Friday, July 27, 2012

Japanese codebreakers of WWII

Imperial Japan entered WWII with three separate codebreaking agencies under the control of the Army, Navy and Foreign Ministry. Due to the hostility that existed between the Army and the Navy these departments did not cooperate but instead often attacked the same problems independently. Both however furnished material to the much smaller department of the Foreign Ministry.

Details of their successes against enemy codes have been hard to find because after Japan’s surrender, in September 1945, they had time to destroy their records and disperse their personnel. Still the few remaining documents in Japan combined with decoded Japanese messages found in the British archives can provide a basis for assessing their operations during the war.

Japanese Army agency
The beginnings of a centralized army cryptologic service date back to 1921 when a study group comprised of Army, Foreign Ministry and Ministry of communications cipher specialists was formed to work towards the solution of US and British codes. In 1922 the Japanese were involved in negotiations with the Soviet Union regarding their forces in Siberia. At that time the crypto service was successful in solving the code of the Soviet delegation. This success proved the importance of codebreaking and more resources were assigned to that department.
Foreign assistance was obtained from the Polish cryptologic service which had an excellent record versus Soviet codes. Captain Kowalewski of the Polish army was invited to Tokyo in 1923 and a small group of Japanese officers were sent to Poland to study. It was this group that formed the basis of the Army’s codebreaking department. By 1936 this department (which often changed designation) had 135 people.

The main effort of the army agency was against Soviet and Chinese codes. This made sense as army units were fighting against the Chinese army and there were a series of border engagements with the Soviet armed forces in Manchuria. Chinese Kuomintang military and diplomatic codes were solved and they gave the Japanese valuable intelligence on upcoming operations and diplomatic initiatives. For example the movement of 54 Chinese divisions in 1940 was detected and followed by solving the Chinese army’s 4-figure code Mi-ma. At the same time the solution of Soviet military and NKVD border unit codes allowed the Japanese to keep a close eye on Soviet dispositions, training and supply in the East. Codebreaking gave them an advantage prior to the Battle of Lake Khasan. From 1943 onwards the Japanese could solve the Soviet diplomatic code used by the embassies/consulates in Seoul, Dairen, and Hakodate for communications with Moscow and Vladivostok. In Harbin, China a spy inside the Soviet embassy gave them copies of intelligence reports coming in from the Soviet embassy in Australia.
The Japanese were helped in their efforts through black bag operations. In the late 1930’s the US diplomatic codes Brown and M-138-A strip were copied by a unit of the Military police. British diplomatic systems Cypher M, Interdepartmental and R code were also physically compromised. These systems allowed the Japanese to read the communications of the US and British ambassadors in the period 1940-41. The Interdepartmental Cypher provided valuable intelligence on the state of British defenses in Malaya.

During WWII Army codebreakers were forced to devote resources to the codes of the United States. Apart from low level codes the M-209 cipher machine was successfully analyzed and decoded in late 1944. Bombing missions of the B-29’s were betrayed through their use of the M-209.
In 1944 a major effort was made to improve performance by recruiting university students from mathematics and foreign language departments. Also in the same period IBM punch card equipment was used for cryptanalysis. These efforts came too late to have an impact on the war situation but they show that the Japanese leadership understood the value of secret intelligence.

Japanese Navy agency
The Japanese Navy’s signal intelligence agency was older than the army’s and its beginnings dated back to the Russo-Japanese war of 1905. A centralized codebreaking department was formed in 1929 to attack US and British communications.
The Navy’s efforts were directed mainly against the United States. The naval codebreakers were able to decode the US diplomatic codes Gray and Brown but unlike their Army counterparts they could not solve the high level M-138-A. Against British diplomatic codes they had very limited success. By reading the Anglo-American diplomatic codes they could see the rising tension in the relations between Japan and the US in the 1930’s.

In order to keep an eye on US fleet movements several monitoring stations were operated prior and during the war. An interesting case was the undercover L agency. In 1938 a small unit called the ‘L Agency’ (L-Kikan) was established in Mexico to monitor US Fleet traffic in the Atlantic and also the commercial RCA radio from New York City.
During the Pacific war most US military codes proved secure. There was only limited success with the US Navy’s CSP-642 strip cipher. However the codes used by merchant ships had been received from the Germans and their enciphering tables were solved in Japan.

The Japanese were able to track the movement and concentrations of merchant ships and thus anticipate major allied operations by reading the Merchant Navy Codes. They also came to rely more and more on D/F and traffic analysis for tracking enemy fleet movements.
Against Soviet systems they were able to solve the codes used by Soviet Merchant Navy ships in Kamchatka and Vladivostok.

In general the performance of the Naval codebreakers versus enemy codes was not as successful as that of their Army counterparts.

Foreign Ministry’s decryption department

Information on the decryption department of the Japanese Foreign Ministry is limited since their archives were destroyed twice during the war. First in a bombing on 25 May 1945 and then in August 1945, when they were ordered by their superiors to burn all secret documents.

According to the recently declassified TICOM report DF-169 ‘Cryptanalytic section Japanese Foreign Office’ this department was established in 1923 and by the end of WWII had approximately 14 officials and 16 clerks. The radio intercept unit supplying it with messages had a station in Tokyo equipped with 10 receivers and 19 operators. They usually intercepted 40-60 messages per day with 100 being the maximum.

The emphasis was on the solution of the codes of the United States, Britain, China and France but some German, Turkish, Spanish, Italian, Swiss, Thailand and Portuguese codes were also read. Despite their limited resources it seems that the Foreign Ministry’s codebreakers were able to achieve their goals mainly thanks to compromised material that they received from their Army and Navy counterparts.

 Japanese radio security services
The identification of agents’ radio transmissions and their location through direction finding was the job of specialized radio security units. In Japan there were two agencies that carried out this mission. One was the Science group of the War Ministry’s Investigation department - Otsu-han. The other was a similar department of the military police Kenpei-Tai.

There were also radio security units with the Japanese armies in China. The military police of the Kwantung Army had a D/F group called ‘Unit 86’. The military police of the Expeditionary Army to China had a similar group named ‘6th Section’. This group was able to locate Soviet radio spies in Shanghai during the war.

Cooperation with foreign powers

An important aspect of the Japanese approach to intelligence was their effort to cooperate with other countries. They were able to gain allies in two ways. One was by offering stolen enemy codes. The other was by spending lots of money.
During WWII the Japanese had huge sums of money available for intelligence operations. German intelligence officers commended on the ease with which their Japanese counterparts could acquire and spend large funds. At the same time copied US and British codes were shared with their German Allies in exchange for other cryptosystems.

Cooperation with the Germans
The Japanese had received valuable material from the Germans in November 1940  when a party from the German auxiliary cruiser Atlantis boarded the British SS Automedon and captured top secret documents. Among them was a copy of the British War Cabinet minutes of August 1940. These files gave a summary of the British Far Eastern strategy and admitted that Thailand and Hong Kong were indefensible. They also indicated that Britain would not go to war with Japan over the fate of French Indochina. These documents were given to the Japanese and allowed them to correctly assess the weakness of the British in the Far East. The captain of Atlantis, Bernhard Rogge, was given a samurai sword for this success!

A Japanese mission headed by Colonel Tahei Hayashi, former head of the Army’s cryptologic agency visited Germany in 1941 and exchanged US and British codes with systems solved by the Germans. This promising start did not lead to closer cooperation as communications between Japan and Germany were problematic. Moreover the Germans did not trust the Japanese with their most recent codebreaking successes. Things changed in summer ’44, when under Hitler’s orders several high level systems (including the latest strips for the M-138-A cipher) were given to the Japanese.

According to Wilhelm Fenner, head of the cryptanalysis department of OKW/Chi, about 200 decoded messages were passed on to the Japanese in 1944-45. For example:

Cooperation with the Finnish cryptologic service
Major Eiichi Hirose was sent to Finland to exchange results with their codebreakers. Also General Makoto Onodera who was military attaché in Stockholm financed the Finnish crypto service in exchange for copies of their work. The Finnish codebreakers were successful in solving several Soviet and US State department codes (especially the M-138-A strip cipher). These were passed on to the Japanese. For example:

In September 1944 Finland signed an armistice with the Soviet Union. Paassonen and Hallamaa anticipated this move and fearing a Soviet take-over of the country had taken measures to relocate the radio service to Sweden. This operation was called Stella Polaris (Polar Star). Roughly 700 people, comprising members of the intelligence services and their families were transported by ship to Sweden. The Finns had come to an agreement with the Swedish intelligence service that their people would be allowed to stay and in return the Swedes would get the Finnish crypto archives and their radio equipment. Their archives were also sold to the Japanese military attaché Makoto Onodera.

Cooperation with the Hungarians
Lieutenant Colonel Shinta Sakurai was sent to Hungary to cooperate with that country’s crypto service.

Use of Polish codebreakers
Cooperation between Japan and Poland in the cryptologic field dated back to the 1920’s. After the fall of Poland some Polish exiles were employed by the Japanese at the attaché office in Rumania, where they worked on Soviet codes.

Overview of major codebreaking successes

Soviet codes
Several military and NKVD border guard figure codes were read (OKK, OK40, PK1), thanks in part to Finnish and German help. These were used to monitor the movements and readiness of Soviet units in the Far East.

Messages from Soviet Merchant Navy ships in the coastal areas of Kamchatka and Vladivostok were read by the Japanese.

The Soviet diplomatic code used in the East by consulates/embassies in Seoul, Dairen, and Hakodate for their communications with Moscow and Vladivostok was read by the Japanese from 1943 onwards. This was not the standard Soviet diplomatic system of codebook plus one time pads but a simpler system

Intelligence reports from Australia were copied by a Japanese spy working in the Soviet embassy in Harbin, China. These messages came from Soviet agents in the Australian government and contained information on Allied political and military plans.

USA codes
The M-209 cipher machine was the US version of the Hagelin C-38 and it was used widely by the US armed forces as a mid-level cryptosystem. The Army used it at division level and the USAAF used it for operational and administrative traffic. The Japanese codebreakers investigated this traffic in 1944, identified the use of a Hagelin type device and were able to solve messages in ‘depth’ (enciphered with the rotors at the same starting position). US Army messages were read, especially during the Philippines campaign of 1944-45. The mathematician Setsuo Fukutomi who worked in the cryptology department wrote: ‘These M209 encoding-machines were in general use at the front divisions of the American armed forces. The General Staff of the Japanese Army had bought a prototype of this machine in Sweden before the war. The military engineer Yamamoto mentioned above guessed that “M209” was an adaptation of this Swedish prototype. He could even determine the details of the modifications of the prototype that had produced M209. At that time, I was serving as a soldier working at the General Staff. I noticed that the keys of the codes of the M209 were so-called “double keys”, and I succeeded in breaking these double-keys. After that, a team including a number of drafted officers including myself were sent to the Philippines. We managed to obtain some good decoding results, but the way the Americans made the daily change of coding keys was such that we were unable to break their codes every day, which strongly restricted the benefits we earned from our work’.

USAAF messages referring to operations of the B-29 bombers were also decoded by the Japanese.
The Merchant Navy Code and the Merchant Ships Code were received from the Germans and the enciphering tables broken in Japan. These were used from 1940 till the end of the war. By reading these codes the Japanese were able to identify the concentration of shipping in specific areas and deduce that major Allied operations would follow.

State Department codes Gray, Brown, A1 and the M-138-A strip cipher were read by the Japanese with varying degrees of success. All these systems had been physically compromised. Through these systems the messages of the US ambassador in Japan Joseph Grew, as well as other embassies abroad, could be solved.

Success with the high level M-138-A in 1943-44 depended on help from the Finns and the Germans.

British codes
The diplomatic systems Cypher M, Interdepartmental cypher and R code were physically compromised. The British Interdepartmental cypher provided intelligence for the Malaya campaign. Through these systems the messages of the British ambassador to Japan Robert Craigie could be read in the years 1940-41. This was stated by General Hideki Tojo (Prime Minister during the period 1941-44).


British diplomatic messages can be found in the archives of the Japanese Foreign Ministry in the file ‘U.S.-Japan Relations, Miscellaneous Diplomatic Correspondence-Special Information File’:

Chinese codes
Chinese Kuomintang military and diplomatic codes were read throughout the war, since the majority was used without additional encipherment.

The code of the Chinese Communist party was much harder to solve but it was read at times and provided advance knowledge for a number of communist offensives.

Japanese Intelligence in World War II by Ken Kotani, ‘Combined Fleet Decoded’, HW 40/29 ‘Exploitation of Russian Civil communications by Axis Powers’, HW 40/75 ‘Enemy exploitation of Foreign Office codes and cyphers: miscellaneous reports and correspondence’, HW 40/85 ‘Exploitation of British Inter-Departmental cypher’, DF-187D ‘Relations of OKW/Chi with foreign cryptologic bureaus’, DF-187FRemarks made by Ministerialrat Fenner in reply to certain questions of a general nature’, DF-169 ‘Cryptanalytic section Japanese Foreign Office’, ‘Hitler's Last Chief of Foreign Intelligence. Allied Interrogations of Walter Schellenberg’, HW 40/7 ‘German Naval Intelligence successes against Allied cyphers, prefixed by a general survey of German Sigint’, HW 40/132 ‘Decrypts relating to enemy exploitation of US State Department cyphers, with related correspondence’, HW 40/221 ‘Poland: reports and correspondence relating to the security of Polish communications’, ‘The Codebreakers – The Story of Secret Writing’, ‘Mathematics and War in Japan’, National Institute for Defense Studies articles: ‘Japanese intelligence and the Soviet-Japanese border conflicts in the 1930s’, ‘Japanese Intelligence and Counterinsurgency during the Sino-Japanese War: North China in the 1940s’, ‘Onodera interrogation vol2-22 - CIA FOIA’, Naval history magazine article: ‘How the Japanese did it’, US Navy report: ‘Japanese Radio Communications and Radio Intelligence’,  Diplomatic records Office, Tokyo, ‘U.S.-Japan Relations, Miscellaneous Diplomatic Correspondence-Special Information File’ (A-1-3-1, 1-3-2) via JACAR (links 1, 2, 3, 4, 5, 6), The Japanese Version of the Black Chamber: (the Story of the Naval Secret Chamber) by Toshiyuki Yokoi

Acknowledgments: I have to thank mr Ken Kotani for answering my questions on WWII Japanese cryptologic history and the staff of the Japan Center for Asian Historical Records for the links to the files in ‘U.S.-Japan Relations, Miscellaneous Diplomatic Correspondence-Special Information File


  1. Dear sir,

    Your Blog is absolutely fantastic!

    Do you any information credible confirmation and reference of the soviet documents intercepted by the Japanese in Harbin related to their plans of a war in Europe (Germany vs Britain and France and in Asia (japan vs US).

    This comes from : ynamics of International Relations, by Ernst B. Haas (U. of California [Berkeley]) and Allen S. Whiting (Michigan State U.), ©1956 the McGraw-Hill Book Co., Inc., New York.

    and a note: These telegrams were intercepted by the Imperial Japanese Government's consul general in Harbin. Japanese copies were examined by A.S. Whiting and accepted as authentic

    Thank you for your help.

    1. Thanks, most of the information presented here is not available anywhere else.
      About the Harbin messages, the Soviet diplomatic code in the East was not the unbreakable one time pad but a simpler system. Files from HW 40/29 mention messages from Harbin being decoded by the Japanese in 1944. The file that i‘ve uploaded here says that the ‘break’ into the diplomatic system of Seoul and Dairen was first achieved in 1943. So from my sources I can’t confirm or disprove that the Japanese were reading the Soviet Harbin code in 1940 (or maybe they had a spy inside the Soviet embassy?)

    2. After checking with Ken Kotani it seems that the Japanese had a spy inside the Soviet embassy. Through him they got the codebooks.

  2. Only just discovered your blog and am finding them invaluable.

    On this topic - do you have any idea what the British R Code and M Cipher were used for?

    I'm trying to pin down what codes/ciphers were used by the British Ambassador in Tokyo when communicating with the Foreign Office in 1941. Japanese codebreakers seem to have deciphered at least some of these communications .

    1. The R code and the Government Telegraph code should have been used only for non confidential traffic as they had limited security.

      Enciphered codes such as Cypher M must have been the main diplomatic cryptosystems for confidential messages.

      There was also the Inter-departmental Cypher, also used by diplomatic authorities.