The Bletchley Park Codebreakers (19 page)

Meanwhile, the Americans were able to push ahead. The US Navy codebreaking unit at Pearl Harbor, under the brilliant Commander Joe Rochefort, enjoyed good reception of the Japanese Navy messages and was breaking JN-25 with ease. As a result, the Americans were able to achieve the extraordinary feat of deciphering the complete Japanese operational orders for Admiral Yamamoto's attempt to draw the US Pacific Fleet into an ambush off the island of Midway, 1,000 miles west of Hawaii. Having obtained the full details of its opponent's plans in advance, the US Navy inflicted a crushing defeat
on the Japanese, destroying four irreplaceable aircraft carriers and putting them on the defensive for the rest of the war.

On the other hand, the British codebreakers working on Japanese Navy material were now at their lowest point of the war. The belligerent Fabian blamed security concerns over Nave for his reluctance to exchange material with the British codebreakers at Kilindini. But they were too far away from the action to intercept anything apart from the loudest signals that FRUMEL, the US Navy's Melbourne site, and FRUPAC, the US site in Hawaii, were already able to receive without assistance. Fabian was also completely open over his belief that his unit had no reason to exchange material with anyone who could not give him anything in return. One senior British officer who visited FRUMEL said that ‘the most notable feature was the inability of the Americans to appreciate the full meaning of the word “co-operation”. The atmosphere was “What is yours is mine, and what is mine is my own”.' Having given the Americans the help they needed before the start of the war, the British were now denied the technical assistance they desperately needed to get back on top of JN-25. With Admiral Sir James Somerville, Commander-in-Chief of the Royal Navy's Eastern Fleet, complaining that Kilindini's problems, and the US Navy's concentration on the Pacific at the expense of the Indian Ocean, meant he was not receiving sufficient intelligence, the Admiralty considered the most drastic of measures.

‘The lack of US intelligence supply to C-in-C Eastern Fleet led the British to consider ditching the Americans on the Japanese side,' said Frank Birch, the head of the Bletchley Park Naval Section. ‘Admiralty was not willing to be dependent on such small scraps as US were willing to provide and the only alternative to sharing all available intelligence between the two countries was for this country to build up independently an organization big enough to provide, without American help, as much intelligence as could be got with American help.'

As a result, the British compromised on their previous ‘Europe first' approach and began to expand the Japanese naval section, Hut 7, as well as sending more codebreakers out to Kilindini. Despite their difficulties, the Kilindini codebreakers did have some significant successes, most notably with the Japanese Merchant Shipping Code, dubbed JN-40. This was believed to be a super-enciphered code similar to JN-25. But in September 1942 a textbook error by the Japanese gave John MacInnes and Brian Townend, another of the civilian codebreakers sent to
Kilindini by GC&CS, the way in. The Japanese operator omitted a ship's position from a detailed message and instead of sending it separately in a different message, re-enciphered the original with the same keys, this time including the longitude and latitude that had previously been missing. A comparison of the two messages made it immediately clear that JN-40 was not a code at all. It was in fact a transposition cipher. It was based on a daily changing substitution table, containing 100 two-figure groups or dinomes, each representing a
kana
syllable, a
romaji
letter, a figure, or a punctuation mark. The operator wrote out the message in
kana
syllables and then substituted the relevant dinomes. This produced a long sequence of figures, which was written into a 10 by 10 square horizontally and then taken out vertically, thereby splitting up the dinomes and making it more difficult to break. “Within weeks, MacInnes and Townend had not only broken it but were able to read all previous traffic and were confident of breaking each message in real time, allowing enemy supplies to be tracked and attacked at will by Allied submarines. What was more, since it was a cipher, there were no code groups to recover and therefore no gaps in any of the messages. Over the next fortnight, they broke two more systems. The first was the previously impenetrable JN-167, another merchant shipping cipher. The second was JN-152, a simple transposition and substitution cipher used for broadcasting navigation warnings.

There had also been a notable British success on a Japanese military code in 1942. The mainline Japanese Army codes had so far evaded the efforts of the Allied codebreakers. But John Tiltman had spent some of the early part of 1942 trying to break the Japanese military attaché (JMA) code. He discovered that it was a digraph code in which the basic
kana
syllables stood for themselves and other two-letter groups stood for certain words or phrases commonly used in military communications, e.g. AB stood for ‘west' and AV for ‘message continued'. The two-letter groups were then set out in a square grid in adjacent squares, sometimes horizontally and sometimes diagonally, and the letters were read off vertically to form the basis for the encrypted text. They were then enciphered using a prearranged ‘literal additive', a series of letters that would be notionally ‘added' to the letters taken out of the grid on the basis of a pattern laid down in advance on a separate table. Reading off the enciphered letter along the relevant horizontal line and the ‘additive' letter down the appropriate vertical column would produce a super-enciphered letter, which would
be transmitted by the operator. ‘By March 1942, I and my section had partially recovered the indicating system and had diagnosed the cipher as a literal additive system with indicators which gave the starting and ending points for messages,' Tiltman recalled. ‘The normal practice was to tail successive messages rigorously through the additive tables, i.e. to start reciphering each message with the additive group following the last group of the preceding message.'

Tiltman set to work on a large number of JMA messages from one particular embassy, where the cipher clerk had used the additive table again and again, giving a large ‘depth' to attack. ‘It was clear from the indicators that the sender had tailed right round his additive table five times and it was this depth that I set myself to resolve.' The solution took a lot of work, but with a depth of five on the cipher additive, he eventually managed to break the system. One of the first JMA messages deciphered revealed the Japanese intentions to construct a Burma Railroad. It was not until several months later that it became clear from another Japanese military attaché decrypt that British prisoners-of-war would be used as slave labour to build the railway.

Tiltman set up a small Japanese military section at Bletchley Park in June 1942. It comprised a codebreaking sub-section and a traffic analysis team, but its main purpose was to handle the JMA material. The staff came mainly from a Japanese course set up by Tiltman in Bedford, which confounded the experts by turning out fluent Japanese linguists within the space of six months.

By the beginning of 1943, there were a number of British sites around the world intercepting or decoding Japanese messages. At Bletchley Park, there was the Japanese naval section Hut 7, under the control of Hugh Foss, and Tiltman's military section. Apart from Kilindini, there were a number of Bletchley Park outposts covering Japanese armed forces traffic. The main outpost was a joint RAF and Army intercept and codebreaking operation at the Wireless Experimental Centre (WEC) just outside Delhi. There was also a Wireless Experimental Depot at Abbottabad on the North West Frontier, but this mainly covered diplomatic links. The WEC had two main outposts: the Western Wireless Signal Centre at Bangalore in south-western India, which was merely an intercept site; and the Eastern Wireless Signal Centre at Barrackpore, near Calcutta. The latter site had its own code-breakers and traffic analysts as well as its own forward intercept and codebreaking outpost at Comilla, covering
tactical air force communications nets. There was also a British officer. Major Norman Webb, based at the Central Bureau, a joint US and Australian Army codebreaking operation in Brisbane, where Nave was also now working, following his disagreement with Fabian.

The Central Bureau was also facing difficulties in its relationship with FRUMEL. It originally attempted to set up an exchange relationship with the US Navy site, but Fabian was not interested and was backed in his reluctance to exchange material by Admiral Joe Redman, the Director of Naval Communications. This was perhaps unsurprising. For much of the war, the US Navy codebreakers and their US Army counterparts were barely on speaking terms. But Fabian insisted that the Central Bureau could give FRUMEL nothing it needed and it had nothing that could help them. The US Navy unit was ‘concerned solely with information on Japanese naval circuits', Fabian said, where ‘the Central Bureau was not'. It was not that simple. Few wars had seen more need for complete co-operation between the army and navy. The Japanese Army was forced by the very nature of the campaigns it was fighting, cut off from its home bases by thousands of miles of ocean, to pass messages on naval communications circuits, often in naval codes and ciphers. Messages would be translated from one system to the other, providing a wealth of potential ‘cribs' if only they could have been followed through the system.

It is impossible to say how much help this might have given the army codebreakers. In the event, the US Army codebreakers at the Central Bureau and the two main British outposts converged on the solution to the first mainline army code at the same time in March 1943. Some credit Wilfred Noyce, a classicist from King's College, Cambridge, and a prominent mountaineer, who was working at the WEC in Delhi, with having broken the Japanese Army Water Transport Code,
senpaku angosho
2, assisted by Maurice Allen, an Oxford don. The difficulties of communication between the various codebreaking bases make it difficult to tell for sure whether they were first. They may well have been beaten by Warrant Officer Joe Richard, a US Army codebreaker at the Central Bureau, Brisbane, who also broke it at around the same time. There is even a possibility that Brian Townend, who remarkably was working on it in his spare time at Kilindini, may have broken it first.

Bletchley Park was not in any way involved in what was perhaps the most controversial use of the ability to break the high-level Japanese
codes – the shooting down, in April 1943, of an aircraft carrying the Japanese Navy's Commander-in-Chief Admiral Yamamoto Isoruku. A JN-25 message giving the itinerary of a tour by Yamamoto of the Solomon Islands was deciphered by the US Navy codebreakers in Hawaii, allowing US fighters to intercept his aircraft and shoot it down, killing all those on board. The Japanese ordered an investigation but fortunately details of the visit had also been sent on low-level nets and it was assumed that the Allies had intercepted these rather than having been able to decipher the original JN-25 message.

Meanwhile, the Japanese military section, where John Tiltman was leading the research, broke the Army Air Force General Purpose Code,
koku angoo-sho 3
, designated 3366, and began to expand to cope with the increased intelligence expected as a result of the recent successes. It acquired its own Army Air subsection in May 1943, presumably as a direct result of the breaking of the 3366 code, and some weeks later, a military intelligence section was added, reporting direct to MI2, the War Office intelligence section covering the Far East.

The breaking of the first two mainline army codes led Tiltman to call a conference at Bletchley Park in July 1943 to allocate coverage among the various Allied outposts. It was agreed that Bletchley Park and Delhi should concentrate their cryptographic resources on the codes and ciphers of the Japanese Army Air Force. Arlington Hall (headquarters of the SIS) would deal with the high-level systems used by the Japanese ground forces, leaving the Central Bureau to concentrate on the low-level material produced by their forward field units and the Army Water Transport code.

A separate Japanese air intelligence section was set up at Bletchley Park in October 1943, with Leonard ‘Joe' Hooper, a future head of the postwar GCHQ, in charge. So many Japanese military messages were now being sent over Japanese naval circuits, and in a number of cases actually using JN-25 and other naval codes and ciphers, that it was decided to co-locate all the Japanese sections. They were moved into Block F, the largest of a series of brick-built blocks designed to cope with the expansion of Bletchley Park. This consisted of a number of wings jutting off a long, central corridor which, as a result of its length and the proliferation of various Japanese subsections in the block, became known as ‘the Burma Road'.

The summer of 1943 saw a marked improvement in the relations between the British and the US Navy codebreakers. Redman attended
a number of conferences at Bletchley and the Admiralty during which he promised to ensure that the Royal Navy codebreakers would receive all urgent Japanese Navy material and all the American JN-25 recoveries. The Royal Navy codebreakers were to return to Colombo in September to improve their coverage and Redman agreed that the new station would have a direct cipher link to Melbourne.

The new Colombo site was on the Anderson Golf Club. The Royal Navy codebreakers were soon swamped with Japanese naval messages to work on. But despite Redman's promises the exchanges with Melbourne did not improve. Commander Malcolm Saunders, a former head of Bletchley Park's Hut 3 intelligence-reporting section, toured Allied naval codebreaking sites in the autumn of 1943. He was very impressed with the FRUPAC centre in Hawaii but less so with FRUMEL, where Fabian continued to block co-operation. ‘The liaison with Colombo is not nearly as good as it should be,' Saunders said. ‘This is partly due to bad communications and insufficient staff at each end, but also due to the present lack of productivity of the Colombo unit, and to lack of a clear-cut statement of policy from Washington in this regard. The security aspect is constantly in mind and there is a constant suspicion of “leakage” to the American military authorities at Brisbane.'