Overview
The document is four pages and its last paragraph is the reason anyone has heard of it.
Arthur McCollum was a naval intelligence officer who had grown up in Japan, spoke the language, and ran the Far East section of the Office of Naval Intelligence. On 7 October 1940 he set out an assessment of the strategic position under the heading "Estimate of the Situation in the Pacific and Recommendations for Action by the United States." Britain was under severe pressure, a German victory in Europe would be catastrophic for American interests, and American public opinion would not support entry into the war.
His framing sentence says what he thought the problem was:
It is not believed that in the present state of political opinion the United States government is capable of declaring war against Japan without more ado; and it is barely possible that vigorous action on our part might lead the Japanese to modify their attitude. Therefore the following course of action is suggested:
The Eight Points
The memorandum is usually paraphrased. It is worth reading as McCollum wrote it, because the paraphrases tend to sharpen it:
A. Make an arrangement with Britain for the use of British bases in the Pacific, particularly Singapore
B. Make an arrangement with the Netherlands for the use of base facilities and acquisition of supplies in the Dutch East Indies
C. Give all possible aid to the Chinese government of Chiang-Kai-Shek
D. Send a division of long range heavy cruisers to the Orient, Philippines, or Singapore
E. Send two divisions of submarines to the Orient
F. Keep the main strength of the U.S. fleet now in the Pacific, in the vicinity of the Hawaiian Islands
G. Insist that the Dutch refuse to grant Japanese demands for undue economic concessions, particularly oil
H. Completely embargo all U.S. trade with Japan, in collaboration with a similar embargo imposed by the British Empire
Then the sentence everything turns on, at paragraph ten:
If by these means Japan could be led to commit an overt act of war, so much the better. At all events we must be fully prepared to accept the threat of war.
Who Actually Read It
The memorandum was addressed within ONI, routed to Captain Walter Stratton Anderson and Captain Dudley Knox. Knox wrote an endorsement on it, and his endorsement is the least-quoted part of the file even though it is the clearest evidence of how the document was received at the time:
It is unquestionably to our interest that Britain be not licked... We should not precipitate anything in the Orient that would hamper our ability to do this... If England remains stable, Japan will be cautious in the Orient. Hence our assistance to England in the Atlantic is also protection to her and us in the Orient. However, I concur in your courses of action.
A reader who concurs in the eight actions while warning against precipitating anything in the Orient is not reading them as a plan to start a war in the Pacific. He is reading them as a way to keep Japan cautious while the real effort goes to the Atlantic.
Why It Reads the Way It Does
The uncomfortable part is that most of the eight actions were subsequently taken. The fleet stayed at Pearl Harbor. Aid to China continued and expanded. The embargo tightened through 1941 until, in the summer, Japanese assets were frozen and the oil stopped. Japan attacked in December.
A document proposing measures that might provoke a war, followed by the adoption of those measures, followed by that war, is not an obviously innocent sequence.
Robert Stinnett built Day of Deceit on it in 1999, arguing that the memorandum was a blueprint, that Roosevelt followed it deliberately, and that American codebreakers had tracked the Japanese fleet across the Pacific while the garrison at Pearl Harbor was left uninformed so that the attack would land and public opinion would change overnight.
What Specialists Say About the Reading
The mainstream assessment does not dispute the document's authenticity, its contents or the fact that the actions were taken. It disputes the causal chain, and it does so on four specific points.
The recommendations were not novel. David Kahn, the historian of cryptology, wrote that Stinnett "does not mention or seem to realize that McCollum's points... reflected longstanding American policies," and that it is "much more likely that McCollum was following national policy and adding his own view on the risk of war than that Roosevelt was taking guidance on American policy from a mid-level Navy officer." The partial embargo on aviation fuel and scrap metal had begun in July 1940, before the memorandum was written.
There is no evidence it reached Roosevelt. Kahn noted that he found "no record of Roosevelt's having seen McCollum's document." No endorsement, routing slip or presidential annotation has been produced, and the two ONI officers who did receive it left no record of acting on it.
The count does not hold. Conrad Crane, reviewing Day of Deceit for the U.S. Army War College journal Parameters, observed that "the author admits that he can link FDR's actions to only six of the eight items on the list," and that those carried out "occurred because of Japanese provocations or from understandable diplomatic or military motivations."
The document's own purpose is contested on the text. Crane's reading is the sharpest statement of the alternative:
A close reading shows that its recommendations were supposed to deter and contain Japan, while better preparing the United States for a future conflict in the Pacific. There is an offhand remark that an overt Japanese act of war would make it easier to garner public support for actions against Japan, but the document's intent was not to ensure that event happened.
McCollum himself is reported to have rejected the later reading of his memorandum in print in 1954, though that account is harder to check than the memorandum is.
Provenance
The memorandum is generally said to have been declassified in 1994 and obtained by Stinnett under the Freedom of Information Act. That date is repeated widely but traces back largely to Stinnett's own account, and no archival confirmation of it turned up in preparing this entry. The document itself is not in question — the full text, with the endorsements, is available and is quoted above.
Provocation Versus Conspiracy
The most defensible version of the argument is also the least dramatic, and it is worth separating from the rest.
The United States did take measures it knew were likely to bring on a war. The oil embargo of July 1941 left Japan with a finite reserve and a choice between withdrawal from China and seizing the resources of Southeast Asia. American planners understood this. Roosevelt believed American entry was necessary and faced a public that did not.
That is a policy of deliberate pressure toward a confrontation the administration expected and thought unavoidable. It is a serious historical claim, and versions of it are argued by historians who reject the conspiracy reading entirely. R. J. C. Butow, writing in the National Archives' own journal, drew the line where the evidence does: the revisionist case, he wrote, tends "to disregard the rules of scholarship and to gloss over the complexities of the historical record," and on the provocation thesis specifically, "archival research does not support these contentions."
What separates the policy claim from the conspiracy version is the distance between expecting an attack somewhere and arranging for one at a specific anchorage while withholding the warning. The congressional joint committee that examined the question in 1946 found no evidence that anyone in the administration "tricked, provoked, incited, cajoled, or coerced Japan into attacking this Nation." That is where the documentary trail on this particular document stops.
The wider question — what was known in Washington, when, and what was passed to Hawaii — is a separate matter with a much larger paper trail, and is taken up in Pearl Harbor Advance Knowledge.