fighting, it was secured, and the remaining defenders retreated into the jungle to try to make their way to bases further west.4
Now came the second phase of the landings, in which the Americans rapidly established workable infrastructure. It was the ‘doctrine of the bulldozer and the grader’, to use a phrase coined by Geoff Ballard, ‘the mechanical trencher, shovel, digger and roller. If there was a hill on the site and a plain was required, that would be provided — or vice versa.’5 These terraforming projects began almost as soon as a particular location was captured. Construction was already underway along the coast at Hollandia while fighting continued around the airfields only a few miles away.
The Nadzab interceptors tracked the battle entirely from the Japanese point of view, again through radio transmissions. The last message they heard from a Japanese transmitter at Hollandia was, ‘The enemy are one mile away.’
33
The Yoshimo Maru
Ever since the discovery of the Sio box, the code-breakers had been reading Mainline Japanese army messages as fast as they could. The fruits of that labour were now seen in the Allied occupation of Hollandia.
But by the time MacArthur’s forces landed at Hollandia, the party was already over. In early April, Central Bureau decrypted a high-level Japanese directive instructing all Japanese army units in the south-west Pacific to switch to a new cipher book of additives. Japanese commanders were of the opinion that the old codebook had been compromised and had fallen into enemy hands. They were right, of course.
On 8 April, all Japanese Mainline army traffic became unreadable to the Allies. Overnight, the translators went from having too much work to having none. The FRUMEL linguists returned to Melbourne, and the ATIS linguists to Indooroopilly.
They could not read the new messages coming in, but they still had a massive backlog of older messages, encrypted in the previous, readable version of the Mainline army code. The analysts worked their way through the backlog, looking for useful nuggets.
To the teams of analysts at Nyrambla and in the huts on Ascot Park, this sometimes seemed futile. Gordon Gibson, an intelligence officer who spent some time in the field at wireless units and some time at Central Bureau, Brisbane, recalled that one member he was working with took every message seriously, as if it had been sent the day before, and as if the outcome of the war depended on it. The man would make a breakthrough, read the long-gone location of a unit months before, and cry excitedly, ‘Oh yes! He’s given away his location now!’1
Old information was useful for enhancing their order of battle estimates and revising old intelligence. And occasionally it would reveal something really new and valuable. This happened when they found an intriguing series of messages about a Japanese cargo ship, the Yoshimo Maru, that was sunk near Aitape on 13 February of that year. The Yoshimo Maru was bombed by US warplanes, caught alight, and burned for 12 hours before sinking near the pier at Aitape.2
What interested the code-breakers was a later message reporting that a sailor had been reprimanded for breaching protocol by leaving a book of additives behind when the ship was abandoned. Despite this reprimand, the messages made it clear that the local commanders were sure that the codebook would have been destroyed in the blaze before the ship sank.
The analysts at Central Bureau had other ideas. Sure, it was possible that the Japanese commanders were right that the ship’s fire had destroyed the codebook. But it was also possible that the book hadn’t burned completely. Thick books don’t always burn all the way through. The outside will char, while pages in the middle can survive relatively unscathed. It was possible that if the book was still somewhere in the wreck, some of its pages might still be readable.
And if this additive book was still aboard the sunken shipwreck, it would be invaluable, as it was the latest issue. It was the missing link — the additive codebook whose absence was stymieing their efforts to read the most recent Mainline messages. They had to at least try.
They told General Akin about the Yoshimo Maru. Akin was naturally excited about the possibility of another codebook discovery, and with Aitape captured, they were able to go and look for it. He ordered a mission to investigate the submerged wreck.
A deep-sea diver was flown from Australia to Finschhafen, and from Finschhafen to Aitape. The diver was given very specific instructions about what to look for. To assist him, the map-making team at Central Bureau drew up a map showing the relevant landmarks and where they believed the Yoshimo Maru had sunk.3
The diver returned to the surface and reported that pretty much everything on the ship had been destroyed by fire. The only object he was able to retrieve was a tin box that had been jammed between two rungs of a ladder. He brought the box to the surface, and it was taken back to Finschhafen and inspected.
The box contained an additive book.
Akin arranged for three of the American cryptanalysts to fly to Finschhafen, inspect the find, and bring it back to Central Bureau. They were Charles E. Girhard, Hugh Erskine, and Sergeant Schokal, who was about to get married and, because of the trip, had to postpone his ceremony.4 At the base in Finschhafen, they discovered that the documents seemed to be, as they hoped, cryptographic. But the documents were in worse shape than the Sio codebook had been. The ship’s fire had left them ‘charred and practically illegible’.5 Nonetheless, they were taken back to Brisbane to see what could be done with them.
The remnants inside the box were handed over to a Mr Holmes, a member of Central Bureau’s nascent photographic section, to see what could be recovered.
Holmes had a background in chemistry. He had an idea about how to retrieve the text, so he ran some experiments first. Various methods were tried, including the use of ultra-violet light, in the course of which he discovered that an alcohol solution would make the writing appear briefly.
The pages were separated with a scalpel, and teams of two were organised. One person would place the page on a soft cotton batting, and then dab it with alcohol. When the ephemeral symbols appeared in the alcohol-soaked, blackened paper, the other person took a photograph of them.6
Everyone available helped out, including Sinkov. Using Holmes’s method, they were able to retrieve almost all of the document from the charred and seemingly illegible remains.7
It turned out to be a conversion square for the army Mainline code. This was not nearly as useful as the Sio box discovery, as it was neither the codebook itself nor the additive book used to encipher the code words, but it helped.
The conversion square was an extra layer of encipherment on top of code words and additives. In a standard code/additive system, like the Japanese army air-to-ground codes, a message was converted into code words, which all happened to be numbers. Then a string of other numbers, ‘additives’, was added to them, which the operator would select at random from a second book, which comprised a long list of numbers for that task.
The conversion square made things even more tricky for the code-breakers. Instead of adding two digits by normal rules, the operator ‘added’ them with a 10-by-10 conversion square that gave the answer. To add 1 and 2, he would look up row one, column two, and that might tell him the answer was 9. New conversion squares could be issued quickly, so even figuring out how the codes and additives were being combined became, for the code-breaker, an exhausting and frustrating pursuit of an ever-shifting target. But, at least with the latest book of conversion squares in their possession, one of the curtains of the Mainline code could be drawn back.
34
Biak
Allied code-breakers in the Pacific War had a spectacular success, and made a spectacular blunder, in April and May 1944. The success — discovering the Take convoy — came first.
Naval code-breakers decrypted messages throughout April about planning for a large-scale convoy to bring Japanese reinforcements to the south-west Pacific. Later decrypts and traffic revealed that the destination of the convoy, known as the Take convoy, was Western New Guinea and Indonesia.
Central Bureau’s intercept units began picking up a massive amount of radio traffic about the convoy as it moved south from Manila. The more they learned about it, the bigger it got. Pappy Clark’s traffic-analysis section identified 30 separate call signs associated with the various units staging from the Philippines, Truk, and other northern locations to the New Guinea area that were on board the convoy. Japanese naval air force planes were relocating to bases in the general vicinity of the Take convoy’s destination.
The convoy was a potential game-changer. When it arrived, the Japanese forces would be replenished, re-armed, and vastly increased in size. Stung by recent losses, and driven out of Hollandia, they were gearing up for a powerful counter-punch. As the convoy drew closer, it cast a shadow over the recent Allied gains at Hollandia and Aitape.
The naval code-breakers discovered the convoy, and traffic analysts at FRUMEL in Melbourne, FRUPAC at Pearl Harbor, and Central Bureau in Brisbane deciphered the convoy’s composition. And then, on 9 May, when it had almost arrived, the Water Transport code gave its position away.
Two noon positions for the convoy were broadcast in Water Transport code for 2 May and 9 May. The United States submarine USS Gurnard moved out into the Celebes Sea to intercept the convoy between those two points, and then waited. On 6 May, the convoy appeared on the horizon. The submarine glided into position and torpedoed the ships as they passed. The torpedoes found their targets, sinking three transports in the space of ten minutes.
The airwaves were suddenly full of frantic messages as escort vessels hurried to rescue the troops from the sunken ships, while other ships in the convoy hunted for the submarine. After the devastating attack, the Take convoy was diverted from its original destination to nearby bases.1
Don Laidlaw, a young Adelaide lawyer, was a Japanese translator at Central Bureau at the time. The translation section, working feverishly on decrypted Water Transport code and army Mainline code messages, was headed by Major Hugh Erskine of the United States army. Laidlaw knew nothing about the Take convoy, but he knew an important message when he saw one. On Saturday evening on 6 May, he was the only officer and therefore head of the translation branch for that shift.
One of the other translators on shift at the time, working on recent army air-ground intercepts, noticed a curious similarity between two messages: they both used a rather idiosyncratic kana spelling for the islands of Halmahera and Celebes, which suggested they had been sent by the same operator. On their own, each message was not overly remarkable; but if they were two parts of the same message, they were of the utmost importance.
Part one referred to an arrival at the harbour port at Manado, while part two referred to five Japanese ships and some landing barges.
Laidlaw contacted the United States duty officer at intelligence headquarters, telling him that he had urgent intelligence. It was, by now, 1.00 a.m., but the duty officer woke General Akin and relayed the message. He then called Laidlaw back and told him to get into the city immediately. There, at the headquarters in the Brisbane AMP building, Laidlaw met General Akin for the first time.
The general was dressed in pyjamas.
After Laidlaw showed him the messages and explained his hunch, Akin said, ‘Major, this may be the remnants of Bamboo No. 1 dispersing into landing barges. Ask Air Command to send a recce plane to Menado at first light.’2
Bamboo Number 1 was the Allied codename for the remnants of the Take Convoy. Laidlaw had no idea what he was talking about, but bit his tongue.
Biak was a little island off the north-west coast of New Guinea. Its airstrips, laid over coral, could be used as staging points into Irian Jaya or towards Indonesian islands. It was strategically important, which was why the Take convoy planned to reinforce Biak. But the destruction of the convoy meant that the reinforcements did not arrive; they had either drowned or had been diverted to other islands far to the west.
General MacArthur decided to seize the moment and snatch the island, which he could then use as a staging point in his advance towards the Philippines. His intelligence officer, General Willoughby, estimated there were only about 1000 Japanese soldiers on the island, making it a pushover. Willoughby’s estimate of Japanese strength was based on advice from Central Bureau’s traffic-analysis section.
However, the estimate was way out of date: there were about 11,000 troops on Biak. Despite the loss of the Take convoy, the Japanese army had moved reinforcements to the island. They had done so cleverly, by not broadcasting any signals about the movements on major channels.
There was a second Allied intelligence failure with Biak, but this one was not Central Bureau’s fault — it was one of geology. They had maps, and they had reconnaissance planes, but nobody in Allied intelligence realised that the entire island was made of coral, with a labyrinthine network of coral caves beneath the surface. The Japanese army commander on Biak, Colonel Kozume, had strategically placed defensive heavy artillery in these caves. The caves also made perfect bomb shelters.
In preparation for the landings, American bombers made a series of raids on the island. General Kenney was confident that such heavy bombardment would have killed most of the defenders. Instead, the Japanese soldiers simply hid safely in the caves.
When the United States 126th Regiment went ashore, they did so unopposed, because the Japanese defenders were waiting to launch an ambush that would inflict maximum damage. The pre-invasion bombing, although it had caused minimal damage to the Japanese army, had created a haze of smoke, making navigation difficult; as a result, the regiment landed in the wrong place, stepping ashore into a mangrove swamp. They moved westwards toward the airfields, in keeping with the plan, but as they stretched out in a long, thin trail along the beach, the Japanese opened fire on them with artillery and heavy mortar from the nearby cliffs, cutting the invaders off from their retreat. Suffering heavy losses, the American soldiers had to be rescued by ship.3
The ambush was a tactical victory for the Japanese army on Biak, but it turned out to be short-lived. Once the element of surprise had passed, MacArthur’s troops re-grouped and re-engaged, although MacArthur still did not realise the extent of the resistance they faced. On 3 June, a week after the initial landings, he announced that the Battle of Biak was won and that all that remained was a mopping-up operation. He was wrong. The fighting was still fierce, and it continued over several weeks. The Japanese defenders, under Colonel Kozume, inflicted heavy casualties in the early stages, but as the battle ground on over several weeks, the Japanese army suffered overwhelming losses. They continued to fight, even as the battle was clearly lost and the Allies, day by day, gained more and more control of the island. Ultimately, the US army suffered about 500 dead and about 2000 wounded in the Battle of Biak. A further 7000 were incapacitated by illness, including from typhus and a fever illness that was never identified.4 Of the 11,000 Japanese defenders, more than 6000 died, and 4000 were unaccounted for. Less than 1000 surrendered.
Sixty miles off the north-west coast of Dutch New Guinea, Biak offered clearer reception of radio signals in the Philippines and East Asia than any of the existing intercept locations. Forward detachments of 1 Wireless Unit and 53 Wireless Section, both currently at Nadzab, relocated to Biak in August. The Japanese troops on the island did not surrender, and there were still skirmishes across the island as the interceptors set up their operations camp. They were given a building to work in — one that was well suited for interception but which, with its corrugated-iron walls, heated up during the day to unbearable levels.
Some of their tent poles were lost in the move from Nadzab, and the tent canvas was frayed and smelt of mildew. They patched the tents as best they could, even while the American troops around them lived in nice new tents. Their uniforms, too, were old and worn; the RAAF seemed uninterested in replacing them. In practice, this did not matter much since, in keeping with Central Bureau’s disregard for uniforms, they took to going shirtless in the heat, but the lack of replacement uniforms for their old, tattered uniforms from Nadzab irked them. They came to suspect that the air force had forgotten that they existed.
The distinctions between the units blurred. The next level of the chain of command was several hundred miles away. The two Australian signals units on Biak increasingly saw themselves as a single unit, accountable only to Central Bureau. They were, after all, stuck there together on that forsaken island with its stench of death.
35
A view of Humboldt Bay
In Port Moresby, Geoffrey Ballard was called into headquarters for a meeting with General Akin in May 1944. Ballard, one of the original members of Central Bureau, had served in Crete under Jack Ryan and Mic Sandford, had spent a year at Coomalie Creek as the intelligence officer-in-charge of the signals unit there, had headed the air-to-ground solution at Central Bureau, and earlier in 1944 had replaced Reg Clements in Port Moresby as the liaison officer to Allied headquarters there (known as ‘Advanced Echelon’ or ADVON).
General Akin, as MacArthur’s chief signals officer, had ultimate oversight of Central Bureau. He told Ballard that ADVON was going to relocate to Hollandia. This was a directive that had come straight from MacArthur himself. What’s more, Akin told him, Central Bureau’s ADVON unit — including Ballard himself — would have to go, too.
Akin explained that it wasn’t just ADVON. This was part of a bigger plan: MacArthur was going to move his entire headquarters from Brisbane to Hollandia. After the meeting, Ballard returned to his office and relayed the news back to Brisbane.
Central Bureau’s three assistant directors in Brisbane — Mic Sandford, Roy Booth, and Abe Sinkov — were incredulous.1 The plan seemed crazy and unworkable. If Advanced Echelon moved to Hollandia, there would be at least 2000 miles between it and Brisbane, creating all kinds of new logistical problems. What about the timeliness of intelligence?
Ballard wrote later, ‘I found it difficult to convince them of the speed of the Allied advance. More importantly, it was difficult to interpret to them the role of GHQ, ‘MacArthur style’, which was markedly different from the normal concept of GHQ — that of a base headquarters, remaining in the ‘home’ country, as far from the front line as possible.’2
Not only was Central Bureau, ADVON to relocate, but many of the wireless units across the south-west Pacific area were to relocate there, too. New battlefronts were opening up on islands further north and further west — targets that were in reach from Hollandia. Getting signals intelligence about those places was now the priority.
As the new headquarters and a hub of radio interception, Hollandia was to be the new Brisbane.
Advanced Echelon departed Port Moresby in September. Ballard said it was ‘a messy affair’. Six planes were involved in the move of headquarters, with the intelligence staff in the last two planes. While they were in flight, storms arrived across the Owen Stanleys, and the planes carrying the intelligence staff diverted to the airfields at Nadzab. There, they camped for three days as the incoming storms raged across the north-east coast. The makeshift camp offered them poor protection against the weather, and as the ground turned to a cascade of rivulets, much of their luggage was washed away.3
When they finally arrived at Hollandia, they camped on the beach near Humboldt Bay. Given the poor state of the roads, the United States army was using the beach as a highway connecting the landing points with the encampments popping up nearby. Jeeps, trucks, and other military vehicles travelled up and down at all times of the day and night, making even a simple walk to the water a hazardous trip.
In Brisbane, the directors of Central Bureau faced a problem. MacArthur’s preference was that they relocate the whole show: the code-breakers, the translators, everything. But there were some obstacles to doing this. For one thing, the IBMs: there was no way those vast, bulky machines at the fire station were going to Hollandia, and if they stayed, Sinkov’s Mainline army solution team stayed. There was also a thorny problem regarding the hundreds of women now working at Central Bureau as WAAAF and AWAS personnel. They were legally not allowed to serve at Hollandia, because of the Australian government’s policy of limiting the total number of women who could serve overseas to 500. That wasn’t Central Bureau’s limit — it was the limit for the entire Australian military forces.
In the end, the traffic-analysis team moved to Hollandia, while the rest of the Brisbane operation stayed put. Since Pappy Clark was head of traffic analysis, he was promoted to major and given the job of running the bureau’s Advanced Echelon at Hollandia. Don Inglis was his second in command.4 Those who were going were started on a program of inoculations against various diseases they might encounter.
The traffic analysis department had a large number of women in its ranks who would have to stay behind, so Akin asked his contacts in the United States for help in meeting the shortfall in personnel. They agreed to send several hundred women from America, members of the Women’s Army Corps — known as WACS.
Sandford, meanwhile, continued to lobby the Australian government to change its mind about restricting women from overseas service.
Central Bureau was split in two, with some personnel ordered to move to Hollandia, while others stayed behind.
All the technical documents — solutions to codes and additive solutions, call signs, and traffic analysis — were duplicated. This had to be done by hand by the clerical staff. It was a huge amount of work, but the directors were worried that material might get lost in transit.5
The contingent departed from Brisbane on board the van Swoll, with a much larger company of Australian and United States troops. Near the Whitsunday Islands, a cyclone forced them to shelter in a harbour for a week before continuing. It took them four-and-a-half weeks to sail from there to Hollandia. There were 800 passengers on board, far more than the small vessel was built to carry, and most of them slept on the deck.
Included in the bulging passenger list was a small contingent of American soldiers who had deserted and been captured, and who were being escorted from a compound in northern Australia to Hollandia, presumably to face court martial. The captain in charge of escorting them disappeared one night, with nobody ever learning his fate. The Australians discussed among themselves the likelihood that he had been thrown overboard by his prisoners. They noted also that there were no lifeboats, but merely one large raft that would have only had room for about a quarter of those on board. Idle conversations centred around who would be allowed onto the raft in the event of the vessel sinking.
When they neared the eastern tip of New Guinea, the captain received a warning about an enemy submarine nearby, and hastily steered the ship into the shelter of a small island. Eventually, a US warship and some planes arrived to hunt for the submarine, but it seemed to have disappeared, so the captain got the go-ahead to continue. Upon disembarking at the wharf at Hollandia, the Central Bureau team climbed into jeeps and were driven to their new home, a camp nearby in the foothills of the Star Mountains. MacArthur’s camp was on the same road, about a mile closer to Hollandia.6
In the hills and mountains nearby, remnants of Hollandia’s one-time Japanese occupants were still in hiding. Many of them lived in the nearby sago swamp, from where they would sneak into the camps and steal food.7 Mice also over-ran the military bases at Hollandia, to the extent that an open jar of lollies, if left unattended for too long, would soon be investigated by a fearless rodent.
The air was often misty and damp, but up on the mountain spur where Central Bureau was encamped, a biting cold wind would blow, whistling through their work tents. They had all been issued with tropical uniforms, which were a poor protection against the wind, so they visited the American equipment depot, where the American servicemen issued them with warm woollen shirts and US army windcheater jackets. These army shirts quickly became standard attire, particularly since the Australian-issue uniforms were in short supply and, apart from being no use in the cold, had become in many cases old and worn.
On one occasion, a contingent of Australian officers, including the director of military intelligence, Brigadier John Rogers, visited the Central Bureau encampment at Hollandia. A high-ranking British officer was accompanying them.8 Rogers was appalled by the attire on display at the hastily assembled Central Bureau ‘parade’, including that worn by his son, Bill. Brigadier Rogers insisted that, henceforth, appropriate Australian military uniform should be worn at all times.9
New recruits, including a contingent of American WACS, arrived in large numbers and needed training by the now-seasoned Central Bureau traffic analysts. Cecil Corey, who arrived from America in late 1944, had been trained in code compilation, but when he arrived in Brisbane, Sinkov assigned him to traffic analysis, since Central Bureau did not undertake code compilation. He was sent to Hollandia, and trained under Mos Williams.
After the war, Corey admitted in an interview that ‘some of the American second lieutenants resented being trained by non-officers. Some of them also felt that US resources were winning the war and we did not need the Australians. Therefore many of the Americans goofed off in Hollandia.’10
Corey himself worked diligently on his traffic analysis, earning the trust of the Australians, and found himself playing a liaison role between the Americans and Australians.
Supply ships and troop ships from America bound for the conflict in the south-west Pacific area now sailed directly to Hollandia, bypassing Australia. On one of these transports, a contingent of WACS duly arrived at Humboldt Bay and reported for duty with Central Bureau. They had no knowledge of Kana or any other aspect of sigint, so Geoff Ballard gave them a crash course in Kana and other basic sigint tasks before flying down to Nadzab to deliver a two-week training course to the newly arrived intelligence staff of the United States Fifth Army air force. Halfway through the course, the air force staff were ordered to relocate to Owi Island at Biak immediately. Ballard travelled with them in a B-24 bomber to Biak, where he resumed the course in the new makeshift base on the coral island, before flying back to Hollandia.
36
The Battle of Morotai
General MacArthur’s moment in the April sun was brief. The spectacular success of the Hollandia invasion, with its 200 ships and 50,000 soldiers, was soon eclipsed by another invasion on an even bigger scale on the beaches of France.
The Axis nations controlled an unbroken arc of the western European coastline from France in the west through to Poland in the north-east, and Italy, the Balkans, and Greece in the south. Spain and Portugal were both officially neutral but had fascist governments, and so did not provide a military entryway to the continent. The Allies had to get a foothold in Europe itself if they were to have any chance of defeating the Axis, but Hitler’s generals had fortified the entire length of their territory on Europe’s west coast in a vast defence network known as ‘The Atlantic Wall’.
The only way through the wall was to launch an enormous attack somewhere along the coast and hope to gain control of some land. Hitler’s generals knew the invasion was coming; the only question was where. The Allied command chose Normandy, a coastal region in northern France quite close to England. On 6 June 1944, 150,000 troops, about half of whom were American, crossed the English Channel, landed on the beaches of Normandy, and established a beachhead.
The exact death toll from the D-Day landings is not known, but it is estimated that about 4400 Allied soldiers, and between 4000 and 9000 German soldiers, were killed.1 With the beachhead established, ground troops moved ashore and pushed further into France. Over the following weeks, the Allies suffered over 200,000 casualties in the Battle for Normandy. For that very high cost, the Allied generals got what they wanted: a piece of the European coast from which they could press towards the German capital, Berlin.
One of the notable things about the Normandy landings on D-Day was the role of signals intelligence and counter-signals intelligence. The British read German messages about their preparations for a possible invasion, using their solution of the Enigma machine. They knew where German defences were and which units were moving into which areas.
They also engaged in counter-sigint deception. The preparations for the invasion were kept a secret, while they pretended to prepare for an invasion further north. British and US personnel communicated by radio, acting as non-existent units.
MacArthur was already planning his next big move. Hollandia was never an end in itself; it was merely a staging post for MacArthur’s true goal, recapturing the Philippines. From his new headquarters overlooking Humboldt Bay, the general was already casting his eyes northward.
The southernmost point in the Philippines was still over 1000 miles to the north-west, well beyond the reach of the long-range bombers stationed on Biak. With such a huge expanse of water for the navy to travel across, air support was vital. One more staging point was needed, closer to the Philippines, for General Kenney to establish an air force presence. MacArthur was planning to invade Halmahera, one of the largest islands in the Indonesian archipelago, but General Akin, his signals chief, suggested that would be a mistake. Pappy Clark, who had also relocated to Hollandia with his traffic-analysis section, had provided Akin with a detailed breakdown of the composition of enemy forces on Halmahera. Clark’s analysts had identified Imperial Japanese Army units there, totalling about 40,000 soldiers. The units were poorly provisioned due to the sinking of provisions on board the Take convoy, but would no doubt fight to the death to protect the airfields on Halmahera. Akin advised MacArthur that capturing the island would likely be a long, hard fight with a high casualty count.2
But Willoughby proposed a solution. There was a smaller island, Morotai, to the north-east of Halmahera, which he had learned from Pappy Clark was lightly defended. What’s more, Morotai had airstrips. Willoughby conceded that Central Bureau’s traffic analysts had recently made the wrong call regarding Biak, but this time they were confident there were no nasty surprises in store for the American troops.
The plans were changed, and MacArthur ordered landings at Morotai. Although he took Akin’s advice, he had Biak fresh in his mind, so he sent a massive force of 50,000 US troops for the invasion on 15 September 1944. What the traffic analyst could not tell them was that the landing sites were a morass of mud flats and coral reefs, forcing the landing craft to stay far from shore while the soldiers waded chest-deep through seawater and mud. But Clark’s intelligence was right: once they were on land, the landing force overwhelmed the small number of defenders. With a beachhead established, MacArthur, who was on board the convoy, came ashore, also having to wade chest-deep through the swampy water.3
One of his staffers later recalled that he stood on the beach, stared out to sea northwards — the direction of the Philippines — and said, ‘They are waiting for me there. It has been a long time.’4
The US army established airstrips, and General Kenney moved aircraft there within weeks. The battle for Morotai was far from over, however. The Allies controlled the eastern part of the island, but much of the coastline and the mountainous, wild interior was still controlled by the Imperial Japanese Army, who immediately made plans to send reinforcements. Naturally, these plans were detected by Central Bureau. Clark’s traffic analysts were able to deduce the planned troop movements three weeks before an intercepted message confirmed it.
American PT boats cruised the strait between Morotai and Halmahera to try to enforce a blockade between the two islands to prevent any attempt by the enemy to bolster its numbers. The boat crews were often supplied with Ultra information about attempts to run the blockade, but even so the Japanese managed to get 50,000 troops on to the island. As a result, the Battle of Morotai dragged on for four months, with fighting only coming to an end in mid-January the following year.5
As hard as it was to secure Morotai, an invasion of Halmahera would have been orders of magnitude costlier in lives, both for Allied and Japanese forces. Ultra intelligence supplied by Central Bureau meant that the planned Battle for Halmahera never occurred. It was a case where Ultra saved lives on both sides of the war. In fact, the Japanese army’s attempts to engage the troops on Halmahera were thwarted twice. The first time was when the convoy taking them to New Guinea — the Take convoy — was attacked, forcing them to take refuge on Halmahera, where they were stranded far from the fighting; the second was when the Allies managed to detour around them via the island of Morotai.6
37
Akin’s secret unit
At the army headquarters overlooking Humboldt Bay, General Spencer Akin called in Major Geoff Ballard, the Central Bureau liaison officer, for a chat. Like everyone else at Hollandia, Ballard was frantically busy with the preparations for the upcoming invasion of Leyte, an island in the Philippines. This was to be an all-American affair.
MacArthur, when he made good on his promise, ‘I shall return’, didn’t want the fanfare of his arrival in the Philippines to be spoiled or complicated by the participation of international forces. It would be him returning, and he would be leading an American force. Australian warships were certainly going to be involved, but they would not be sending Australian soldiers to make landfall.
Akin explained his problem to Ballard. It was to do with the United States interception units, now known as Radio Squadrons Mobile. Their commander, Lieutenant Colonel Howard Brown, was not being exactly cooperative.
Since MacArthur wanted the invasion to be all-American, they needed to use their radio squadrons, but they were not under Akin’s control any more, and refused to take orders from Central Bureau or even Akin himself. Brown wasn’t even prepared to negotiate about what frequencies his units would monitor. The bottom line was that they weren’t prepared to work for Central Bureau, and so Akin needed to create a new unit that was. He told Ballard to make arrangements for a new wireless unit, Number 6, to accompany him on the invasion.
Akin told him, ‘Get them on an assault ship and make sure they are fully equipped for operations for the journey. A detachment from Biak will be the nucleus of the unit, and they’ll be back here tomorrow. Their OC will be Flight Lieutenant Foster.’1
Ballard immediately made arrangements to find space on one of the ships. The logistics section laughed at his request, telling him that there was no extra space anywhere, forcing him to invoke Akin himself as the source of the request.
As for this mysterious new unit, it was to be composed of the forgotten personnel of 1 Wireless Unit and 53 Wireless Section on Biak.
Twenty-three wireless personnel on Biak were given orders to leave Biak that day. These orders came from the Americans on the island, not through Australian command. They were instructed to hand over their Australian rifles, and were issued with United States army weapons, including Thompson submachine guns, carbines, .45 revolvers, and American jungle knives. They then boarded a US Dakota and were flown to Hollandia.2
On the aircraft’s approach, they could see hundreds of ships in Humboldt Bay. This fleet had been assembled to transport an invasion force from Hollandia to the Philippines. Once on the ground, the wireless personnel were driven to their campsite — not with the local Central Bureau wireless units, but with a unit of elite US Rangers. There they were provided new, spacious tents, already assembled with comfortable stretchers and mosquito nets, and they dined not on the meagre rations of Nadzab and Biak but on plentiful servings of fresh meat, vegetables, ice cream, and bread.
They reported to General Akin, who introduced them to senior officers from naval intelligence. The 24 of them — they had had an officer added — were told that they were the nucleus of a new group, 6 Wireless Unit. Their mission was to accompany the invasion convoy, and as the voyage proceeded, to scan the airwaves for any clues that the convoy had been discovered. They were, in effect, to provide signals cover for the convoy, just as they had done for the invasion of Hollandia. But this time, rather than doing so from a distant land base, they would be aboard one of the ships in the invasion fleet.
The convoy included 470 ships, more than twice the number used in the invasion of Hollandia earlier that year — although not as many as the nearly 7000 vessels used in D-Day at Normandy.3 Based in part on Ultra information from Central Bureau, MacArthur intended to bypass the southernmost islands of the Philippines and instead invade Leyte.
Leyte sat right in the middle of the archipelago between the major southern island of Mindoro and the Philippines’ most populous and most heavily defended island, Luzon — the home of the capital, Manila, and MacArthur’s previous base at Corregidor, from where he had escaped three years earlier.
The Americans had lost the Philippines to the Japanese, and they wanted it back. For this reason, the landings were going to be ‘American-only’, but with support from Australian ships such as the heavy cruiser HMAS Australia. But in the supposedly all-American landing party were the 24 Australians — hand-picked, experienced members of 6 Wireless Unit, chosen by General Akin himself. Six of them were on his own communications craft, and they sailed with the invasion force.4 Their job was to monitor Japanese air-to-air and air-to-ground frequencies, so that the fleet would not be ambushed by enemy attack. The invasion force left Hollandia on 13 October 1944 for the eight-day voyage by sea.
The Kana operators were split into three groups, covering all frequencies that were likely to indicate that the fleet had been discovered.5 Two days before they arrived at Leyte, they intercepted a signal from an aircraft frantically sending messages about the massive fleet off the east coast of the island, However, no defensive air raid resulted.
Another ship in the invasion force, the Wasatch, was carrying a signals team from FRUMEL, which had the specific task of protecting the vessel they were travelling on.6
While 1 Wireless Unit had the run of their vessel and dined with General Kenney himself, the first FRUMEL team had a harder time of it. From the outset, things went wrong. They were flown to Hollandia to join the Princeton before she set sail for Leyte, but arrived too late. The ship had already departed, so they were assigned to the Wasatch instead.
The Wasatch was dominated by its US army-communications functions, so the wireless unit had to work in with the standard operation of the ship, and even had trouble getting access to antennas. They would be instructed when they could connect, and when they had to disconnect.
On 20 October, the American fleet arrived at Leyte, pouring from the transport ships onto the east coast of the island. The Kana operators were not combat troops, and Kenney did not want them on the front line, so they were ordered to stay aboard on the first night, out of harm’s way from any hand-to-hand fighting on the land. They continued their shifts throughout the night, monitoring Japanese radio frequencies, and picking up many messages about Japanese aircraft within bombing range of Leyte.7
At eight o’clock the following morning, on 21 October, the Kana operators witnessed a new kind of warfare: a kamikaze attack. One of the Kana teams was working on a ship near the HMAS Australia when a Japanese Aichi D3A ‘Val’ dive bomber appeared, flying directly toward the ship. The gunners fired at it, but failed to bring it down, and the bomber crashed into the Australia.8 Alfred Bobin, one of the Central Bureau Kana operators in the fleet at the time, recounted, ‘When the plane hit it burst into pieces and was like a fire ball as it scattered along one side of the ship.’9
Thirty Australian sailors were killed by the dive bomber, and 60 more were wounded. The attack caused confusion and panic among Allied ranks. At noon, it was decided that the fleet was not as safe as thought, and that the Kana operators should go ashore after all.10
They set up a listening base about three miles south of the town of Tacloban, camping out in the open, with their radio receivers in the back of trucks.
Four days later, a typhoon hit Leyte, flattening their campsite and destroying their intelligence set-up. Much of the direction-finding equipment was lost.11 The radio receivers, being on the back of the trucks, were spared, but flooding now cut them off from the rest of the invasion force, and they had to hike back through deep water to retrieve their supplies.12
MacArthur waded ashore through the waves of the beach, a cameraman in front of him, filming his arrival.13 He travelled immediately to Tacloban and declared the reinstitution of civil government, and delivered a speech that was recorded and broadcast.
MacArthur declared, ‘People of the Philippines, I have returned. By the grace of almighty God, our forces stand again on Philippine soil. Soil consecrated in the blood of our two peoples. We have come dedicated and committed to the task destroying every vestige of enemy control over your people. The hour of your redemption is here.’
As Jack Brown of 6 Wireless Unit noted, this triumphant announcement was somewhat premature, since most of the Philippines — including the capital, Manila — was still under Japanese control.
While the US army fought a land battle against the Japanese army on Leyte, American and Australian warships became engaged with Japanese ships in a naval battle in Leyte Gulf.
On 24 October, a single Japanese Judy bomber strafed the deck of the Princeton and dropped a bomb on it, which landed in the centre of the deck. The bomb ignited a petrol fire that eventually spread throughout the vessel, sinking it. Explosions rent the Princeton as it went down, damaging the nearby ships that had come to assist, and killing an additional 200 sailors.14
It had been purely bad luck — a plane arriving late — that had resulted in the FRUMEL intercept team being reassigned from the Princeton to another ship. That bad luck now appeared in hindsight to have been good luck.
Not long after the invasion, more Kana operators from 1 Wireless Unit were also posted to Leyte. They set up a listening station nearby at Tolosa, a small town about 40 miles south of Tacloban.
Now that the wireless units were established, their work got back to normal: three shifts of Kana operators worked around the clock, monitoring the movements of Japanese aircraft, warships, and troops, and passing the information back to Central Bureau and on to local commanders.15
In November, a United States army unit discovered Japanese documents, left behind by Japanese army forces, which were provided to Pappy Clark at the new Central Bureau headquarters on Leyte. The documents turned out to be air-to-ground additive sheets. This code had been broken some time before by Eric Nave and Thomas Room in Brisbane, but additive-sheet discoveries were always welcome.
The Japanese army issued new additive sheets quite frequently, and each time a new one came into use the code-breakers had to solve it all over again. The newly found sheets were copied and issued to wireless field units, who immediately put it to use decrypting traffic.
The win was short-lived, however. The Japanese army air force must have suspected their codes were compromised, because they issued a special order to all units to begin using the next month’s additive sheet immediately, and to burn all copies of the November sheet.16
During the Leyte landings and the Battle of Leyte Gulf, three aircraft had crashed into American and Australian warships, including the first suicidal attack on the Australia, causing massive damage and loss of life. Disturbingly, soon after these events, the interceptors heard a broadcast from Tokyo praising the pilots of the crashed planes, naming each in turn as if they were heroes. They were being glorified.
A new word started soon appearing in Japanese radio messages: kamikaze.17 It was a reference to a Japanese legend, in which Japan was under attack from invaders but a strong wind — the kamikaze — came and blew the invaders away. The metaphor was clear.
Using a kamikaze pilot was a new tactic. A smaller or older plane, of limited military value, would be loaded up with explosives, turning it into a flying bomb. An inexperienced pilot would take control, and would fly it into an American ship or land base on a suicide mission. The reward was honour, and the chance to defend the homeland.
The Japanese historian Shunsuke Tsurumi has called this fervour for death in battle ‘the mass mania for glorious self-destruction’,18 noting that there were precursors during the progression of the war. When the island of Saipan was invaded by the Allies in mid-1944, most of the 25,000 Japanese there committed suicide. And as the Allied soldiers had discovered, taking control of small, relatively insignificant outposts such as Biak or Manus Island would end up as a gruelling, drawn-out fight against defenders who preferred to fight rather than surrender, even in the face of almost certain defeat.
An American journalist on Leyte discovered 6 Wireless Unit, referring to it in an American newspaper report. This was how the Royal Australian Air Force commanders learned of the existence of 6 Wireless Unit and its role in the invasion. They were furious.
They contacted Wing Commander H. Roy Booth, the air force’s representative at Central Bureau, for an explanation. To his great embarrassment, Booth had to admit that he knew as much as they did, but promised to find out what had happened. He asked the other two assistant directors what the story was. Mic Sandford was similarly in the dark, but Abe Sinkov had been informed about the secret plan by General Akin in October. The wireless units were outside Sinkov’s jurisdiction, and he had not thought to tell the others.
Booth, still furious, wanted to get to the bottom of it. He ordered Major W.E. ‘Nobby’ Clarke, head of the air-to-ground solution, to investigate what had happened and report back.
Clarke’s report gave Booth and Sandford little joy. General Akin himself was responsible for the mess. But if heads had to roll, there was somebody else to blame: Geoffrey Ballard, the Central Bureau representative at Hollandia headquarters. Apparently, Ballard was in it up to his neck. Booth and Sandford were itching to give Ballard a verbal lashing next time they saw him.
That opportunity came quite soon. Ballard and Sandford engaged in a dispute via a series of messages between Leyte and Brisbane. Some of Ballard’s messages were very long, explaining how miserable he was and insisting that he be given leave. He got it. Captain Neil Evans arrived on Leyte in early December to relieve Ballard of his duties as liaison officer, and he returned to Brisbane a few days later on a Douglas C-47 Skytrain.
It was Sandford, as his commanding officer, who gave Ballard the dressing down.19
‘Well, Geoff, you seem to think that this war is being fought for your convenience,’ Sandford said to him.
Ballard quipped in response, ‘Really? I can’t say I’ve noticed the convenience.’
Ballard was put on leave for six months, after which he was sent to Colombo to take up the position of Central Bureau liaison officer with the British code-breaking unit there.20
1945
38
Spies
The head of Ultra distribution at Bletchley Park, Group Captain Frederick Winterbotham, arrived in Brisbane in December as part of a worldwide tour of Allied Ultra centres. Bletchley Park had a series of liaison teams at various code-breaking centres known as SLUs (Special Liaison Units). Winterbotham had planned, in his visit, to focus on administrative issues with the local unit, SLU9, making sure they had access to all the appropriate intelligence channels. He was, for the most part, impressed:
I found the Ultra set-up in Brisbane was most efficient, but it suffered a little from some restrictions placed by MacArthur, on ‘who should have what’. Nevertheless I was able, with the cooperation of Colonel Sandford, the young Australian officer in charge whom I had known in London when we were teaching him the job, to sort out some of the distribution problems so that Melbourne and Stripe’s SLU’s in Delhi could get more of what they needed.
But now Australian and American forces were moving northward again, distances were great and security in the ‘island hopping’ was going to be dicey.1
However, upon his arrival he discovered that there was a far more serious problem to deal with than bureaucratic wrangling with American intelligence officers. Australia had become a target of espionage: top-secret information, including Ultra information derived from code-breaking, was reaching the enemy.
Australian intelligence officials had, for some time, been concerned about Chinese intelligence-gathering, and in particular the activities of Lieutenant Colonel Wang Chi, the Chinese army attaché in the south-west Pacific area.
Colonel Wang had arrived in Australia in 1942 under an agreement by Prime Minister Curtin to allow the appointment of two Chinese military attachés. Colonel Wang Chi came representing the Chinese army, and was attached to Australia’s army land headquarters; his counterpart, Captain Wang Chi-Kuang, was attached to the navy.
Colonel Wang Chi, from China’s Hunan province, had lived for years in the United States, where he had studied at the West Point Academy, after which he had returned to China and served as an officer in the conflict between China and Japan. In 1941 he was appointed to the role of liaison officer with the United States army in the Philippines.
Colonel Wang was gregarious and extroverted, throwing parties with invitation lists that were a who’s who of senior Allied officials in Australia. He drove an expensive car, and enjoyed the attention of local women. He had been causing headaches for Australian intelligence officials since his arrival. Within months, he managed to get himself admitted to an Australian military staff conference in Melbourne, causing General Blamey to issue a stern warning that foreign nationals were not to attend staff conferences.
Wang was warned that Allied experts believed that the Japanese had broken Chinese military and diplomatic codes, and that he should therefore refrain from sending sensitive military information in those codes. Instead, Wang started using his own private cipher when communicating with his leaders in Chungking.
Cocky Long’s naval intelligence department viewed Wang as untrustworthy, noting that, at times, the colonel ‘drank to excess, and under the influence of drink was very loose in his talk’.2 Others believed that he was developing a case of ‘Anglophobia’, because he was not treated with the respect he believed he deserved. This may have been a cultural clash, as Wang’s ostentatious manner had unintended consequences in an Australian culture where ‘tall poppies’ were not well received.
Wang had access to some intelligence reports, but did not have access to Ultra. While China was a member nation of the Allies, it was not part of the inner circle of Allied nations consisting of the United States, Britain, and Commonwealth countries (including Australia) that shared Ultra intelligence.
Early in 1944, Central Bureau decoded a message from Wang to Chungking containing Japanese order-of-battle estimates. General Willoughby, MacArthur’s intelligence chief, compiled and regularly updated his estimates of the Japanese order of battle — the strength and location of enemy forces at any given time — and he relied extensively on Ultra decrypts from Central Bureau to do this. The information that Wang had secretly sent to the Chinese capital was classified as Ultra.
This was worrying, for two reasons. First, Wang was not supposed to have access to Ultra intelligence. Second, if Japanese eavesdroppers intercepted the message, they might figure out that their codes and ciphers were broken, thereby ruining the entire Ultra project.
It turned out that Wang had not stolen Ultra information, but had come across it in the course of his business in Brisbane. He had paid a visit to Major George Batchelor at the army’s land headquarters at the University of Queensland, where he had found Batchelor’s map of the south-west Pacific area open on the wall. The map, like similar maps at Brisbane and Melbourne headquarters, could be secured by a metal roller-blind, but Batchelor had left it open, giving Wang the opportunity to study the markers over the map that showed the estimated locations of enemy troops.
In August, Bletchley Park sent alarming news to Central Bureau. The British code-breakers had intercepted part of a message from Tokyo to Japanese headquarters across east Asia that listed the code names and locations of what appeared to be Japanese spies.
The first part of the message was missing (although it was later recovered). The second part said:
Their extent of use must be reduced to a minimum. Moreover, particular care should be taken to see that the source of information is not recorded in Intelligence files, especially in areas which are likely to be involved in battle in the near future. The following is a list of such Intelligence.
Para 1. Information from the military reporting officer in London is D1.
Para 2. From the military officer in New Delhi – D2 Intelligence.
Para 3. From the military reporting officer in Ceylon – D3 Intelligence.
Para 4. From staff officer Noguchi in Sydney – D4 Intelligence.
Para 5. From the military reporting officer in Melbourne – D5 Intelligence.
Para 6. From the military reporting officer in Washington – D6 Intelligence.
Para 7. From the military reporting officer in Ankara – D7 Intelligence.
Para 8. From the general staff (Gunreibu) in Chungking – D8 Intelligence.
This message had been transmitted by radio, sent to overseas locations such as Manila and Rabaul, and, as such, was within range of the high-frequency intercept site at Mornington, Victoria, manned by 52 Wireless Section. Diplomatic radio traffic picked up at Mornington was passed directly to the diplomatic code-breaking unit, Special D Section, at Park Orchards. Professor Dale Trendall had returned to the University of Sydney, passing command of Special D section to his promising code-breaking protégé, Ronald Bond. Bond had been Trendall’s star student, graduating with first-class honours in Classics at the start of 1942, after which Trendall recruited him as a code-breaker.3 The section played an active role in breaking Japanese diplomatic codes throughout the war, remaining in close contact with Bletchley Park, where many of the intercepted messages were decrypted.
In all likelihood, the message had been intercepted at Mornington, decrypted at Bletchley Park, then sent to Central Bureau. The immediate questions were: Who were ‘D4’ and ‘D5’? Were these the Chinese consulates in Sydney and Melbourne, passing on information obtained by Wang?
Also in August, Wang provided Chungking with information about Australian troop locations, using a cipher known to have been broken by the Japanese. This was not, on its face, an act of deliberately sending secrets to Japan, but of carelessness that had the same effect. It was not clear whether Wang was a Japanese spy, or whether, as the latest breach seemed to suggest, the Japanese were getting information from Wang by decrypting Chinese radio traffic, and also, perhaps, from their agent D8 in Chungking, who might be someone in the Chinese military with access to Wang’s reports.
Wang’s access to intelligence was further restricted. Yet throughout 1944, Allied estimates of the order of battle, originating in Australia, continued to appear in decrypted high-level Japanese messages. In particular, the Japanese seemed to be obtaining information from the Australian Military Forces’ Weekly Bulletin, an intelligence briefing with a select readership of about 200 individuals. Blamey was now so worried about the situation that he ordered that production of the bulletin be stopped completely. As this was going to hamper the dissemination of intelligence to those who needed it, the director of military intelligence, John Rogers, suggested to Blamey that he allow the bulletin to recommence, but with a smaller list of recipients.
The same day that he cancelled the bulletin, Blamey contacted Amalgamated Wireless, the Australian government-owned entity that ran cabling services, and told them to start sending all messages from the Chinese consulate in Melbourne to the Australian Security Service. But none of this stopped the flow of Ultra intelligence coming out of Tokyo in diplomatic messages.
Mic Sandford touched down at Hollandia airfields on 19 November. He was there to meet General Blamey’s chief of staff, Lieutenant General Frank Berryman, to discuss with him the growing problem of the intelligence leaks. Ultra was now reaching Tokyo quite quickly after it was disseminated within Australia. Sandford asked Berryman how it was that Colonel Wang was still receiving Ultra information. Wang was supposed to have been excluded from all Ultra distribution lists, and General Blamey himself had gone to some trouble to restrict Wang’s access to intelligence.
Berryman told Sandford that Colonel Wang had not been supplied with any Ultra information. He wasn’t on the list, and Berryman was sure he did not have access to any of the leaked briefings. Whoever was supplying the Japanese with Australian intelligence, it was not Wang.4
Five days later, Central Bureau was informed about a diplomatic cable that shed new light on the leaks. It had been sent from the Japanese consulate in the Manchurian city of Harbin to Tokyo. Manchuria was controlled by Japan through the puppet government installed during the occupation of 1931, and although Changchun was the Manchurian capital and headquarters of the armed forces, the intelligence service was headquartered further north at Harbin, a city that had since become a hotbed of espionage activity.
This report, titled ‘Harbin Special Spy Report’, informed Tokyo that ‘Australian forces have taken over from the Americans in the Solomons and New Guinea’.
This was true. As American troops were deployed northwards to engage in the invasion of the Philippines, Australian troops assumed responsibility for maintaining control of territories further south.
The report also informed Tokyo that,
Although the United States Army command has been urging commencement of operations against Ragoru Bay at the eastern tip of Bacol Peninsula, Pacific Fleet HQ on the pretext of bad weather conditions is delaying execution of this operation.5
The Harbin Special Spy Report claimed that the source of this highly detailed, accurate information was the Soviet ambassador in Australia. The leaks were coming not from the Chinese, but from another ‘friendly’ nation: the Russians.
Soviet Russia, like China, was a member of the Allies, but was similarly excluded from the circulation of Ultra material. It was not clear how the Soviet ambassador was obtaining this information.
The leaks were having a real effect on the war. During the battle for Leyte, General Willoughby’s order-of-battle estimates for Leyte were received by Tokyo in a Harbin Special Spy Report, and appeared in intercepted traffic. Willoughby and the code-breakers knew this. Even so, the importance of the leak was not understood.
The Allies knew that their secret estimates of the strength and location of Japanese forces had fallen into the hands of the Japanese, but could not yet know the extent of the damage that had caused. What they did not know, but the Japanese army did know, was that Willoughby’s order-of-battle estimates for the Japanese forces were wrong. Estimating the strength of enemy forces is an inexact science, even with good intelligence, and for various reasons Willoughby had significantly underestimated the number of Japanese troops on Leyte.
The Japanese commanders, realising his mistake, maximised their advantage by sneaking large numbers of reinforcements on to Leyte while making a show of withdrawing some forces from the island. In this way, they increased the disparity between Willoughby’s estimates and the truth. That, in turn, led to poor planning decisions and an under-resourcing of the battle to secure the island. Willoughby advised Macarthur that Leyte was being abandoned, when in fact the Japanese forces were being strengthened. The result was a longer, costlier battle for the island than it might have been had the estimate of the Japanese order of battle been accurate, and had the Japanese not been informed of the Allies’ mistake.
On Christmas day, Mic Sandford received another decrypted Harbin Special Spy Report. In notifying Robert Little at Military Intelligence headquarters in Melbourne, he wryly treated the intercept as a Christmas present from the Japanese:
Dear RAL,
On this auspicious day Imperial General Headquarters feels compelled to make an appropriate present to the Honourable General Staff of the Australian Military Forces. I enclose the text, for you to pass on to the CGS and the C. in C, of our MBJ 31258.
Winterbotham and I had conversations at some length with the DMI and Winterbotham has been asked to bring out a special Section 5 man to deal with the problem. It is most imperative of course, that no action whatever be taken until his arrival.
Norman Webb and the staff here join me in sending our very best wishes for Christmas and the New Year to you and your family and to the staff at Victoria Barracks.
Yours ever,
Mic Sandford6
With the letter, Sandford included a second sheet of paper: a decrypted diplomatic communiqué from Tokyo, sent at 11.00 am on Christmas Eve. It was now clear, given the restricted circulation of Ultra and the speed at which it was reaching Japan via the Harbin Special Spy Reports, that the Soviet ambassador was getting the information from the Australian cabinet. How he was doing this was not clear.
The Soviet embassy did not have a direct line of communication with Moscow, as the distance was too great. Instead, all messages were sent via Harbin, where the Soviet consulate acted as a relay post. The messages were reaching the Japanese when they passed through Harbin, but how that was happening was not clear either.
One thing was now apparent. The damaging Ultra leaks through the Harbin Special Spy Reports were not being sent by Wang or other Chinese consular officials. The Chinese attaché and Chinese consulates were engaging in zealous intelligence-gathering, but were not deliberately undermining the Allies.
Mic Sandford, John Rogers, and General Blamey met on 2 January 1945. They decided to take no further action over the activities of Wang, or the Chinese consulates in Sydney and Melbourne, other than ensuring that sensitive information did not reach them. Sandford was of the opinion that if the Japanese continued to read Chinese military and diplomatic traffic, this would keep their code-breakers busy and give them less incentive to break Australian or American codes.
The leaks from the Soviet ambassador were more dangerous, as he seemed to have access to the most exclusive and secret information within Canberra. On that front, there was nothing they could do other than further restrict the dissemination of Ultra.
Somehow, the ambassador was accessing top-level information that had a very limited circulation to select federal ministers and their staff. The foreign minister, Herbert V. ‘Doc’ Evatt, and the minister for supply and shipping, John Beasley, and their staff, had access to all the leaked information. Perhaps the ambassador’s source was someone in the office of one of those two ministers.7
The question of how Ultra information was getting to the Japanese consulate in Harbin remained unresolved for the rest of the war. Further allegations of espionage against Evatt’s staff arose in 1954 following the defection of Vladimir Petrov, a Soviet diplomat in Australia, but a subsequent royal commission did not find sufficient evidence for anyone to be charged. It was not until 1998 that a solution to the mystery was provided by the Australian historians Desmond Ball and David Horner in their book, Breaking the Codes: Australia’s KGB network, 1944–1950. Ball and Horner concluded that some individuals among Evatt’s staff were in regular contact with a Russian diplomat, Feodor Nosov, who was probably the Soviet ambassador’s connection for receiving classified information.8
As for the Harbin connection, Ball and Horner revealed that the Japanese had not been spying on the Soviet consulate, nor getting information from broken Soviet codes. The solution was far more simple: the Soviet consulate in Harbin had simply been handing the information over. The reason was because Joseph Stalin, Russia’s ruler, had concluded by the end of 1945 that the Japanese were going to lose the war, one way or another. He wanted to declare war on Japan before that happened so that he would have a seat at the table during peace negotiations. But Hitler had not yet been defeated, and Stalin’s troops were occupied in Europe. By prolonging the Pacific War, Russian troops would have time to defeat Germany and capture swathes of Eastern Europe while doing so, and then relocate to the east in order to wage war on Japan.
Stalin was providing the Japanese with Allied intelligence, not to help them, but to delay their defeat for his own strategic reasons.9
39
Lingayan Gulf
Seven Australian men sat in a small, hot ship’s cabin, headphones over their ears, hunched over their radio receivers. They were all members of the secret Australian air force unit, 6 Wireless Unit, and they were on board General Akin’s command ship, PCE 848 for the second time. Their officer in charge was Flight Lieutenant ‘Alf’ Davis, one of the original wireless operators trained by Jack Newman in 1941, part of the team that started monitoring Japanese messages from Darwin before the war in the Pacific began.
Eleven weeks earlier, in October 1944, the same men had sat in this same room as the ship glided toward Leyte. This was Akin’s communications command ship for the first massive assault on the Japanese-occupied Philippines, scanning the airwaves for signs that the convoy had been discovered. Now they were performing the same task as the ship headed north as part of a large fleet to make landfall on Luzon, the largest and most populated island of the Philippines.
On their mission six weeks earlier, there was no sign that the Japanese defenders knew the fleet was coming. Today, on 8 January 1945, it was a different story. The fleet had been discovered, and it was suffering a series of aerial attacks. The Australians’ job was to listen for when the next attack would come, and where from.
After the war, Alf Davis described the harrowing journey. ‘Kamikaze planes fell all around us and we saw a number of American warships hit. We landed at Lingayen Gulf after the fleet bombarded the shore area for some hours. Even under fire, we were able to continue operations at all times.’1
In addition to kamikaze planes were kamikaze boats — makeshift human torpedoes that slammed into the side of the ships. In the journey northward, seven American ships were sunk, and numerous others damaged.
Nine other members of 6 Wireless Unit were on another ship in the fleet. Meanwhile, the remainder of the unit, still on Leyte, were also tracking Japanese air-to-ground radio traffic and sending their findings to Akin’s ship, using an American encryption machine, a SIGABA.
Also on the fleet was the United States army signals unit, 112 Signal Radio Intelligence Company, now re-formed as a Radio Squadron Mobile and reporting to the United States army air force. They were a veteran intercept unit that had operated on Guadalcanal, where three of their number had been killed, and later on Manus Island. They, too, scanned the airwaves to monitor incoming attacks.
MacArthur had insisted since the start of the war that the Philippines should be recaptured. Other generals didn’t agree. As the Allies captured territory closer and closer to Japan itself, a battle to retake the Philippines seemed like another big, nasty land war that they didn’t need. Military planners argued that the Allies could simply capture some islands further north — such as Okinawa, a string of islands very close to Japan — and use those positions as a launching pad for an assault on the mainland Japanese islands. That would be easier and quicker, in the way that MacArthur himself had bypassed and isolated hundreds of thousands of Japanese troops in the south-west Pacific.
MacArthur wouldn’t hear of it. After all, had he not promised ‘I shall return’ in 1942 at a train station in South Australia? That promise had been relayed across the world, including to the Filipino people themselves. And now, having recaptured Leyte, he promised that liberation was at hand. Bypassing the Philippines was out of the question.
MacArthur’s strategy of making surprise invasions where the enemy was not expecting it had paid off at Hollandia and Morotai. He wanted to do the same thing with Luzon. Instead of making landfall at the nearby beaches at the south of the island, the invasion force was sent all the way up the west coast to Lingayen Gulf in the north.
All the same deception techniques were used as before. In the lead-up to the invasion, the air force made reconnaissance flights over the south of the island, and bombers launched air raids against Japanese army positions along the south coast. American minesweepers cleared mines from the waters near the southern beaches. But Lingayen, far to the north, the place where the Japanese themselves had landed three years earlier, was the true target.2
This time, none of it worked. The Japanese army had seen it all before. Rather than trying to guess where the Americans were going to land, the strategy of the Japanese commander of the Philippine forces, General Tomoyuki Yamashita, was to let the Americans land wherever they pleased, and to mount a defence further inland.
The American invasion force made landfall at Lingayen the next day, unopposed on land, but suffering a barrage of air attacks, including from more kamikaze planes. Soldiers poured onto the beaches and established a beachhead. In the space of three days, 175,000 troops came ashore — more than in the first wave of the Normandy landings in Europe a year earlier. As the Americans started pushing south toward the capital, Manila, they came up against Yamashita’s inland defences.
Within twenty minutes of the first landings, 6 Wireless Unit was on shore and operating. However, the American radio unit ran into operational difficulties. The commander of the United States 37 Division had reassigned them to duties as a shore party. They unloaded ammunition, petrol, and other supplies, and continued to work as general labourers for another week, until they were released and were able to begin their operations.
As the American forces moved south, the signals detachment led by Alf Davis travelled with them. The United States 14th Corps recaptured their former airbase, Clark Field, on 23 January, and the First Cavalry Division reached Manila on 3 February. MacArthur had personally issued specific instructions to try to spare as much of the city as possible.
General Yamashita decided to withdraw, and fled the city, leaving orders with Vice Admiral Denshichi Okochi to destroy the city’s port facilities and to then get out. If that had happened, capturing Manila would have been quick. But Okochi and about 10,000 Japanese troops did not obey those orders, deciding instead to stay and fight. When the Americans entered the city, the Japanese resisted their advance street by street. The result was a protracted campaign of urban warfare that dragged on for four weeks and left much of Manila in rubble.
The Japanese defenders, in defying their commanders, were now effectively operating on their own, no longer part of the military chain of command. They had gone rogue. Even as they stalled the American advance, they knew they were doomed. During lulls in the fighting in Manila, the Japanese soldiers went on looting and killing sprees across the city, taking out their anger on the civilians. In nearby occupied villages, similar events occurred. Civilians were rounded up and killed en masse in various ways. Hospitals were burned to the ground, with the patients and staff trapped inside. Local girls were rounded up, the most beautiful among them selected and taken away to be raped. It is estimated that between 100,0000 and 500,000 Philippine civilians died directly at the hands of this relatively small number of renegade Japanese soldiers in that period.
One of the soldiers later said to the Japanese historian Jintaro Ishida, ‘In the beginning, we could not kill even a man, but we managed to kill him. Then we hesitated to kill a woman. But we managed to kill her, too. Then we could kill children. We came to think as if we were just killing insects.’3
MacArthur established a field headquarters about 60 miles north of Manila at Tarlac, near San Miguel, and the wireless units established themselves nearby at a local sugar mill. The Radio Squadrons Mobile unit reaped the benefits of travelling with an American force, establishing themselves in the mill’s office building, while Alf Davis’s Australian detachment had to operate from a tent.
When reports came through in late February that most of Manila was secure, Pappy Clark sent the remaining contingent of 6 Wireless Unit at Leyte to Manila. The city proved to be a perilous place to work, so the unit moved north to join the detachment at the sugar mill at Tarlac.
Pappy Clark stayed on at Leyte, which had been designated the headquarters of a new Central Bureau branch, Forward Echelon. Four Wireless Unit arrived in December, and Central Bureau’s army air-to-ground solution section, headed by Major Nobby Clarke, relocated from Central Bureau in Brisbane.4
The Allied occupation of the Philippines spelled the end of FRUMEL in its previous form. Captain McCollum, the Seventh Fleet head of naval intelligence, transferred most of the Japanese linguists to the Allied Translator and Interpreter Service, and deployed them in the Philippines. Most of the resident American code-breakers were also transferred elsewhere. Control of FRUMEL was handed over to Jack Newman.5
Capturing the Philippines put the Allies in a powerful position to threaten Japanese-held territory closer to the Japan. Meanwhile, Admiral Nimitz had waged an island-hopping campaign across the northern Pacific, capturing strategic islands such as Iwo Jima and Saipan. The Allied command hoped that, before long, Japan itself would be within striking distance.
40
The expanding web
As MacArthur had moved, Central Bureau had followed him — first from Melbourne to Brisbane, then to Port Moresby, then to Hollandia, then Leyte, and now Luzon. The result was a strung-out network of headquarters and units across thousands of miles of ocean. Communication lines had become stretched between Brisbane, Hollandia, Leyte, and Luzon, complicating the transmission of intercepted material and the coordination of work.
The wireless units were also scattered across the south-west Pacific area, with units in Perth, Darwin, Townsville, Port Moresby, Hollandia, Biak, Morotai, Leyte, and elsewhere. As wireless units multiplied, so did the daily flow of intercepts arriving for the Central Bureau code-breakers; and as more messages came in, more personnel were needed to sort them, file them, copy them, decrypt them, feed them into IBMs, strip routing codes, and generally suck as much information out of them as possible.
Morotai, being the staging point between the south-west Pacific area and the Philippines, had quickly become a major Allied base. Several personnel with 4 Wireless Unit on Morotai learned that in June they would form part of an intercept unit on the island of Labuan, to the west.
Australian artist Donald Friend was appointed to the role of ‘War Artist’, and was deployed to Morotai immediately before the unit left for Labuan. Friend arrived on the island of Morotai on 27 May, and, being a compulsive diarist, recorded his impressions of the island:
It’s just what you’d expect. All tropical islands are alike; the jungle is international. In this place, our forces squat on one end of it, and the Japs on the other.
All is mad excitement. Jeeps and trucks and aeroplanes tear about like mad. Busy little ants running everywhere. And mad muddle …
Because any day now, perhaps tomorrow, the AIF will invade the island of Tarakan. A very important attack. It’s all rather alarming, this nearness to the battle. All the time overhead the coming and going of flocks of planes never ceases. Everyone seems to be a bit tense — excited and expectant.1
Friend’s plane had touched down only hours earlier, so his confusion about the battle preparations was understandable, but he had it slightly wrong: the Australian 26th Brigade had captured the island of Tarakan three weeks earlier, in a prelude to taking the large island of Borneo. A week later, 16,000 grim-faced Australian soldiers stationed on Morotai boarded transport ships, bound for Labuan, on the far side of Borneo. They were the soldiers of the 9th Australian Infantry Division. At Labuan, they went ashore without meeting any opposition, but when they moved to secure the airfields, the Japanese put up some resistance before retreating.
Sailing to Labuan with the 9th Australian division was a detachment of 4 Wireless Unit. Among others, they included Hugh Melinksy, a British translator who was sent to Central Bureau from Bletchley Park in 1944; Ron Warlow; and Keith Carolan, who was a child welfare officer before the war.2
The Australians landed and established a base there, and the wireless unit set up direction-finding equipment in communication with the main contingent on Morotai.3 The Japanese army retreated inland, their snipers making occasional forays to take shots at the Australian infantrymen. One wireless operator was injured by sniper fire; Keith Carolan was standing in line for dinner when the digger two places in front of him was shot dead. On another occasion, a Kana operator was hit in the arm by a bullet.4
The unit operated at a building near the airstrip, and had been there for some time when a local man of Chinese descent appeared on the runway, asking for help, as his wife was very ill. After introducing himself as Cheng Lee, he took them to his family, who lived in a clearing in the jungle a few hundred yards away, where the unit discovered that his wife had malaria.5 The wireless operators organised for her to get some Atebrin tablets, which were in plentiful supply at the Australian base.
The curious thing was that the family lived so close to the base, but the Australian wireless unit had no idea of their presence until the man made his presence known. Two of the Australian operators, Keith Carolan and Ron Warlow, stayed in contact with Cheng Lee and his family long after the war.
The Australian army’s 53 Wireless Section also began operations on Morotai. They had been in eastern New Guinea at Finschhafen since mid-1943, but now, with Allied forces on Morotai, Borneo, and the Philippines, tracking enemy circuits around eastern New Guinea was no longer of strategic interest. The commander of the section was Captain William Hill, a veteran of the 1941 Crete campaign, where, as a lieutenant, he had shared his water bottle with a thirsty Mic Sandford as they escaped the German advance.
Vic Lederer — the British-born German speaker who had fled Germany due to his Jewish heritage — was attached to one of the American Radio Squadrons Mobile on the island before joining 53 Wireless Section as the intelligence officer in charge. An incident on Morotai became burned into Vic’s memory.
One night, Japanese soldiers threw a grenade into their campsite before running away. The unit hastily grabbed their weapons and pursued the grenade lobbers, tracking them down to a thicket. They opened fire on the thicket, and were about to depart when Lederer insisted that they search for the bodies — if there were any. As it turned out, they came across the bodies of two Japanese soldiers in the nearby jungle. Searching their belongings, Lederer and his companions found a letter that one of the soldiers had written to send home, which they collected and delivered to an American intelligence unit stationed on Morotai. 6
In June, around the time of the Labuan invasion, movement orders for the direction-finding team of 4 Wireless Unit came through. They had to relocate yet again, this time to the Philippines, to join Central Bureau’s Forward Echelon at Tarlac. The Labuan detachment continued to operate at the Labuan airstrip, and another small detachment was formed to stay on Morotai itself.
While Australian and American wireless units moved in hops and jumps northwards up the Pacific islands, chasing the advance of MacArthur’s forces, one intercept unit moved south … from Canada to Australia.
On his visit to North America in 1944, Pappy Clark had arranged for a Canadian intercept unit, ‘Number 1 Canadian Special Wireless Unit’ (or ‘1CSWG’) to be deployed in Australia.
By the time the Canadian intercept unit arrived in Brisbane in 1945, the city presented a stark contrast to the frenetic bustle there three years earlier, when it had been a garrison town close to the war. Since then, the war had moved far north. Troop ships were no longer unloading thousands of freshly recruited young American soldiers before their deployment to the battlefields of the south-west Pacific. Those soldiers were taken directly to the Philippines, or to Guam, or other territories recently wrested from the Japanese forces.
Gil Murray, one of the Canadian intercept operators, described what it was like:
The Brisbane docks were devoid of life. We hadn’t expected crowds of well-wishers to greet us, but we did at least expect to see a busy waterfront, teeming with wartime dockyard activity, as we’d witnessed in the war zone at Hollandia. There we had seen huge cargo cranes swinging ominous loads of war materiel onto ships, stevedores sweating, swearing, and heaving boxes off trucks; bosses shouting and cursing their charges; quiet men in shirts and ties carrying clipboards and carefully checking off war goods as they moved hither and yon. Here, apart from the tramp-tramp of Canadian parade drill boots and the snarl of the sergeant-major, there was peace and tranquillity.7
The Canadians soon discovered that a large part of the reason for the ‘peace and tranquillity’ was that the Brisbane dockworkers were on strike. That being the case, the Canadians unloaded their equipment themselves from their frigate at the deserted docks. The unit camped at Chermside barracks for several weeks while they undertook training at the Central Bureau ‘I School’ before being moved out to the Northern Territory in mid-April.
When they arrived at their site at MacMillan’s Road outside Darwin, a local brass band, the Australian Northern Territory Band, was there to greet them. As the Canadian wireless operators jumped down from the trucks, the band burst into a rendition of ‘O Canada.’8
The Canadians erected antennas and radio receivers, and began monitoring the seas to the north. Gil Murry described the operation:
Almost from the moment of our arrival, the Number One Section operators, their skin turned deep brown from the sun, sat in khaki shorts, sneakers, and headphones in a big, steamy, tin-roofed hut, their fingers jabbing at typewriter keys. They got down on paper a typed record of the frenetic beeps picked out of the air by short-wave radio sets piled on a long bench … Each monitored two radio sets, one tuned to a Japanese sender, the other to a receiver.9
An hour’s drive south, the 51 Wireless Section was stationed at Coomalie Creek, where it had been since September 1942. The section was now ordered to wind up its operation and return to Brisbane, but it could not leave until the Canadians were up and running, and covering their circuits. The unit continued until Central Bureau was satisfied with the Canadians’ performance, and then packed up and relocated to Brisbane.
The gaze of the Allied commanders had shifted north, but intercept units were still needed on the Australian mainland. The Canadians monitored traffic from Singapore and parts of Indonesia and Irian Jaya that were still under Japanese occupation.
41
The paper war over women
MacArthur entered Manila on 4 March and declared it liberated. He then established a new headquarters there. This was bad news for Central Bureau: they had to move again.
Advanced Echelon, the branch of Central Bureau attached to MacArthur’s headquarters, had already moved to the Philippines, but it was decided that a bigger move was needed. Central Bureau now had about 4000 personnel. As much of the operations as possible needed to be relocated to Luzon in preparation for the coming invasion of Japan.
Lieutenant Colonel Mic Sandford was of the opinion that they should move the entire operation to the Philippines, and anyone who refused to go, they would get ‘rid of’.1 Given that Eric Nave had a long history of refusing to travel overseas on medical grounds, this was likely a reference to Nave.
The problem still remained, though, that Australian women were not allowed overseas on military service, and Central Bureau employed hundreds of female servicewomen at Nyrambla, at the Cipher Office in the garage behind Nyrambla, at Ascot fire station where the IBMs were installed, and in the tent city of Ascot Racecourse.
The director of military intelligence, John Rogers, had raised the issue with MacArthur on Sandford’s behalf in late December, suggesting that MacArthur, with his influence at the very top levels, might convince the Australian government to change its policy. MacArthur told Rogers that he didn’t want to raise it again with the Australian government just then. He added that once the Philippines were invaded and he moved his headquarters to Luzon, things would change. The women, as a critical part of the war effort, would naturally be needed in the Philippines, and at that time the Australian government would surely relent.
That time was now here, and Sandford made a renewed push to get permission for the Central Bureau women to relocate to the Philippines. In late January, he broached the topic with Lieutenant Colonel Little from the Military Intelligence office in Melbourne. He told Little about the conversation between Rogers and MacArthur, and about how MacArthur was basically assuming that the Australians would come to the party on this.
There was just a small problem, Sandford explained: ‘If we are to move the AWAS however, certain amendments to our war establishment in the matter of house keeping personnel for our female members will be necessary. It will also be necessary to make one or two minor amendments to our own W/E [war establishment].’
He added, ‘I suppose you are sick and tired by this time of the constant amendments we are requesting of the War Establishments Committee, and I feel I owe you an apology in this matter, but our situation has changed so greatly from time to time that it has been very difficult to predict our needs as much as three months in advance.’
Sandford wasn’t asking Little to make the decision, since Little didn’t have that power. He just wanted Little to do a bit of back-room politicking to soften the politicians up for the formal request. ‘If you could spare the time to give it your consideration and perhaps discuss it informally with a selected member of the War Establishments Committee, I should be very grateful.’
But Sandford’s hope that Little would help him out was misplaced. Little responded tersely, sending Sandford a copy of the original decision not to allow large numbers of women to serve overseas, and, commenting on the memo, said that ‘it is considered most unlikely … that approval will be given by cabinet for AWAS to go beyond New Guinea’.
Little replied that he would not help, formally or informally. He further suggested that if Sandford really wanted the women to go to Luzon, he should make a formal request higher up the chain of command.2
Sandford had tried to get it to happen through contacts at Victoria Barracks, but had been thwarted by Little. However, he had another card to play. It was time for General Akin himself to approach the government on MacArthur’s behalf.
Akin visited the minister of defence in person for a one-on-one conversation. He explained what Central Bureau was doing, why it was so important to MacArthur and needed to therefore relocate to Luzon, and how critical it was that the AWAS women serving in Central Bureau be included in the relocation to Luzon.
Akin added that the expertise that Australia was developing through Central Bureau was of enormous value and would provide the basis for post-war expertise in sigint. Therefore, sending the women to Luzon would not only help the Allied war effort, but would also be a great long-term investment for Australia’s defence.3
At the next meeting of the War Cabinet, the minister reported his conversation with Akin. The politicians did not want to do it, but Akin had put them in a tight situation. After some discussion, they decided to defer making a decision, and instead to ask for the opinions of the heads of the Australian army and air force.
By mid-May, both the army and air force had responded. The chiefs of staff for both services told the War Cabinet that, yes, the women were needed overseas and should be sent.
The chief of the general staff on behalf of the army wrote a submission to cabinet, saying,
I am of the opinion that the operational requirements in the South West Pacific Area require the best information regarding the enemy that can be obtained and if the AWASs (and WAAAFs) at present on the strength of Central Bureau are not permitted to accompany the unit when it moves to the Philippine Islands the flow of vital information will be adversely affected. I therefore desire to endorse most strongly the representations by General Akin to the Minister of Defence that an exception be made in the case of Central Bureau and that the AWASs (and WAAAFs) at present on strength be permitted to accompany Central Bureau beyond New Guinea.
With the submissions now in, the War Cabinet discussed the matter again on 18 May. Clearly, the Americans wanted them to let the women go to the Philippines, and the Australian army and air force agreed. The cabinet had been informed by Central Bureau as well as by their own military chiefs that the women were a vital part of the war effort, and their advisors were telling them that 500 women should be allowed to go.
The politicians considered all this, and yet they refused. The reason they gave was that it was ‘wrong in principle that civilian women be allowed to travel abroad in these numbers’.
The secretary of the War Cabinet, Frederick G. Shedden, issued a ruling:
The proposal that members of the Australian Women’s Services, at present on the strength of the Allied Central Bureau, be transferred to Manila with this establishment was not approved. A definite limit has been laid down by the Government on the number of women to be permitted to serve outside Australia and War Cabinet did not consider that this should be increased.’4
And that was the final decision. No Central Bureau women could go to Luzon — or, for that matter, anywhere else outside Australia.
42
Tarlac
At Central Bureau Brisbane, preparations were made for the move to the Philippines. Everyone who could go was going, which meant most of the men but none of the women. Sections that were slated to move included the air-to-ground solution centre, still working on air-to-ground three-figure additive codes, and the wireless units that worked closely with them.1 A medical team arrived to dispense inoculation shots against the various tropical diseases they might be exposed to.
Wireless intercept units across the south-west Pacific that were now too far from the action were also given orders to relocate to Luzon.
The IBMs couldn’t move, though; assembling them once had caused enough headaches. They would continue to run at Ascot Fire Station.
Central Bureau’s new home, dubbed Forward Echelon, was established on the grounds of a golf course by a burned-out sugar refinery at Tarlac in the countryside near the city of San Miguel.
One Wireless Unit worked in the sugar refinery itself, a corrugated-iron factory that had been left with bullet holes in the walls after the recent fighting between the Japanese and MacArthur’s invasion force. Here in the refinery, an American intercept team joined the Australians, and set up their listening stations next to the unit.2
The Americans brought typewriters with them, and they arrived with extras — enough to supply the Australians. No longer would the interceptors have to write messages down in shorthand; they could type them straight onto a keyboard.3
The team from Central Bureau’s Forward Echelon lived in tents in a flat field on the grounds of the golf course. The course had been unused for some time, and was overgrown with weeds and long grass. It was the wet season, and so, to prevent flooding, the team dug trenches around the tents. This was not a sure-fire preventative measure, and sometimes the rain was so torrential that the tents would flood and their clothes would be saturated. But the wet season was also the hot season, and there would be periods of hot sunlight in which their clothes would quickly dry.
The living conditions were as squalid as those in the jungles of New Guinea. Disease was common. Many — perhaps most — of them spent time in the nearby army hospital at Tarlac.4
Near the tent city were five makeshift huts for the Forward Echelon: three to work in, a mess hall, and a cookhouse. The commanders — Mic Sandford, Abe Sinkov, Roy Booth, Pappy Clark, and a few others — did not live with the teeming masses of workers in the fields that flooded, but in rented houses in the nearby hills. They were removed from the waves of malaria and other tropical diseases that swept back and forth through the sodden campsite.
The wireless stations were nearby. 5 Wireless Unit was camped just across the creek, and a short way along a road from them was a brothel in a nearby house, which was frequented by Australian and American troops. The Philippines madame ran the brothel in a strict, business-like manner. Troops would line up outside, and she would step out of the door and announce in a loud, crisp voice, ‘Next!’5
The mess tent quickly attracted Filipino children, who would stand outside the tent with empty bowls and other containers, hoping for food. The soldiers would oblige by scraping their plates into the bowls rather than throwing the leftovers away.
This led to more children coming with more empty bowls, and soon there were crowds of children with bowls at every meal. The children would jostle for the best positions, and the jostling turned into fighting. The situation became increasingly unmanageable. The commanding officers issued an order that no food was to be given to children outside the tent: all food was to be scraped into rubbish containers. The children then took to going through the discarded containers of rubbish, but at least the problem had been moved away from the mess tents.6
The local children would also arrive, selling mangoes, avocados, star fruit, and bananas, in exchange for pesos or the cigarettes that were US regulation issue. They would then take the cigarettes to the local market and sell them.7
Some of the local boys would also wander about the camp offering to do jobs for money, such as cleaning and tidying tents. They would steal food from the camp and take it to the hungry Japanese soldiers hiding and living nearby. In exchange, the Japanese soldiers would give them military items such as army caps, which could then be sold to the Australians and Americans as souvenirs. Thus there was a thriving black market in food and Japanese items at the base.8
Doug Pyle, an enciphering-machine mechanic, travelled from San Miguel to Manila, and was shocked by the devastation:
In spite of what I had heard, I was not prepared for the almost-total devastation I saw … Many enemy soldiers remained buried in the rubble and the stench was appalling. Worse was to come as I made my way to Intramuros (Spanish for ‘within the walls’), a once-beautiful walled city built by the Spaniards in the 1570s. Part of Manila’s heritage, it had been destroyed by the American bombardment. When I was a couple of hundred yards from the walls I came to an old aquarium, where 115 Japanese had fought to the end and had all been killed. Great dugouts, which had been their last refuge, were now deep wells with their rotting bodies floating in the water…
There were no schools for children to attend; no police stations left standing; hospitals had been replaced with makeshift temporary huts; roads were impassable; bridges were in ruins and the harbor was full of sunken ships.9
Gordon Gibson, a Central Bureau veteran, observed that as the work expanded it became increasingly pointless: ‘It seemed like there were more and more people doing less and less.’10
They were reading army air-to-ground, naval air-to-ground, and Water Transport code messages, among others. But the high-level Mainline code was unreadable again. The Japanese army had switched to a new codebook, Rikugun Angosho 5 (Army Code 5) on 1 January. Central Bureau’s long-running streak, which had been kicked off by the Sio box discovery, had come to an end: the Sio codebook was out of date.11
But although it might have seemed that way, the code-breakers’ efforts were not pointless. In three years, Pappy Clark had built a sprawling, monstrous traffic-analysis empire. All the personnel, the piles of messages moving back and forth, the banks of antennas manned by rows and rows of Kana operators, were all components of a huge machine, watching and listening to the hum of activity to the north and west. They were close enough now to pick up messages from mainland Japan, and as the American bombers destroyed the mainland phone lines, the Japanese army was increasingly relying on radio to communicate, even on its home turf.
Pappy Clark was making every radio message count, and was feeding MacArthur and his generals large amounts of detailed, reliable intelligence every day. The generals were eating it up.
43
The Garage: a cryptological love story
Back in 1943, when Central Bureau’s IBMs were relocated to Ascot Fire Station, a division known as the Cipher Office had moved into the garage behind 21 Henry Street where they had previously been installed. The office was the communications hub for other sigint centres such as Bletchley Park, Arlington Hall, and FRUMEL, and so on, as well as other Allied command centres. Every message it sent was top secret, and therefore had to be enciphered using a TypeX machine, the Allies’ equivalent of the German’s Enigma machine.
The TypeX worked a bit differently from the German Enigma. There was one keyboard; the operator would type in a message, and then the encrypted version would be printed out on a paper tape. (Or, conversely, the operator would type the encrypted version, and the plain version of the message would be printed out).
The doors were industrial baffle ones, which meant that to enter you had to turn left or right — you could not go straight ahead. The walls were fibro, and the roof was made of corrugated iron.
The floor was bare earth, covered with sheets of linoleum. To a visitor, the grimy, close interior would have looked like a very run-down newspaper office, with 12 typewriters on desks, piles of papers everywhere, and filing cabinets. But there were never any visitors.
With no air conditioning and a corrugated-iron roof, it was baking hot in summer and cold in winter. Inside were 12 cipher machines, staffed by 12 cipher operators working around the clock, in three shifts of eight hours. The shifts were synchronised to Greenwich time, not Brisbane time, so inside the garage the time was always whatever it was in London. The operators had punch cards, and would ‘punch on’ at the start of every shift.
The ‘Garage Girls’, as they called themselves, resided at the AWAS barracks in nearby Chermside. Half an hour before their shift, they would climb into an army lorry outside their barracks that would take them to 21 Henry Street, Ascot. At the end of the shift, they would again climb onto a waiting army lorry that would transport them back to Chermside. And so it went, day after day, night after night, between the barracks and the garage.
The incandescent lights dangling from the rafter were always on. Messages would arrive throughout the night from other parts of the world, and the TypeX operators would faithfully transcribe each one, stripping the message of the starting ‘filler text’, then setting the message in a stack to be delivered to important people within the walls of Nyrambla. They were allowed inside the main building to use the toilets down a corridor on the ground floor, and to collect or deliver batches of TypeX messages, but, otherwise, the interior of the building was a mystery to them.
According to one TypeX operator, Helen Kenny, ‘most of the messages were deadly dull’. Often they would be in code, even after the TypeX machine had decrypted them. In many cases, these were intercepted messages in JN-25, or some variant of it, that needed to be deciphered. But whether in plain text or code, interesting or dull, they were delivered.
At the end of every shift, the operators would sweep the floor, making sure to collect any loose pieces of paper that might have fallen down. Then all paper from that shift was taken outside to an incinerator and burned.
Madeleine Chidgey, a Garage Girl, would talk to her sister on the phone, but was vague about what she was doing. Her sister told her, ‘You don’t do anything, you just sit there and talk!’ She also wrote in a letter to Madeleine, ‘What are you doing up there all day? It just sounds as if you’re just doing letters.’1
The officer in charge was Lieutenant Archibald Ian Allen, who had been at Tobruk as part of the Australian pigeon unit. (Some army messages were still sent by carrier pigeon in the Second World War.) Lieutenant Allen, who went only by his middle name, ‘Ian’, was a towering six feet and seven inches tall. Helen Kenny remembers that, ‘He spoke little, but his voice was memorable for its diction and quality.’ After the war, Allen became an announcer for the ABC in Melbourne before establishing the radio station 2NU in Tamworth, New South Wales.2
Corporal Sandy Hinds also worked for the Cipher Office from the beginning. He had fallen in love with Coral Osborne, one of the Garage Girls. Coral was from Orange, a small country town in New South Wales, and was the daughter of a carpenter. Her parents came from Cornwall, and were raised, in her words, by ‘poor Cornish farmers’.
When she originally enlisted, the army sent Coral to the signals course at Bonegilla, and from there to Melbourne to do a cipher course. She worked for a short time for Jack Newman at Monterey Flats before being sent on a troop train to Brisbane and then being escorted from the train directly to 21 Henry Street. It was a charming old mansion, and seeing it lifted Coral’s spirits. After the oppressive regimentation of Monterey Flats, Coral thought this would surely be a lovely place to work. Two uniformed officers met her at the gate. One officer introduced the other, saying, ‘This is Corporal Hinds. He will tell you what to do.’ Then he left, and she was in the hands of Corporal Hinds.
But Coral never set foot inside that beautiful old house. Instead, Corporal Hinds took her around the side of the house, to the garage. This building, not the lovely house beside it, was to be her workplace for the next two years.
It turned out that Corporal Hinds was the officer in charge at the garage. His cheerful and pleasant style made the gruelling work more bearable. Coral learned over time that he was a Melbourne lad from a well-to-do family, and that he had attended the elite Scotch College, where he had learned to play the church pipe organ. Whether in the stifling heat or the biting cold of the garage, with the TypeX machines quietly rattling like insects, he would sometimes sing church hymns quietly as he worked.
‘Jesu, joy of man’s desiring,’ was a favourite, and Coral wondered whether the ‘joy of man’s desiring’ he was referring to was her. She hoped so.
In early 1944, he asked her to accompany him to dinner in Brisbane city. At the start of the date, he told her to stop calling him ‘Corporal Hinds’ and to call him Sandy instead. There was a second date, but no chance for a third, because Corporal Sandy Hinds was transferred to the signals unit in Darwin.
Coral received a hand-written letter from Sandy, telling her that he loved her, and that they might be able to pass messages to each other more easily soon. They corresponded frequently, and got to know each other by letter.
When Pappy Clark moved Central Bureau’s Advanced Echelon to Hollandia in 1944, four TypeX machines were transported as well, and four TypeX operators were selected to go with them. Since the Australian government would not allow women to leave Australia on military service, only male operators could serve in Hollandia. Corporal Hinds had been selected to be part of the Advanced Echelon TypeX unit.
Then, in early 1945, Advanced Echelon moved to the Philippines. The TypeX machines moved with it, and with the machines went Sandy Hinds.
Four TypeX machines arrived at Tarlac, and a small cipher unit was installed on the upper floor of the golf clubhouse in a room at the end of a long hallway, with Doug Pyle as TypeX mechanic, and four men from the Australian cipher unit, including Sandy Hinds, as operators. An American signals unit also moved into the building, with many more staff and machines than the Central Bureau contingent.
In contrast to the collegiate atmosphere at Central Bureau, the Australian and American units upstairs in the clubhouse never interacted. On one occasion, Doug Pyle walked into a room in the clubhouse occupied by the Americans and was ushered away at gunpoint by a scowling American guard.3
Coral and Sandy, separated by several thousand miles, were able to send each other messages on the TypeX machines by exploiting a security procedure.
Allied cryptographers believed that TypeX was probably secure, but they could not be sure. After all, the Allies had cracked the Axis’s high-level cipher machines, Enigma and Purple, so precautions needed to be taken, just in case.
Having cracked the Enigma, British cryptographers understood the weaknesses in such machines, and developed countervailing procedures accordingly. As we have seen, one entry point into an enemy code or cipher was finding patterns that were repeated over many messages, such as greetings or formalities at the start, or stock-standard weather reports. Another was to look at the length of the message, in case it provided a clue as to whether it was important and what it might be about. Both of these could be thwarted by making every message the same length, and by padding the start with nonsense. This became standard TypeX procedure.
The TypeX operators in Central Bureau’s garage and at other intelligence centres were told to type in nonsensical ‘free text’ before the start of each message. Then they were to type the letter ‘Q’ five times to indicate the start of the real message. All the messages were on loose sheets of paper that were arranged in stacks on tables and little desks around the garage.
Thus a TypeX operator might have to send a message that said, ‘Two enemy frigates have been identified near Java.’ The operator would first need to pad the message with nonsense to bring it up to the right length, so the message that she sent might end up being something like, ‘THE BLACK CAT CROSSED THE ROAD QQQQQ TWO ENEMY FRIGATES HAVE BEEN IDENTIFIED NEAR JAVA.’
In October 1944, Sandy wrote a letter and asked a friend to deliver it personally when he went on leave to Australia. In the letter, he asked Coral to marry him.
Sandy Hinds was the TypeX operator at Hollandia, and then in the Philippines, but, due to staffing limitations, he was both the sender and receiver of messages.
Soon after that, one of the women working with Coral received a short message from the Philippines addressed to Central Bureau. The message itself was relayed to recipients at Central Bureau, but the filler text — which she removed, as she always did — said ‘love Sandy.’ The TypeX operator brought it to Coral and said, ‘I think this is for you.’
Coral’s job was to encrypt messages, not decrypt them. She could not receive messages, but she could send them. Soon, an order came to send a message by TypeX to the Philippines. Coral entered the message into the TypeX, but added the following filler text at the beginning: LOVE CORAL.
She added five Qs to indicate that this was filler text and was to be removed before delivery to anyone important. And so they began a correspondence, using the filler space at the front of the enciphered messages to write short letters to each other. Sandy wrote that he was not well. Coral wrote that she missed him. When Coral became sick, her friend Joy, another operator on her shift in the garage, sent a message on the machines: CORAL IN HOSPITAL BUT RECOVERING.
A message came back from the Philippines with the filler: GOT YOUR MESSAGE ABOUT CORAL.
Once she was well again, they resumed sending messages to each other directly. These were always short and to the point, as they were aware that others could read them and scrutinise them. But the important people, the codebreakers and military commanders who received the official messages, never saw the abbreviated love letters between Carol and Sandy, because the filler text was always stripped away before being passed on.
Meanwhile, living in tents in the jungles of New Guinea took its toll on Corporal Hinds. He contracted amoebic dysentery, hookworm, hepatitis, and jaundice. The move to the Philippines did not help, because accommodation there was also a tent in a large military base. The food was poor quality. The war was nearby, and on more than one occasion, when Japanese bombers arrived overhead, he and his unit had to run and hide in nearby trenches to survive.
Corporal Hinds worked the TypeX machine as long as he could, but eventually his health deteriorated to the point where he could not work at all. He was discharged and sent back to Australia in May.
Coral received a message on the TypeX machine from the Manila cipher office: SANDY COMING HOME.
He arrived one evening at her barracks at Chermside, thin and sickly. They went for a private walk near the barracks, and he asked her to marry him as soon as possible.
Coral was given ten days’ leave to get married. The wedding was on 1 June 1945, in Orange, Coral’s hometown. The war had caused shortages and scarcities in everything, including wedding dresses. It was either impossible or prohibitively expensive to buy one new, so Coral wore a wedding dress that had recently been worn by three other brides. The next day, they travelled to the Blue Mountains for their honeymoon; by the end of the following week, Coral was again at work in the garage.
When Sandy Hinds came back from the honeymoon, he was sent immediately to a military hospital in Melbourne, where he remained until the end of the year. Immediately upon discharge from the army, Coral moved to Melbourne to be by his hospital bedside.
After the war was finished, and after Sandy’s health recovered enough for him to be discharged from hospital, they found a home in Melbourne’s south, started a family, and remained married until Sandy’s death in 2007. Coral, in her 90s at the time of the writing of this book, said that Sandy was ‘a wonderful man who gave me 68 wonderful years’.
44
All eyes on Kyushu
By early 1945, the war in Europe had turned in the Allies’ favour. The Normandy landings in June 1944 on D-Day had come at a huge cost of human life, but the Allies had established a beachhead and held it. Through that beachhead, Allied tanks and troops poured into France and pushed forward into Axis-occupied territory. On 25 August, Paris was liberated with the aid of the French Resistance, an underground movement that had waged a guerrilla campaign against Nazi control throughout the war.
From Africa, Allied troops launched a series of campaigns into southern Europe. They invaded Italy, one of the original Axis nations, and, on 5 June 1944, Allied forces entered Rome. In October 1944, they captured Athens. In Europe’s east, Russia and Germany engaged in a series of back-and-forth offensives, with the Soviet army gradually gaining the upper hand. As the new year began, the Allies continued to capture Axis nations, pushing closer to the heart of the Axis alliance — the German capital, Berlin.
In late January, the leaders of Britain, America, and Russia met secretly on the Mediterranean island of Malta. They knew that the Germans would resist the Allies’ advance every inch of the way to Berlin. Even so, it was clear that the Axis would be defeated in Europe within a matter of months. Berlin would be reached. When that happened, the focus would shift to Japan.
The Allied leaders decided to take the Japanese mainland using the same basic strategy that they had employed on D-Day — a massive landing force to establish a beachhead, followed by a land war. The invasion date would be designated as X-Day, with the landings to take place on the south coast of Kyushu, a large island in the south of Japan.1 They assigned overall command to General MacArthur.
Planning commenced for X-Day, which was scheduled to take place on the first day of November. The invasion plan was called Operation Downfall, and was split into two parts, Operation Coronet and Operation Olympic.
Operation Olympic would comprise an invasion force launched from the Mariana Islands, comprising 25 army divisions (each division having 15,000 men), 42 aircraft carriers, 400 destroyers, and 24 battleships, with well over half a million personnel involved.2
Operation Coronet, involving a force twice the force of Operation Olympic, would launch on 1 March 1946.
Strategists tried to estimate how many casualties might be incurred. Estimates varied wildly. Macarthur’s staff initially estimated about 105,000 losses within the first 90 days, but, as the date approached, other strategists provide much higher estimates. The more optimistic strategists thought there might be as few as a quarter of a million Allied deaths, while more pessimistic voices expressed the view that the total number could reach close to a million, with the estimated number of Japanese deaths ranging between one and five million.
MacArthur decided that Central Bureau should play a central role in the invasion. Its Kana operators had provided valuable intelligence in the invasion of Hollandia, Morotai, Leyte, and Luzon. He had come to believe that they were indispensable.
There were by now numerous proficient American signals units operating in the Pacific. From the original Special Radio Intelligence Company started by Howard Brown in Melbourne, and trained by Eric Nave and other experts at Central Bureau, the Radio Squadrons Mobile had been born. They would also play a role in the invasion, but MacArthur had developed a sentimental attachment to the signals-intelligence agency that he had created and over which he had total control. Preparations began for the wireless units at Tarlac and on Leyte to take part in the invasion of Japan.
Soviet forces entered Berlin on 21 April, and on 2 May located Hitler’s bunker, the final hiding place for Hitler and his closest personal friends. They discovered that Hitler had committed suicide, as had everyone else in the bunker, except for one person.
The remaining German defenders in Berlin surrendered on the same day, ending the war in Europe. In what became known as ‘VE’ day (‘Victory in Europe’), celebrations broke out across the Allied nations, and the Australian soldiers stationed throughout the Pacific were each issued with a bottle of beer. But the war raging across Asia and the Pacific did not stop; it did not even pause. From May onwards, Allied troops in Europe and Africa began moving in large numbers to Pacific bases.
On 1 April, one month before the capture of Berlin, United States army soldiers waded ashore onto the shallow, sandy beaches of Okinawa, the largest of the Ryukyu islands near the southern end of Japan. It was the first day in April, both Easter Sunday and April Fools’ Day. This landing on Okinawa was the biggest yet — bigger than Leyte, bigger than Normandy, bigger than Lingayen.
The islands of Okinawa, so close to the Japanese mainland, had become an enemy stronghold, yet the landing troops did not encounter much resistance. From the landing barges came troops, equipment, ammunition, and tanks. They swept across the northern part of the island and sections in the south, securing them easily.
An admiral radioed Admiral Nimitz with the message, ‘I may be crazy, but it looks as if the Japanese have quit the war, at least in this sector.’ Nimitz replied, ‘Delete everything after “I may be crazy.” ’
As they had done on Luzon earlier in the year, the Japanese army had decided to wait for the Americans to choose a landing site, let them land, and then mount a well-prepared resistance further inland. The Okinawan hills are riddled with caves, and the Japanese army used them as natural bunkers, in addition to concrete bunkers constructed under orders from General Mitsuru Ushijima
As US troops advanced into the interior, they ran into layer after layer of defences. The Americans and Japanese both suffered huge casualties in the Battle for Okinawa, ultimately losing 65,000 soldiers and 77,000 soldiers respectively.
A week after the first landings on Okinawa, American soldiers discovered partially burned codebooks in a burned-out pile of documents. They rescued the books from the still-smouldering embers and passed them to military intelligence. The discoveries turned out to be the code and additive books for Rikugun Angosho No. 5, the successor to the codebooks discovered at Sio. They were transported to the Philippines, where they were delivered to Central Bureau’s Forward Echelon at San Miguel.3
With radio signals coming in from East Asia, and the Japanese mainland loud and clear, Pappy Clark could monitor enemy troop movements from Central Bureau’s new home at Tarlac as clearly and as crisply as if he were perched above them. Similarly, Abe Sinkov’s Mainline code-breakers were putting the recently discovered Okinawan code books to good use, breaking open the latest high-level communications within Japanese army command.
By the middle of the year, the overwhelming theme in the traffic analysis had to do with a massive build-up of Imperial Japanese Army defences along the Kyushu coast. Units were being pulled out of China and other locations in East Asia, and redeployed to the Japanese mainland. It appeared that the Japanese commanders had correctly guessed the location of the impending invasion, and were digging in.
There were three Japanese armies stationed along the coastline by July, with a fourth one on the way. Deciphered messages revealed that an additional division was on Kyushu. The total number of defenders was estimated to be just over half a million.4 Then, in late July and the first days of August, as military historian Edward Drea put it, ‘New units sprang up seemingly overnight.’5
Suddenly, there were two more divisions on Kyushu, raising the estimated total number of defenders to 560,000. Two days later, there were another two divisions, raising the estimate yet again. Although it played a key role in these revelations, traffic analysis wasn’t the only source of information for the Allies about the defensive build-up in southern Japan. Ultra intercepts across various channels provided details about bunkers and other fortifications that were being constructed apace.
The defensive build-up in southern Kyushu that Central Bureau was detecting caused alarm among Allied commanders, who were already shaken by the severe losses incurred on Okinawa. Macarthur’s estimate of 105,000 casualties, based on his recent operations in Leyte and Luzon, was looking increasingly unrealistic. The possibility of a scenario similar to Okinawa, or Iwo Jima, where the defending Japanese inflicted heavy casualties, loomed as more likely. The early estimates had assumed a ratio of three attackers (Allied personnel) to every one Japanese defender, but the build-up meant that the ratio was now one-to-one.
Questions moved beyond the death toll to whether the operation would even be successful. The United States army chief of staff, George Marshall, asked MacArthur if he could change the invasion target, given the defensive strength along the southern Kyushu coast. MacArthur told him it was not possible, because the alternative sites were out of range of the airfields needed to provide air support.
As for the build-up, Macarthur did not believe the reports. He turned on Central Bureau as well as his own intelligence chief, General Willoughby, in a written response to Marshall, telling him he did ‘not, repeat not, credit the heavy strengths reported to you in southern Kyushu’. Intelligence reports, MacArthur claimed, often erred by overestimating enemy strengths. It suited him to say this, but while the intelligence he received from Central Bureau and elsewhere was not always accurate, errors tended to be underestimates rather than overestimates. Regardless, MacArthur told Marshall to ignore the intelligence reports, and insisted that preparations for Operation Olympic continue.
Meanwhile, word came through diplomatic channels that Japan might be open to a peace deal. Allied planners wondered what was going on.
Was Japan preparing for peace, or for yet another showdown? In fact, both were true. A rift was opening up between Japanese political and military leaders. Some politicians were hopeful of ending the war, but the military leaders were not interested. The Japanese army and navy had an alternative idea: they would prepare their defences and wait for the Allies to attempt an invasion, and then extract such a toll on them that they would be forced back to the bargaining table. The name of this plan was Ketsu-Go, meaning Decisive Operation.
According to historian Richard Frank, the idea was to make the costs of the war too great for the Allies to continue, driving them to the bargaining table:
The premise of Ketsu-Go was that American morale was brittle. Japanese leaders believed with great conviction that a great deal of bloodletting in an invasion of Kyushu would compel American politicians to negotiate out an end to the war on terms that the Japanese would find acceptable. Those terms essentially involved preservation of an old order in Japan, an old order in which the militarists and the imperial institution were dominant.6
While Japanese leaders argued about their strategy, Allied leaders met in the Berlin suburb of Potsdam to hammer out what sort of peace deal they were prepared to propose to Japan. After lengthy negotiations, they drafted a peace agreement, known as the Potsdam Agreement, that was conveyed to the Japanese government via mediators in Switzerland, a country that had remained neutral throughout the war. Word came through on 4 August that Japan had rejected the deal.
American code-breakers monitored Japanese diplomatic channels via their ‘Purple’ machines — their replicas of the Japanese diplomatic cipher machines, similar to the German Enigmas. It was clear that, behind the scenes, more was going on than the blanket rejection had revealed. There was some appetite in Japanese political and industrial circles for the deal, but the fact that it did not mention the emperor — leaving his fate open to interpretation — concerned them deeply.7
The United States army air force conducted air raids on Tokyo and other Japanese cities from China, the Mariana Islands, and newly acquired airfields on Okinawa. General Curtis LeMay, learning that many Japanese buildings were made of paper, ordered that incendiary bombs be used.
The Australian wireless units on Leyte and Luzon monitoring Japanese army air-to-ground radio traffic could tell that the Japanese pilots were struggling now. Col Brackley, a wireless unit operator on Leyte, recalled, ‘We were picking up pilots trying to get back to carriers and airfields when they were in strife. It was a toss-up whether they would make it or not. Sometimes we’d hear the messages as they got back to safety. Other times their signals would just stop when they were still miles out across the ocean, and we would hear no more from them.’8
45
The mushroom cloud
In a remote part of the New Mexico desert, the United States government had tested a powerful new weapon on 16 July. The atomic bomb, as it was called, exploited recent breakthroughs in theoretical physics to cause a nuclear chain reaction. It was far more powerful than any other weapon in existence — a single one of these bombs could devastate a city. The secret project to develop such a bomb, the Manhattan Project, had been given priority in part because the Allies believed that the two major Axis powers, Germany and Japan, were conducting research into atomic bombs themselves. It later turned out that this was true, but that the atomic-research projects of the Axis nations had stalled and were far behind that of the United States.
The massive, devastating explosion caused by the atomic bomb produced a column in the air with a high-altitude bloom that bore a passing resemblance to a mushroom. This formation became known as a ‘mushroom cloud’.
The United States president, Harry S. Truman, did not tell General MacArthur, who was busy making planes for the Kyushu invasion, or other military leaders, about the New Mexico test or even about the existence of the two bombs.
Nor did he tell them that he and his civilian military advisers were intent on using the bomb to end the war as quickly as possible. The strategists in Washington wanted to choose a location that would have maximum effect, to force Japan to surrender. They wanted to cause both a massive loss of civilian life as well as heavy military disruption.1 They wanted to demoralise the Japanese people.
Many factors went into the decision about the bomb’s first target, some of them arbitrary — including the site’s military value, whether there was reliable enough weather for the project to go ahead, and whether it contained a large-enough population hub.
Hiroshima and a handful of other cities were shortlisted. In terms of military value, one of the factors that made Hiroshima attractive to the strategists was that it was a troop-transportation hub. As one Japanese report said, ‘Probably more than a thousand times since the beginning of the war did the Hiroshima citizens see off with cries of “Banzai” the troops leaving from the harbor.’2
Hiroshima’s role as a water-transport hub had become clear ever since the breaking of the Water Transport code at Central Bureau, Brisbane, two years earlier. Decrypted messages, and above all comprehensive traffic analysis of the code, pointed time and time again to the location of water-transport headquarters in the Japanese city of Hiroshima. And so Hiroshima was chosen as the target.
General MacArthur and Admiral Nimitz were only informed two days before the bomb was to be dropped, and after the decision to do so had already been made.3
When they were told of the plan, the military leaders were horrified. MacArthur, Nimitz, and other military commanders advised the president not to use the weapon, but Truman had made up his mind.
The key components of the bomb were shipped from the west coast of the United States to Tinian Island aboard the United States cruiser USS Indianapolis. On the cruiser’s return journey, a Japanese submarine torpedoed the Indianapolis and sunk it; almost the entire crew of 1200 drowned.4
On 6 August 1945, a bomber named the Enola Gay left North Field, an airstrip on the tiny island of Tinian, one of the Mariana Islands, about 2500 km east of the Philippines in the Pacific Ocean. It flew six hours straight to Hiroshima, where it dropped a single bomb directly above the city.
To coincide with the Enola Gay’s journey, several wireless intercept sites in the Philippines, including Australian wireless units, were instructed to monitor Japanese air-to-ground traffic. They were alert for Japanese reconnaissance planes discovering the approach of the Enola Gay, in which case the aircraft would turn around and abandon its mission. Nobby Clarke, head of the air-to-ground solution section on Leyte, coordinated the activity. No such intercept was heard, and the Enola Gay reached the skies above Hiroshima without incident.
The atomic bomb, named ‘Little Boy’, detonated about 800 metres above the city with a devastating sonic boom, and caused a tall column to rise into the stratosphere, where it blossomed out into a wide, symmetric cloud. The bomb killed about 80,000 people, most of them instantly.
At San Miguel, Kana operators noticed that certain frequencies of Japanese radio traffic, usually buzzing with messages, suddenly fell silent. Then, after a lull, they resumed, with a flurry of urgent messages about the bomb that had been dropped on Hiroshima.5
The interceptors received numerous messages describing the details of the bomb.6 What soon became clear was that the military were trying to figure out how to defend against it. They discovered from eyewitness reports that the bomb had been dropped by parachute, which suggested that a defence against it might involve firing at parachuted objects. For example:
Imperial headquarters is investigating a new type bomb used by the enemy (Atomic bomb) and devising counter measures against it. Since this type bomb is dropped by parachute, each group will immediately issue orders to have the anti aircraft guns fire on all bombs dropped by parachute.7
The following day, 7 August, a statement was issued by President Truman, announcing that the United States had detonated an atomic bomb over Hiroshima. Detailing the long period of its development and the massive expenditure of resources it had involved, he described the new weapon as harnessing ‘the basic power of the universe’, saying, ‘The force from which the sun draws its power has been loosed against those who brought war to the Far East.’ Truman threatened to drop further bombs if there was no peace.
On 8 August, the Japanese foreign minister, Shigenori Tōgō, met with the emperor of Japan, Hirohito, and told him it was time to surrender.
The dropping of the bomb on Hiroshima stirred Russia’s leader, Joseph Stalin, into action. He was planning to join the war against Japan, but was now concerned that a peace deal was imminent and that Russia would be excluded from it. The Russian ambassador informed Hirohito that their nations were now at war. Russian forces immediately launched an attack on Manchuria.
On 9 August, the Japanese Supreme War Leadership Council held a crisis meeting to discuss the new developments: the new type of bomb, and Russia’s declaration of war. There was furious argument about whether Japan should surrender. While they argued, a second bomb was dropped on the smaller Japanese port city of Nagasaki. A second mushroom cloud ballooned into the sky above Japan, killing a further 45,000 people. This bomb, ‘Fat Man’, had been scheduled to be dropped on 11 August in the event that the Japanese did not surrender, but inclement weather was threatening the operation, so Truman had ordered it to be dropped two days early.
Despite the reality of atomic warfare and a new war with Russia, the leadership council failed to reach a consensus. The war minister and the heads of the Japanese army and navy were opposed to any surrender. They were planning to wait for the Allies to try to invade, so that they could fight a decisive battle on the homeland.8 The military leaders would agree to peace terms only under strict conditions, the central pillars of which were no occupation of Japan; Japanese control of disarmament; and Japanese control of any war-crimes trials. The conditions they had in mind effectively amounted to a truce rather than a surrender. Prime Minister Kantarō Suzuki left the meeting and met with his cabinet. At this meeting, too, no consensus emerged about a peace deal.9
Suzuki confronted Hirohito, telling him that Japan’s political and military leaders were deadlocked, and that he, the emperor, would have to choose a course of action.
A Japanese offer to accept the Potsdam terms was made on 11 August via Switzerland, although an amendment was inserted that preserved the ‘prerogatives of His Majesty as a sovereign ruler’. The wording was the result of a bureaucratic tussle within Japan over the precise crafting of the peace offer. President Truman rejected it. Given that Japan had no real bargaining power left, modifications to the agreement at this point were out of the question. On 14 August, Hirohito sent a message through Swiss diplomats, who were acting as intermediaries, that he would unconditionally accept the terms of the Potsdam Agreement.
46
A message for the emperor
On the morning of 15 August, a teletype machine was halfway through printing out a routine message at MacArthur’s headquarters in Manila when it switched to a completely different message. The teletype printed this:
STAND BY FOR IMPORTANT MESSAGE
****
FROM MARSHALL TO MACARTHUR
****
YOU ARE HEREBY NOTIFIED OF JAPANESE CAPITULATION
It was clear what this meant. It was the end of the war.
Outside the Manila headquarters, someone put up a hastily made sign saying, ‘War Is Over.’ Word quickly spread across Manila and throughout the military bases on the Philippines.
At the Central Bureau campsite outside Tarlac, a man ran through the tents, dressed only in his underpants, shouting, ‘The war is over! The war is over!’ He was initially met with gruff shouts of ‘Bullshit!’1 But soon the message arrived from other sources, proving him right.2 In America, President Truman announced that he had received a peace offer from the emperor.
The mood at the Central Bureau base at Tarlac was euphoric. Back in Australia, people came out into the streets, celebrating and cheering, giddy with relief and happiness.
There was a slight hitch, though. The war wasn’t actually over. Certainly, Hirohito had offered Japan’s surrender, but until the offer was acted upon, it was just words. Nothing had been signed. No order had been given to Japanese troops to surrender, or to come home, or to in any way stop the war. Japanese forces remained in occupied positions across Asia and the Pacific, in China, Singapore, Indonesia, Rabaul, and elsewhere. And, indeed, the Allies had not even accepted the offer of surrender. The opposing forces were still at war, and people were still dying.
A reminder that the war was not really over came that same day from an intercepted message from the Southern Army headquarters at Saigon to all headquarters under its command:
Although an Imperial Rescript concerning the acceptance of the Potsdam Declaration has been issued, you are to continue to execute your present duties and to continue to repel the enemy.3
Truman gave MacArthur the job of working out the details of the peace deal. The first step was to make direct contact with the emperor, and then to figure out the logistics of initiating the agreement. MacArthur didn’t want to drag this out for possibly several more days while envoys contacted the Swiss, who contacted the Japanese, who contacted the Swiss again, and so on. With both sides now agreed in principle to signing a peace deal, he needed to get in touch with the emperor directly, to bring the war to a formal end as quickly as possible.
That was not an easy thing to do, because both sides were, and had been for several years, engaged in total war. There were no direct lines of communication between them.4 MacArthur and his chief signals officer, General Akin, had anticipated this problem and had enlisted some expert help in the event of a peace deal.
Nine days earlier, the same day that the first atomic bomb was dropped on Hiroshima, a small plane had flown the short distance from San Miguel to Manila carrying a sole passenger, a man who knew more about Japanese radio circuits than anyone else in the Allied forces. The man was Major Stan Clark, or, as he was known to his colleagues, Pappy Clark. General Akin had summoned him on special duty to General Headquarters in Manila.
As soon as Hirohito’s message came through, General Akin gave Pappy Clark the order — make contact. He had at his disposal the radio operations room in Manila, staffed by American operators. The first thing he tried was to contact Radio Tokyo, which had handled American commercial correspondence before the war. Radio Tokyo was the obvious first port of call.
The message was sent on 15 August, the same day that the emperor offered peace. It was from General MacArthur, and it was addressed to the Japanese emperor, the Diet (the Japanese parliament), and the commander-in-chief of the Japanese military. It was along the following lines: ‘You will send envoys without side-arms [no swords] in aircraft carrying the following markings … and will arrive at [location]. You will be met by 535 Thunderbolt Fighters which will escort you to Clark Field, Manila.’
A master sergeant outside the operations room said to one of the operators, ‘Don’t snafu that message. I’ve got a kid brother up on the perimeter, and I’d hate to lose him now.’
There was no response from Radio Tokyo.
Pappy Clark came up with an alternative plan. They would broadcast on known Japanese army frequencies. They would first wait for two Japanese operators to open a communication channel, then interrupt the exchange by broadcasting the message for the emperor. It would be sent in Japanese using Japanese signal procedures, including Japanese Morse code — the Kana code.
Clark tried circuit after circuit, but to no avail. Each time he broke in with the message, the Japanese operator would say to the recipient, ‘There’s some station interfering with us. I’ll call you later.’ They invoked standard procedure as if Clark’s message was nothing but radio noise.
At eleven o clock that night, Clark suggested that they give up on the military circuits and try the diplomatic ones. Eventually, at five in the morning, the Japanese diplomatic station at Singapore acknowledged the message and advised that they open up a schedule of communication at Radio Tokyo. The Singapore operator provided a call sign and frequency for Clark to communicate with them on.5
Once communication had been established, the Japanese immediately organised for a secret delegation to travel to Manila to plan the details of the peace deal.
The intercept base at San Miguel was near Clark airfield, and Jack Bleakley recalls many of the Central Bureau and wireless operators standing outside, scanning the skies, looking for the plane from Japan.6
The Japanese delegation returned home the next day, having been told that the conditions for peace were Japan’s unconditional surrender. The victory in Europe had been dubbed VE Day, and the day of Japan’s surrender was known as VJ Day.
A plain-language message came through on high-grade Japanese frequencies. Robert Christopher, the senior linguist on duty, recalled later:
We knew the message had to be important because it was sent in the clear. It contained words that I had never seen before. For instance, the word Chin. Eventually I found the right dictionary and discovered that Chin stood for I. The only person allowed to use this particular expression was the Emperor of Japan. I was translating Emperor Hirohito’s 14 August message telling the troops to surrender.7
The Australian government took a particularly hardline view on the peace deal, and was opposed to any leniency toward the emperor. The new Labor prime minister, Ben Chifley, made no official pronouncement, but other cabinet ministers did, including the foreign minister, ‘Doc’ Evatt.
The Mercury newspaper reported, ‘The view taken by senior ministers is that Emperor must be regarded as a war criminal, that he should not be immune from trial, and that he cannot be regarded as a mere puppet — a suggestion that has been put forward in some quarters.’8
General Akin later contacted Pappy Clark with an offer from ‘The General’, meaning General MacArthur: ‘The general wants to know if you will accept a direct commission into the United States army. Come to Japan with us to help police internal communications.’
Clark declined. ‘I’ve had enough,’ he told Akin. ‘I want to go home.’
General Akin then said, ‘What can we do to help you get home?’ Clark said, ‘You can probably do a lot.’9
Pappy Clark was given a priority flight back to Australia. He arrived in Brisbane on 9 September, only a week after the official signing of the peace agreement on the Missouri in Tokyo Bay.10 The rest of the Central Bureau Advanced Echelon remained stranded in the Philippines for several more weeks.
47
Peace in the Pacific
In the days following the formal surrender, Japanese soldiers arrived at Allied military bases across Luzon, including the Central Bureau camp at Tarlac. Tired and hungry, these were the scattered remnants of the Japanese occupying force left behind when the Imperial Japanese Army had retreated and abandoned the Philippine islands. The presence of Japanese soldiers hiding in the nearby hills had been known for some time, but it was not until after the surrender, when they emerged from hiding, that it became apparent how many of them there were.
Central Bureau continued to operate in the days following Japan’s surrender. Many intercepts provided insights into the reaction of Japanese forces around the Asia-Pacific region to the order that they surrender. Many were devastated by the news. Some commanders initially refused to surrender, and indeed some of them simply did not believe that the war was over.
To inform the Japanese troops strewn across Asia and the Pacific that the war had ended, Allied planes dropped leaflets on them instead of bombs. These were not always effective. After one such leaflet bombing-run, an army radio broadcast was detected on 24 August from the Japanese army base commander at Kokas, a Papua New Guinea village in the Owen Stanley Range, to his subordinate units:
(a) The enemy has been distributing propaganda leaflets — as expected they are cunningly phrased … This is no more than an enemy trick … the sugar-coated thoughts will turn out later to prove our expectations reversed … take particular care to guide the thinking on this point to all those under your command.
(b) Take precautions to prevent Allied aircraft spotting our troop dispositions.1
An intercepted directive from Tokyo gave instructions to Japanese officers in charge of prisoner-of-war camps that Allied prisoners of war should all be given appropriate rations, and that they must no longer be put to work.2 The clear implication was that some Allied prisoners of war had not been getting their food rations, and that they had been put to work as slave labour. This was one of the first glimpses of the scale of the mistreatment of prisoners of war in Japanese internment camps. The full picture emerged as Allied prisoners were freed across the Asia-Pacific.
One of the most notorious examples was the Burma Railway, a rail line built by Asian civilian labourers and Allied prisoners of war under appalling conditions of forced labour.3 Thousands died, including 2815 Australian prisoners of war. Other intercepted messages, particularly in the Mainline army code, revealed attempts by Japanese army commanders to cover up any evidence of their war crimes. For instance, the Southern Army headquarters at Saigon issued the following directive, which was decrypted by Central Bureau:
1. Work involving the use of prisoners will cease.
2. Food, clothing, sanitation, lodgings will be same as for Japanese troops.
3. We would particularly like to count on harmony in executing this.
This was clear evidence that prisoners had been mistreated, but the intercepted message, being a product of Ultra, would never be revealed in a war-crimes trial as evidence of wrongdoing, because the Allies intended to keep Ultra secret even after the end of the war.
American forces poured northwards from the Philippines for the occupation of Japan. Clark Field, the major military airfield near Manila, hummed with activity as planes departed, packed with troops and equipment, while others returned empty, ready for another load.
Mic Sandford, returning from a high-level delegation to Tokyo, encouraged Central Bureau personnel to find a way to get to Tokyo if they could and see it for themselves. This was not an official order, and he did not make transportation available. He merely promised — or hinted that he was promising— to turn a blind eye.
Australians from Central Bureau did travel to Tokyo, while others visited China and Korea by hitching rides on United States planes. Two of those who went to Tokyo were Gordon Gibson and Brian Walsh. According to Gibson, it was Walsh’s idea.
Gibson was reluctant, but Walsh explained that if they went soon they might beat MacArthur to Tokyo (which was what in fact happened). They applied for and were granted one week’s leave, headed to Clark Airfield, and asked around for a ride among the various crews preparing to leave. Eventually, they found one that obliged.
‘Come aboard,’ the soldiers said to them. ‘There’s room.’
The plane, a DC3, was not a long-range aircraft, and it stopped twice on the way to Japan. One stop was on the deck of an aircraft carrier, where the sailors, no longer preoccupied by war, were playing a game of baseball.
The second stop was Okinawa, from where they had to find a further ride to Tokyo. They eventually found a flight willing to take them to Tachikawa, an airport just outside Tokyo. They were awestruck by the spectacular view of Mount Fuji as the plane descended to the airfield.
From Tachikawa, they caught an electric train to Tokyo. The landscape was a blackened desert, the suburbs between the two stations having been annihilated by the Allied bombing. There were places where not a single tree or building had been left standing.
Gordon Gibson recalled his visit many years later, saying, ‘A lot of Tokyo was absolutely flattened. They’d bombed the hell out of it. That was in certain areas. I really hope it was strategic bombing. I don’t know how much of it was.’
They arrived in the heart of the city to find that central Tokyo was relatively undamaged. There, they found lodging at a newly established United States army barracks.
At a cafeteria set up by the Red Cross, they met up with some American soldiers, and were joined by some young local ‘geisha girls’. As Gibson recalled, ‘We sat at the table with a beautiful-looking Japanese lass all done up in her kimono, and chatted away with her — just enjoyed the coffee. Her name was Kazumi Shirakawa. Nothing went any further than that. It was all very proper and in hindsight it was quite amusing.’
They then met a group of Japanese people who were out drinking, and who invited the two Australians to join them. After drinking with the group, the pair made their way back to the American base, where they slept on the floor.
Brian Walsh, exploring the city on his own, met a man who introduced himself as Kim Sato, the editor of the Nippon Times, Japan’s only English-language newspaper. Sato had approached Walsh because his distinctive slouch hat identified him as being Australian.
Sato took Walsh for a drink at the Imperial Hotel opposite the palace, where they talked about motorbike dirt-racing and the latest exploits of the Australian motorcycle racing champion, Lionel von Prague — a topic that Sato was surprisingly familiar with. From there, Sato invited Walsh back to his home, where he introduced the Australian to his wife and son.
After staying in Tokyo for a few days, Gibson and Walsh’s welcome at the United States army barracks was wearing thin, and their leave was running out. They caught the train back across the charred wasteland to Tachikawa airfield. From there they hitchhiked on an United States army air force plane back to Clark Field, and from there back to Tarlac.
By the time they returned to camp, they had been gone for longer than a week, and their disappearance had been noted. On their return, they were charged with being AWL — absent without leave. They were ordered to report to a disciplinary officer. When they did so, a young man in the office berated them for their insubordination.
But they noticed that in the room was also Keith ‘Zero’ Falconer, the man who had been in charge of 3 Wireless Group in the Northern Territory when Gordon was there. Falconer recognised Gibson, gave a wink, and had a quiet word to the disciplinary officer. The charges were soon dropped, and they heard no more about it.4
The surrender agreement was signed by Allied and Japanese representatives on board the USS Missouri in Tokyo Bay on 2 September. MacArthur was appointed supreme commander of the Allied powers, with complete authority over the Allied occupation of Japan. Upon taking the position, he granted immunity to the emperor and his family from war crimes’ prosecution. Four weeks later, MacArthur created the Civil Communications Section to demilitarise and modernise Japan’s communications networks, and appointed General Akin as its chief.
A small contingent from Central Bureau relocated to Tokyo. This was the organisation’s final move, having started in Melbourne and successively relocating to Brisbane, Hollandia, Leyte, and Luzon. Americans in the Tokyo operation included Abe Sinkov and Joe Richard. Australians in the contingent included Major W.E. ‘Nobby’ Clarke, Wing Commander H. Roy Booth, and probably also Professor (now Colonel) Room.5
They were there to visit and study Japanese cryptological centres, to unravel the final secrets of the codes they been living with for the past three years. From a cryptological point of view, there was still valuable information to be obtained from doing this. They planned to find out how and where the codes were created, maybe learn where the Allies had gone wrong or could have made faster progress, and generally resolve various lingering questions.
They arrived at the building where the Japanese cipher unit had worked, but there was no equipment on site. There were no codebooks, no manuals, no Purple or Red machines. Nor were there any personnel who seemed to know anything about codes. Certainly, there were staff inside the building who claimed to be cryptographic personnel, but it quickly became apparent that they knew nothing about codes. The impersonators knew so little about cryptology that they could not even lie well about it. In one interview, Sinkov asked how they enciphered high-grade messages. Having spent the past three years studying such messages, Sinkov had a pretty good idea of how this was done, but wanted to ask anyway. The staff told him that they used to throw pencils into the air, and would then look at the patterns the pencils made as they landed, to use the patterns to make codes.
‘It was ridiculous,’ Sinkov said later.
The team also wanted to find out about Japanese code-breakers, and how much luck they had had against Allied codes. But although they knew there was some code-breaking going on in Japan during the war, they could find absolutely no trace of any such organisation.
Many documents were destroyed across Japanese institutions after the surrender, including all records of special signals organisations. Nothing remained. Whatever story there was to tell about code-breaking from the Japanese side of the war, it was erased.
48
Radio silence
Around Tarlac, the roads were no longer military highways thrumming with the movement of troops, and were becoming once again the quiet back roads used by local farmers and townspeople. The vast populations of American personnel who had been amassed for Operation Downfall departed for Japan or returned on transports back to the United States.
Central Bureau stayed at Tarlac, remaining operational to monitor the Japanese demobilisation, but before long there was little radio traffic to intercept. It was a welcome change from the relentless 24-hour cycle during wartime — but as the intercept operations wound down, nobody had anything to do. The senior officers tried to keep them busy by putting them to work on non-signals-related tasks such as helping army engineers to build a road.
Three of the traffic analysts — Bill Rogers, Hugh Dunn, and Peter Elkin — came up with the idea of starting a newspaper. All the Australians stationed in the Philippines were starved of news from home. But the intercept stations were still operating, and could be tuned to pick up news broadcasts from afar, which could then be used in their newspaper. The three of them met with Mic Sandford in the control hut, and explained the idea to him. He replied that he didn’t have a problem with it, and gave his permission. And so the newspaper Union Jack Up! was born, providing news for the campsite. Union Jack Up! came out once a week. There were six issues published, which they distributed by driving around the campsites in a jeep.1 Bill Rogers kept a copy of each issue, donating them many years later to the Australian War Memorial, where they can be viewed today.
On 9 October, the Australian contingent at Tarlac departed for home, boarding the liberty ship, Francis N. Blanchet, a ‘rickety craft with cracks in the hull’.2 The ship was in poor shape. Additional toilets had been hastily fitted out to accommodate the large number of passengers, but their construction was unfinished. It was the only transportation back to Australia, so they all crowded on board. A few lucky individuals had alternative transport: some members of 6 Wireless Unit scored flights home with the RAAF, and a small Central Bureau contingent were selected for an assignment in Japan.
In addition to the Australians at Tarlac, there were over one thousand United States army personnel, many of whom were also stranded as they awaited demobilisation. Before departing for Tokyo, Abe Sinkov gave the remaining Americans at Tarlac an assignment to keep them busy. He instructed them to collectively compile a book about their wartime experiences.
Many of them had collected newspaper cuttings and taken photographs of the various locations; these were contributed to the work, which was titled SIS in the Far East (or alternatively, Special Intelligence Service in the Far East, named after the United States army component of Central Bureau).
The book was printed the following year in New York, with the author and publisher listed as ‘SIS Record Association’, an organisation that did not exist. It was full of photographs of the various locations that Central Bureau had operated in, and of some of the people involved, peppered with light-hearted anecdotes about those places.
The book was unclassified because it revealed no wartime secrets. A casual reader of SIS in the Far East would have no idea what the American personnel of SIS had done in these exotic places. There was no mention of code-breaking or intelligence. It did not even mention the Second World War.
In Brisbane, most people employed by Central Bureau were demobilised in August and early September, except for twelve personnel who remained under Eric Nave’s command. Professor Room and Judy Roe were two of the twelve. Nyrambla was to be returned to its owner, so Nave relocated the remaining staff to Mic Sandford’s former residence at Eldernell Terrace in Hamilton.
Their final task was to compile a set of reports on the activities of Central Bureau, which they would name the Central Bureau Technical Records. In a way, this complemented the book being compiled by the Americans in Tarlac. While SIS in the Far East gave no account of signals intelligence, the Central Bureau Technical Records were only about the signals intelligence.
Of the eleven reports, five dealt with the different kinds of code systems they had attacked: army air-to-ground; naval air-to-ground; three-figure; weather; and Mainline codes (which included Water Transport). There was a report on traffic analysis; on the field sections; on studying code books; and on organisational processes, and there was an overview. The final report was a critique written by Mic Sandford and co-signed by Roy Booth and Abe Sinkov.
The IBMs at Ascot Fire Station were silent, with the technicians led by Zach Halpin preparing to return to the United States. Despite the efforts of both Mic Sandford and Professor Room, the machines could not stay.
Sandford had written to the secretary of defence in March urging the government to approve the purchase of some of these machines, so that the Central Bureau Australian component could be ‘entirely self-contained and independent of US resources’. He acknowledged that Australia lacked the expertise to operate the machines, but suggested that the government ask Canada to lend it some suitable personnel to assist:
It is proposed to approach the Canadian government through the Department of External Affairs, to ascertain whether they would make available to Australia the necessary personnel, to attachment to the AMF [Australian Military Forces], for as long as Australia requires them.
Sandford helpfully attached an itemised shopping list, which included several machines as well as 40 million punch cards.3
At about the same time, the director of military intelligence, John Rogers, wrote to the chief of the general staff supporting Sandford’s position, and further emphasising how important it was that Australia invest in this new technology.
The acting minister agreed on two conditions; first, that the exact cost of purchase be established, and second that he could be assured that Australia would have the expertise to use them. Zach Halpin confirmed to his Australian Central Bureau colleagues that, sure, he would be happy to train the Aussies to use the IBMs. This news was relayed to the chief of the general staff in May.4
Professor Room was particularly intrigued by the machines and their ability to perform computations. He made enquiries about the possibility of Sydney University acquiring one or more such machines — perhaps those being used at the fire station, if the United States army had no need for them.
From a bureaucratic point of view, the machines were now orphaned. The United States army had written off their large investment in Central Bureau, Brisbane. Their office complained that they had originally expected to spend about 150,000 Australian pounds on equipment, but had instead spent over 2 million pounds, the biggest expense being the IBMs.5 The money had been spent without going through proper channels, and the United States army didn’t have an inventory of what they’d sent. In theory, they could employ clerks to search through archives of receipts and purchase orders, but this itself would be arduous and time-consuming, so they wrote the equipment off as a loss.
It initially looked like the IBMs would be available for free, but, once the Australian army realised the cost of staffing and running the machines, they lost interest. The minister asked General Blamey for his opinion on the IBMs. Blamey suggested, rather cleverly, that the army go along with Sandford’s request for the time being, keeping in mind that it had the right to cancel the deal once the war was over.6
On 23 August, just over a week after Hirohito’s announcement of surrender, the assistant director of military intelligence, Robert Little, fired off a brief memo:
In view of the recent developments in the war with Japan, it is desired to inform you that the International Business Machines and the operatives therefore will not be required now.7
Sandford didn’t give up. From Luzon, he wrote to Little:
Irrespective of whether or not IBM machines are obtained for use by CB, it is considered absolutely essential to obtain all possible information about the development and application of these machines without delay.
Professor Room has already suggested that the mathematics staff of Sydney University would be interested in technical investigations in this connection, and Major Z Halpin, US Army, would be pleased to offer his services for this purpose.
It is hoped that Major Dennis Ayre (Australian Corps of Signals) will shortly be joining this unit with the purpose of making a study of IBM machines, and it is suggested that with Major Halpin’s co-operation, it may be possible to undertake a course in the maintenance and technical exploitation of this machinery. The installation at Brisbane will be available for use in this connection. It is understood that the US Army will be happy to give us all the assistance they can. 8
Thomas Room’s dreams of organising a training course from the Americans before they departed, so he could understand the machines, were not to come to fruition, despite Mic Sandford’s enthusiastic support.9
Three weeks after Sandford’s letter, Little received a memo from Army Landforces:
ALL IBM MACHINERY NOW REMOVED FROM CENTRAL BUREAU BRISBANE10
Meanwhile, army engineers dismantled the huts at Ascot Racecourse and cleared all the equipment out of Nyrambla, including the TypeX machines in the garage.
At the end of November, the RAAF’s head of intelligence, Group Captain Julius Allain Cohen, convened a secret meeting at Brisbane land headquarters to discuss the disbandment of Central Bureau. Mic Sandford and Roy Booth had both arrived in Brisbane two weeks earlier from Manila via Morotai. Lieutenant Colonel Robert Little had flown from Melbourne for the conference. Eric Nave, currently in charge of Central Bureau’s Brisbane operations, was there to report on the completion of the technical records.11 Wing Commander H.W. Berry of the RAAF was also in attendance. A notable absence was John Rogers, as he had stepped down from his role as Australia’s director of military intelligence a week earlier to return to civilian life.
Minutes of the meeting were taken, but unfortunately only the first page has survived, and it is rather tattered.12 It reveals that the group decided to destroy all of Central Bureau’s records, keeping only the technical records that Nave’s team had recently completed and 50 Ultra decrypts that might be useful for ‘training purposes’ in the future. Nave was assigned the task of selecting 50 suitable messages, and destroying the rest.
It is clear from other official records that the senior officers at the meeting also assigned Nave the task of delivering the technical records by safe hand to the army’s archives in Melbourne. Nave delegated this to Geoff Ballard, who was not at the meeting but had returned from Colombo and was now stationed nearby at army headquarters in central Brisbane. Ballard, no doubt, would have appreciated a work assignment to visit his hometown.13
The intercept units had mostly been disbanded, with some units still in transit to Australia from Morotai, and had not yet been discharged. They decided to keep a single wireless unit going, which would be named 1 Special Wireless Unit, to be based in Darwin, so that the Australian army could maintain and develop the expertise in radio interception acquired during the war.
The identity disks also had to be disposed of, because they referred to units and organisations whose existence was still secret. After the cessation of hostilities, the disks had been collected from personnel stationed in Central Bureau’s sigint and intercept units across the south-west Pacific. Their disposal was Nave’s responsibility, too, as the operational officer in charge. Once that was done, Central Bureau would cease to exist.
A few individuals, including Geoff Ballard and Don Inglis, were retained by the army and were transferred to the Directorate of Military Intelligence, which was now at Albert Park Barracks — the same location as FRUMEL’s home after it was evicted from Monterey Flats.14
On a hot Queensland December day, Nave incinerated all the records, saving only 50, as he had been instructed to do. Thousands of documents went up in flames; load after load of paper wads, all meticulously saved and filed over the previous three years.
Nave faced the question of what to do with the disks. They could not be burned; they would have to be buried. But where?
The AWAS barracks at Chermside, where many of the women at Central Bureau had lived, was now vacant. The main barracks to the north had been converted to a demobilisation area for troops as they arrived in Australia prior to discharge, but as Australian women had not been allowed to serve overseas, the smaller women’s barracks to the south had not been given a similar interim role. Access to the barracks was by a dirt road that passed through a gully and across a cattle-grid, which would have given Nave confidence that the location was sufficiently rural and remote for his purposes.15
The suggestion to dispose of the name tags at the barracks most likely came from Annette Steele, an AWAS sergeant who had helped compile the technical records at Eldernell and was one of the few remaining residents at the Chermside women’s barracks.
Nave, and probably Steele, as well as kana operator Clarrie Hermes, another Central Bureau person still retained, took the identity disks to the Chermside barracks, dug a hole in the ground, and buried them.16 Those who had not yet been discharged were transferred elsewhere for demobilisation.
Nave assigned Hermes to clean out Eldernell and organise its return to its owner. Hermes had been in signals intelligence since 1941, when he was trained by Jack Newman as a Kana operator and sent to Darwin. He claimed later that, as he was also physically the last person at Eldernell, he was, in a sense, the ‘first and last’. Hermes cleared out the remaining equipment, locked the door, and headed for Melbourne.17
1946 and beyond
49
Secret medals
About a year after the end of the war, the US army informed the Australian government that it wished to award three Australian officers with the Medal of Freedom with a Bronze Palm for their services in the Second World War. The Medal of Freedom is an award given by the US secretary of state to individuals who are not members of the United States armed forces, the Bronze Palm being an additional honour. The medal recipients were to be Roy Booth, Mic Sandford, and Stan ‘Pappy’ Clark. Usually, a medal would come with a citation — a short explanation of why it had been awarded — but in this instance the US army said it could not provide citations for secrecy reasons.1
This caused a flurry of activity in Canberra, where bureaucrats discovered that they had scant operational records for the three proposed recipients, who had not been part of the normal chain of command. Without some knowledge of what the code-breakers had done, and in particular without an assurance that the men had been engaged in ‘operational’ service, the bureaucrats could not agree to the awarding of medals by a foreign power. Stewart Jamieson from the Department of External Affairs (himself a veteran intelligence officer who had served in New Guinea, although for a different organisation), pursued the matter with both the army and the air force, trying to find out exactly what these men had been up to during the war.
In regard to Booth, the air force replied in September:
I desire to advise that, during the whole of the period in question, Wing Commander Booth was engaged in intelligence duties and no exact records of his movement are available in this department. It is known that, in January 1943, he was serving in New Guinea for two periods totalling two months approximately, and that during 1944 he was in New Guinea for a further three weeks approximately. In September 1944, he proceeded to the United Kingdom, returning in January 1945, and in March 1945 he visited the Philippines for approximately two weeks.2
That was a patchy description, at best, but the air force admitted that it did not have the whole picture of Booth’s military service. The air force added that it believed Booth was on ‘operational’ duty, which qualified him for a medal.
As for Sandford and Clark, the army was able to dig up the various places where the two had served, but had no details at all about what duties they performed there, or whether they were ‘operational’. Under these circumstances, Jamieson concluded that the Australian government could not let them accept the medals from the United States.
He wrote:
It is noted that citations for the proposed awards has been withheld for security reasons. Consideration has been given to the proposal in consultation with the Australian army and air force authorities, but, in view of the policy of the government that only operational awards be accepted for bestowal on members of the Australian forces, it is not possible to finalise these cases, in the absence of the citations containing the particulars of the services for which the awards are proposed.3
This forced the Americans’ hand. They provided three citations, which gave a one-page synopsis of what Sandford, Clark, and Booth had done at Central Bureau.
Yet even this did not settle the matter, because the citations, being brief and general in nature, still did not say whether the achievements had been operational. Jamieson wrote once more to the secretary of defence about the matter on 17 April, suggesting that the secretary ask the men themselves about their wartime movements:
I cannot at the moment think of any way of discovering such movements other than direct discussion with the persons concerned. In the difficulty which faces you in this respect, my own suggestion would be (and I put it forward with some hesitation since the matter is not one really of my concern) to approach them themselves on this matter. They were all senior officers who are not likely to be misleading on the subject of their own movements and there is no necessity to raise with them the reason for your enquiries.4
There is no record of whether the acting secretary, Major P. E. Coleman, took Jamieson’s advice to ask the men personally about their wartime movement, and if he did, what he learned. By that time, Sandford, Clark, and Booth had all been discharged and had returned to civilian life, but all three would have been easy for Coleman to locate. Pappy Clark was once again working in commercial radio in Melbourne; Roy Booth had recommenced his legal practice in Crow’s Nest, Sydney; and Mic Sandford was embarking on a new career as a lawyer in London.
At any rate, Coleman decided that no medal should be awarded to Roy Booth or Pappy Clark. For unknown reasons, Mic Sandford was the only one of the three to be given a Medal of Freedom, but it was downgraded so as not to include the additional honour of the Bronze Palm. In keeping with the secrecy provisions surrounding the award, Sandford was provided with the medal, but it came with no accompanying citation to explain why he received it.
A week after Jamieson sent his suggestion to Coleman, he received another letter, this one from the United States navy. It said:
The work of fleet radio unit, Melbourne, was not highly publicised because of its secret nature, but the results obtained were of immeasurable importance in the successful prosecution of the late war. The successful accomplishment of its mission was in no small measure due to the unfailing devotion to duty of the 318 personnel of the armed forces of Australia who were attached to that unit and who worked side by side with the United States Personnel.
It is requested that you express to the Government of Australia the appreciation of the Navy Department for services of the Australian personnel attached to the Fleet Radio Unit, Melbourne.5
Jamieson forwarded this on to Coleman as well, noting, ‘It does not appear to require any action unless it is decided to convey its contents to those concerned.’ No action seemed to be required of Australia in regard to this letter, and he had already replied to the US navy, acknowledging receipt.6 Coleman filed the letter, and took no further action.
The Australian personnel of FRUMEL who had worked around the clock for the duration of the war, and many of whom had assisted in vital code-breaking activities during the Battle of Midway and the Battle of the Coral Sea, were never informed that the US navy had written to thank them for their service. For that matter, the Royal Australian Navy did not take it upon itself to thank them, either.
Only one Australian code-breaker was given an official honour by the Australian government after the cessation of the war. In March 1947, Geoffrey Ballard was gazetted as having been ‘Mentioned in Dispatches’ for ‘Exceptional service in the field in S.W.P. Area’ (the south-west Pacific area). This was the belated result of Mic Sandford having nominated him for that honour in 1945. The nomination had missed the lodgement deadline of August of that year, causing the two-year delay.
Mic Sandford had also recommended Don Inglis, who had been Pappy Clark’s second-in-command of traffic analysis, for a Mention in Dispatches. But, despite high praise from Sandford as having ‘quickness and originality of approach combined with great skill and experience’, Inglis was denied the honour.
Eric Nave, who had founded the diplomatic section and had broken the air-to-ground codes, was overlooked, partly because Mic Sandford declined to nominate him. When nominating Inglis and Ballard in 1945 for Mentions in Dispatches, Sandford wrote to Little, expressing his ambivalence about Nave:
I have recommended nothing for Nave since I thought I should like to have your advice on this point. Technically, Captain Nave is the by far the most brilliant officer in the unit, but he is so lacking in initiative and appreciation of changing operational requirements of our forces that his efforts must be constantly guided by Major Clarke or myself. One cannot forget however that it was he who ‘broke’ the first Japanese cipher code by any of the military units during this war, and on that account alone he should perhaps be specially considered. Can you give me your advice on this.7
Little’s response to Sandford’s request for advice isn’t on file, but Nave was not recommended for any honours.
In contrast to the Australians, whose achievements were almost universally unrecognised by their own government, the American code-breakers were lauded. Abe Sinkov helped establish the National Security Agency in 1949 (originally named the Armed Forces Security Agency), and became its first head of communications security. Later, when the NSA founded a Hall of Fame for great figures in the United States’ intelligence community, Sinkov was one of the original set of eight people inducted.
Joe Richard, who had played such a critical role in the breaking of the Water Transport code, was awarded the Legion of Merit before the war was over, was later awarded the NSA’s Cryptologic Service Award, and was inducted into the US army’s Military Intelligence Corps Hall of Fame.8
Clarence Yamagata, the Nisei translator from the Philippines, was awarded the Legion of Merit.
The achievements of United States navy personnel at FRUMEL were widely recognised within the American military establishment, although the lion’s share of the glory for the breaking of JN-25 and the resultant victory at Midway went somewhat unjustly to their sibling unit in Hawaii, FRUPAC. Even so, their contributions were recognised and honoured. Rudy Fabian and Ralph Cook were awarded the Legion of Merit. Duane Whitlock was awarded the Bronze Star for his code-breaking contributions in relation to the battles of Midway and the Coral Sea.
Only Eric Nave, despite no official commendation, attained some level of public recognition for his wartime experience, and even for him recognition, was slow in coming. In the post-war years, Nave helped establish Australia’s post-war security agency, the Australian Security Intelligence Organisation, becoming one of its assistant directors. With the declassification of some intelligence activities thirty years after the war, he became known as an accomplished Australian code-breaker. When he died, he received an obituary in The New York Times — although, sadly, it focussed overly on his association with Rusbridger’s book Betrayal at Pearl Harbor.
The Australian code-breakers had to wait until 2011 for official acknowledgement of their service, and even then it came from Britain. Every living Australian who had served at Central Bureau, in ASWG, or at FRUMEL received a Bletchley Park medal from the British government, along with a signed letter of thanks from the British prime minister.
By then, only a fraction of those who had participated were still alive.
50
Return to Nyrambla
On a crisp, cloudy winter morning in July 2015, people mingled in the gardens of Nyrambla. Many were in uniforms of various kinds: some were army, some navy, some were Australian, and some United States — a variety of services reminiscent of the mix of personnel who worked there seventy years before. Most of them were very young.
Caterers set up trellis tables along the driveway, covered them in tablecloths, and laid them with salmon sandwiches, scones, and coffee urns. The invitees arrived and took their places in the seats under a marquee. The front row was reserved for a small number of special guests: Australians who had worked for Central Bureau during the Second World War.
There were eleven of them: Helen Kenny (Garage Girl TypeX operator and supervisor); Colin Brackley (Kana operator with 1 Wireless Unit); Gordon Gibson (ASIPS with 51 Wireless Section, and also spent some time as a traffic analyst at Nyrambla); Bill Rogers (traffic analyst, and the son of the director of military intelligence, John Rogers); Frank Hughes (traffic analyst); Madeline Chidgey (TypeX operator); Diana Parker (Garage Girl); Clarrie Millar (of 3 Wireless Unit); Mac Jamieson (1 Wireless Unit); Dorothy Morrow (Central Bureau as well as the Australian Advanced Land headquarters on the University of Queensland campus); and Ross Gwyther (whose role in Central Bureau is not known). All aged over ninety, they were the stars of the event.
Other guests included spouses and children of Central Bureau people who had since died; local historian Peter Dunn; members of the Chermside historical society; and journalists. The Australian Signals Directorate had organised a plaque, commemorating Central Bureau, to be placed at the front gate of the house.
Steve Meekin, Australia’s deputy secretary for intelligence, stood at the front of the marquee and gave a speech in honour of the veterans of Central Bureau.
Central Bureau was first established in Melbourne. When MacArthur moved General HQ to Brisbane in September 1942, Central Bureau followed and set up operations at Nyrambla and the former fire station in Kitchener Road.
Depending on rank and service Central Bureau staff lived in camps at Chermside, Yeronga and at the racecourse. Others were billeted with local families.
But it was here that the Bureau really began to come into its own. In this house and the garages at the back. For it was the garages that housed the forerunners of what today we would call computers.
A mix of experience and inexperience, mainly very young and very talented. The men and women of Central Bureau were dedicated, idealistic and determined.
There was a tremendous spirit of camaraderie based on shared tasks and what we call today the mission imperative, as the veterans’ accounts attest. Rank and national identity took second place to getting the job done: which was — through the timely provision of signals intelligence, to facilitate the defeat of an implacable enemy.
And the task was too vast and demanding to be tackled in any other way. The Japanese were acutely communications security aware. They used high-grade ciphers and complex codes which required intense and concentrated intellectual and technical efforts to break.
Meekin recounted some of the achievements of Central Bureau, including the bombing campaign against Wewak and Lae, the breaking of the Mainline army codes, and the invasion of Hollandia. He closed by saying, ‘To the men and women of Central Bureau — we honour your legacy and we salute you.’1
Meekin then introduced the director of security and future capability at the Australian Signals Directorate, Derek Dalton, who said, ‘Their weapons were not rifles and guns. Their weapons were their incredible talent for solving puzzles, their imagination and the sheer determination to prevail in the face of incredible technical challenges.’2
Many of the veterans of Central Bureau and the ASWG had made contact with each other in 1975, thirty years after the end of the war, when the existence of the signals-intelligence work became declassified. There were arguments when they gathered in Sydney because some of those present believed they should not break their oath of silence. During the war, nobody had said there was a thirty-year limit. They had sworn to keep the secrets forever, and believed they should honour that promise.3
But, despite the objections, an association was formed thirty years later, and they got to know each other. They were no longer code-breakers or Kana operators; just normal people living normal lives. They started a newsletter to share memories and try to piece together what exactly had happened during the war years. That newsletter is still running today. Even as late as the year 2000, while other Second World War veteran’s associations dwindled, the Central Bureau Intelligence Corps Association was still growing as other veterans continued to discover its existence and make contact with their wartime colleagues.
Resentment continued to burn for many of them, decades later, about the lack of acknowledgement from the Australian government of their wartime roles.4 Jack Brown related in his book Katakana Man how, after the war, he attended an ANZAC day march in 1948. He had sustained lifelong injuries from crippling illnesses acquired during operations, but was not deterred. He was proud of his service, and wanted to march:
I managed to walk quite well but, to my amazement, some supposed Air Force aircrew friends, who had enlisted with me and had received rank, referred to me as an ‘erk’ and said, ‘Just an LAC’. This name erk had been given to RAAF ground staff. Little did these rude friends realise how vital my work had been and, of course, I did not tell them what I had done. Following the march, we went to the Returned Services League clubrooms in Adelaide and one of the guys grabbed my tie and bit it off at the knot and said, ‘Now you have been operational.’ What an insult! It was 50 years before I went into an Anzac Day march again.5
Brown was unusual in even attending. Veterans of signals intelligence knew they were not welcome, and for the most part stayed home, watching the ANZAC day marches on the television.
If any of them hoped for recognition after the organisations they worked for became declassified, they were to be disappointed. By the time Steve Meekin unveiled the plaque and gave his words of thanks, few remained, and fewer still were able to attend.
The plaque unveiling was not the first time that the Central Bureau veterans had returned to Nyrambla, their former secret workplace in suburban Brisbane. They had gone back once before, in 1988, when Abe Sinkov, having retired from the NSA, visited Australia to reunite with the Australians he had known during the war.
To honour his visit, the veterans privately organised a plaque to be made, and gained permission from the owners of Nyrambla to install the plaque at the front entrance and to hold a small event with Sinkov. There was no fanfare, and there was no media.
The plaque said,
Central Bureau, an organisation comprising service personnel of Australia, U.S.A., Britain, Canada and New Zealand, both men and women, functioned in this house from 1942 till 1945. From intercepted enemy radio messages the organisation provided intelligence which made a decisive contribution to the Allied victory in the Pacific War.
Madeline Chidgey, one of the remaining Garage Girls in attendance that day, wrote about the plaque in the next newsletter, saying, ‘I wonder how many young eyes will see it when we are gone, and ponder on its meaning.’6
Bibliography
Books
Admiralty Historical Section. 1955. Naval Staff History Second World War, Battle Summary No. 14 (Revised): Loss of H.M. Ships Prince Of Wales and Repulse: 10th December, 1941. B.R. 1736 (8)/19 (revised). United Kingdom: Admiralty Historical Section.
Ball, Desmond, and David Horner. 1998. Breaking the Codes: Australia’s KGB network, 1944–1950. St Leonard’s, NSW: Allen & Unwin.
Ball, Desmond, and Keiko Tamura. 2013. Breaking Japanese Diplomatic Codes: David Sissons and D Special Section during the Second World War. Canberra: ANU Press.
Ballard, Geoffrey St. Vincent. 1991. On Ultra Active Service: the story of Australia’s signals intelligence operations during WW2. Richmond, VIC: Spectrum Publications.
Beevor, Antony. 1991. Crete: the battle and the resistance. London: John Murray.
Bleakley, Jack. 1991. The Eavesdroppers. Canberra: Australian Government Publishing Service.
Bliss, James 2015. The Fall of Crete, 1941: was Freyberg culpable? Pickle Partners Publishing.
Bou, Jean. 2012. MacArthur’s Secret Bureau: the story of the Central Bureau, General MacArthur’s Signals Intelligence Organisation. Loftus, NSW: Australian Military History Publications.
Brown, Jack. 2006. Katakana Man: I worked only for generals. Canberra: Air Power Development Centre.
Cannon, M. Hamlin. 1993. United States Army in World War II: the war in the Pacific: Leyte: the return to the Philippines. Washington, DC. Office of the Chief of Military History, Department of the Army.
Cohen, Graeme L. 2006. Counting Australia In: the people, organisations and institutions of Australian Mathematics. Broadway Bay, N.S.W.: Halstead Press.
Dexter, David. 1961. The New Guinea Offensives. Australia in the War of 1939–1945, Series 1—Army. Volume VI. 1st ed. Canberra: Australian War Memorial.
Donovan, Peter, and John Mack. 2014. Code Breaking in the Pacific. Cham [Switzerland]: Springer.
Drea, Edward J. 1992. MacArthur’s Ultra: codebreaking and the war against Japan, 1942–1945. USA: University Press of Kansas.
Gallehawk, John. 1998. How the Enigma secret was nearly revealed. Report No. 11. The Bletchley Park Trust Reports. Milton Keynes [England]: Bletchley Park Trust.
Hall, Timothy. 1983. The Fall of Singapore. Melbourne: Mandarin Australia.
Halmos, Paul R. 1985. I Want To Be a Mathematician. New York: Springer Verlag.
Hamm, Diane L. 1987. Military Intelligence: its heroes and legends. Arlington Hall Station, Virginia: US Army Intelligence and Security Command.
Hanyok, Robert J. and David P. Mowry. 2008. West Wind Clear: cryptology and the winds message controversy: a documentary history. National Security Agency Center for Cryptologic History.
Hasegawa, Tsuyoshi. 2005. Racing the Enemy: Stalin, Truman, and the surrender of Japan. Belknapp. Cambridge, Massachusetts: Harvard University Press.
Herman, Arthur. 2016. Douglas MacArthur: American warrior. New York: Random House.
Hetherington, Paul. 2003. The Diaries of Donald Friend, Volume 2. Canberra: National Library of Australia.
Hewitt, Admiral H. Kent. 1945. Pearl Harbor Attack: report of Admiral H. Kent Hewitt to Secretary of Navy, dated July 12, 1945. Washington, DC: Congress of the United States.
Hillier, Jean. 1996. No Medals In This Unit. Mundulla, S.A.: Privately published.
Howard, Ann. 1990. You’ll Be Sorry! How World War II Changed Women’s Lives. Newport, NSW: Big Sky Publishing.
Huie, Shirley Fenton. 2000. Ship’s Belles: the story of the Women’s Royal Australian Naval Service in war and peace, 1942–1985. Sydney: Watermark Press.
Laidlaw, Donald H. 2001. Anecdotes of a Japanese Translator. Adelaide: Privately published.
Lewin, Ronald. 1982. The Other Ultra: codes, ciphers and the defeat of Japan. London: Hutchinson.
MacArthur, Douglas. 1994. Reports of General MacArthur: the campaigns of MacArthur in the Pacific, Volume I. Washington, D.C: Department of the Army.
MacEachin, Douglas J. 1998. The Final Months of the War with Japan: signals intelligence, U.S. invasion planning, and the A-bomb decision. Washington, D.C: Central Intelligence Agency, Center for the Study of Intelligence.
McNaughton, James C. 2006. Nisei linguists: Japanese Americans in the Military Intelligence Service during World War II. Washington, D.C.: Department Of The Army [United States].
Maffeo, Capt. Steven E. 2015. US Navy Codebreakers, Linguists, and Intelligence Officers against Japan, 1910–1941. Maryland [USA]: Rowman and Littlefield.
Maneki, Sharon A. 1996. The Quiet Heroes of the Southwest Pacific Theater: an oral history of the men and women of CBB and FRUMEL. Fort George G. Meade, Maryland, USA: National Security Agency.
Masuda, Hiroshi. 2009. MacArthur in Asia: the general and his staff in the Philippines, Japan and Korea. London: Cornell University press.
Melinsky, Hugh. 1998. A Code-Breaker’s Tale. Dereham, UK: Lark’s Press.
Miscable, Wilson D. 2011. A Most Controversial Decision: Truman, the atomic bombs and the defeat of Japan. Cambridge, Massachusetts: Cambridge University Press.
Murray, Gil. 2001. The Invisible War: the untold secret story of Number One Canadian Special Wireless Group: Royal Canadian Signal Corps, 1944–1946. Toronto: The Dundurn Group.
Odgers, George. 1968. Australia in the War of 1939–1945. Volume II – Air War Against Japan, 1943–1945. Canberra: Australian War Memorial.
Palenski, Ron. 2013. Men of Valour: New Zealand and the Battle for Crete. Auckland, New Zealand: Hachette New Zealand.
Penny, Bruce. 2007. The Army at Bonegilla, 1940–71. Wodonga, Vic: Parklands Albury-Wodonga: Wodonga.
Perlberg, Miriam F. 1992. Intelligence Lessons Learned from the Battle of Crete, 1941. Newport, Rhode Island: Naval War College.
Perrett, Geoffrey. 1996. Old Soldiers Never Die: the life of Douglas MacArthur. New York: Random House.
Pfennigwerth, Ian. 2006. A Man of Intelligence: the life of Captain Eric Nave, Australian codebreaker extraordinary. Dural, NSW: Rosenberg.
Pfennigwerth, Ian. 2008. Missing Pieces: the intelligence jigsaw and RAN operations from 1939–71. Canberra: Sea Power Centre.
Pike, Francis. 2015. Hirohito’s War: the Pacific War, 1941–1945. Bloomsbury Publishing.
Prados, John. 1995. Combined Fleet Decoded. New York: Random House.
Prados, John. 2012. Islands of Destiny: the Solomons campaign and the eclipse of the Rising Sun. Penguin Publishing Group. Kindle Edition.
Pyle, Doug, and Nell Pyle. 2006. The Ultra Experience: service with Central Bureau Intelligence Corps. Loftus, NSW: Australian Military History Publications.
Rusbridger, James, and Eric Nave. 1991. Betrayal at Pearl Harbor. New York: Summit Books.
SIS Record Association. 1946. Special Intelligence Service in the Far East 1942–1946. New York: SIS Record Association.
Smith, Michael. 2001.The Emperor’s Codes: Bletchley Park and the breaking of Japan’s secret ciphers. London: Bantam.
Stubbings, Leon. 2003. Simpson SO-in-C. Blackburn, Victoria: PenFolk Publishing.
Symonds, Craig L. 2011. The Battle of Midway. New York: Oxford University Press.
Willoughby, Charles C. 1948. A Brief History of the G-2 Section, GHQ, SWPA and Affiliated Units. Delaware, USA: United States Center for Military History.
Winter, Barbara. 1995. The Intrigue Master: Commander Long and naval intelligence in Australia, 1913–1945. Moorooka, QLD: Boolarong Press.
Winterbotham, Frederick W. 1974. The Ultra Secret: the inside story of Operation Ultra, Bletchley Park and Enigma. London: Orion Books.
Winton, John. 1993. Ultra in the Pacific. Annapolis, Maryland: Naval Institute Press.
Book chapters and reference entries
Axelrod, Alan and Jack A. Kingston. 2007. Encyclopedia of World War II, Volume 1. New York: Infobase.
Drea, Edward, and Joseph Richard. 1999. ‘New Evidence on Breaking the Japanese Army Codes.’ Allied and Axis Signals Intelligence in World War II, edited by David Alvarez. London: Frank Cass.
Ferris, John. 1993. ‘Worthy of Some Better Enemy? The British estimate of the Imperial Japanese Army 1919–41, and the fall of Singapore.’ Canadian Journal of History 28 (2).
Hirschfeld, J.W.P. and G.E. Wall. 1998. ‘Thomas Gerald Room 1902–1986.’ Australian Academy of Science: biographical memoirs of deceased Fellows.
Johnson, Richard. 1995. ‘Trendall, Arthur Dale (1909–1995).’ ANU Reporter, 13 December: 11.
Mack, John. 2012. ‘Room, Thomas Gerald (1902–1986).’ Australian Dictionary of Biography. National Centre of Biography, Australian National University.
Nelmes, Michael. 2012. ‘McKenzie, Florence Violet (1890–1982).’ Australian Dictionary of Biography. National Centre of Biography, Australian National University.
Primrose, B. N. 1981. ‘Colvin, Sir Ragnar Musgrave (1882–1954).’ Australian Dictionary of Biography. National Centre of Biography, Australian National University.
Journals and periodicals
[author name withheld by journal]. 1989. ‘The Origination and Evolution of Radio Traffic Analysis: World War II.’ Cryptologic Quarterly 7(4): 23–38.
[author name withheld by journal]. 1992. ‘Now Playing: Churchill as Pearl Harbor Villain.’ Cryptologic Quarterly 11(2) :1–26.
Budiansky, Stephen. 2001. ‘Codebreaking with IBM Machines in World War II.’ Cryptologia 25(4): 241–55.
Cain, Frank. 1999. ‘Signals intelligence in Australia during the Pacific War.’ Intelligence and National Security 14 (1): 40–61.
Donovan, Peter, and John Mack. 2002. ‘Sydney University, T.G. Room and Codebreaking in WW II.’ Gazette of the Australian Mathematical Society. Section 11.
Drea, Edward. 1995. ‘Were the Japanese Army Codes Secure?’ Cryptologia 19 (2): 113–16.
Fenton, Corinne. 2007. ‘Arthur Cooper and his pet gibbon ape.’ Australian Nursing Journal 15(6): 3.
Ham, Paul. 2015. ‘The Bureaucrats Who Singled Out Hiroshima for Destruction.’ The Atlantic, 6 August.
Inches, Roy. 2002. ‘Some Serious Eavesdropping (Sigint).’ The Whisperer: The newsletter of the Beaufighter and Boston Association of Queensland, March: 9–12.
Maloney, Shane, and Chris Grosz. 2006. ‘Thomas Blamey & Douglas MacArthur.’ The Monthly, October.
McKenzie-Smith, Graham. 2012. ‘The Other Dick Smith and the Sio Code Books.’ Sabretache: The Journal of the Military Historical Society of Australia 53 (1): 13-16.
Nichols, Robert. 2004. ‘The first kamikaze attack?’ Wartime 28.
Richard, Joseph E. 2004. ‘The Breaking of The Japanese Army’s Codes.’ Cryptologia 28 (4): 289-308.
Sontheimer, Morton. 1947. ‘The Radio Drama of V-J Day.’ Coronet February: 66.
Whitlock, Duane L. 1993. ‘Station “C” and Fleet Radio Unit Melbourne (FRUMEL) Revisited.’ Cryptologic Quarterly 12(2): 131–139.
Wilford, Timothy. 2002. ‘Decoding Pearl Harbor: USN cryptanalysis and the challenge of JN-25B in 1941.’ The Northern Mariner/Le Marin du Nord 12 (1): 17–37.
Personal interviews
Bleakley, Jack. October 2015 and November 2015. Telephone.
Brackley, Colin. August 2016. Telephone.
Brown, Robert. August 2016. Telephone.
Chidgey, Madeline. 9 July 2015. Ascot, QLD.
Fairbridge (nee Duff), Joan. October 2014, 8 May 2016, and several phone calls between 2014–2016. Albury, NSW.
Falconer, Keith. 7 May 2016. Castlemaine, Victoria.
Field, Don. 10 November 2014. Melbourne, VIC.
Gibson, Gordon. 8 June 2015. Hornsby, NSW.
Hughes, Frank. 9 July 2015. Ascot, QLD.
Isaksson (nee Seward), Margot. August 2012. Crow’s Nest, NSW.
Kenny, Helen. April 2015. Sydney, NSW.
Lederer, Victor. 2015–2016 (numerous visits and conversations). Curtin, ACT.
Norton, Allan. 3 August 2015. Telephone.
Osborne, Coral. November 2015. Frankston, VIC.
Pyle, Doug. 22 March 2015. Maitland, NSW.
Pyle, Doug and Nell Pyle. 1 October 2015. Maitland, NSW.
Roe, Judy. 3 October 2015. Manly, NSW.
Rogers, Bill. October 2015. Telephone.
Walsh, Brian. November 2015. Adelaide, SA.
Central Bureau newsletter articles
Carolan, Joan. 2012. ‘Vale Keith Carolan.’ Central Bureau Intelligence Corps Association Newsletter, December: 8.
Carolan, Keith. 1994. ‘Labuan Revisited.’ Central Bureau Intelligence Corps Association Newsletter, March: 3–4.
Chidgey, Madeline. 1988. ‘Bicentennial Activities.’ Central Bureau Intelligence Corps Association Newsletter, May: 1–4.
Clarke, William E. 2002. ‘The Quiet Heroes Make Some Noise.’ Central Bureau Intelligence Corps Association Newsletter, September: 4–5.
Dalton, Derek. 2015. ‘Contemporary Signals Intelligence: Speech for the Central Bureau Commemoration.’ Central Bureau Intelligence Corps Association Newsletter, September/October: 1–4.
Davis, Alf. 1987. ‘A few words from Alf Davis, ex Fl/Lt RAAF.’ Central Bureau Intelligence Corps Association Newsletter, March: 4.
Hermes, Clarrie. 1990. ‘First and last but not oldest.’ Central Bureau Intelligence Corps Association Newsletter, August: 5–6.
Jamieson, Mac. 2006. ‘Mac Jamieson Writes.’ Central Bureau Intelligence Corps Association Newsletter, June: 5.
Kenny, Helen. 2000. ‘Vale Captain Ian Allen.’ Central Bureau Intelligence Corps Association Newsletter, March: 5.
Kenny, Helen. 2001. ‘Book Review: Anecdotes of a Japanese Translator, 1941–1945, By D.H. Laidlaw, A.O.’ Central Bureau Intelligence Corps Association Newsletter, December: 7.
Kenny, Helen. 2004. ‘Taking off for Tokyo.’ Central Bureau Intelligence Corps Association Newsletter, December: 7–8.
Kenny, Helen. 2009. ‘Vale Alan Marsland.’ Central Bureau Intelligence Corps Association Newsletter, December: 7–8.
Kenny, Helen. 2002. ‘The MacArthur Museum.’ Central Bureau Intelligence Corps Association Newsletter, June: 6–7.
Kenny, Helen. 2012. ‘Vale Dorothy Bleakley.’ Central Bureau Intelligence Corps Association Newsletter, June: 8.
Kenny, Helen. 2013. ‘Chooks at Chermside (A CB Mystery).’ Central Bureau Intelligence Corps Association Newsletter, August: 6–7.
Long, Lynda. 1989. ‘Ballroom has Harmony Restored.’ Central Bureau Intelligence Corps Association Newsletter, September: 8. [reprint of newspaper article, source unknown].
Meekin, Steve. 2015. ‘Address by Mr Steve Meekin.’ Central Bureau Intelligence Corps Association Newsletter, August: 8–12.
Richard, Joe. 1994. ‘Inside Central Bureau: A First Person Account.’ Central Bureau Intelligence Corps Association Newsletter, June: 1–2.
Richard, Joe. 1995. ‘The Yamamoto Saga.’ Central Bureau Intelligence Corps Association Newsletter, June: 6–7.
Richard, Joe. 2004. ‘Memories of Central Bureau (Part 1).’ Central Bureau Intelligence Corps Association Newsletter, December: 3–5.
Richard, Joe. 2005. ‘Joe Richard’s Written Legacy.’ Central Bureau Intelligence Corps Association Newsletter, December: 3–6.
Australian War Memorial
4 Special Wireless Section, War Diary 1940–1941. AWM 52 7/39/41.
Australian War Memorial, Honours and Awards: Robert Arthur Little, R1601287.
Bobin, Alfred. 1986. Bobin, Alfred (Leading Aircraftman, 1 Wireless Unit, RAAF). AWM PR86/366. Letters and photograph.
Hartley, Robert (Bob) Walter, AM. 2015. The Antecedent of the Defence Signals Bureau. Unpublished Manuscript.
Hartley, Robert (Bob) Walter, AM. 2015. The History of the Australian Special Wireless Group, 1939 to 1947. MSS2343. Unpublished Manuscript. Australian War Memorial.
Hill, William Charles Fitzmaurice. 1964. A Wireless Section of Greece and Crete (Hill, William Charles Fitzmaurice; VX19513, Major). MSS1915. Unpublished Memoir. Australian War Memorial.
Lederer, Victor Edgar. 1957. A Span of Years. AWM MSS1155. Unpublished manuscript.
Nave, Eric. An Australian’s Unique Naval Career. MSS 1183. Unpublished Memoir, Australian War Memorial.
Audio-visual
Clark, Stanley Robert Irving. 1974. Stan Clark Oral History. Cassette recording. Sydney: Australian Broadcasting Commission, January 1. [available through the Australian National Archives as item no. NAA C101, 1343517]
CriticalPast. Allied forces capture Hollandia in New Guinea during World War 2 HD Stock Footage. https://www.youtube.com/watch?v=6YAh5DVr40o [Accessed January 23, 2017].
Freyne, Catherine. 2008. Signals, currents, and wires: the untold story of Florence Violet McKenzie. Hindsight radio program. ABC Radio National, March 16.
The History Channel. ‘I Have Returned’ MacArthur Returns to the Philippines. https://www.youtube.com/watch?v=pM8ktDuCuLc
WGBH. 2005. The American Experience: Victory in the Pacific. The American Experience, Documentary, PBS, May 2005.
National Security Agency oral histories
Electronic copies can be obtained at https://www.nsa.gov/news-features/declassified-documents/oral-history-interviews/
Fabian, Rudolph T. 1983. Interview. National Cryptologic Museum Oral History Number 9 [NSA-OH-09-83]. Fort George G. Meade, Maryland, USA: National Security Agency.
Sinkov, Abraham. 1979. Interview. National Cryptologic Museum Oral History Number 2 [NSA-OH-02-79]. Fort George G. Meade, Maryland, USA: National Security Agency.
Whitlock, Duane L. 1983. Interview. National Cryptologic Museum Oral History Number 5 [NSA-OH-05-83]. Fort George G. Meade, Maryland, USA: National Security Agency.
Webpages
Australian War Memorial. ‘The Burma–Thailand Railway.’ https://www.awm.gov.au/exhibitions/stolenyears/ww2/japan/burmathai/. Accessed 7 January 2016.
Australian War Memorial. ‘Australian Women’s Army Service (AWAS) and Women’s Royal Australian Army Corps (WRAAC).’ https://www.awm.gov.au/atwar/structure/awas/. Accessed 6 October 2016.
Avalon Project, Yale Law School. ‘The Atomic Bombings of Hiroshima and Nagasaki.’ http://avalon.law.yale.edu/subject_menus/mpmenu.asp. Accessed 14 August 2015.
Chen, C. Peter. ‘Invasion of the Philippine Islands: 7 Dec 1941 – 5 May 1942.’ World War II Database. http://ww2db.com/battle_spec.php?battle_id=46. Accessed 9 August 2015.
Chen, C. Peter. ‘Philippines Campaign, Phase 2: 12 Dec 1944 – 15 Aug 1945.’ World War II Database. http://ww2db.com/battle_spec.php?battle_id=27. Accessed 22 January 2017.
Chermside and Districts Historical Society. ‘Chooks and Tags.’ http://www.chermsidedistrict.org.au/01_cms/details.asp?viewMode=printable&ID=295#2476. Accessed 1 August 2016.
D-Day Museum. ‘D-Day and the Battle of Normandy: Your Questions Answered.’ http://www.ddaymuseum.co.uk/d-day/d-day-and-the-battle-of-normandy-your-questions-answered. Accessed 14 January 2017.
Dictionary of Sydney. ‘Violet McKenzie.’ http://dictionaryofsydney.org/entry/mckenzie_violet. Accessed 10 October 2016.
Dunn, Peter. 2003. ‘“Tighnabruaich”, 195 Clarence Road, Indooroopilly used to Interrogate Japanese Prisoners During WW2 by the Allied Translation and Interpretation Section.’ Ozatwar. http://www.ozatwar.com/ozatwar/tighnabruaich.htm.
Dunn, Peter. 2015. ‘126 Signal Radio Intelligence Company: Us Army Signal Corps in Australia During WWII.’ Ozatwar. http://www.ozatwar.com/usarmy/126sric.htm.
Dunn, Peter. 2015. ‘Combined Operational Intelligence Centre (COIC) in Australia during WW2.’ Ozatwar. http://www.ozatwar.com/sigint/coic.htm.
Dunn, Peter. 2015. ‘RAN/USN Fleet Radio Unit, Melbourne – FRUMEL.’ Ozatwar. http://www.ozatwar.com/sigint/frumel.htm.
Dunn, Peter. 2016. ‘51 Wireless Section: Australian Special Wireless Group, in Australia During WW2.’ Ozatwar. http://www.ozatwar.com/sigint/section51swg.htm.
Dunn, Peter. 2016. ‘Central Bureau in Australia during WWII.’ Ozatwar. http://www.ozatwar.com/sigint/cbi.htm.
Dunn, Peter. 2016. ‘Eagle Farm @ War.’ Ozatwar. http://www.ozatwar.com/eaglefarm.htm.
Dunwall, Stephen W. 1992. ‘World War II Code-Breaking Exploits Unveiled By Retired IBM Engineer After 50 Years Of Secrecy.’ IBM. 12 February. http://www-03.ibm.com/ibm/history/exhibits/builders/builders_dunwell2.html.
Hellenic Foundation. ‘Battle of Maleme, Crete.’ http://www.hellenicfoundation.com/Maleme.htm. Accessed 15 January 2017.
History.com Staff. 2009. ‘The Battle of Okinawa.’ History.com. http://www.history.com/topics/world-war-ii/battle-of-okinawa. Accessed 8 November 2016.
History.com Staff. 2009. ‘The Battle of Stalingrad.’ History.com. http://www.history.com/topics/world-war-ii/battle-of-stalingrad. Accessed 31 August 2016.
Holloway, Don. 2012. ‘Death by P-38.’ HistoryNet. 4 March. http://www.historynet.com/death-by-p-38.htm.
Lenihan, Denis. 2016. ‘The Battle of Crete 1941: The Poverty of Ultra.’ Academia.com., http://www.academia.edu/22144185/The_Battle_of_Crete_1941_The_Poverty_of_Ultra. Accessed 3 August 2016.
MacKinnon, Colin. 2013. ‘Bletchley Park Diary: William F. Friedman.’ Colin MacKinnon. http://www.colinmackinnon.com/files/The_Bletchley_Park_Diary_of_William_F._Friedman_E.pdf.
Military Factory. ‘Kawanishi H8K (Emily) Reconnaissance / Bomber Flying Boat Aircraft (1942).’ http://www.militaryfactory.com/aircraft/detail.asp?aircraft_id=594. Accessed 23 January 2017.
Military History Now. ‘Operation Downfall — The Campaign to Conquer Japan Would Have Dwarfed the D-Day Landings.’ 6 November 2013.
Naval History and Heritage Command. ‘Casualties: US Navy and Coast Guard Vessels, Sunk or Damaged Beyond Repair during World War II, 7 December 1941–1 October 1945.’ http://www.history.navy.mil/research/histories/ship-histories/casualties-navy-and-coast-guard-ships.html. Accessed 9 August 2015.
Pelvin, Ric. ‘Battle of the Coral Sea, 4–8 May 1942.’ Australian War Memorial Encyclopedia. https://www.awm.gov.au/encyclopedia/coral_sea/doc.asp. Accessed 6 November 2016.
Pfennigwerth, Ian. 2012. ‘No Contest! The US Navy Destroys Australia’s Special Intelligence Bureau.’ Military History and Heritage Victoria. http://www.mhhv.org.au/?p=2754.
Rickard, J. 2015. ‘Battle of Biak Island, 27 May–29 July 1944.’ History of War. http://www.historyofwar.org/articles/battles_biak.html.
Royal Australian Navy. ‘HMAS Harman.’ Royal Australian Navy Establishments. http://www.navy.gov.au/biography/admiral-sir-ragnar-musgrave-colvin http://www.navy.gov.au/establishments/hmas-harman. Accessed 13 January 2017.
Royal Australian Navy. ‘Admiral Sir Ragnar Musgrave Colvin.’ Royal Australian Navy Establishments. http://www.navy.gov.au/biography/admiral-sir-ragnar-musgrave-colvin.
Scott, Eve. Interview. http://evescott.blogspot.com.au/. Accessed 10 November 2016.
Stone, June. 2003. ‘Women’s Auxiliary Australian Air Force (WAAAF) (1941–1947).’ The Australian Women’s Register. http://www.womenaustralia.info/biogs/AWE0400b.htm. Accessed 13 January 2017.
Straczek, Joe. ‘Listening for the Empire.’ Royal Australian Navy. http://www.navy.gov.au/sites/default/files/documents/Straczek_-_Listening_for_the_Empire.pdf. Accessed 23 January 2017.
Straczek, Josef. ‘Battle of the Coral Sea. Navy Feature Histories.’ Royal Australian Navy. http://www.navy.gov.au/history/feature-histories/battle-coral-sea. Accessed 6 November 2016.
University of Sydney Faculty of Arts. 2014. ‘Dr Paul Scully-Power AM, DSM, NSM.’ Alumni and Friends. http://sydney.edu.au/arts/alumni_friends/profiles/scully-power.shtml
Wukovit, John. 2016. ‘The Ill-Fated Goettge Patrol.’ Warfare History Network. http://warfarehistorynetwork.com/daily/wwii/the-ill-fated-goettge-patrol/. Accessed 6 November 2016.
National Archives of Australia (NAA)
Central Bureau (CB); B5436, Technical Records, B5436; PART A-M, Central Bureau Technical Records.
Defence Division, Department of the Treasury [I]; A649, Correspondence files, multiple number series, Classes 600-602 (unclassified).
Department of Defence [III], Central Office; B883, Second Australian Imperial Force Personnel Dossiers, 1939–1947.
Department of Defence; A816, Correspondence files, multiple number series.
Directorate of Military Intelligence; A6923, Army Headquarters/Army Office record sets; Technical Records.
Fleet Radio Unit, Melbourne; B5555.
Naval Board; A2585, Naval Board minute books.
Notes
Prologue
1 Helen Kenny, Chooks at Chermside (A CB Mystery); Chermside & District Historical Society, Chooks and Tags.
2 Earlier histories of the code-breaking effort in the Pacific gave the credit to breaking JN-25 entirely to the United States Naval units at Hawaii and the Philippines, but it is now clear that the original break came from John Tiltman, and was delivered by Malcolm Barnett. Michael Smith makes this case in The Emperor’s Codes.
Chapter 1: The fires of a distant war
1 Spain and Portugal were also ruled by fascist governments, but they did not join the Axis alliance and remained neutral until the end of the war. Thanks to Peter Donovan for suggesting that I clarify this point.
Chapter 2: Tropical sickness
1 The Far East Combined Bureau, like the Middle East Combined Bureau, was controlled by the Government Code and Cypher School (GCCS), which for most of the war was based at Bletchley Park, England. The GCCS is often informally referred to as ‘Bletchley Park’.
2 Nave’s involvement in more traditional espionage activities including human intelligence in Hong Kong is clear from his memoirs, which are in the Australian War Memorial.
3 John Ferris, Worthy of Some Better Enemy? The British Estimate of the Imperial Japanese Army 1919–41, and Timothy Hall, The Fall of Singapore, p. 252.
4 Timothy Hall, The Fall of Singapore, pp. 6–9.
5 Eric Nave, Personal Memoirs, p. 405.
6 Nave left Singapore in February 1940, according to Peter Donovan and John Mack, Code Breaking in the Pacific, p. 72; see also David Jenkins, ‘Our War of Words’, Sydney Morning Herald, 18 September 1992.
Chapter 3: Code club
1 John Mack, ‘Room, Thomas Gerald (1902–1986)’; Paul Halmos, I Want to be a Mathematician, pp. 280–1; University of Sydney Faculty of Arts, ‘Dr Paul Scully-Power AM, DSM, NSM’.
2 J. Hirschfeld & G. Wall, Thomas Gerald Room 1902–1986.
3 Peter Donovan & John Mack, 2014. Code Breaking in the Pacific, pp. 52–3.
4 Graeme Cohen, Counting Australia In: the people, organisations and institutions of Australian mathematics, pp. 136–7.
5 Richard Johnson, Trendall, Arthur Dale (1909–1995).
6 Desmond Ball & Keiko Tamura, Breaking Japanese Diplomatic Codes: David Sissons and D Special Section during the Second World War, p. 6.
7 David Jenkins, ‘Our War of Words’, Sydney Morning Herald, 18 September 1992.
8 NAA: A6923 37/401/425, CGS Branch — Military Intelligence — Special Intelligence Section, pp. 278, 283, 286.
9 The couple in question were Phyllis Malley, married to RAAF officer and former WWI pilot, Garnet Malley. The British officer in Shanghai was Sir Frederick Maze, and was the Shanghai inspector-general of customs.
10 Ball and Tamura, Breaking Japanese Diplomatic Codes, p. 22.
Chapter 4: Sailor Jack
1 Jean Hillier, No Medals in This Unit, p. 17.
2 Bob Hartley, The History of the Australian Special Wireless Group, 1939–1947, p. 47.
3 Barbara Winter, The Intrigue Master: Commander Long and naval intelligence in Australia, 1913–1945, p. 52; Stanley Clark, Oral History.
4 Stanley Clark, Oral History.
5 Bob Hartley, The History of the Australian Special Wireless Group, 1939–1947, p. 56.
6 Stanley Clark, Oral History.
7 Victor Lederer, personal communication.
8 Bob Hartley, The History of the Australian Special Wireless Group, p. 44.
9 ibid., p. 57.
Chapter 5: A wireless unit in Greece
1 Geoffrey Ballard, On Ultra Active Service: The Story of Australia’s Signals Intelligence Operations During World War II, p. 3.
2 Geoffrey Ballard, in On Ultra Active Service, refers to this as the British Air Ministry W/T Section in Heliopolis; Peter Donovan informs me that it was also the headquarters for Combined Bureau Middle East (CBME). The CBME would have been responsible for the training that Ballard and Inglis received.
Chapter 6: A purple jacket on Crete
1 John Gallehawk, How the Enigma secret was nearly revealed. Report No. 11. The Bletchley Park Trust Reports, p. 10.
2 Miriam Perlberg, Intelligence Lessons Learned from the Battle of Crete, 1941, Naval War College, Newport, RI. On page 4, Perlberg describes the dire situation that Freyberg found himself in, which I have described. Geoffrey Ballard, On Ultra Active Service had published his book in 1991, the year before Perlberg’s article, but the role of Australian signals on Crete, as elsewhere, was not well known among historians at the time.
3 Antony Beevor, Crete: The Battle and the Resistance, p. 40.
4 See, for example, Ron Palenski, Men of Valour: New Zealand and the Battle for Crete: ‘The original liaison officer appears to have been George Beamish, the senior air officer, but he lasted only until a specialist arrived.’
5 National Archives of Australia (henceforth NAA), B883, SX11231, Sandford Alastair Wallace, Service Record.
6 Australian War Memorial (henceforth AWM) 52 7/39/41, 4 Special Wireless Section War Diary, December 1940 to December 1941.
7 William Hill, A Wireless Section of Greece and Crete, p. 27.
8 ibid., p. 28.
9 This was told to me by Vic Lederer, who worked in Four Special Wireless Section with Sandford.
10 Denis Lenihan, ‘The Battle of Crete 1941: the poverty of Ultra’, p. 16.
11 Antony Beevor, Crete: The Battle and the Resistance, p. 161.
12 Denis Lenihan, ‘The Battle of Crete 1941: the poverty of Ultra’, p. 41.
13 The entire message, codenamed OL 2/302, can be found in Bliss, Major James, The Fall of Crete 1941: was Freyberg culpable? pp. 145–6.
14 Hellenic Foundation, ‘Battle of Maleme, Crete’.
15 Lieutenant Hill describes this and other experiences on Crete in a lively and readable way in his memoir, A Wireless Section of Greece and Crete. This memoir is, sadly, unpublished but is available in physical format at the Australian War Memorial in Canberra.
16 John Gallehawk, How the Enigma Secret was Nearly Revealed, p. 13.
17 ibid., p. 14.
Chapter 7: Mrs Mac and her girls in green
1 Newman’s visit occurred in late January 1941; the exact date is unknown. I’ve added some colour to this section by briefly describing the weather, based on Bureau of Meteorology historical data. There was only one day of heavy rainfall, 26 January, a Sunday. As Mrs Mac’s school did not run on Sundays, Newman did not visit in the rain. Late January was otherwise mild and sunny.
2 NAA: A2585, 1939/1941/ Reference Copy, Naval Board Minutes, 1939–1941, p. 270.
3 Ann Howard, You’ll be Sorry! How World War II changed women’s lives, p. 8.
4 Dictionary of Sydney, ‘Violet McKenzie’.
5 ibid.
6 Michael Nelmes, ‘McKenzie, Florence Violet (1890–1982)’.
7 ‘Women’s Emergency Signalling Corps’, Sydney Morning Herald, 4 May 1939.
8 Catherine Freyne, Signals, Currents, and Wires: the untold story of Florence Violet McKenzie.
9 ibid.
10 Shirley Huie, Ships Belles: the story of the Women’s Royal Australian Naval Service in war and peace 1942–1985, p. 27.
11 Admiral Sir Ragnar Musgrave Colvin was First Naval Member of the Australian Naval Board. See Royal Australian Navy, ‘Admiral Sir Ragnar Musgrave Colvin’.
12 NAA: A2585, 1939/1941/ Reference Copy, Naval Board Minutes, 1939–1941, p. 270.
13 ibid., p. 276.
14 B. N. Primrose, Colvin, Sir Ragnar Musgrave (1882–1954); NAA: A6769, Durnford John Walter.
15 NAA: A2585, 1939/1941/ Reference Copy, Naval Board Minutes, 1939 – 1941, p. 284.
16 ibid., p. 293.
17 Royal Australian Navy, ‘HMAS Harman’.
18 NAA:A2585 1939/1941/Reference Copy, Naval Board Minutes 1939–41, pp. 306–307
19 June Stone, ‘Women’s Auxiliary Australian Air Force (WAAAF) (1941–1947)’.
20 Australian War Memorial. ‘Australian Women’s Army Service (AWAS) and Women’s Royal Australian Army Corps (WRAAC)’.
21 Catherine Freyne, Signals, Currents, and Wires.
Chapter 8: Fabian’s tunnel
1 Sharon Maneki, The Quiet Heroes of the Southwest Pacific Theater: An Oral History of the Men and Women of CBB and FRUMEL, p. 75.
2 Rudolph Fabian, Interview, pp. 9–10
Chapter 9: Stone frigates
1 Joan Duff interview with David Dufty in Albury, NSW, October 2014.
2 Barbara Winter, The Intrigue Master, p. 4.
3 ibid., p. 7.
4 NAA: A3978, NEWMAN J B Officers (RAN) personal record — Jack Bolton Newman. Virtually all of his six-monthly reviews in the inter-war years make mention of his superior intellect.
5 Ian Pfennigwerth, A Man of Intelligence: the life of Captain Eric Nave, Australian codebreaker extraordinary, p. 159.
6 ibid., p. 156.
7 Barbara Winter, The Intrigue Master, p. 26.
8 ibid.
9 ibid.
10 NAA A6923 SI/10 p. 310 for the full Menzies rejection letter.
11 Barbara Winter, The Intrigue Master, p. 49. See also Ian Pfennigwerth, A Man of Intelligence and Ian Pfennigwerth, ‘No Contest! The US Navy Destroys Australia’s Special Intelligence Bureau’, both of which refer to Nave meeting Long and Newman.
12 The seven RAAF personnel were Clarrie Hermes, S. ‘Snow’ Bradshaw, G. ‘Taffie’ Davis, Ted Cook, Alf Towers, B. ‘Bing’ Crosby, and J. Wilson. The Army personnel were Robert ‘Bob’ McGhie and one other individual, whose name has not been located. Snow Bradshaw was the NCO in charge of the unit when they went to Darwin. They would go to receivers installed at nearby Mont Park, tune it to the Japanese Navy HQ frequency and ‘pip’ it through to Victoria barracks, where the others could practise their interception skills. (Clarrie Hermes, ‘First and Last but Not Oldest’; and Jack Bleakley, The Eavesdroppers, p. 7).
13 Clarrie Hermes, ‘First and Last but Not Oldest’.
14 Jack Bleakley, The Eavesdroppers, p. 11.
15 Shirley Huie, Ships Belles, pp. 69–70.
Chapter 10: The Special Intelligence Bureau
1 Peter Donovan and John Mack, ‘Sydney University, T.G. Room and Codebreaking in WW II’.
2 ibid.
3 Ian Pfennigwerth, A Man of Intelligence, p. 154.
4 Eric Nave, An Australian’s Unique Naval Career, p. 409.
5 ibid.
6 J.W.P. Hirschfeld and G.E. Wall, ‘Thomas Gerald Room, 1902–1986’.
7 See Desmond Ball & Keiko Tamura, Breaking Japanese Diplomatic Codes, p. 1, and Joe Straczek, ‘Listening for the Empire’.
8 Barbara Winter, The Intrigue Master, p. 52.
9 Ian Pfennigwerth, A Man of Intelligence, p. 166; Frank Cain, ‘Signals intelligence in Australia during the Pacific War’, p. 43.
10 Timothy Wilford, ‘Decoding Pearl Harbor: USN cryptanalysis and the challenge of JN-25B in 1941’. See page 27, referring to the 45-sign code as the latest flag officer’s code, which the Americans could not break either.
11 The Merchant Shipping Code was known to the Americans as JN-40.
12 Peter Donovan & John Mack, Code Breaking in the Pacific, pp. 45, 73. Barbara Winter, The Intrigue Master, p. 52.
13 David Jenkins, ‘Our War of Words’, Sydney Morning Herald, 18 September 1992.
14 Peter Dunn, ‘Combined Operational Intelligence Centre (Coic) in Australia During WW2’.
15 Eric Nave, An Australian’s Unique Naval Career, p. 411.
Chapter 11: East wind, rain
1 ‘US Not Hopeful About Talks’, The Mercury, 20 November 1941.
2 Eric Nave, An Australian’s Unique Naval Career, p. 331. For the United States navy translation, see Robert Hanyok & David Mowry, West Wind Clear: Cryptology and the Winds Message Controversy: a documentary history, p. 120.
3 Eric Nave, An Australian’s Unique Naval Career, p. 334.
4 H. Hewitt, Pearl Harbor Attack: Report of Admiral H. Kent Hewitt to Secretary of Navy, dated July 12, 1945, p. 37.
5 Franklin Roosevelt, ‘Pearl Harbor Speech: Day of Infamy’, http://www.digitalhistory.uh.edu/disp_textbook.cfm?smtID=3&psid=1082.
6 Robert Hanyok & David Mowry, West Wind Clear, p. 225.
Chapter 12: West wind, clear
1 ‘Tim Vigors Obituary’, The Telegraph, 19 November 2003.
2 http://www.ibiblio.org/hyperwar/UN/UK/RN/BS-14_POW+Repulse/index.html
3 John Curtin, ‘The Task Ahead’, The Herald (Melbourne), 27 December 1941.
Chapter 13: Escape from Manila Bay
1 Joe Richard, ‘Memories of Central Bureau (Part 1)’.
2 Sharon Maneki, The Quiet Heroes of the Southwest Pacific Theater, p. 88.
3 Naval History and Heritage Command, ‘Casualties: US Navy and Coast Guard Vessels, Sunk or Damaged Beyond Repair during World War II, 7 December 1941–1 October 1945’.
4 Peter Chen, ‘Invasion of the Philippine Islands: 7 Dec 1941–5 May 1942’.
5 Sharon Maneki, The Quiet Heroes of the Southwest Pacific Theater, p. 89.
6 [Author name redacted]. ‘The Origination and Evolution of Radio Traffic Analysis: World War II’, p. 26.
7 Diane Hamm, Military Intelligence: its heroes and legends.
8 ‘I came through, I shall return’, The Advertiser, 31 March 1942.
9 Geoffrey Perret, Old Soldiers Never Die: the life of Douglas MacArthur.
Chapter 14: Beirut bookshops
1 Geoffrey Ballard, On Ultra Active Service, p. 136.
2 ibid.
3 Shane Maloney and Chris Grosz, ‘Thomas Blamey & Douglas MacArthur’.
4 Hiroshi Masuda, MacArthur in Asia: the general and his staff in The Philippines, Japan and Korea, p. 14.
5 NAA: A6923, SI/2, p. 285.
6 Leon Stubbings, Simpson SO-in-C, pp. 60–5, 71.
7 NAA: A6923, SI/2, p. 249.
8 Peter Dunn, ‘RAN/USN Fleet Radio Unit, Melbourne – FRUMEL’; Corinne Fenton, ‘Arthur Cooper and his Pet Gibbon Ape’; David Jenkins, ‘Our War of Words’, Sydney Morning Herald, 18 September 1992.
9 It has been suggested that the named may have been inspired by Franz Kafka’s novel The Castle. However the Central Bureau technical records are clear that the reason was primarily to obscure the purpose.
10 Rudolph Fabian, Interview.
11 NAA: B5436, Central Bureau Technical Records Part A, p. 6.
Chapter 15: The secrets of the Coral Sea
1 Sharon Maneki, The Quiet Heroes of the Southwest Pacific Theater, p. 56.
2 ibid., p. 90.
3 Joan Duff interview with David Dufty.
4 Sharon Maneki, The Quiet Heroes of the Southwest Pacific Theater, p. 48.
5 [Author name withheld by journal]. ‘Now Playing: Churchill as Pearl Harbor Villain’, p. 20.
6 Duane Whitlock, ‘Station “C” and Fleet Radio Unit Melbourne (FRUMEL) Revisited’.
7 [Author name redacted]. ‘Now Playing: Churchill as Pearl Harbor villain’, p. 14. It turned out that 20 per cent was the tipping point, after which ‘book building’ sped up and became much easier.
8 NAA: B5555, 3, FRUMEL records (incomplete) of communications intelligence relating to the Coral Sea Battle, p. 5.
9 Sharon Maneki, The Quiet Heroes of the Southwest Pacific Theater, p. 90.
10 NAA: B5555, 3, FRUMEL records (incomplete) of communications intelligence relating to the Coral Sea Battle.
11 Jack Bleakley, The Eavesdroppers, p. 38.
12 Peter Donovan & John Mack, Code Breaking in the Pacific, p. 343.
13 Joan Fairbridge interview with David Dufty. Mr Moto was a Japanese character in a comedy show in America in the 1930s; thus a Japanese male was often referred to as Mr Moto.
14 Patrick Lindsay, The Coast Watchers, pp. 224–5.
15 Jozef Straczek, ‘Battle of the Coral Sea’.
16 Ric Pelvin, ‘Battle of the Coral Sea, 4–8 May 1942’; Josef Straczek, ‘Battle of the Coral Sea’.
17 Peter Donovan & John Mack, Code Breaking in the Pacific, p. 343.
18 For intercepted messages, see NAA: B5555, 3.
19 Craig Symonds, The Battle of Midway, p. 388.
20 Duane Whitlock, Interview, p. 68.
21 Navy Department Library: Naval History and Heritage Command. 2015. Battle of Midway: 4–7 June 1942. https://www.history.navy.mil/research/library/online-reading-room/title-list-alphabetically/b/battle-of-midway-4-7-june-1942.html. Accessed 5 November 2016..
22 David Jenkins, ‘Our War of Words’, Sydney Morning Herald, 18 September 1992.
23 ibid.
24 David Jenkins, ‘Kokoda Blunder Cost Thousands of Lives’, Sydney Morning Herald, 13 August 1992.
25 David Jenkins, ‘Invasion Warnings Were Ignored’, Sydney Morning Herald, 13 August 1992.
Chapter 16: Cranleigh
1 NAA A6923 /SI/2 Staffing, p. 229.
2 Desmond Ball & Keiko Tamura, Breaking Japanese Diplomatic Codes, p. 22.
3 James McNaughton, Nisei Linguists: Japanese Americans in the Military Intelligence Service during World War II, pp. 80, 245
4 Abraham Sinkov, Interview. On page 31, Sinkov says that Nave solved the air-to-ground codes. Joe Richard, ‘Inside Central Bureau: a first person account’ states that Nave broke the army and navy air-to-ground codes.
5 Geoffrey Ballard, On Ultra Active Service, p. 164.
6 NAA: B5436, Central Bureau Technical Records Part B - Naval Air - Ground communications, p. 5.
7 ibid., pp. 7, 9. Page 7 displays the burnt, recovered cage, while page 9 shows the picture of a reconstructed cage.
8 NAA: B5436, Central Bureau Technical Records Part B - Naval Air - Ground communications, p. 8.
9 There was also a third kind of code — Romaji codes — that spelled out words using Japanese representations of English letters. But Romaji codes were not used as a complete code system; they were typically incorporated into other codes and ciphers to represent placenames.
Chapter 17: Morse and Kana code
1 Bob Hartley, The History of the Australian Special Wireless Group, 1939 to 1947, p. 180.
2 Having the numbering of Army wireless sections start at 51 lessened confusion with other units, including the Four Special Wireless Section and the RAAF’s ‘wireless units’, which were numbered consecutively from 1 onwards.
3 Jean Hillier, No Medals In This Unit, pp. 24–5.
4 Jean Bou, MacArthur’s Secret Bureau, pp. 12, 13.
5 Bruce Penny, The Army at Bonegilla, 1940–71.
6 Jean Hillier, No Medals In This Unit, p. 13.
7 The exact number of characters in the Kana code is difficult to ascertain. A copy of Keith Falconer’s Kana Code sheet, created in WWII, has 75 characters plus ten ‘short symbols’ for Universal Morse code numbers, making 85 characters. Geoffrey Ballard, On Ultra Active Service included two images of the Kana code; one has 69 characters and one has 72. The reason for the variations is not clear at present.
8 NAA: A69233, SI/2 Staffing, p. 218.
9 Geoffrey Ballard, On Ultra Active Service, p. 216; Bob Hartley, The History of the Australian Special Wireless Group, 1939 to 1947, p. 212.
10 The practice is still operating with his name in Crow’s Nest, North Sydney.
11 Jack Bleakley interview with David Dufty.
12 ibid.
13 NAA: B5436, Central Bureau Technical Records Part A — Organisation, p. 6.
14 Allan Norton interview with David Dufty.
15 Jack Bleakley, The Eavesdroppers, p. 58.
16 Allan Norton interview with David Dufty, 3 August 2015.
17 Geoffrey Ballard, On Ultra Active Service, p. 223.
18 Colin Brackley interview with David Dufty.
19 By coincidence, one of Bletchley Park’s celebrated code-breakers was also named William Clarke and also went by the nickname ‘Nobby’, a situation caused in part by the fact that, in those days, ‘Nobby’ was the default nickname for a man with the surname ‘Clarke’.
To add to the confusion, there was another William Clarke in Central Bureau, Squadron Leader W. J. Clarke, known as Bill. Bill Clarke established the Signals for Number 4 Fighter Sector in Port Moresby, and was commanding officer of signals at Milne Bay during the battle of Milne Bay, and received an MBE for his efforts. Bill Clarke’s obituary can be found in Central Bureau Newsletter, 2006, June, p. 7.
Chapter 18: Traffic
1 John Prados, Islands of Destiny: the Solomons campaign and the eclipse of the Rising Sun, p. 17.
2 John Prados, Combined Fleet Decoded, p. 382. Vandergrift’s intelligence officer, Lieutenant Colonel Frank B. Goettge, liaised closely with MacArthur’s headquarters in the planning. Stan Clark, in his Oral History, says that the Tulagi landings, in particular, being completely free of bloodshed were what earned him the MBE on the recommendation of Admiral Halsey — although, as Halsey took command in October, the most likely scenario is that either Vandergrift or Turner, or both, told Halsey about the high-quality intelligence they had been receiving.
3 Military Factory, Kawanishi H8K (Emily) Reconnaissance / Bomber Flying Boat Aircraft (1942).
4 Stanley Clark, Oral History.
5 John Wukovit. ‘The Ill-Fated Goettge Patrol’.
6 Derek Dalton, ‘Contemporary Signals Intelligence: Speech for the Central Bureau Commemoration’.
7 Although it was not of interest to the traffic analysts, another feature of the preamble was of great interest to the code-breakers — the indicators. This was a key that gave the recipient specific information about how to unlock the message using the codebooks supplied. Once the operator had used a codebook to encode the message before sending, there would often be a second level of obfuscation using an additive book. The indicators told the recipient which section of the additive book, down to the page number, row and, column that the message had been encrypted with.
8 Example courtesy of Doug Pyle & Nell Pyle, The Ultra Experience, p. ix.
9 [Author name redacted], ‘The Origination and Evolution of Radio Traffic Analysis: World War II’.
10 ibid., p. 27.
11 Edward Drea, MacArthur’s Ultra, pp. 37–8.
12 ibid., p. 40.
13 ibid., p. 42.
14 ibid., pp. 46–7.
Chapter 19: Leaks
1 Sharon Maneki, The Quiet Heroes of the Southwest Pacific Theater, p. 29.
2 ibid., pp. 69–70.
3 Desmond Ball & Keiko Tamura, Breaking Japanese Diplomatic Codes, p. 23; Peter Donovan, personal communication.
4 Jess Bravin, ‘Echoes From a Past Leak Probe: Chicago Tribune reporter targeted after World War II scoop on Japanese navy codes’. Wall Street Journal, 7 August 2013; see also Peter Donovan & John Mack, Code Breaking in the Pacific, pp. 322–3.
5 ‘Breaking the Code on a Chicago Mystery from World War II’, Chicago Tribune, 21 November 2014.
6 Peter Donovan & John Mack, Code Breaking in the Pacific, pp. 318–21.
7 ibid., p. 311.
8 Ian Pfennigwerth, ‘No Contest! The US Navy Destroys Australia’s Special Intelligence Bureau’. Pfennigwerth writes, ‘Evidence has now been discovered in British and US archives clearly showing that up until the day before he was removed from ‘Monterey’ Nave and his staff were contributing ideas on solutions to this major code.’
9 NAA: A6923, SI/2, p. 12.
10 ibid., p. 11.
Chapter 20: Nyrambla
1 Peter Dunn, ‘Central Bureau in Australia during WWII’; Peter Dunn, ‘Eagle Farm @ War’. Ascot is a confusingly similar name to the Melbourne suburb Ascot Vale, where the Royal Australian Air Force had been running their Kana Code training in the showgrounds.
2 F. Lord, Brisbane’s Historic Homes: XCIX — Nyrambla.
3 Lynda Long, ‘Ballroom Has Harmony Restored’.
4 Desmond Ball & David Horner, Breaking the Codes, pp. 39–40; Australian War Memorial, Honours and Awards: Robert Arthur Little.
5 Gordon Gibson interview with David Dufty.
6 Geoffrey Ballard, On Ultra Active Service, p. 139.
7 NAA: B5436, Central Bureau Technical Records — H Traffic Analysis. See section 8.b on p. 21 (p. 23 of the electronic version), in which it is clear that Central Bureau was the central repository for Allied call signs. Given the amount of work done with paper documents and that Central Bureau was the main recipient of intercepts, no other arrangement would have been feasible. See also on the following page of the same document, ‘technical studies were handled mainly by CB’. While this is in the context of the contribution of field sections, it nonetheless names a location that the technical studies occurred in.
8 Edward Drea, MacArthur’s Ultra, pp. 33–5.
Chapter 21: Wet boots
1 Doug Pyle & Nell Pyle, The Ultra Experience, p. 27.
2 Bob Hartley, The History Of The Australian Special Wireless Group 1939 to 1947, Annex A, p. 8.
3 ibid.
4 NAA: B5436, Central Bureau technical records J – field sections.
5 Thanks to Gordon Gibson for this amusing snippet.
6 Geoffrey Ballard, On Ultra Active Service, p. 201.
Chapter 22: Sausages and sandshoes
1 History.com Staff, ‘Battle of Stalingrad’.
2 Edward Drea, MacArthur’s Ultra, p. 63.
3 Geoffrey Ballard, On Ultra Active Service, p. 176.
4 NAA: B5436, Central Bureau Technical Records – H Traffic Analysis, p. 11.
5 ibid., p. 6.
6 Stanley Clark, Oral History.
7 NAA: B883, VX21132, Clark, Stanley Robert Irving.
8 Sandford was ‘mentioned in dispatches’ for his work in the Middle East, and received a medal from the American army in 1947, described in Chapter 49.
Chapter 23: Planes in daylight
1 NAA: A6923 16/6/289, Australian Military Forces — Central Bureau — Administration of, pp. 67–8.
2 ibid.
3 Doug Pyle & Nell Pyle, The Ultra Experience, on page ix, mention Room’s contribution.
4 NAA: B5436, Central Bureau Technical Tecords Part C — Army Air-Ground Communications, p. 13.
5 NAA: B5436, Central Bureau Technical Records Part E — Naval Short Weather Synoptic weather reports — JN36, pp. 5, 17.
6 Judy (Roe) Carson interview with David Dufty.
7 Peter Dunn, 51 Wireless Section: ‘Australian Special Wireless Group, in Australia During WW2’.
8 NAA: A11093 334/81J Part 1, RAAF Command Headquarters GHQ – Committee ‘I’ – Signal Intelligence, p. 68.
9 ibid., p. 66.
Chapter 24: Computing machines
1 Joe Richard, ‘Inside Central Bureau: a first person account’.
2 Sharon Maneki, The Quiet Heroes of the Southwest Pacific Theater, p. 31.
3 Doug Pyle & Nell Pyle, The Ultra Experience, p. x.
4 Eve Scott, Interview.
5 Stephen Dunwall, ‘World War II Code-Breaking Exploits Unveiled By Retired IBM Engineer After 50 Years Of Secrecy’.
6 ibid.
7 Doug Pyle interview with David Dufty.
8 Stephen Budiansky, ‘Codebreaking with IBM Machines in World War II’.
Chapter 25: Killing Yamamoto
1 The letter E designated the codebook being used; the E codebook first came into use two months earlier.
2 Katsuhiro Hara & Shinzo Kitamura, Ango ni yabureta nihon: Taiheiyo senso no meian o waketa beigun no ango kaidoku.
3 John Winton, Ultra in the Pacific, p. 107; Joe Richard, ‘The Yamamoto Saga’; NAA: B5436, Central Bureau Technical Records PART J — Field Sections, p. 15. The section on 1 Wireless Unit makes it clear that 1 Wireless Unit was exclusively intercepting Japanese naval air-to-ground signals until August 1943, when it took responsibility for army air-to-ground from 55 Wireless Section based nearby.
4 Jack Bleakley, The Eavesdroppers, p. 95; Mac Jamieson, ‘Mac Jamieson Writes’; See also NAA: A1969/100 No 1 Wireless Unit 1/2/Air Part 3 Advance Unit Port Moresby, which states that the personnel in Port Moresby were 12 kana operators and two officers; no I staff were attached. See also Jack Brown, Katakana Man, p. 29, who confirms this detail.
Although it is not certain, the person who processed the message at Townsville was probably Alan Marsland. See Helen Kenny, Vale Alan Marsland.
5 NAA: B5436, Central Bureau Technical Records — Part A Organisation, p. 10. Section 4, ‘Intelligence Production in the Field,’ states that field sections could bypass Central Bureau and pass important messages directly to the top of command. In addition, 1 Wireless Unit had a close working relationship with the United States Army signals corps stationed there, so that there was a direct channel to the American chain of command.
6 Sharon Maneki, The Quiet Heroes of the Southwest Pacific Theater, p. 75.
7 Jack Bleakley, The Eavesdroppers, p. 95.
8 William E Clarke, ‘The Quiet Heroes Make Some Noise’; Helen Kenny, ‘The MacArthur Museum’; and Roy Inches, ‘Some Serious Eavesdropping (Sigint)’; ‘Use of outdated code led to ambush that killed Yamamoto, U.S. files show’, Japan Times, 29 September 2008.
9 Don Holloway, ‘Death by P-38’.
10 Steven Maffeo, US Navy Codebreakers, Linguists, and Intelligence Officers against Japan, 1910 –1941.
11 In Japanese, the Navy Kō Incident is rendered as 海軍甲事件
12 Jean Bou, MacArthur’s Secret Bureau, pp. 6, 10.
13 Peter Donovan & John Mack, Code Breaking in the Pacific, p. 157.
14 John Prados, Combined Fleet Decoded, p. 459.
15 Duane Whitlock, Interview, pp. 67–8.
16 Jack Brown, Katakana Man, p. 39.
Chapter 26: Noon positions
1 Joe Richard, ‘Joe Richard’s Written Legacy’.
2 Doug Pyle interview with David Dufty.
3 Joseph Richard, ‘The Breaking of The Japanese Army’s Codes’, p. 296.
4 Peter Donovan, personal communication. See also Peter Donovan & John Mack, Code Breaking in the Pacific.
5 ibid., pp. 233–40. In the chapter The Indicators of Cipher 2468, Donovan and Mack explain how the Water Transport code worked, what its weaknesses were, and exactly how Central Bureau broke it.
6 Doug Pyle & Nell Pyle, The Ultra Experience, p. 8.
7 Sharon Maneki, The Quiet Heroes of the Southwest Pacific Theater, p. 32.
8 Edward Drea & Joseph Richard, ‘New Evidence on Breaking the Japanese Army Codes’.
9 Sharon Maneki, The Quiet Heroes of the Southwest Pacific Theater, p. 36.
10 Edward Drea, MacArthur’s Ultra, p. 62.
11 ibid., p. 76.
Chapter 27: Kaindi
1 NAA: B5436, Central Bureau Technical Records J — Field Sections, p. 10.
2 Victor Lederer, A Span of Years.
3 NAA: B5436, Central Bureau Technical Records — J Field Sections, p. 3.
4 NAA: A705 132/12/20, Visit of Squadron Leader Pask RAF to SWPA.
5 Jack Bleakley interview with David Dufty.
Chapter 28: Joe Sherr’s final flight
1 NAA: B5436, Central Bureau Technical Records — Part A Organisation, p. 12.
2 Peter Dunn, ‘126 Signal Radio Intelligence Company: Us Army Signal Corps In Australia During WWII’.
3 Gil Murray, The Invisible War, p. 118.
4 Military Intelligence Network Hall of Fame (United States Army), 1988. Colonel Joe R Sherr US Army (Deceased); Stanley Clark, Oral History..
5 Doug Pyle & Nell Pyle, The Ultra Experience, p. x.
6 [Author redacted], The Origination and Evolution of Radio Traffic Analysis: World War II. On page 26, in a footnote, this article says ‘SRH 045, p. 56. As a Lieutenant Colonel, Brown was later in charge of radio intelligence in the Southwest Pacific Theater as a member of MacArthur’s signals staff.’ See also Ronald Lewin, The American Magic: codes, ciphers and the defeat of Japan (New York: Farrar Straus Giroux, 1982), p. 126.
7 Ernest Sidney L Goodwin was his (almost) full name.
8 Stanley Clark, Oral History.
9 Rudolph Fabian, Interview, pp. 26–7.
10 ibid., p. 62.
11 ibid., p. 27.
12 Sharon Maneki, The Quiet Heroes of the Southwest Pacific Theater, p. 81.
13 Pfennigwerth, Missing Pieces, p. 308.
Chapter 29: Cartwheels
1 Edward Drea, Macarthur’s Ultra, pp. 83–5.
2 George Odgers, Australia in the War of 1939–1945. Volume II – Air War Against Japan, 1943–1945, p. 70.
3 Jack Brown, Katakana Man, p. 40.
4 ibid. p. 48.
5 Jack Bleakley, The Eavesdroppers, p. 122.
6 Jack Brown, Katakana Man, p. 55.
7 ibid., p. 48.
8 Jack Bleakley, The Eavesdroppers, p. 123.
Chapter 30: The Sio box
1 David Dexter, The New Guinea Offensives: Australia in the War of 1939–1945, Series 1—Army. Volume VI, p. 735.
2 ANGAU is an abbreviation for Australian New Guinea Administrative Unit (thanks to Steve Meekin for this information).
3 David Dexter, The New Guinea Offensives, pp. 736–7.
4 Graham McKenzie-Smith, ‘The Other Dick Smith and the Sio Code Books’.
5 Jack Bleakley, The Eavesdroppers, p. 110.
6 Peter Dunn, ‘Tighnabruaich’.
7 Sharon Maneki, The Quiet Heroes of the Southwest Pacific Theater, p. 28.
8 Edward Drea and Joseph Richard, ‘New Evidence on Breaking the Japanese Army Codes’, p. 72.
9 ibid., p. 73
10 Peter Donovan & John Mack give a detailed description of the process in Code Breaking in the Pacific, p. 190.
11 Sharon Maneki, The Quiet Heroes of the Southwest Pacific Theater, p. 26.
12 ibid., p. 28.
13 ibid.
14 Jack Bleakley, The Eavesdroppers, p. 126.
15 ibid.
16 All these intercepts come from Jack Bleakley, The Eavesdroppers, p. 128.
Chapter 31: Pappy Clark goes to Washington
1 Colin MacKinnon, ‘Bletchley Park Diary: William F. Friedman’.
2 James Bamford, The Puzzle Palace, p. 315.
3 NAA: B5436, Central Bureau Technical Records — Part A Organisation, p. 11.
4 Stanley Clark, Oral History.
5 ibid.
Chapter 32: Hollandia
1 Jack Bleakley, The Eavesdroppers, p. 141.
2 CriticalPast, Allied forces capture Hollandia in New Guinea during World War 2 HD Stock Footage, https://www.youtube.com/watch?v=6YAh5DVr40o
3 Geoffrey Ballard, On Ultra Active Service, p. 239.
4 Jack Bleakley, The Eavesdroppers, p. 143.
5 Geoffrey Ballard, On Ultra Active Service, p. 239.
Chapter 33: The Yoshimo Maru
1 Gordon Gibson interview with David Dufty.
2 Sharon Maneki, The Quiet Heroes of the Southwest Pacific Theater, p. 29; Hyperwar, The Official History of the US Navy in World War II, Chapter VI, 1944. There has been some confusion about the identity of this ship, due to misspellings in various sources. The Central Bureau Technical Records misspell the name of the ship as the ‘Yoshino Maru’, which is the name of a different ship that was sunk near the Philippines later in the year. Sharon Maneki’s book, The Quiet Heroes of the Southwest Pacific Theater, also misspells it, using different spellings at different places in the book.
3 Sharon Maneki, The Quiet Heroes of the Southwest Pacific Theater, p. 41.
4 ibid., p. 29.
5 B5436, Central Bureau Technical Records — Part A Organisation, p. 12.
6 Sharon Maneki, The Quiet Heroes of the Southwest Pacific Theater, p. 37.
7 ibid., p. 29.
Chapter 34: Biak
1 Edward Drea, MacArthur’s Ultra, pp. 129–30.
2 Don Laidlaw, Anecdotes of a Japanese Translator 1941–1945.
3 J. Rickard, ‘Battle of Biak Island, 27 May–29 July 1944’.
4 Alan Axelrod & Jack Kingston, Battle of Biak Island.
Chapter 35: A view of Humboldt Bay
1 Geoffrey Ballard, On Ultra Active Service, p. 243.
2 ibid.
3 ibid.
4 Sharon Maneki, The Quiet Heroes of the Southwest Pacific Theater, p. 19.
5 ibid., p. 30.
6 The anecdotes about this voyage were told to me by Bill Rogers, a Central Bureau traffic analyst on board the van Swoll. They also appear in Doug Pyle & Nell Pyle, The Ultra Experience.
7 Doug Pyle & Nell Pyle, The Ultra Experience, p. 46.
8 ibid., p. 55.
9 Anecdote told to me by Bill Rogers.
10 Sharon Maneki, The Quiet Heroes of the Southwest Pacific Theater, p. 19.
Chapter 36: The Battle of Morotai
1 D-Day Museum, ‘D-Day and the Battle of Normandy: Your Questions Answered’.
2 Arthur Herman, Douglas MacArthur: American warrior, pp. 519–20.
3 ibid., p. 520.
4 Francis Pike, Hirohito’s War: the Pacific War, 1941–1945, p. 786.
5 Even then, many of the Japanese defenders retreated into the island’s rugged interior. The last remaining soldier, Teruo Nakamura, was captured on Morotai in 1974, his last orders being to ‘fight on,’ which he did.
6 Thanks to Professor Peter Donovan for making this point, about how Central Bureau prevented several battles, and thus, in a sense, saved lives — on both sides. Donovan described Halmahera as ‘The Battle that Never Happened’.
Chapter 37: Akin’s secret unit
1 Geoffrey Ballard, On Ultra Active Service, p. 246.
2 Jack Brown, Katakana Man, pp. 69, 73. A ‘Dakota’ is the nickname for the ‘Douglas C-47 Skytrain’ aircraft, a fast passenger aircraft used extensively during WWII.
3 Cannon, M. Hamlin, Leyte: The Return to the Philippines, https://www.ibiblio.org/hyperwar/USA/USA-P-Return/USA-P-Return-4.html
4 Jack Bleakley, The Eavesdroppers, pp. 176–7.
5 Alfred Bobin, Bobin, Alfred (Leading Aircraftman, 1 Wireless Unit, RAAF).
6 Sharon Maneki, The Quiet Heroes of the Southwest Pacific Theater, p. 57.
7 Alfred Bobin, Bobin, Alfred.
8 Robert Nichols, ‘The first kamikaze attack?’
9 Alfred Bobin, Bobin, Alfred.
10 ibid.
11 Alf Davis, ‘A few words from Alf Davis, ex Fl/Lt RAAF’.
12 Alfred Bobin, Bobin, Alfred.
13 The History Channel, ‘I Have Returned’ MacArthur Returns to the Philippines.
14 Harry Popham, ‘Eyewitness to a tragedy: Death of USS Princeton’, HistoryNet,http://www.historynet.com/eyewitness-to-tragedy-death-of-uss-princeton-may-97-world-war-ii-feature.htm. Accessed 8 August 2015.
15 Alfred Bobin, Bobin, Alfred.
16 NAA: B5436, Central Bureau Technical Records Part C - Army Air-Ground Communications, p. 15.
17 Jack Bleakley, The Eavesdroppers, pp. 202–3, has an early intercepted message about this.
18 David Jenkins, ‘Discreet Reminder of Japanese Zeal, And Valour’, Sydney Morning Herald, 8 September 1988.
19 Geoffrey Ballard, On Active Ultra Service, p. 251.
20 ibid., p. 298.
Chapter 38: Spies
1 Frederick Winterbotham, The Ultra Secret, p. 172.
2 Desmond Ball & David Horner, Breaking the Codes: Australia’s KGB network, 1944–1950, p. 77.
3 ibid., p. 21.
4 ibid., p. 84.
5 NAA: A6923, SI/3, pp. 63–5.
6 NAA: A6923, SI/8, p. 82.
7 Desmond Ball & David Horner, Breaking the Codes, p. 105.
8 ibid., p. 132.
9 ibid., p. 343.
Chapter 39: Lingayan Gulf
1 Alf Davis, ‘A few words from Alf Davis, ex Fl/Lt RAAF’.
2 C. Peter Chen. ‘Philippines Campaign, Phase 2: 12 Dec 1944 – 15 Aug 1945’.
3 Seth Mydans, ‘Japanese Veteran Writes of Brutal Philippine War’, New York Times, September 2.
4 AWM Number 4 Wireless Unit History, p. 1 (see also Hartley History of No 4 Wireless Unit, p. 44); NAA: B5436, Central Bureau Technical Records Part C — Army Air-Ground Communications, p. 16.
5 NAA: MP150/1, 404/201/691.
Chapter 40: The expanding web
1 Paul Hetherington, The Diaries of Donald Friend, Volume 2, p. 247.
2 Joan Carolan, ‘Vale Keith Carolan’.
3 ibid.
4 ibid.
5 Keith Carolan, ‘Labuan Revisited’.
6 Victor Lederer interview with David Dufty; Victor Lederer, A Span of Years.
7 Gil Murray, The Invisible War, p. 19.
8 ibid., p. 132.
9 ibid., p. 135.
Chapter 41: The paper war over women
1 Bob Hartley, Central Bureau Complete History, p. 682.
2 ibid., pp. 681–2.
3 ibid., pp. 687–8.
4 NAA: MP729/8, 41/431/18, Allied Central Bureau —Transfer to Manila, p. 3.
Chapter 42: Tarlac
1 B5436, Central Bureau Technical Records Part C — Army Air-Ground Communications, p. 16.
2 Allan Norton interview with David Dufty.
3 ibid.
4 Every single Central Bureau veteran I spoke to who was posted at San Miguel spent time in the hospital for dysentery, malaria, or some other serious illness.
5 The anecdote about the brothel was told to me by Bill Rogers.
6 Doug Pyle & Nell Pyle, The Ultra Experience, p. 64.
7 ibid., p. 66.
8 Anecdote from Bill Rogers.
9 Doug Pyle & Nell Pyle, The Ultra Experience, pp. 71–2.
10 Gordon Gibson interview with David Dufty.
11 Edward Drea, ‘Were the Japanese Army Codes Secure?’
Chapter 43: The Garage: a cryptological love story
1 Alexandra Patrikios, ‘WWII Codebreakers Were “Slent Heroes” Operating out of an Ascot House’, Brisbane Times, 11 July 2015.
2 Helen Kenny, ‘Vale Captain Ian Allen’.
3 Doug Pyle & Nell Pyle, The Ultra Experience, p. 62.
Chapter 44: All eyes on Kyushu
1 Douglas MacArthur, Reports of General MacArthur: the campaigns of MacArthur in the Pacific, Volume I, p. 410.
2 Military History Now, Operation Downfall — The Campaign to Conquer Japan Would Have Dwarfed the D-Day Landings.
3 Edward Drea & Joseph Richard, ‘New Evidence on Breaking the Japanese Army Codes’, p. 79.
4 Edward Drea, MacArthur’s Ultra, p. 219.
5 ibid.
6 WGBH, The American Experience: victory in the Pacific.
7 Tsuyoshi Hasegawa, Racing the Enemy: Stalin, Truman, and the surrender of Japan, p. 270.
8 Col Brackley interview with David Dufty.
Chapter 45: The mushroom cloud
1 Paul Ham, ‘The Bureaucrats Who Singled Out Hiroshima for Destruction’.
2 Avalon Project, Yale Law School, ‘The Atomic Bombings of Hiroshima and Nagasaki; 29 June 1946’.
3 Edward Drea, MacArthur’s Ultra, p. 219.
4 Wilson Miscable, A Most Controversial Decision: Truman, the atomic bombs and the defeat of Japan, p. 86.
5 Jack Bleakley, The Eavesdroppers, 219.
6 There are many vivid examples in Jack Bleakley, The Eavesdroppers.
7 Jack Bleakley, The Eavesdroppers, p. 221.
8 Wilson Miscable, A Most Controversial Decision, p. 96.
9 ibid., pp. 96–7.
Chapter 46: A message for the emperor
1 Bill Rogers interview with David Dufty.
2 ibid.
3 Jack Bleakley, The Eavesdroppers, p. 223.
4 Morton Sontheimer, ‘The Radio Drama of V-J Day’.
5 Stanley Clark, Oral History.
6 Jack Bleakley, The Eavesdroppers, p. 222.
7 Sharon Maneki, The Quiet Heroes of the Southwest Pacific Theater, p. 45.
8 ‘Australia Wants Japanese Emperor Tried As War Criminal’, The Mercury, 14 August 1945.
9 Stanley Clark, Oral History.
10 NAA B883, VX21132, Clark, Stanley Robert Irving, p. 18.
Chapter 47: Peace in the Pacific
1 Jack Bleakley, The Eavesdroppers, p. 225.
2 ibid., p. 222.
3 Australian War Memorial, The Burma–Thailand Railway.
4 This anecdote about the trip to Tokyo was told to me by Gordon Gibson and Brian Walsh in separate interviews, and parts of it can also be found in Helen Kenny, ‘Taking Off for Tokyo’.
5 Booth’s presence in Japan is given by a letter from Sandford to Rogers, see Bob Hartley, Central Bureau Complete History, p. 807.
Chapter 48: Radio silence
1 Told to me by Bill Rogers.
2 Helen Kenny, ‘Book Review: Anecdotes of a Japanese Translator, 1941–1945, by D.H. Laidlaw, A.O.’
3 NAA: A6923 SI/2, Staffing, p. 77.
4 ibid., p. 63.
5 NAA: A649, 82/605/25, pp. 6–7 and 9. The IBMs were the most expensive item installed at Central Bureau’s headquarters.
6 NAA A6923, SI/2, Staffing, p. 47.
7 ibid., p. 33.
8 NAA: A6923, SI/2, Staffing, p. 21. Sandford misspelt Ayre’s first name: it was Denys (not Dennis).
9 Jean Bou, MacArthur’s Secret Bureau, p. 112.
10 NAA: A6923, SI/2, Staffing, p. 22.
11 Julius Allain Cohen, who changed his name to Richard Kingsland in 1947, had a fascinating Second World War career that is unable to be covered by this book.
12 The first page of the minutes can be found in the Australian National Archives in A6923 SI/2, Staffing. It is page 4 of the electronic version that can be viewed online through National Archives record search.
13 NAA: A6923 SI/7, p. 5.
14 Two other CB originals that stayed on were ‘Mos’ Williams and Bob Botterill.
15 Helen Kenny, ‘Chooks at Chermside (A CB Mystery)’.
16 A comparison of a map of the Chermside barracks from 1945 with a modern map of Chermside shows that the discs were buried right in the middle of the AWAS grounds.
17 Clarrie Hermes, ‘First and Last but Not Oldest’.
Chapter 49: Secret medals
1 NAA: A816, 66/301/232, Proposed Awards of US Decorations to Lt.Col. A.W. Sandford and Maj. S.R.J. Clark, AMF & to W/Cdr. H.R. Booth, RAAF.
2 ibid., p. 12.
3 ibid., p. 11.
4 ibid., p. 4.
5 ibid. pp. 6–7.
6 ibid., p. 2.
7 NAA: A6923, 16/6/289, Australian Military Forces — Central Bureau — Administration of, p. 4.
8 ‘Obituary: Joseph Eugene Richard’, Washington Post, 13 April 2005.
Chapter 50: Return to Nyrmambla
1 Steve Meekin, ‘Address by Mr Steve Meekin’.
2 Derek Dalton, ‘Contemporary Signals Intelligence’.
3 Doug Pyle interview with David Dufty.
4 ibid.; Brian Walsh interview with David Dufty.
5 Jack Brown, Katakana Man, p. 150.
6 Madeline Chidgey, ‘Bicentennial Activities’.
Acknowledgements
In the course of my research on this topic, several individuals involved in wartime signals intelligence spoke with me at length, in many cases inviting me into their homes. Their first-hand accounts contained the vibrancy of human experience that is so hard to capture from archival sources alone. These conversations shaped my understanding of wartime events, guided my investigations, and influenced the tone and content of the final book.
The veterans I spoke to were (in chronological order), Margo Isaksson (née Seward) and her husband Zac Isaksson, Joan Fairbridge (née Duff), Don Field, Doug Pyle, Helen Kenny (née Frizell), Gordon Gibson, Victor Lederer, Frank Hughes, Madeline Chidgey (née Bell), Di Parker, Allan Norton, Jack Bleakley, Judy Carson (née Roe), Bill Rogers, Brian Walsh, Coral Hinds (née Osborne), Keith Falconer, Colin Brackley, and Robert Brown.
While I am grateful for the assistance of all these individuals, three deserve special mention (although in saying so I do not intend to diminish the contribution or hospitality of others). Doug Pyle’s uniquely detailed knowledge of the inner workings of Central Bureau was invaluable. He and his wife, Nell, have hosted me at their house on several occasions, and they introduced me to the Central Bureau veterans’ community.
Victor Lederer has been similarly gracious over numerous conversations both in person and by phone. Vic provided me with a copy of his memoirs, invited my partner and me as guests to his medal ceremony in 2016, and regaled me with a large store of entertaining anecdotes. His wife, Tina, and sister, Arena, deserve my thanks as well.
Joan Fairbridge, in conversations during my two visits to her at home, as well as in numerous phone conversations, provided vivid accounts of key figures — particularly Jack Newman, with whom she worked closely.
The Central Bureau Intelligence Corps Association has been supportive, providing contacts, documents, invitations to events, and generally helping however they can. The association’s treasurer, Annette Salmon, in particular, has been hugely supportive and helpful. Rob Moore, Mark Brackley, Katy Denis, and Bruce Goudge also deserve thanks, not just for assisting me, but for the voluntary work they do for the association.
Peter Dunn is the creator and curator of Australia’s best independent resource on Australian military history, the website Ozatwar.com. Peter made available to me additional resources that are not publicly available on his website.
Bob Hartley has spent several years compiling information about Australian signals operations during the war, and generously provided me with the detailed documents on Central Bureau and the Australian Special Wireless Group, both of which he has lodged with the Australian War Memorial.
Several people read early drafts, or portions of drafts, and provided feedback, including Charlotte Harper, Maria Hawthorne, Mary Cunnane, Jenny Edwards, Barbara Dufty, Steve Meekin, and Professors Peter Donovan and John Mack. Charlotte Harper was an early supporter of this project and helped me make the decision to make the plunge in taking it on. Mary Cunnane helped the book on its journey toward publication. Professor Donovan deserves special mention. He provided extensive feedback, and over the course of several interactions and conversations has given insights, tips about obscure references, and clarifications of various points of fact. His knowledge of the Second World War extends far beyond his nominal area of expertise of code breaking.
My editor, Henry Rosenbloom, has my thanks both for taking on the book, and for the amount of time and effort he has personally spent on it. I also want to thank everyone at Scribe for helping make it a reality.
Finally, I want to thank my partner, Jenny Edwards, who has supported and encouraged me throughout the long journey from inception to completion.