Ever since the Japanese attack on Pearl Harbor on December 7, 1941, there has been much debate as to how and why the United States had been caught unawares, and how much and when American officials knew of Japanese plans and related topics. Some have argued that various parties (in some theories Roosevelt and/or other American officials, or Churchill and the British, in others all of the above, or additional players) knew of the attack in advance and may even have let it happen, or encouraged it, in order to force America into war.

For instance, one position — prominently discussed in Stinnet's recent book — suggests that a memorandum prepared by Office of Naval Intelligence Lieutenant Commander Arthur McCollum for its Director, and passed to Captains Walter Anderson and Dudley Knox , two of Roosevelt's military advisors on October 7, 1940 was central to US Government policy in the immediate pre-war period. The memo suggests that only a direct attack on US interests would sway the American public to favor direct involvement in the European war, specifically in support of the British. Anderson and Knox offered eight specific plans to aggrieve the Japanese Empire, writing: "If by these means Japan could be led to commit an overt act of war, so much the better." There is no evidence available to suggest that Roosevelt ever saw the memo. And, of the eight "plans" offered in the memo, only one was ever implemented in any fashion, and there is considerable doubt that the memo was the source of the inspiration for that implementation. The text of the memo is available here.

Examination of information released since the War has revealed that there was considerable intelligence information available to US, and other, officials. It was the failure to process and use this information effectively that has led some to invoke conspiracy theories rather than a less interesting mix of mistake and incompetence. The US government had nine official enquiries into the attack – the inquiry by Secretary of the Navy Knox (1941), the Roberts Commission (1941–42), the Hart Inquiry (1944), the Army Pearl Harbor Board (1944), the Naval Court of Inquiry (1944), the Hewitt investigation, the Clarke investigation, the Congressional Inquiry (1945–46) and the top-secret inquiry by Secretary Stimson authorized by Congress and carried out by Henry Clausen (the Clausen Inquiry (1945)).

One of the main considerations making an attack against Pearl Harbor unimaginable was the depth of Pearl Harbor of generally less than 40 feet. Depths of less than 150 feet were generally acknowledged as insufficient to make torpedo attack feasable; at the time, torpedoes from planes dove deeply before attaining running depth because of their front-heavy design and in water that was not deep enough (like Pearl Harbor) would dive straight into the bottom, detonating, or just coming to rest on the sea floor. But the British had proved that torpedoes modified for shallow water could be effective during their attack on the Regia Marina at Taranto on November 11, 1940 (Battle of Taranto). The US Navy discussed this new development (documents to be linked soon), but as Taranto was about 75 feet deep and Pearl was largely less than 40, it was not considered that the British attack method could be used in Hawaii. The RN had used Swordfish torpedo planes, and their slow speed was part of the reason the Taranto attack succeeded. The IJN had no torpedo planes that could fly that slowly and stay in the air. The Japanese had independently developed shallow water torpedo modifications during the planning and training for the raid in 1941. Daniel Martinez published drawings of the wooden fins attached to the torpedoes in the Dec. 1991 issue of Naval History Magazine. The fins kept the torpedo level in the air and broke off when they hit the water. The level flight of the torpedoes helped keep them from "porposing" and sticking their nose in the mud. This very simple modification of the Japanese torpedoes was not anticipated by the USN.

US signals intelligence in 1941 was both impressively advanced and uneven. The US MI8 cryptographic operation in New York City had been shut-down by Henry Stimson (Hoover's newly appointed Secretary of State), which provoked its now ex-director, Herbert Yardley, to write a book (The American Black Chamber ) about its successes in breaking other nations' crypto traffic. Most responded promptly by changing (and generally improving) their cyphers and codes, forcing other nations to start over in reading their signals. The Japanese were no exception. Nevertheless, US cryptanalytic work continued after Stimson's action in two separate efforts by the Army's Signal Intelligence Service (SIS) and the Office of Naval Intelligence's (ONI) crypto group, OP-20-G. The US was able, in the period just before December 1941, to read several Japanese codes and ciphers. By late 1941, those groups had broken several Japanese cyphers (mostly diplomatic ones, eg "PA-K2 " and the "Purple Code") and had made some progress against some naval codes/cyphers (eg, the pre-December version of JN-25), but very little against Japanese Army traffic. In fact, the break of the Purple cypher was a considerable cryptographic triumph and proved quite useful later in the War. It was the highest security Japanese Foreign Office cypher, but prior to Pearl Harbor carried little information about future events planned by the Japanese; the military, who were essentially determining policy for Japan, didn't trust the Foreign Office and left it, "out of the loop". Unfortunately, the two US groups generally competed rather than cooperated, and distribution of intelligence from the military to US civilian policy-level officials was poorly done (capriciously chosen and distributed), and furthermore done in a way that prevented any of its recipients from developing a larger sense of the meaning of the decrypts.

Japanese intelligence efforts against Pearl Harbor included at least two German Abwehr agents, one of them, Otto Kuhn , a sleeper agent living in Hawaii with his family; he and they were essentially incompetent. The other, Dusko Popov a Yugoslavian businessman, was thought quite effective by the Abwehr, but was actually a double agent whose loyalty was to the British. He worked for the XX Committee of MI5. In August 1941 he was tasked by the Abwehr with specific questions about Pearl (see John Masterman 's book on the Double Cross operation for the text of the questionnaire), but the FBI seems to have evaluated the effort as of negligible importance. There has been no report that its existence, or even Popov's availability as a double agent, was passed on to US military intelligence or to civilian policy officials. J. Edgar Hoover dismissed Popov's importance noting that his British codename, Tricycle, was connected with his sexual tastes. In any case, he was not allowed to continue on to Hawaii and to develop more intelligence for the UK and US.

Throughout 1941, the US, Britain, and Holland collected a considerable range of evidence suggesting that Japan was heading for war against someone new. But the Japanese attack on the US in December was essentially a side operation to the main Japanese thrust south against Malaya and the Philippines — many more resources, especially Imperial Army resources, were devoted to these attacks as compared to Pearl. Many in the Japanese military (both Army and Navy) had disagreed with Yamamoto's idea of attacking the US Fleet at Pearl Harbor when it was first proposed in early 1941, and remained reluctant through the Imperial Conferences in September and November which first approved it as policy, and then authorized the attack. The Japanese focus on South-East Asia was quite accurately reflected in US intelligence assessments; there were warnings of attacks against Thailand (the Kra Peninsula), against Malaya, against French Indochina, against the Dutch East Indies, even one against Russia. There are reports of concern at the Pentagon and in the White House about Japanese plans for the SE Asian region. There had even been a specific claim of a plan for an attack on Pearl Harbor from the Peruvian Ambassador to Japan in early 1941. (The source for this bit of intelligence has been traced to the Ambassador's Chinese cook.) Since not even Yamamoto had yet then decided to even argue for an attack on Pearl Harbor, discounting US Ambassador Grew's report to Washington about it, in early 1941, was quite sensible. Later reports from a Korean organization also seem to have been regarded as unlikely. There has been no report of a serious conviction by anyone in US or UK military intelligence or among US civilian policy officials, prior to the attack, that Pearl Harbor or the US West Coast would be attacked. The so-called "Winds Code" announcing the direction of new hostilities remains a curious and confused episode, demonstrating the uncertainty of meaning inherent in most intelligence information, and in this case, even uncertainty about the existence of some intelligence information, especially some years after the event. It is known, however, that the Winds system was to be implemented only if the communications between Japan and Washington were cut, and they were never out of touch with their embassy prior to the attack. Given this any talk of intercepted "Winds" messages would appear to be specious.

Nevertheless, in late November, both the US Navy and Army sent war warnings to all Pacific commands. Although these clearly stated the possibility of war, they did not mention the significant likelihood of an attack on Pearl Harbor itself, instead focusing on the probability of an attack on the Far East. The Washington officials also forwarded none of the intelligence available to them to ascertain the views of the local Hawaii commanders, Admiral Kimmel and General Short. The messages have also been criticised by some for containing "conflicting and imprecise" language. The mistakes made here by both Hawaii and Washington meant that little was done to prepare for an attack. Inter-service rivalries between Kimmel and Short did not improve the situation.

As Nagumo's attacking force neared Hawaii, there is claimed to have been a flurry of later warnings to US intelligence and, even, directly to the White House or to White House connections. For instance, the SS Lurline, heading from San Francisco to Hawaii on its regular route, is said to have heard and plotted unusual radio traffic. That traffic is further said to have been from the approaching Japanese fleet. There are problems with this. All surviving officers from Nagumo's ships claim that there was no radio traffic to have been overheard by anyone; their radio operators had been left in Japan to fake traffic for the benefit of listeners (ie, military intelligence traffic analysts in other countries), and all radios aboard Nagumo's ships were claimed to have been physically locked to prevent inadvertent use and thus remote tracking of the attack force. Unfortunately, neither the Lurline's log (seized by either Navy or Coast Guard officers in San Francisco on its return), nor the original reports to the Navy in Hawaii, have been found, so contemporaneous written evidence of what was recorded about what was heard aboard the Lurline is not now available. ONI is further said to have been aware of the eastward movement of Japanese carriers from other sources (eg, the Dutch), but nothing in the way of credible, compelling evidence on this point has yet turned up. Each of these points remains controversial. No Japanese officer would disobey orders lightly, and to disobey the radio silence orders would mean that the entire attack force would have been jeopardized.

Closer to the moment of the attack, the attacking planes were detected and tracked as they approached by an Army radar installation being used for a training exercise. The Opana Point radar station, operated by two enlisted men (Pvts. Lockard and Elliot) ploted the approaching force and then their relief team plotted them returning to their carriers. These radar returns were thought, by the junior officer in charge at the barely operational information center at Pearl Harbor, to be incoming bombers from the mainland, and in fact the bombers did arrive -- in the middle of the attack. Additionally, mini-subs were sighted and attacked outside Pearl Harbor and at least one was sunk -- all before the planes came within bombing range. It has been argued that this failure saved the USS Enterprise. If she had been directed to investigate she may have run into a six-carrier Japanese strike force.

Japanese consular officials in Hawaii, including spy and Naval officer Takeo Yoshikawa, had been sending information to Tokyo about conditions in Hawaii, and in Pearl Harbor, for some months. Some of this information was hand-delivered to intelligence officers aboard Japanese vessels calling at Hawaii, but some was transmitted back to Tokyo. Many of the messages in this last group were overheard and decrypted; most were evaluated as the sort of intelligence gathering all nations routinely do about potential opponents and not as evidence of an attack plan. None of those currently known, including those decrypted later in the War when there was time to return to those remaining undecrypted, explicitly stated anything about an attack on Pearl; the only exception was a message sent from the Hawaiian Consulate on 6 December, which was not decrypted until after the 7th and which thus became moot. No cable traffic (the usual communication method to/from Tokyo) was intercepted in Hawaii until after David Sarnoff of RCA agreed to assist during a visit to Hawaii immediately before the 7th; such interception was illegal under US law, though it had been going on in New York for some time.

The Japanese had, during the days prior to the attack, asked Yoshikawa to divide the waters of Pearl Harbor into sections to save time and telegraph bills. He could say "six DDs section AA" instead of "six destroyers in the area northwest of Ford Island". Some people have referred to this as the "bomb plot" message. However, the Japanese Navy did not use the "bomb plot", the exact locations of the ships in Pearl were not forwarded to them prior to the attack. Examination of an aerial photo of Pearl Harbor at the time of the attack makes it plain that a pilot could easily determine the locations of individual ships and their classes without reference to a guide. This photo, from the Naval Historical Center, is a perfect example. Pearl Harbor from the air It also handily debunks the claim that Japanese residents of Hawaii chopped hundred-foot-long arrows in nearby sugar cane fields to point the way to Pearl. Even in the absence of any documentary evidence of this claim, it is immediately apparent that a hundred-foot arrow would appear negligibly small compared to the 400-foot-long capital ships clearly visible in the harbor.

Locally, Naval Intelligence in Hawaii had been tapping telephones at the Japanese Consulate before the 7th, and overheard a most peculiar discussion of flowers in a call to Tokyo (the significance of which is still publicly opaque and which was discounted in Hawaii at the time), but the Navy's tap was discovered and was disconnected by the Navy in the first week of December. They didn't tell the local FBI about either the tap or its removal; the local FBI agent in charge later claimed he would have had installed one of his own if he'd known the Navy's had been disconnected.

Further reading

• John Toland, Infamy: Pearl Harbor and Its Aftermath (Berkeley, reissue edition 1991) is an account of the various investigations of the US failure to be prepared at Pearl. He claims that Roosevelt had advance knowledge of the attack, which he deliberately did not use to warn the commanders at Pearl. Note that some of Toland's sources have since said that his interpretation of their experiences is incorrect.
• James Rusbridger and Eric Nave, Betrayal at Pearl Harbour: How Churchill Lured Roosevelt into WWII (Summit, 1991) which posits that while the Americans couldn't read the Japanese naval code (JN-25), the British could, and Churchill deliberately withheld warning because the UK needed US help. Nave was an Australian cryptographer whose diaries were used in writing this book; he later distanced himself from its content. A check against them has made clear that some of the charges Rusbridger makes here are unsupported by Nave's diaries of the time.
• Robert Stinnett, Day Of Deceit: The Truth About FDR and Pearl Harbor (Free Press, 2001) is a recent examination which begins with the conviction that Roosevelt deliberately steered Japan into war with America, and ends with the same conclusion. Stinnett's understanding of cryptography/cryptanalysis is quite limited and his conclusions regarding the cryptographic evidence are accordingly unreliable. He also sees a great deal of meaning in short analysis memos which support his thesis and less meaning in other information that does not. A distinctly limited account, though including an impressive research record.

External links

• Did Roosevelt know in advance about the attack on Pearl Harbor yet say nothing? — The Straight Dope, Straight Dope Science Advisory Board, February 28, 2001
• The Independent Institute: Pearl Harbor Archive — Mostly a Stinnett site, but also has Pearl Harbor articles, debates, interviews, transcripts, book reviews, books, and Pearl Harbor documents
• Too Late for Pearl Harbor — An article detailing the amount of Japanese code that was readable to allies before the attack, refuting key points in Stinnett's "Day of Deceit"
• Communism at Pearl Harbor — An article proposing that the Russians maneuvered the US into war