In a March 1941 “Professional Notes” profile on Japanese naval power, the U.S. Naval Institute’s Proceedings reported Japan to be reorganizing its air forces along German lines. Three months later, the same journal credited the Imperial Japanese Navy (IJN) with 330 carrier-based aircraft, and in November, Proceedings tabulated Japanese naval air strength at 1,000 planes. Both of those numbers were close to what U.S. naval attachés had been reporting from Tokyo. The “Professional Notes,” however, were replete with observations that the Japanese were imitative and their aircraft comparable to older Western technologies.
The Proceedings articles were symptomatic of U.S. intelligence reports on Japanese naval assets in the run-up to the Pacific war—quantitative accuracy but qualitative underestimation.1 Why that should be, in the years and then months leading up to Pearl Harbor, continues to be puzzling, because the U.S. Navy had built an effective intelligence organization in Tokyo through its naval attaché office.
The Desk at ONI
In the U.S. Navy, all matters of foreign information were the province of the Office of Naval Intelligence (ONI). It had a system of regional “desks” responsible for various parts of the globe. Japan was under Desk D in the 1920s, then OP-16-B-12 (later changed to B-11) after 1931—all part of the Far East Section. Japan had been a main collection target for ONI since the Russo-Japanese War.
Desk D had an advantage within ONI—attachés in Japan and China were its only responsibilities. The Chinese fleet was minimal. The IJN remained the real point of interest. Ancillary Japanese interests—the empire, relations with Thailand, the impact on the United States in the Philippines, the Mandated Islands across the Pacific, and implications for U.S. facilities such as Pearl Harbor or Guam—were limited in terms of formal ONI-directed intelligence requirements.
The desk’s task also was pretty simple. The Navy maintained monographs on foreign powers. They were a kind of basic intelligence product similar to a country-specific encyclopedia. Content included government, population, economy, ports, shipping and trade routes, imports and exports, critical commodities, naval forces—all manner of information that would be relevant in a conflict. The D, or B-12, Desk might ask an attaché to pay special attention to one aspect or another, but in general, Washington levied few formal intelligence requirements. The function of the two or three officers, the enlisted sailors, and civilian staff on the desk at the Navy Department was to incorporate information from attaché reporting into the monograph. The monograph would be updated in 1924 and periodically thereafter. The dimensions of the task are suggested by the fact that an ONI annual report for 1926 indicates the monograph revision “is almost completed.”2
With fighting between Japanese and Chinese troops erupting in 1931, things began to change. Though the Far East Section still did not demand the kinds of data that would later be included in key intelligence questions, the conflict afforded Americans an opportunity to observe the Japanese in action, and it made available additional information sources—the U.S. Asiatic Fleet and its intelligence officers and U.S. Marines on the ground in Shanghai and elsewhere. The struggle for China would be episodic—in 1931 in Manchuria, in 1932 at Shanghai, and then from 1937, starting at Shanghai and Beijing, spreading across the country.
The hostilities propelled ONI in new directions. The loose-leaf collections of at-sea surface silhouette views of IJN ship types that became familiar as ONI 41-42 I started in 1935. ONI sought the assistance of the Washington Navy Yard, Naval Research Laboratory, Bureau of Aeronautics, and Bureau of Construction and Repair in configuring the images displayed.3 It was also during this period that ONI tasked ships deploying to join the Asiatic Fleet to collect photographs of the Japanese Mandated Islands. At that time, data from signals intelligence hardly found its way into ONI reporting. The Royal Navy and ONI began exchanging information on the Japanese in about 1936.4 But if you had to point to a single source of U.S. intelligence on Japan’s navy it would be the Tokyo attaché.
Captain Edward H. Watson was the Tokyo attaché in the wake of World War I. Before taking the post, he spent time with Desk D mastering the monograph in its current form and reading messages from his predecessors. ONI also had a manual on information collection that assigned special numbers to every sort of intel, helping analysts integrate attaché reports into the monograph.
This was the time of the IJN’s “8–8” Program, a shipbuilding scheme that would have added eight superdreadnought battleships and an equal number of matching battle cruisers to the fleet, making the Japanese Navy an even more serious challenge. The Washington Naval Treaty of 1922 led Japan to cancel that initiative, converting the would-be battleship Kaga and battle cruiser Akagi, then under construction, into aircraft carriers and canceling at least five other vessels.
While the treaty encountered political headwinds in Japan, it led to a certain détente among navies. The then–U.S. attaché, Captain Lyman A. Cotton, was allowed to tour Japanese shipyards—an ideal way to verify that treaty restrictions were being followed. As assistant attaché from 1928 to 1930, Commander Arthur McCollum visited the IJN’s yards at Kure, Sasebo, and Yokosuka. In 1932, with Japanese opinion turning against the two-year-old London Naval Treaty and naval arms limitation in general, attaché Captain Isaac C. Johnson encountered increasing resistance when seeking visits to naval facilities.
The Business of Intel Collection
Both sides played games of deception, but each took precautions. One important one was to have language officers who were fluent in the language of the receiving country and aware of naval terminology, technology, etc., and could translate and tease out meanings. Assistant attaché Commander George Courts, one of the Tokyo language officers, was treated to a long disquisition on IJN thinking by the commander of the Kure Naval Shipyard, so that the allotted time for his visit practically was used up. Conversely, attaché Captain Harold M. Bemis is said to have been told quietly in about 1938 that the Navy would prefer not to let Japanese officers visit U.S. shipyards. Bemis told his successor, Commander Henri Smith-Hutton, not to insist on yard visits so as to preclude Japanese attachés from demanding reciprocity.
Most of the information U.S. attachés garnered about the IJN came from the Japanese themselves. Aside from the shipyard visits, the Japanese legislature, or Diet, debated naval construction, and its proceedings were published in the official Tokyo Gazette. Japanese officers exchanged copies of the final budget tallies for copies of U.S. naval appropriation bills. The sides also exchanged “Navy Lists” of officers in the order of their seniority. The IJN ones also noted officers’ current assignments.
On occasion, the Japanese Navy entertained foreign attachés with visits to naval bases or to warships in port. Attachés visited the battleships Fuso, Yamashiro, Haruna, Nagato, and Mutsu—exemplars from three of the four existing battleship classes prior to the Yamatos.5 The IJN’s defense contractors also held public exhibitions that foreign attachés were free to attend. Even encounters with members of public organizations that existed to support the IJN were mined for data. Ship launchings, Navy Day festivities, speeches, official declarations, articles by maritime journalists such as Masanori Ito, Japanese warnings against foreign spies, Tokyo air drills—all were reported.
Japanese political events also informed U.S. thinking. The putsch of February–March 1936, in which army hotheads assassinated senior officials, severely wounding Admiral Kantaro Suzuki, was revelatory concerning relations between Japan’s army and navy. Ambassador Joseph C. Grew, who spent the decade before the Pearl Harbor attack in Tokyo, credited attaché Captain Fred Rogers and others for working in complete cooperation during those stark days. About 18 months later, on the occasion of the Japanese bombing of the gunboat USS Panay (PR-5), newly arrived Captain Bemis sat with Grew in his study as Vice Admiral Isoroku Yamamoto, then serving as Deputy Navy Minister, and other IJN and Imperial Japanese Army officers presented their version of events.6
Fred Rogers, who as a lieutenant in 1913 had been the Navy’s only Japanese language officer, enlarged the dining room in his home to entertain bigger groups more frequently, collecting information and gossip from Japanese guests. Commander Courts met and liked Yamamoto and invited him to his home several times. Other informal espionage opportunities came through social events and recreational activities such as golf—there was a course right next to the main experimental naval air station—tennis, amateur photography, and even vacationing.
Sailors who served as couriers bringing pouched documents to the attaché asked the captains of the ships they were traveling on board to route their vessels past Japanese shipyards, where they could take pictures. This was also a technique employed by professional baseball player Moe Berg, who dabbled in spying for the United States in the mid-1930s.7 As with diplomats presenting their credentials, arriving attaché personnel, even junior officers, customarily met with IJN counterparts. The senior attaché saw the Japanese Navy minister. Such contacts became the basis for reports to ONI on thinking at the highest levels of the Japanese Navy.8
Sharing with Colleagues
Another way to augment information was through the “Attachés Club.” This was a circle of various nations’ naval attachés who lunched together periodically. The French attaché, in Tokyo the longest, was doyen in the mid-’30s. British attaché Captain Guy Vivian had an assistant who was an engineering specialist adept at estimating the power of a warship’s plant just by looking at her propellers.
Japan began designing the 72,000-ton Yamato-class battleships in 1934 and adopted a final design in 1937. Meanwhile, the Attaché Club spent a lot of time discussing IJN plans for capital ships. Most participants were convinced Japan would break the Washington Naval Treaty limit of 35,000 tons for a battleship. In 1936, Captain Rogers predicted that IJN ships might displace 45,000 to 55,000 tons. Two years later, Captain Bemis estimated the Japanese were designing ships over the tonnage limit and with 16-inch guns. He added that Italian sources thought (correctly) the tonnage would be considerably higher and that the Japanese had two such ships under construction and were planning to add a third and possibly a fourth. Bemis confirmed that view in another report a month later. The second report, amazingly, carries a date five weeks before the keel was laid for the Musashi, the second of the Yamato-class battleships.9
The China Incident, as the Japanese termed their war on the mainland, was a major opportunity for U.S. attachés. In China, the IJN made its first use of carrier airstrikes, mass aerial bombardment, large-scale amphibious landing operations, ground missions by naval infantry, and blockades. During the navy’s Shanghai operations in 1937, it set up an airfield just south of the city. The Asiatic Fleet sent an aviator, Lieutenant J. P. Walker, to lead a nearby U.S. guard force so that he could observe Japanese carrier aircraft operations for several weeks. Walker came away convinced the Japanese were highly efficient. Most attachés passed through Shanghai periodically to keep their finger on the pulse of the war. Henri “Hank” Smith-Hutton, who had been communications officer on board the Asiatic Fleet flagship and then fleet intelligence officer before rising to Tokyo attaché in 1939, spent several months a year there. He liked to read the Shanghai Evening Post & Mercury.
The Final Attaché
When the Tokyo naval attaché post had become vacant in 1939, Smith-Hutton was a Japanese language officer but only a lieutenant commander—junior for the position. Ambassador Grew had given him a glowing review after an earlier tour at the Tokyo embassy and asked that he be sent back, but ONI thought otherwise. Another of the early language officers, Captain Ellis M. Zacharias, who had been at the embassy in 1927–28 and ruffled feathers there, was selected. Hearing of his forthcoming appointment as attaché, both Ambassador Grew and his counselor, Eugene H. Dooman, protested. The Navy scratched the Zacharias assignment, then changed Smith-Hutton’s orders and sent him to Tokyo with a promotion to full commander.10
Smith-Hutton moved into a house next door to the embassy that had served the last two attachés and shifted his reading to the Japan Times and Advertiser. He found the Japanese more somber and subdued than during his earlier time in Tokyo. Shortages in cloth, metals, and construction materials had developed. Sorting through attaché files, Smith-Hutton learned that Captain Bemis had a poor opinion of the Japanese and that Lieutenant Commander Ralph Ofstie, an assistant naval attaché and a naval aviator, provided excellent reporting on the IJN air establishment.
Japanese police, meanwhile, became much more aggressive, interrogating citizens or officers who interacted with the Americans, and the attaché’s sources increasingly dried up. The last luncheon given by the IJN for the foreign attachés came in late 1939, after the German invasion of Poland, on the occasion of Vice Admiral Zengo Yoshida’s accession to Minister of the Navy.
Despite difficulties, the attaché office produced voluminous reports from open sources and what the Americans otherwise could glean. The cost of sending cables was measured by the word, so most reports went by pouch as memoranda. During Smith-Hutton’s tenure, the attaché sent information on the Zero fighter, the “Long Lance” torpedo, and the upgunned Mogami-class heavy cruisers. Navy experts at ONI, the Bureau of Ships, and other U.S. agencies were not always receptive to the information.
In March 1939, the Tokyo embassy reported the IJN increasing its budget for ship construction by 50 percent through 1944, but doubling spending on shore stations and quadrupling expenditures for aviation. There were reports on other attachés’ opinions on the Japanese building program and on the capacities of Japanese shipyards, as well as on the British assistant attaché’s estimate that the vessel being built on Mitsubishi Nagasaki way no. 2 (the Musashi) must be at least 706 feet long. One significant overestimate, based on the Attaché Club, was that the Japanese were building eight (rather than three) battleships. The U.S. attaché’s Report 6-40 in January 1940 discussed 16-inch guns for the new vessels and added, “one 18-inch gun was also attempted but this gun was never actually completed.” On 29 November 1941, Smith-Hutton reported the Thai naval attaché’s information that the IJN had recently commissioned a 45,000-ton battleship with 16-inch guns and expected to add another before the end of the year. Those ships were the Yamato (commissioned on 16 December) and Musashi. The latter vessel would not join the fleet until 5 August 1942.11
The Last Days
The stream of reporting continued as relations between the nations deteriorated. On 17 November 1941, while the Japanese government deliberated secretly on commencing hostilities, Prime Minister Hideki Tojo delivered a speech at the Diet demanding the end of sanctions against Japan and of efforts to obstruct the country’s actions in China. Commander Smith-Hutton sat in the diplomatic box of the gallery observing. As Tojo ended, the Navy officer leaned over to a colleague and whispered, “Well, he didn’t declare war, anyway.”12 Three weeks later, Japan attacked Pearl Harbor.
Commander Smith-Hutton had expected the worst and prepared for it. He procured Japanese-language textbooks and shipped them to the United States for future language training. He sent home linguists, including men who would be of critical importance to U.S. intelligence during the war. They included Navy Lieutenants Rufus L. Taylor (a future ONI director), Allyn Cole, Forrest Baird, and Lieutenant (junior grade) Gilven Slonim, plus Marine Captain Bankston T. Holcomb. By 6 December (5 December in the United States), the attaché office had finished burning its codes and documents.
On the day of the Pearl Harbor attack, Smith-Hutton got up a little early to listen to shortwave radio. He had heard that Japanese warships had been sighted off the Malayan coast. He could not pick up San Francisco, so he tuned in to Shanghai. The announcer said the consul general advised Americans to stay off the streets. When the attaché arrived at the embassy, he learned of the Hawaiian attack. Ambassador Grew sent him to the Japanese Navy Ministry to check the report, and the minister’s aide confirmed it.
Americans in Tokyo soon were interned, to be repatriated in the summer of 1942. The embassy staff sailed on board an Italian passenger ship, and Smith-Hutton managed to smuggle out a diary and a card index of IJN officers above the rank of commander. They sailed to Lourenco Marques, Mozambique. There they boarded the Swedish liner Gripsholm, which had arrived with Japanese diplomats from North and South America. An article in the Lourenco Marques Guardian of 17 July 1942 noted the exchange of American and Japanese staffs. The reporter mentions Henri and Jane Smith-Hutton and their eight-year-old daughter, Cynthia, as “among the charming people I met.”13
1. U.S. Naval Institute Proceedings 67, no. 3 (March 1941) through 67, no. 11 (November 1941); John Prados, Combined Fleet Decoded: The Secret History of American Intelligence and the Japanese Navy in World War II (New York: Random House, 1995), 37–39.
2. United States Naval Administration in World War II: Naval Intelligence, vol. 1 (Washington, DC: Department of the Navy, 1946, 1976), 592–600.
3. Naval Intelligence, vol. 1, 595. In 1987, the U.S. Naval Institute published the 1942 version of ONI 41-42 as Japanese Naval Vessels of World War II: As Seen by U.S. Naval Intelligence (Annapolis, MD: Naval Institute Press, 1987).
4. Arthur Marder, Old Friends, New Enemies (London: Oxford University Press, 1981), 70.
5. Thomas G. Mahnken, “Gazing at the Sun: The Office of Naval Intelligence and Japanese Naval Innovation, 1918–1941,” in William B. Cogar, ed., New Interpretations in Naval History: Selected Papers from the Twelfth Naval History Symposium (Annapolis, MD: Naval Institute Press, 1997), 255.
6. Joseph C. Grew, Ten Years in Japan (New York: Simon & Schuster, 1944), 177, 237.
7. Nicolas Dawidoff, The Catcher Was a Spy: The Mysterious Life of Moe Berg (New York: Vintage Books, 1994).
8. A compendium of this material exists at National Archives and Records Administration (hereafter NARA), RG 38, Office of Naval Intelligence, Far East Section, Foreign Intelligence Branch, Reference Material, Box 22, “Serial Reports” folders.
9. Prados, Combined Fleet Decoded, 21–22. That date reference is to Tokyo attaché report 45-38, “Third Replenishment Program,” 18 February 1938, Franklin D. Roosevelt Library, President’s Secretary’s File, Confidential File, b. 66, “Estimates of Military Potential: Naval Attache Tokyo, v. 1” folder. The keel for the Musashi was laid on 29 March 1938.
10. Naval Institute oral history with Captain Henri Smith-Hutton, Tape no. 23, 264. On page 102 of his memoirs, Secret Missions: The Story of an Intelligence Officer (New York: Paperback Library, 1961), Zacharias makes the barest mention of his presence in Tokyo in 1928 and says nothing of a controversy.
11. Prados, Combined Fleet Decoded, 29–32; gun data: Tokyo Attaché report 6-40, “16-Inch Guns on New Capital Ships,” 15 January 1940, Record Group 38, ONI, FE series, Foreign Intelligence Branch, Reference Material, b. 22, “Serial Reports” file, NARA. The Thai attaché is cited in a 28 November 1941 report (b. 22, f. “Special Survey Natal Attaché Tokyo Reports”).
12. Grew, Ten Years in Japan, quoted, 480.
13. “Transfer of Diplomats: This, That, and the Other,” Lourenco Marques Guardian, 17 July 1942.