The Capture and Exploitation of Japanese Records during World War II

Today’s post is by Dr. Greg Bradsher, Senior Archivist at the National Archives at College Park, Maryland.

U.S. Military forces began capturing records almost as soon as the war began and started exploiting them immediately. Documents were first captured from a Japanese plane downed in the Pearl Harbor attack. These provided the first clues to breaking the Japanese Navy’s operational codes. American military leaders knew that while the number of prisoners (and thus information) taken in the Pacific would be relatively small, compared to the war in Europe, Japanese records would become all that more important as an intelligence source. “Fortunately,” one American officer wrote in 1944, “the enemy as a nation is addicted to keeping diaries, and converting everything into writing.” 

To ensure that all involved in captured records activities had an appreciation for records and information, the Allied Translation and Interpreter Section (ATIS) (Southwest Pacific Area [SWPA]) published, at the specific direction of the War Department, Publication No. 6, “The Exploitation of Japanese Documents,” dated December 14, 1944. The publication was intended as a manual for the training and indoctrination of intelligence personnel and as a reference book for the exploitation of intelligence documents. The work contains a complete study on the collection, translation, and processing of captured documents. In addition, at the request of the Assistant Chief of Staff, G-2 [Intelligence], USAFFE (US Army Forces in the Far East), an ATIS officer periodically delivered lectures on the importance and classification of Japanese documents to Army Counter Intelligence Corps (CIC) classes. In 1944, a CIC officer was detailed for liaison with ATIS for eventual duty as an instructor for CIC units. It is important to note that all ATIS units maintained close relations with the CIC units and Australian Army Field Security Service, since these units were largely responsible for the collection and dispatch of captured documents in forward areas to the language personnel stationed with tactical units.

CIC personnel were constantly engaged in providing lectures to soldiers about the importance of captured Japanese documents. Before the operation against the Japanese at Hollandia, Dutch New Guinea, 41st CIC Detachment Special Agent in Charge Duval Edwards at Finschhaven during March and April 1944 gave many lectures on the great importance of soldiers turning in any captured documents. He told soldiers that ATIS personnel had told him that they had seen Japanese documents held as souvenirs of earlier battles in New Guinea, which contained information of tactical value which if had been turned in at the time, would have saved lives and shortened battles. Edwards also spoke with Graves Registration units about the importance of acquiring from the bodies of enemy soldiers everything with Japanese writing.

It was not just ATIS that was engaged in captured Japanese records operations. Other organizations were established throughout the Pacific Theater to translate and exploit the records.[1] Among them were the Southeast Asia Translation and Interrogation Center (SEATIC)[2]; the Sino Translation and Interrogation Center (SINTIC)[3]; and, the Joint Intelligence Center, Pacific Ocean Areas (JICPOA).

In many instances these organizations were staffed with translators trained in military and naval language schools in the United States. The War Department Intelligence Service established a secret school at the Presidio of San Francisco to teach individuals to be Japanese-language interpreters and translators. Classes began November 1, 1941, with four instructors and 60 students in an abandoned airplane hangar at Crissy Field. The students were mostly second-generation Japanese-Americans (Nisei) from the West Coast.[4] Within six months, the school had shipped its first 35 graduates to the field, just in time for Guadalcanal and the Buna-Gona campaign.[5] During the Guadalcanal campaign a large quantity of documents were captured, including ones retrieved from the Japanese submarine I-1, just offshore.[6]

During the war, the Military Intelligence Service Language School (MISLS), as it came to be called, grew dramatically. When Japanese Americans on the West Coast were moved into internment camps in the late spring of 1942, the school moved to temporary quarters at Camp Savage, Minnesota. By 1944 the school had outgrown these facilities and moved to nearby Fort Snelling. More than 6,000 graduates served throughout the Pacific Theater during the war and the subsequent occupation of Japan. The US Navy had similar language programs.

SEATIC and SINTIC operating in Southeast Asia and China received and translated relatively large quantities of captured documents during the war. Although the quantity of documents captured in South East Asia and China were not as voluminous as those found elsewhere, nevertheless there were major collections captured. For example, in the fall of 1944, Task Force Galahad, commanded by Brig. Gen. Frank D. Merrill, captured 2 ½ tons of documents at Myitkyina, Burma.[7] The volume would have been more but members of the 414th CIC unit learned that Chinese soldiers through ignorance destroyed many documents. Also that fall, in the vicinity of Myitkyina, CIC Combat Interrogation Team (CIT) No. 4, to acquire Japanese records, staged a contest, making awards to Burmese or Chinese turning in the most documents. This contest produced a number of valuable documents and propaganda leaflets. In early 1945, in the vicinity of Bhamo in northern Burma, CIC CIT No. 5 “captured” many military documents. In July and August 1945 CIC agents gathered more documents at Kweilin in southern China on the Gui River.

Once the war ended, Southeast Asia Command Field Security Sections were assigned to seize records that, among other things, could be used for the prosecution of war criminals. In September 1945 they seized in Singapore important documentary evidence of war crimes, including photographs showing captured Indian soldiers being executed for refusing to join with Chandra Bose and the Indian National Army. Office of Strategic Services (OSS) Counter Intelligence (X-2) personnel at Rangoon Burma seized, in the former Japanese Embassy, a mass of documentation on the Kempei Tai (Japanese Military Police), Japanese political intelligence organizations, spy schools, and other political and intelligence organizations.

The Joint Intelligence Center, Pacific Ocean Areas (JICPOA), had its origins in the Intelligence Center, Pacific Ocean Areas (ICPOA) which had been established on July 14, 1942 in Hawaii as Admiral Chester W. Nimitz’ intelligence center.[8] At the start of 1943, ICPOA was basically dealing with intercepted messages because not that many prisoners of war or documents had been captured. When the first thirty-five prisoners of war arrived in June 1942, after the Battle of Midway, Japanese interrogators had to be borrowed from other activities.

The first major collection of captured Japanese documents in the Pacific Theater was made in August 1942 when the 1st Marine Raider Battalion, under Col. Evans Carlson and Lt. Col. James Roosevelt, made a harassing raid on Makin Island in the Gilberts. [9] The documents were quickly brought back to Hawaii. They included plans, charts, air defense details on all Japanese-held Pacific islands, and battle orders. 

In February 1943, the first contingent of twenty graduates from the Navy’s Japanese Language School at Boulder, Colorado arrived at ICPOA and began interrogating prisoners of war and translating captured documents. On September 6, 1943, ICPOA was designated a joint Army-Navy-Marine organization by a CINCPAC directive and was given the name Joint Intelligence Center, Pacific Ocean Areas (JICPOA). It was placed under the direction of the Assistant Chief of Staff for Intelligence (J-2), Joint Staff, CINCPAC, and CINCPOA (Commander in Chief, Pacific Ocean Areas). Joining JICPOA once it became an inter-service organization were some 50 US Army Military Intelligence Service (MIS) Nisei linguists. By 1944 there were over 200 translators serving with JICPOA.

Two months after JICPOA was formed US forces invaded the Gilbert Islands. During the landing, the first JICPOA team accompanied the invasion forces. Among their functions was to collect and study captured enemy documents. Similar JICPOA teams participated in succeeding amphibious assaults to examine prisoners and documents for intelligence of immediate tactical value. Instructions were issued to the assaulting forces personnel not to pocket captured documents as souvenirs but to turn them over to JICPOA personnel.[10]

During the so-called atoll campaign in the Pacific, US Navy, Marine Corps, and Army personnel captured many valuable documents on various islands. In November 1943, during Operation Galvanic, marines of the 2nd Marine Division captured important documents at Tarawa Atoll (Betio). The most important find was a set of plans and specifications for some of the defenses encountered on the island. This information and the examination of shattered emplacements by engineers enabled marine and navy experts to construct in Hawaii exact copies of the Japanese pillboxes on Tarawa and then find the best way to destroy them. Such experiments led to improvements in naval gunfire techniques and infantry tactics in time for the Marshalls operation. In February 1944, marines and soldiers from the 27th Infantry Division captured important documents at Engebi Island. These documents, contrary to American intelligence, indicated that the Japanese were strongly entrenched on Parry and Eniwetok islands. This discovery resulted in a hurried revision of the assault plans regarding these islands. At Anguar Island in the Western Carolinas in early September 1944, agents from the 81st CIC Detachment, with the 81st Infantry Division, captured a large volume of records, including blueprints, books, miscellaneous documents, files, 40 pounds of mail, and Japanese currency and coins.

It was not just on the islands that important information was captured. A Japanese carrier pigeon landed on a US transport on the way to Kwajalen Atoll in the Marianas. It was captured and was found to be carrying a Japanese message. Captured and sunken Japanese ships and boats also provided large quantities of documents, many of immediate value. After the Battle of the Bismarck Sea, in March 1943, an abandoned lifeboat at Goodenough Island (northeast of New Guinea) from the Teiyo Maru was recovered and found in it was “The Japanese Army List,” dated October 15, 1942. This document provided a complete list of approximately 40,000 Japanese Army officers together with their assignments.[11] This document was quickly translated and published as ATIS Publication No. 2 in May 1943. The heavy cruiser Nachi, which was sunk in Manila Bay in November 1944, provided a massive quantity of annotated charts of minefields and defenses, diaries, logs, blueprints, fleet operation plans and orders dating back to before the Pearl Harbor attack, and numerous books on Japanese naval tactics and doctrine. This material was translated by ATIS in May 1945 and provided Allied naval commanders with immediate intelligence regarding a variety of topics.

JICPOA personnel also served, beginning in January 1945, at the Advance Intelligence Center (AIC), established at the CINCPAC Advance Headquarters at Guam. While captured records were quickly evaluated in the field, almost all were eventually sent back to Hawaii, and some of those on to the Washington Document Center in Washington, D.C.[12] Although JICPOA was a major player in the captured documents intelligence business, it was ATIS, however, during the war years that handled the most documents.

ATIS was established on September 19, 1942, and was headquartered in a suburb of Brisbane, Australia.[13] Eventually it would grow to over 2,500 personnel, some of who served with Advanced Echelons and combat units. The translation of documents captured in the Southwest Pacific Area began with those taken in the Milne Bay operation in August 1942. Urgent information was extracted before rushing the documents on to the Advanced Echelon where they were sorted, stamped, examined, and translated as necessary. The documents were then sent on to ATIS, SWPA, for final examination. By December 27, 1942, 1,100 Japanese documents had been received at the Advanced Land Headquarters, Brisbane, from the New Guinea area.

The quantity and type of documents captured from the Japanese varied widely. During the early days of the war the Japanese forces were advancing. Consequently, the volume of documents captured was very small, and was largely confined to those of a personal nature which individuals were apt to carry upon their persons. When the Allied forces began to advance, more documents were captured and a much higher proportion was official. At Kokoda [between Port Moresby and Buna] 268 documents were captured, at Buna 1,349, at Lae [eastern New Guinea] 1,562, while at Saipan in July 1944 the figure reached at least 27 tons.[14]

Many of the captured documents provided significant intelligence to General Douglas MacArthur’s forces in the SWPA. In early January 1943, Japanese documents taken from the body of a sniper killed near Soputa, New Guinea, revealed the entire standing operating procedure of the Japanese in that area. This information was given to the 163rd Infantry Regiment of the 41st Infantry Division who used it in subsequent offensive operations. In March 1943, a document was captured showing the submarine schedule between Lae, New Guinea, and New Britain. Interrogation of a prisoner confirmed the fact that supplies were being unloaded at Lae from enemy submarines. The Allied Air Forces acted promptly on this information and sank both Japanese submarines and supply barges. In early April 1943, a Japanese map was captured showing hidden positions of 87 barges at Labu, New Guinea. As a result of immediate translation of the map, the 5th Air Force was informed and proceeded to destroy practically all of the barges. During the Lae-Salamaua Operation in the summer of 1943 the Allies relied heavily on captured documents for intelligence information and planning. In December 1943, an operational order indicating the times and dates at which Japanese submarine were scheduled to appear in designated spots in the Arawe area, New Britain, was translated by ATIS and immediately forwarded to Naval Intelligence where prompt action was taken.

In January 1944, during the New Britain-New Guinea operations, captured Japanese code books enabled radio intelligence staff to determine the intentions of Lt. Gen. Hatazo Adachi, commander of the Japanese 18th Army. Admiral Nimitz exploited the same intelligence advantage when he planned the next stage of the Navy’s campaign in the central Pacific. On March 1, 1944, soldiers found on the body of the commander of Baba Battalion a copy of a field order issued by him in which he ordered an attack on American positions for that same afternoon. This information was immediately translated, relayed to naval and air units, and, coordinated with the translation of a captured map showing enemy positions, resulted in the repulse of the enemy attack by naval and aerial bombardment. Three weeks later, on March 21, 1944, a captured field order disclosed the Japanese strength at Rossum, New Britain. This information was put to immediate tactical use and resulted in the capture of the position by the US 7th Cavalry Squadron. A map, also captured on March 21st, and quickly translated, proved to be more accurate than maps possessed by the attacking forces. 

ATIS received and translated in April 1944 the diary of prisoner of war Hiroshi Horikoshi, a civilian employee (interpreter) with the Japanese 14th Army, who was captured at the same time. The diary covered the period January 1942-January 1944. Horikoshi was in the Philippine Islands from May 1942 to August 1943, and the diary contained a good coverage of that period, depicting atrocities, conditions in Allied prisoner-of-war camps, and conditions in the Philippine Islands in general. Horikoshi, upon arrival at ATIS, at first denied all knowledge of any atrocities but on being confronted with his diary, admitted that such things had occurred. This diary along with other documents relating to atrocities was used in the trials of Japanese war criminals.

In the spring of 1944, ATIS received a document which, after being translated, proved to be of exceptional value and probably considerably shortened the war. This was the so-called “Z Operation” document which gave the Japanese air and naval plan of defense against Allied attacks on their South Pacific possessions, giving their solutions for the defense against Allied attacks in three sectors of the South Pacific. This translation aided materially in speeding up the execution of the subsequent attack on Saipan and other Japanese bases in the Pacific, which occurred shortly thereafter. The story of the capture and “return” of the Z documents is detailed in Appendix II.

When Hollandia, Dutch New Guinea, was captured in late April one of the first places CIC agents seized was the post office. It held what turned out to be a “gold mine” of valuable documents, including battle plans, codes and letters. Late the next month at Biak, an island in Geelvink Bay, New Guinea, CIC agents seized the records of the finance office, post office, bank, and Japanese headquarters. They immediately sent back to Australia approximately 3,500 pounds of records, letters, and other documentary material. Among this cache were code books and a list of Japanese and German agents in the United States. For his action during the Biak operation, Jack Y. Cannon, the commanding officer of the 41st CIC Detachment received the Silver Star. Excerpts from the citation indicate that he, “with great risk to his life made reconnaissance in a number of caves which had been occupied by Japanese, approaching dangerously close to enemy fire” and “recovered more than 11 cases of enemy documents vitally needed for the successful conclusion of the operation.”

A complete list of the names of all officers and noncommissioned officers of the Japanese 222nd Infantry Regiment was captured on May 28, 1944. It showed the units to which they belonged. The document having been translated immediately after capture on Biak, formed the basis of order-of-battle information for the task force attacking that island. 

During mid-May American forces intercepted a Japanese landing craft near Arare that carried material for reinforcing the Japanese-held islands offshore of New Guinea that were to be attacked as part of the Wakde-Sarmi Bay Operation. Found on the vessel were Japanese documents, including field orders and prisoner interrogation reports that indicated the Japanese knew when and where the American operation was to take place. This information resulted in the Americans speeding up their invasion convoy by a day.

Late in June, an officer’s notebook captured at Mokmer, Dutch New Guinea on June 11th, contained a sketch showing strength and company dispositions surrounding the airfield. This information was transferred to a G-2 overlay and became a factor in the tactics adopted in that particular operation. In mid-July 1944, near Moemi, soldiers recovered three cases of buried records, including seven important documents that a Japanese deserter had led them to. 

Documents captured during the Philippine operations also proved useful. On October 22, 1944, X Corps captured four sketches, one of gun positions north of Dulag, Leyte, and three of San Roque, Catmon Hill Area, Leyte, containing gun and coastal defense positions. Exact tracings and translations were supplied to XXIV Corps prior to attack on these positions. At the same time, two sketches were captured at Tacloban, Leyte, which showed the disposition of the Japanese 16th Division. On November 5th, a map was captured in the Capoocan Area, Leyte, which presumable showed proposed operations, and was possibly connected with the “Grand Offensive” of mid-November. On November 4th, a Japanese 16th Division Operations Order, dated October 31st, was captured. It contained details of the proposed landing of Tama Group (full strength of one division) at Ormoc, Leyte, on November 1st. It stated also that the land offensive was scheduled to commence some time in the middle of November. Combat boundaries were listed. Miscellaneous identifications taken from documents captured in early November in the Pinamopoan Area, Leyte, gave the first indication of the Japanese 1st Division’s presence in this area. Documents recovered from the bodies of dead Japanese, members of a Special Suicide Penetration Unit, killed near San Fabian, Luzon, on January 19, 1945, gave full accounts of the units and personnel involved. Written orders including route, objective of raid, and extent to which enemy intended to rely on these new tactics were also included. Also captured on January 19th was a radio chart that was used by I Corps Signal officers to gain highly satisfactory results in the monitoring of Japanese radio communications. Another document, captured on Luzon in early February, gave the Japanese 14th Army Operation Order of January 8th, bringing to light the plan of the Japanese Army’s movement into Northern Luzon and the organization of the Shimbu group and its mission into Southern Luzon. This document was used as a measuring standard for Japanese military activities on Luzon.

Possibly the most important translations published by the 6th Army ATIS Advanced Echelon in February 1945, were files of orders of the Japanese 58th Independent Mixed Brigade. Through this, a complete picture of the organization, strength and disposition of this force was gleaned. Having been organized along lines completely unorthodox, these were invaluable documents to the G-2 Section, especially as this Force was a major enemy unit on the left flank near La Union, Luzon, at that time.

During the first week of March 1945, I Corps ATIS Advanced Echelon on Luzon translated four top secret Japanese operational orders made between February 26th-March 2nd. I Corps’ Commanding General was informed in detail of a major enemy operation involving several divisions and embracing the entire Corps front from Rosario to Puncan. Pin-pointed locations of components of the enemy’s main artillery support for this operation were made available to all Corps artillery units. They subsequently neutralized the Japanese positions, as well as interdicted a portion of the Japanese movements, and anticipated Japanese defensive position and strengths. Interestingly, one of the Japanese operational orders provided the instructions “Utmost precautions will be taken to conceal the plan.”

In mid-March agents of the 40th CIC Detachment captured on Panay Island and Negros Island incriminating documents of Panay’s puppet governor. Earlier, during the Leyte campaign, important documents and diaries were captured by CIC detachments indicating the collaboration of a prominent police officer. He was thrown into his own jail.[15]

Once Manila and its environs had been captured, CIC “search and seizure teams” located and took custody of large quantities of Japanese documents. At the Kempei Tai (Japanese Military Police) headquarters they found numerous lists of names and evidence of collaboration and disloyalty to the Philippines and the United States.

Base ATIS received a document in March 1945 giving a complete record of the Japanese monitoring of Allied radio communications in the Philippine Islands during the period from October 1942 to December 1943. 

Even before the war ended, ATIS was exploiting captured records for war crimes purposes. On April 29, 1944, ATIS Research Report No. 72 was published as an accumulation of documentary evidence for the “Commission Regarding Breaches of the Rules of Warfare by the Japanese Forces” of the Australian Commonwealth. The report contained 28 pages of translations, each translation accompanied by a photostatic copy of the original document and authenticated under oath by the translation. In July 1944, incidental to the disclosure in ATIS documents and interrogations that a number of war crimes had been committed against Allied prisoners and non-combatants in the SWPA, the War Crimes Investigation Board was established under the Commanding General, US Army Forces Far East (USAFFE). ATIS was directed to make available to the board any and all information having to do with the identification of Japanese war criminals. For this purpose, liaison was established and during July and August ATIS furnished the Board with approximately 1,200 pages of translations. In addition, ATIS officers collaborated in the accumulation of evidence from prisoners and testified before the Board. After the cessation of hostilities, the War Crimes Echelon, a separate part of ATIS, was established.

When very few documents were captured and relatively little was known about the enemy forces in the SWPA, it was imperative to translate all documents in full. As their number grew, and the volume of available intelligence increased, such a procedure became unnecessary, and also impossible due to the limited number of linguists available. In early 1943, it became apparent, as the Allies assumed the offensive, that the volume of documents captured would far exceed the capacity of personnel available to translate each and every document in full. As a result, a system of thorough “Screening,” i.e., the rapid examination of documents and the extraction (partial translations) therefrom of the more important material only, was given added prominence. Later, the procedure was altered again to cope with the tons of documents captured at main Japanese bases. This procedure called for all documents to be briefly examined, and those of operational value segregated from those having probable or general value and those having no apparent military value, and provisions were made for the translation on a priority bases of those sections of documents containing information of operational value. An “A” “B” and “C” priority system was established; with “A” being documents of operational value; “B” being documents of probable or general value; and, “C” being documents containing information of no apparent value.

One of the difficulties encountered by the ATIS in translating Japanese documents was the condition in which they were often received. Coming from battle fields, crashed aircraft, graves, sunken ships and foxholes, many of them torn, defaced, water-soaked, soiled and charred, making them difficult or impossible to read. Only 30 percent of the captured documents needed no treatment; the rest needed cleaning, drying, and/or other conservation treatment. To alleviate this difficulty, in July 1944, an officer was assigned for duty with ATIS for the purpose of organizing a sub-section to clean and restore documents making them more readily legible.[16] That summer a Document Restoration Sub-Section, staffed by six WACs (Women’s Army Corps), including one officer, was established. ATIS also published a “how-to” handbook on conservation treatment of captured records and produced a Document Restoration Kit for units in the field. Also that summer, the 441st CIC unit established a clinical laboratory, which, among other things, restored charred documents.

In March 1944, plans were developed for ATIS to be located in closer proximity to combat operations. Accordingly, the Assistant Chief of Staff, G-2, directed on September 22, 1944, that Advanced Allied Translator and Interpreter Section (ADVATIS) be established in the immediate vicinity of General Headquarters at Hollandia. The unit was in effect a miniature ATIS, with various sections, coordinating the production of translation and interrogation reports of immediate operation value. In May 1945 ADVATIS followed the advance of General Headquarters into Manila. Base ATIS was closed at Brisbane on June 4, 1945, and established several weeks later in Manila. At war’s end it moved to Tokyo. By the end of the war, ATIS had processed over 350,000 documents (or 1,680 cubic feet of records).[17]

Appendix I. ATIS Publications

The ATIS Information Section supplied information derived from interrogations; translations; and, situation reports, intelligence summaries, maps, photos, and other outside publications. During the period of October 1942-July 1943, the work of indexing, abstracting and collating information from captured documents and prisoners of war, answering internal queries, and providing information to assist translators and examiners, was carried on by a staff consisting of six officers and ten enlisted personnel. In May 1943, external requests for information available from ATIS sources led to the development of Information Request Reports published only in answer to specific requests for information. As the body of available material continued to grow, individual studies based on information available to ATIS were produced as Information Bulletins on subjects of general interest. Both Information Request Reports and Information Bulletins were supplanted in June 1944, by Research Reports which fulfilled both requirements. The information contained in these reports were bibliographically indexed. Reports were issued when sufficient information on any subject had been collated to warrant publication. Over 120 of these Research Reports were published.

Some of the research reports dealt with military and naval matters, such as No. 67 (Japanese Warships and Merchant Vessels Sunk, Damaged or not Previously Listed); No. 87 (Japanese Mines and Minesweeping); and, Nos. 99-108 (Japanese Place Names-Philippines). Others included information about the Psychology in the Japanese Armed Forces (No. 76) and Japanese efforts to fight Plague and Cholera (No. 92). Report No. 84 dealt with The Japanese and Bacterial Warfare. It included excerpts from Japanese captured documents on their research and use of bacterial warfare. Report No. 117, Infringement of the Laws of War and Ethics by the Japanese Medical Corps, contains information on violations of the Geneva Convention on the rules of warfare and points out how, time and again, medical personnel put to death their own patients. Report No. 119 deals with the Japanese Military Police Service and Report No. 126 is “Hoko: The Spy-Hostage System of Group Control-The Clue to Japanese Psychology.” On April 29, 1944, Research Report No. 72 (formerly ATIS Information Bulletin No. 14) was published and entitled “Japanese Violations of the Laws of War.” The report contained 28 pages of translations, each translation accompanied by a photostatic copy of the original document and authenticated under oath by the translator. Subsequently, Supplements No. I and II were published on March 19, 1945 and June 23, 1945, respectively.

In late 1943, the Information Section was given the task of writing “Briefs” consisting of a summary and highlights of Enemy Publications and Current Translations. Current Translations were publications containing complete translations of documents classified A, B, C, or D in ATIS Bulletins. Over 170 were published, including many extracts from diaries and notebooks. Current Translation No. 141, for example, contained random poems of a Prisoner of War. Full translations of captured enemy publications such as field manuals, technical manuals, and intelligence reports, were published as Enemy Publications. Often, they consisted of combined translations of several documents relating to the same subject, such as (No. 255) “Procedure in interrogating and handling [Allied] prisoners of war.” Over 420 of these were published.

Also produced were ATIS Publications. Twelve of these were scheduled to be produced, beginning in March 1943. They were special works, compiled for general reference purposes. They were prepared and distributed as a result of a specific need, and represented a form of publication for matters outside the usual range of translations and reports. They included: No. 1, List of Japanese Military Conventional Signs and Abbreviations (March 4, 1943); No. 2, Alphabetical List of 40,000 Japanese Army Officer (May 1943); No. 3, Glossary of military terms encountered in Japanese documents; No. 4, Bibliographic Index used for all ATIS publications; No. 5, Bibliographic Subject Index for Enemy Publications 1-200 (November 30, 1944), with a supplementary index from 201-300 (March 1945); No. 6, “The Exploitation of Japanese Documents” (December 14, 1944); No. 8, “Kanji Abbreviations, Variants, and Equivalents; No. 9, Japanese-English Medical Dictionary; No. 10, “Restoration of Captured Documents” (June 28, 1945). Publication No. 7 was cancelled and no record is held that No. 11, “Factors in Japanese Military Psychology” was ever completed, although the material intended for this publication could have been used instead for Research Report No. 76, Part 4, “Prominent Factors in Japanese Military Psychology.”

ATIS Inventories were also prepared. They were numerical inventories under 17 principal categories of documents considered to be of probable or general value. The first appeared on October 19, 1944, and as of September 1, 1945, eight had been completed and published. Document numbers and a brief description including authority, title, date, area of reference and similar essential data were set forth under seventeen headings, such as 1) Diaries, Field; 5) Letters, Postcards; and 16) Technical Documents. Philippine Series Bulletins represented special reports of items pertaining to the Philippine Islands. The timely publication of 18 of these reports afforded a wealth of information preparatory to the invasion. They were discontinued with the dissolution of the Philippine Island Research Section of ATIS on October 9, 1944. Limited Distribution Reports were special reports, highly classified, consisting of translations of documents possessing information of the highest intelligence value or of immediate importance, issued from time to time as directed. They totaled 104 in number. No. 16 dealt with interrogation of captured American B-24 air crews; No. 17 with Allied and Japanese Operations Among Natives of Dutch New Guinea; No. 25 with Anti-Japanese activities in Java; and No. 39 with Navy Operations, Plans and Orders (1941-1944).

There were also other ATIS publications, based on captured documents and interrogations, such as Advanced Echelon Reports, Philippine Series Translations, and Interrogation Spot Reports.

Copies of these ATIS publications can be found at the National Archives at College Park, the Australian National Archives, as well as other archival repositories. To assist researchers interested in World War II-era research regarding the Pacific and Far East, I prepared a 1,700-page finding aid entitled “Japanese War Crimes and Related Records: A Guide to Records in the National Archives,” which is searchable and available online. Researchers should, of course, use the National Archives Catalog.

Appendix II. The First Return of Captured Japanese Records, 1944

In February 1944, the Japanese devised a plan known as “Z Plan” to counter the American naval offensive and destroy the U.S. Pacific fleet. It would commit all the remaining Japanese naval power to one last major battle with the Allies. The “Z Plan” [issued as Admiral Koga’s Combined Fleet Secret Operations Order No. 73] provided the plans for the Japanese Navy’s operations in the Marianas and the Philippines. The “Z” Plan would fall into the hands of the Allies and be translated by ATIS during the early summer of 1944 and immediately be returned to the Japanese. The story begins on March 31, 1944, when two Japanese Kawanishi flying boats were enroute to Mindanao in the Philippines. They were carrying Admiral Mineichi Koga, commander in chief of the Imperial Japanese Navy, and his staff, including Vice Admiral Shegeru Fukudome, who was carrying the “Z plan” documents and the associated cipher system. The plane in which Koga was flying crashed at sea, with no survivors. The plane in which Fukudome was flying also crashed into the sea, near the island of Cebu. The survivors were rescued by local villagers and handed over to Lieutenant Colonel Jim Cushing, US Army, the leader of the Philippine Guerrillas. Fukudome was still carrying the “Z Plan” and the cipher codes. Cushing, realizing the possible significance of the documents, notified his superiors who in turn notified the Allied Intelligence Bureau in Brisbane. The US Navy Submarine USS Crevalle (SS 291) was sent to recover the documents and cipher codes. The submarine picked up the documents on May 11th and sailed to Darwin. From there the documents were immediately sent to Brisbane for translation by ATIS personnel. Copies of the documents were made in Brisbane and the original documents were returned to the aircraft crash site by another submarine. This was done to fool the Japanese into believing that the documents had not been discovered by the Allies. The Allies made good use of the information in the naval campaigns that followed. Historians acknowledge that the deciphering of the “Z Plan” was one of the greatest single intelligence feats of the war in the South West Pacific Area. [18]

ATIS-translated copy of the Z Plan. (Records of General Headquarters, Far East Command, Supreme Commander Allied Powers, and United Nations Command, RG 554).

[1] All of the various organizations widely disseminated the information contained in the captured records.  This was usually done in the form of listings (usually termed bulletins) that provided a brief description of the records and various types of publications containing full or partial translations of specific documents and publications containing full or partial translations of documents relating to a general or specific topic. The National Archives at College Park as well as other United States and foreign archival institutions hold copies of these publications. See Appendix I for information about the ATIS publication program.

[2] SEATIC was part of the South East Asia Command, established at New Delhi, India in November 1943 and moved to Kandy, Ceylon, on April 15, 1944. 

[3] Headquartered in unoccupied China.

[4] See “The Beginnings of the United States Army’s Japanese Language Training: From the Presidio of San Francisco to Camp Savage, Minnesota 1941-1942,”

[5] The first Nisei linguists were tested when the Marines invaded Guadalcanal on August 7, 1942 and flew prisoners of war and captured Japanese documents were sent a short distance away to New Caledonia for processing by the Nisei language team attached to Task Force 6814.

[6] See “The Sinking of the Japanese Submarine I-1 off of Guadalcanal and the Recovery of its Secret Documents.”

[7] See “A Letter from ‘Somewhere in Burma,’ June 1944”

[8] ICPOA’s first officer in charge was Cmdr. Joseph J. Rochefort (of the Battle of Midway code-breaking fame). He was succeeded in September 1942, by Capt. Roscoe H. Hillenkoetter, who later became the director of the Central Intelligence Agency.

[9] See “Seventy Years Ago: The Makin Island Raid, August 1942.”

[10] The Army’s Counter Intelligence Corps faced similar problems with souvenir hunters. To help ensure soldiers turned in any souvenirs of intelligence interest, the CIC established a “souvenir grab bag.” This contained items of no intelligence value, such as Japanese postcards, stationery, pictures, and clothing, and any soldier who handed over a souvenir needed for intelligence analysis was allowed to take an item from the “grab bag” in exchange. For the military souveniring problem that first began at Guadalcanal see “The Marines and Japanese Souvenirs on Guadalcanal August-October 1942.”

[11] See “From Rabaul to Stack 190: The Travels of a Famous Japanese Army Publication.”

[12] Initially, the intelligence product of JICPOA received no CINCPAC-CINCPOA authentication. After July 1944, however, documents, including published translations, were prepared under the imprint of CINCPAC-CINCPOA, and the title JICPOA was used only for administrative purposes.

[13] See “Seventy Years Ago: Colonel Sidney F. Mashbir and the Allied Translator and Interpreter Section (ATIS), September-October 1942.”

[14] Some sources indicate the figure was 50 tons.  However, using 27 tons, at 40 lbs. per cubic foot, this works out to 1,350 cubic feet of records. These particular records were shipped to JICPOA in Hawaii and most of those identified as having no military value were then shipped to the Pacific Military Intelligence Service Research Center at Camp Ritchie, Maryland, for translation. Interestingly enough, among these records was a complete listing of the Japanese Imperial Army Ordnance Inventory.  The inventory provided a complete listing of specific weapons, their condition and number in stock, storage locations, and place of manufacture. This document was immediately translated and subsequently provided new bombing targets for the B-29s over Japan and during the early occupation provided a means of quickly locating and seizing armaments.

[15] Under a memorandum of August 27, 1945, CIC was ordered to cease its investigation of wartime collaborators in the Philippines on September 2nd and turn over all records, together with over 5,000 interned Filipinos charged with treason, collaboration and subversive activities, to the Department of Justice of the Philippine Commonwealth Government.

[16] See “The National Archives’ Arthur Evarts Kimberly and the Allied Translator and Interpreter Section’s Document Restoration Sub-Section, 1944-1945.”

[17] The Allied Translation and Interpreter Section (ATIS) estimated the average size of a captured document at 12 pages. I am estimating that a cubic foot of records is 2,500 pages.

[18] For more information regarding the Z Plan see my article “The Z Plan Story: Japan’s 1944 Naval Battle Strategy Drifts into U.S. Hands,” Part I and Part II in Prologue, Vol. 37, No. 3 (Fall 2005).

One thought on “The Capture and Exploitation of Japanese Records during World War II

Comments are closed.