Pearl Harbor after a Quarter of a Century
Harry Elmer Barnes
VI: The Blackout of
Hawaii on the Eve of
Pearl Harbor
In
the confession of the Russian spy in
Tokyo, Richard Sorge, he stated that in October,
1941, he had informed Stalin that the Japanese intended to attack
Pearl
Harbor within sixty days. Stalin may well have passed this on to
Roosevelt in return for Sumner Welles’ helpful gesture in informing him
of Hitler’s plan to attack Russia. One of the last things that Stalin
would have wished to have happen at this time in the Far East was the
destruction of the American Pacific fleet. . . . [A] very prominent
American Army Intelligence officer in service in the Far East during
1941, whose name I am not yet free to mention, had gained knowledge of
the Yamamoto plan to send a task force to attack Pearl Harbor and sent
three separate messages to Washington revealing this information, and at
least two of these reached the Army files well before the attack on
Pearl Harbor.”
—Harry Elmer Barnes
We
may now deal with the problem of why, how and by whom Short and Kimmel
were, during the more than a week before the Pearl Harbor attack,
deprived of the large and varied mass of information that had been
accumulated in Washington and demonstrated, surely by October, 1941,
that war with Japan was now definitely in the making, that by November
27, 1941 it might start at any time, but most likely when Japan
submitted its reply to Hull’s ultimatum of November 26th, that by
December 1st and 2nd it was at hand, that by December 4th Japan would
declare war against the United States and Great Britain, that by the
early afternoon of the 6th war could come at any moment, and that by the
morning of the 7th the Japanese would in all probability attack Pearl
Harbor about 1:00 P.M. Washington time, or 7:30 A.M. Pearl Harbor time.
This leaves out of consideration the Kita message, which had been
processed by 2:30 on the afternoon of the 6th and definitely indicated
that the Japanese would arrive off Hawaii by the evening of the 6th and
be prepared to attack Pearl Harbor on the morning of the 7th.
The
blacking out of Short and Kimmel relative to the Japanese threat at
Pearl Harbor
is a highly complicated situation involving many facts, issues and
changes of policy and operations, especially during the year 1941. The
only consistent item and unvarying policy in all the tortuous maze of
developments from October 5th, 1937, to the surprise attack on Pearl
Harbor was the determination of Roosevelt from the autumn of 1939 to put
the United States into the war, but his conception of the enemy to be
fought at the onset of war changed markedly throughout this period. At
the outset, it was Japan, as revealed by his suggestion at the first
Cabinet meeting, the largely secret strengthening of the American navy,
the
Chicago
Bridge
speech of October 5, 1937, and Captain Ingersoll’s mission to
Britain in the winter of 1937-1938. After the outbreak of war in
Europe in September, 1939, it became
Germany, where Churchill was exerting his main pressure for
collaboration. It was not until this seemed almost impossible to
accomplish by mid-summer of 1941, because Hitler provided no casus
belli, that
Roosevelt finally decided that he would have to enter war through
the back door of Japan.
No
other prominent American official except Stimson was clearly determined
to support war with Japan at this time. Stimson first became publicly
very influential in this policy only by the summer of 1941 when
Roosevelt
decided that Japan would probably have to become the main initial target
of his bellicosity. After this date Stimson, already appointed
Secretary of War in June, 1940, logically became the most undeviating
member of Roosevelt’s entourage so far as upholding the war motif with Japan
was involved.
There
are a number of relevant questions which have to be raised, some of
which have not been entirely resolved even today and may never be wholly
cleared up. The first one is how and why many of the top military
officials in both the Army and Navy at least appeared to ignore at the
most crucial period, November and December, 1941, the basic Japanese
strategy of a Pacific war—an initial attack on the American Pacific
fleet—which had been demonstrated to be sound and practical and had been
given special relevance after the Pacific fleet had been based at Pearl
Harbor in the spring of 1940? How could they have disregarded the
numerous Bomb Plot messages and the Martin-Bellinger Report, both of
which clearly pinpointed Pearl Harbor as the inevitable target of any Japanese air attack if war
came?
How
could this top military personnel appear to be unaware of the special
threat to Pearl Harbor when all the operating groups in the lower
echelons, who were devoted to matters of Magic and intelligence, were
discovering and emphasizing this danger and were persistently seeking to
get this evidence presented to their superiors and have Short and Kimmel
properly warned while there was still an abundance of time in which to
alert Hawaii and avert an attack there? Why did the most concerted
effort to blackout
Hawaii
begin when Roosevelt’s shift of policy to direct bellicosity toward
Japan obviously increased the danger of an attack there? Short
was blacked out as to negotiations with
Japan after the latter part of July and Kimmel after the
Argentia meeting in August.
How
were the top military echelons able to keep the impressive evidence of
danger to
Hawaii suppressed? Were they ordered by Roosevelt to suppress
this material and withhold it from
Hawaii? If so, how many were so ordered, and who were those who
suppressed the evidence without any order to do so? Why, when the
threat to
Hawaii
became more clear and evident, did most of the top military echelons
turn their attention to the Far East and apparently neglect
Hawaii?
Who
in the upper civil and military echelons in Washington wanted the
United States to go to war, and if they did, was it to be war against
Germany or
Japan? Neither Marshall nor Stark really wanted any kind of war
at the moment, with either
Germany or
Japan, because they believed that this country needed to get
better prepared to wage a world war; they were especially opposed to war
with
Japan in 1941. Hull
was apparently satisfied to continue feeding his banalities and
platitudes to Nomura and assuring the probability that no peaceful
settlement could be made with Japan. He hated both the Germans and the Japanese and, as an
old Tennessee feudist, was hardly opposed to a little killing on
principle, but he was surely not a leading protagonist of open
hostilities although he knew that they would almost surely result from
his operations as Secretary of State.
Secretary
of the Navy, Frank Knox, as one of the leading warmongers of the time,
was eager to get us into any available war, although he preferred one
with Germany, but he wished to have Hawaii well prepared for war and
seems to have played no decisive role in precipitating war with Japan or
blacking out Hawaii. By the latter part of November, when the Japanese
began to send extensive forces southward and it seemed possible that the
Japanese would make their first attack in the southwest Pacific, on the
islands or mainland, Knox was especially vigorous in maintaining that
the United States must stick by the arrangements in ABCD and Rainbow 5
and resist the Japanese by force even though there was no attack on
American territory and forces.
When I
was teaching at the University of
Colorado in 1949, one of my mature students was a nephew of Knox.
Learning of my interest in
Pearl Harbor,
he brought up the subject of Knox in relation to this question. He said
that the Knox family had always believed that the Secretary’s death was
hastened by his sense of shame and humiliation over what he had
discovered to be the deliberate failure of Washington to warn Short and
Kimmel about the coming surprise attack, and the subsequent attempt to
make Short and Kimmel the scapegoats for the quasi-criminal neglect by
the guilty parties.
Since
Knox died on April 28, 1944, he did not live to learn the revelations
brought forth in most of the post-Pearl Harbor investigations, but the
Naval Intelligence and Communications experts of 1941 knew and resented
the failure to warn Short and Kimmel, and Knox may have called them in
for questioning. Indeed, all he would have needed to do was to talk to
his friend, the distinguished Admiral William H. Standley, about the
“kangaroo court” conducted by Justice Owen Roberts, where this
disgraceful smearing of Short and Kimmel, especially the latter, got off
to a running start. A similar impression was given to me by Admiral Ben
Morreell, who was closely associated with Knox and travelled thousands
of miles with him between Pearl Harbor and Knox’s death. He assured me
that Knox was “clean as a hound’s tooth” with respect to any complicity
in blacking out Pearl Harbor in December, 1941, and became increasingly
suspicious of Roosevelt’s role in this matter.
Only
Stimson, who had been brought into the Cabinet in June, 1940, clearly
stood with Roosevelt in strongly favoring war with both Germany and
Japan. He had been one of the leaders in the interventionist
group in the East from 1937 onward who had urged our entering the
European war; but he had also been the outstanding Japanophobe among the
top civilian figures in the
United States for a decade. How were Roosevelt and Stimson able to
steer the country into war in the face of the great strength of
non-interventionist sentiment in the country at large?
The
year 1941 brought all these confused policies and personal attitudes to
a head, partly due to new international developments and partly as a
result of the unexpected responses of leading personalities involved,
notably Hitler. Although keeping
Japan
as a martial ace-in-the-hole, Roosevelt started out the year with his
interventionist policy mainly centered on
Germany, an attitude which was supplemented by strong pressure
from Churchill. Hitler was to be provoked into starting war by
challenging American unneutral action in convoying supplies to Britain
and Russia on the Atlantic, but Hitler refused to rise to the bait as he
had earlier declined to do in the case of the Destroyer Deal of 1940
with Britain and the lavish shipment of arms to Britain. By the end of
June, 1941, the prospect of provoking Hitler had greatly dimmed and it
seemed likely that the most effective way in which to get into the war
was to incite
Japan to take some action which would inevitably mean war. At
this moment, Roosevelt, most appropriately, brought Secretary Stimson
into direct action to implement the Japanese policy that he had “sold”
to
Roosevelt
with great ease on
January 9, 1933.
Although
there is no doubt that after September, 1939, Roosevelt definitely
preferred to get into the war directly in Europe, he had always kept
Japan as an ace in his sleeve ever since his meeting with Stimson in
January, 1933, and the first meeting of his Cabinet in March, 1933, as
we have been told by then Postmaster-General James A. Farley. He had
secretly built up the American navy, and our only likely naval enemy was
Japan. His Quarantine Speech in
Chicago in October, 1937, straight Stimson doctrine, emphasized
Japan
more than Germany. In the winter of 1937-1938, he sent Captain Royall E.
Ingersoll to Europe to consider possible American operations with the
British in the event that they became involved in a war with
Japan. Roosevelt early adopted measures aiding the Chinese in
their war with Japan, and there is much evidence that the financial and
diplomatic policies of the United States played a very considerable role
in bringing about the renewal of war between Japan and China in July,
1937.
The
outbreak of war in Europe in 1939 only provided a temporary interlude
which distracted Roosevelt from his underlying aggressive program
relative to
Japan. The Roosevelt-Stimson policies and actions with
reference to
Japan after June, 1941, that led to the outbreak of war on
December 7, 1941, were summarized early in this article and need not be
repeated here.
Much
has been written on the possible communist influence on Roosevelt’s
decision to make war on Japan, but even revisionist historians have
concentrated mainly on that exerted by Chiang kai-Shek and Owen
Lattimore through pro-communist officials among Roosevelt’s associates
at the White House, such as Lauchlin Currie and Alger Hiss, in leading
Hull to kick over the modus vivendi and send his ultimatum to
Japan on November 26, 1941. It was a far more complicated and
far-reaching operation than this, but to deal with it adequately would
require much more space than is available here. Moreover, it does not
require extensive treatment here, for Roosevelt, Stimson, and
Hull
did not need any encouragement and support from the Communists in their
determination to pressure Japan into war with the
United States.
Most
basic, perhaps, was the fact that Litvinov sold his doctrine of
“collective security” to the Popular Front politicians in Europe, and
this was adopted by the American Liberals as the dominant consideration
in their pro-war propaganda in the
United States. This matter has been treated in detail by Professor
James J. Martin in his American Liberalism and World Politics,
1931-1941. The liberal propaganda was most potent in supporting
American intervention in the European War until Hitler failed to provide
the expected provocation to war on the
Atlantic.
In
Asia, the predominant motive of the Communists in supporting war against
Japan
was provided by the fact that Japan was the main bulwark against Communism in the
Far East.
But
Russia left this propagandist operation chiefly in the hands of
the Communists of Asia, mainly those in
China,
since
Russia had to move cautiously to avert vigorous Japanese
defensive movements against
Siberia. The Chinese Communists pressured Chiang kai-Shek to act
aggressively against
Japan,
and they were encouraged by the pro-communist figures in Roosevelt’s
entourage in
Washington. After
England became involved in war in Europe, and especially after
Hitler attacked
Russia, the latter stepped up its pressure on the Chinese
Communists to involve the
United
States in war with
Japan.
But it
was not until the Russian spy in Tokyo, Richard Sorge, informed Stalin in mid-October 1941 that
Japan
would move southward and not molest Siberia, that
Russia began in earnest to influence American action against
Japan. Prior to Hitler’s attack on
Russia on June 22, 1941, the most publicized Soviet attitude in
the
United
States had been anti-interventionist. American Communists sought to
line up with the America First organization until they became
embarrassing to the latter and its leaders repudiated any communist
support, which evaporated after Hitler’s attack. But
Russia had never abandoned its previous cautious support of
pressure against
Japan in the
Far East.
The
Lauchlin Currie-Owen Lattimore episode was only a dramatic item in this
broader campaign of the Communists against Japan in the
Far East.
Lauchlin Currie, an assistant-President in the White House circle, was
a strong pro-communist sympathizer, perhaps a member of the party. Owen
Lattimore, who was similarly pro-Communist, but not personally a
Communist, occupied the somewhat curious position of American adviser to
Chiang kai-Shek in
China. When Roosevelt, Hull, and even Stimson, at the
insistence of Marshall and Stark, were considering a modus vivendi
with Japan to gain time in order better to prepare for a Pacific war,
Lattimore sent a strongly worded cablegram to Currie protesting against
any such temporary truce with Japan. The cablegram was vigorously
supported by Currie and it has been regarded by many historians as
constituting the final item which induced
Hull to kick over the modus vivendi and send his
ultimatum to
Japan.
There
were other far more basic, communist influences on items with regard to
Hull’s ultimatum to Japan which have been overlooked even by
many revisionist historians. The most interesting of these is the
extent to which the terms of
Hull’s ultimatum reflected the views of Harry Dexter White, the
pro-communist brains of the Treasury Department, Felix Frankfurter
having once observed that secretary Morgenthau did “not have a brain in
his head.”
On
November 18, 1941, Morgenthau sent to Hull a memorandum drafted by White setting forth proposed terms
that should by presented to Japan by
Hull. They were so drastic that it was obvious that
Japan would never accept them. Nevertheless, Maxwell Hamilton,
the chief of the Far Eastern division of the State Department, read the
Morgenthau-White memorandum and said that he found it the “most
constructive one which I have yet seen.” He revised it slightly and
filed it with
Hull, who had this
Hamilton revision before him when he drafted his ultimatum of
November 26th to
Japan. Actually, no less than eight of the ten points in
Hull’s ultimatum to
Japan embodied the drastic proposals of the Morgenthau-White
memorandum.
Despite
all this volume of evidence of communist pressure in the Far East for
war between the United States and Japan, I remain unconvinced that it
exerted any decisive influence upon Roosevelt, who, after all,
determined American policy toward Japan. Roosevelt had made up his mind
with regard to war with Japan on the basis of his own attitudes and
wishes, aided and abetted by Stimson, and he did not need any persuasion
or support from Communists, however much he may have welcomed their
aggressive propaganda. If he had desired to preserve the modus
vivendi he would have had no hesitation in repudiating
Hull’s action. Hence, it remains my conviction that the
contention that Soviet
Russia exerted any preponderant influence in pushing the
United States into war with
Japan must be discarded. This also applies to the belief that
Churchill, who was then working hand-in-glove with the Russians, exerted
decisive influence on
Roosevelt in his pressuring the Japanese into war. Roosevelt was in no way dependent on Churchill’s support; the
reverse was the case. The responsibility for the final action in
pressing
Japan
into war was that of Roosevelt, and this must be judged solely on the
basis of its wisdom with respect to the national interest of the
United States at this time. The apologists for Roosevelt, from Thomas
A. Bailey to T. R. Fehrenbach, have contended that our national interest
required our entry into the war and justified
Roosevelt’s “lying” the country into the conflict to promote our
public welfare.
For
at least fifteen years after the attack on Pearl Harbor, most
revisionist historians still believed that by December 4th or 5th, at
the latest, virtually all the top officials in
Washington,
civilian and military, were convinced that, in the event of war, the
Japanese would first attack Pearl Harbor. They based this conclusion chiefly on the whole broad
historical background and the traditional naval strategy in the Pacific:
the assumption that
Japan would never start a war without making her first move an
attempt to destroy the American Pacific fleet, wherever it was
stationed. This was necessary to protect the Japanese flank before they
could safely move into the southwest Pacific and the East Indies or go
north to attack Siberia, unless they could be assured of American
neutrality, and nothing in Roosevelt’s foreign policy gave the Japanese
any reason to expect American neutrality. By mid-summer of 1941 it
seemed evident that Roosevelt and Stimson were determined to wreck
Japan by either economic pressure, military operations, or both.
These
revisionist historians were also familiar with the series of Bomb Plot
messages which clearly pinpointed Pearl Harbor as the target of any
Japanese surprise attack on the
United States. They were also well acquainted with the fact that our
Navy had been holding maneuvers for years off
Hawaii, long before the Pacific fleet was retained there in the
spring of 1940, to discover the nature and prospects of a surprise
Japanese attack on
Pearl
Harbor. Unfortunately, the prospect of success for Japan seemed
very good indeed, and hence it was taken for granted for years that any
evidence of imminent hostilities between the United States and Japan
would bring with it prompt action on the part of Washington to keep
Pearl Harbor on the alert for a prospective Japanese attack, and ready
to anticipate and repel one when it came. When this Japanese action did
not take place before December, 1941, it was logically assumed that the
top officials in Washington, acquainted with all the evidence that war
was at hand, must have been personally prevented from warning Short and
Kimmel, and only one man could give such an order and have it obeyed.
That person was Franklin D. Roosevelt. Hence, he must have ordered all
these top officials not to warn Short and Kimmel until it was too late
for them to detect and repel the attack.
We
now know that this interpretation needs some qualification, even though,
as presented by Admiral Robert A. Theobald and other informed experts,
it seemed to be soundly based upon both the factual historical
background and sound logical inference. In the first place, any such
general order by
Roosevelt
might have been very difficult to sustain. There were just too many
important officials to be restrained by such an order without a
considerable possibility that there would be a leak or disobedience
somewhere. This is one point on which the views of Admiral Samuel E.
Morison, in his article in the Saturday Evening Post, of
October 28, 1961, are in my opinion worthy of consideration, although the
reasons he gives for it are in part erroneous. It would obviously have
been rather risky for
Roosevelt.
Some of these numerous officials who had to be warned to keep silent
might reveal Roosevelt’s order to black out Pearl Harbor, and this would
have been disastrous to both Roosevelt’s political career and military
plans.
It
is only fair, however, to present Commander Hiles’ defense of Admiral
Theobald’s contention that Roosevelt could have ordered that Short,
Bloch and Kimmel were not to be warned of the threat to Pearl Harbor without any great personal risk of exposure. The
so-called chain-of-command procedure would have made this possible
without too much risk. Roosevelt did not have to reach all of his important subordinates
personally. The Joint Board of Command was the highest military
authority in the land, except for the President. It was made up
exclusively of the armed forces, with Marshall and Stark at its head.
Even the Secretaries of War and the Navy were not members and had no
voice in the deliberations of the Joint Board, although as a matter of
routine its reports to the President were submitted through the
Secretaries and the latter could add such comments as they wished to
make for what value they might have from the military point of view. No
person except
Roosevelt
had any jurisdiction over the Joint Board. Consequently, it is not at
all difficult to discern how
Roosevelt
could control the situation with no great difficulty or risk; from the
Joint Board on down it was solely a matter of the chain-of-command.
Certainly, there might be some minor leaks and some disobedience, as in
the “contact Rochefort” message in connection with Winds Execute, the
“October (1941) revolution” in the Office of Naval Intelligence, and in
the Sadtler-Akin “pipeline” arrangement, not to mention the efforts of
Sadtler, Bratton and McCollum to get past the Marshall barrier.
Roosevelt was well covered up because he would almost never place
any orders in writing—they were nearly invariably verbal.
At
any rate, Roosevelt appears to have kept Hawaii in the dark about the
threat to Pearl Harbor without any blackout orders of which we have any
definite evidence save those to Marshall, Arnold and Stark, and then not
until December 4th.
Finally,
and most important, it appears that Roosevelt may not have needed to
order many of his leading subordinates against warning Short and Kimmel.
These top officials seem to have become unduly absorbed by the fact
that all the known Japanese military movements, and these were on a
grand scale and rather conspicuously displayed, indicated that Japanese
task forces were moving down into the southwest Pacific and the East
Indies, and there were no known Japanese fleet movements that appeared
to threaten Pearl Harbor. Some writers believe that this virtual
parading of Japanese power moving southward was in part deliberately
designed to distract attention from
Pearl Harbor.
This is doubtful. The extensive movement southward was a basic part of
the campaign connected with the attack on
Pearl Harbor,
and had to be timed accordingly.
To
this, and a very important consideration, was added the concentration of
the top brass naval authorities on the strategic implications of the
ABCD agreement and the Pacific War Plans, Rainbow 5 (WPL 46), drawn up
in April, 1941, and approved verbally by Roosevelt in May and June,
which envisaged the launching of the first Japanese attacks in the Far
East. The extensive Japanese task force movements southward in
November, 1941 appeared to confirm this assumption. The top naval
officers, Stark and Turner, had warned that the economic strangulation
of Japan in late July would certainly mean that Japan would have to move
southward to get, by force if necessary, the indispensable vital
supplies that were denied to her by the July embargo imposed by the
United States, Britain and Holland. Both the navy and the army leaders
were fully aware that Rainbow 5 (WPL 46) provided that the United States
would make war on Japan if the latter went too far in this quest, even
if there was no Japanese attack on American forces or territory.
Very
significant evidence of this concentration on the Far East, especially
by the Navy, on the eve of
Pearl Harbor
is provided by Admiral Beatty, the aide of Secretary Knox in 1941. He
recalls that, at the last meeting of the top officers of the Navy with
Knox on the afternoon of December 6th, Knox inquired as to whether the
Japanese were about to attack the United States. Turner, who, as usual,
spoke for Stark, answered rather dogmatically in the negative, and went
on to say that he believed Japan would first strike the British in the
Far East.
Beatty asserts that there was no dissenting voice from any of the navy
officers present, from Stark down. Perhaps more conclusive as evidence
of the shift of interest and concern from Pearl Harbor to the Far East
is provided by the agenda and discussions of Roosevelt’s “War Cabinet,”
made up of Roosevelt, Stimson, Knox, Marshall and Stark, on November
28th, and of the final conference of Stimson, Knox and Hull on the
forenoon of December 7th. In both cases the main subject and problems
discussed were the movements of Japanese forces to the southwest
Pacific, the obligations of the United States under ABCD and Rainbow 5
to check these by war, if necessary, and the question as to whether the
country would unite to support a war which had not been started by an
attack on American territory or forces.
It
is desirable to point out, however, that the newer Revisionism on Pearl
Harbor, which is based on the assumption that most of the top civilian
and military authorities in Washington expected that the Japanese would
almost surely begin their aggressive action in the Far East, also needs
qualification, just as does the older view that Roosevelt specifically
ordered them all not to send any warnings to Pearl Harbor.
This
newer interpretation, stressing the Far Eastern fixation of most top
Washington officials from early November to the Pearl Harbor attack,
does not account for the failure to supply Short, Bloch, and Kimmel with
the planes and other equipment which they had requested early in 1941
to enable them to carry out the necessary reconnaissance to detect and
repel any Japanese attack; the failure in the summer of 1941 to provide
Pearl Harbor with a Purple machine or even to assign Commander Rochefort
and his large and capable cryptanalytical group the task of
intercepting, decoding, and reading the other Japanese diplomatic
messages in J-19 and PA-K2; the blacking out of Short after the economic
strangulation of Japan in July and of Kimmel after Argentia with respect
to the nature of American negotiations with Japan; or the reasons why
Stark and Turner, as well as the responsible army officials, refused to
permit the Bomb Plot messages to be sent to Pearl Harbor in October
1941, and later on.
Their
concentration on the Far East may account for the attitude and
operations of the top echelons in the Army and Navy after the extensive
ship movements of the Japanese into this area in November, 1941, but it
fails to provide an adequate explanation of the obvious efforts to keep
Short and Kimmel from getting the essential information available in
Washington long before that time or of sending them bogus
“warnings” on November 27th.
Pending
a better explanation, which has never been provided by Roosevelt’s
defenders, it must be assumed that this long continued and unbroken
effort to keep Short and Kimmel in the dark as to the tense diplomatic
situation between the United States and Japan was keyed to Roosevelt’s
persistent recognition that he must have an attack by Japan, once it
became rather clear that Hitler would not rise to the provocative bait
provided by American convoying on the Atlantic. The situation surely
calls for something more fundamental than the trivial and impersonal
“noise,” which is offered by Roberta Wohlstetter in her defense of
Roosevelt and his bellicose collaborators in Washington.
As
late as December 1st, it is very possible that Roosevelt himself feared
lest Japanese aggressive action might start in the southwest Pacific and
the East Indies and not provide any prior and direct attack on the
United States. On that date, he sent a note to Admiral Hart at Manila
ordering three “small vessels” to be fitted out at Manila, each manned
by Filipino sailors, commanded by an American naval officer, flying the
American flag, and carrying a machine gun and a visible cannon. They
were to be sent out to specified positions where they could be fired
upon by the Japanese task forces that were moving southward. This would
give him the attack on American ships that he vitally needed to
get the United States into the war by the back door of Japan, unite the
country behind him, and also save the Pearl Harbor fleet if the Japanese
attacked this bait in the Far East before Nagumo reached Pearl Harbor.
The
Democratic platform of 1940 had declared that the
United States would not enter the war unless attacked. The
anti-interventionist sentiment in the United States was so overwhelming
in 1940 that, during the campaign of that year, Roosevelt thought it
necessary repeatedly and vigorously to assure the American public that
he would avoid war, culminating in his famous speech in Boston on
October 30, 1940, in which he told American mothers and fathers, “again
and again and again” that their sons would not be sent into any foreign
war.
But
on the heels of his victory in the election of 1940, Roosevelt, as noted
earlier, started military conferences with the British which, in April
1941, ended at
Singapore with the ADB agreement, to include the Dutch. It was all
implemented by ABCD and Rainbow 5, which specified that if the Japanese
went beyond a certain arbitrary line in the Southwest Pacific-100˚E and
10˚N—and even threatened the British and Dutch possessions there, the
United States would enter the war against Japan even if American
territory, forces and flag were not attacked by the Japanese.
Roosevelt actually desired, above all, to avoid having to enter the war
in this manner. If this happened, he would have to reveal that he had
deceived the American public in his campaign promises and would not have
anything like a united country behind him.
This
was obviously what induced Roosevelt to order the three “small vessels”
to move out from
Manila into the path of the Japanese task forces as they sailed
southward. Aside from a futile trip by the dispatch ship, Isabel,
which was not even repainted, only one of the small vessels” had left
Manila harbor before the Japanese struck at Pearl Harbor, and this ship,
the little schooner Lanikai, was not able to proceed beyond
Manila harbor into the path of the Japanese task forces before the
attack on Pearl Harbor. This so-called Cockleshell ship stratagem of
the three “small vessels,” first noted among revisionist writers by Dr.
Frederic R. Sanborn in his Design for War (1951) has been vividly
described by Admiral Kemp Tolley, commander of the Lanikai, the
second ship that was ready to leave as “bait” for the Japanese, in the
U.S. Naval Institute Proceedings of September, 1962, and October
1963.
Commander
Hiles, a close and well-informed student of the Pearl Harbor episode,
believes that, although Roosevelt was in all probability convinced
before December 1st that the Japanese planned to attack Pearl Harbor, he
devised the three “small vessels” scheme to get a prior attack which
would start the war in a politically satisfactory manner without
sacrificing the Pearl Harbor fleet. This is undoubtedly true, but if
this was his motive Roosevelt thought up the plan some days too late. The Japanese hit
Pearl Harbor
before even one of the three “small vessels” could get fired on. The
order to equip and dispatch them should at the latest have been sent
coincidental with
Hull’s ultimatum on November 26th. Indeed, it should have been
sent by November 5th, when it was evident that the Japanese proposals
for settling American-Japanese relations peacefully which were to be
offered in November were the final Japanese gesture that could preserve
peace, and Roosevelt knew that the situation built up by Stimson, Hull
and himself precluded the possibility of accepting any Japanese
proposals short of a virtual surrender. The memory of the sinking of
the
Panay
on December 12, 1937, and the bellicose excitement caused by the
accidental attack on one small American vessel should have inspired an
order identical with that he sent to Admiral Hart on December 1, 1941.
Roosevelt should not have needed the report on the Japanese hostility
to the gunboats passing
Formosa on November 29th and 30th to inspire the note to Hart.
Secretary
Henry Morgenthau tells of a conversation with Roosevelt as late as the
morning of December 3rd in which the latter seemed frustrated,
despairing of any Japanese attack, and feared that he and Churchill
might have to plan and strike the first blow, an emergency which
Roosevelt desperately wished to avoid for political reasons, as Stimson
has revealed in his Diary and was stipulated in the messages to
Short on November 27th and to Kimmel on November 29th.
On
December 4th, everything seemed changed. Roosevelt appeared assured
that the Japanese had decided to attack Pearl Harbor as their first
stroke, and he now seemed convinced that all possible emphasis and
effort in Washington must be placed on keeping Short and Kimmel from
being warned of an impending attack, although he was still hoping for an
attack on one of the three “small vessels” before the Japanese could
reach Pearl Harbor.
There
is no definitive documentary evidence which has thus far been revealed
and fully proves that Roosevelt had been explicitly informed by
Decemer 4th that
Japan
would attack Pearl Harbor as the first act of war. There may be none until the
voluminous secret correspondence between
Roosevelt and Churchill, which began in September, 1939, is opened
to reputable investigators. Even in this event, it is likely that so
incriminating a document will have been removed from any American copy
of the files, following the pattern of the removal of so much
incriminating material from the American Army and Navy files dealing
with
Pearl Harbor.
There
are three reputable reports from British intelligence in the Far East
that, between November 30th and December 7th, London was informed that the Japanese would attack
Pearl Harbor
on December 7th. If these reports, or any one of them, are accurate,
then there is little doubt that Churchill would have passed the
information on to Roosevelt. General Bonner Fellers, who was in Army Intelligence in
the Near East and located at
Cairo, has given me personally and by letter the following
relevant information. Here, quoting from his letter of March 6th, 1967:
About
10:00 A.M.,
Saturday,
December 6, 1941, I walked into the Royal Air Force Headquarters in
Cairo. The Air Marshal who was then in command of the RAF
Middle East sat at his desk. Immediately, he opened with: “Bonner, you
will be at war within 24 hours.” He continued: “We . . . have a secret
signal
Japan
will strike the U. S. in 24 hours.” . . . I had been in
Egypt for about fifteen months. During that time no word
whatsoever had been sent to me from G-2 in
Washington that Japanese-American relations were strained.
In
the confession of the Russian spy in Tokyo, Richard Sorge, he stated that in October, 1941, he had
informed Stalin that the Japanese intended to attack
Pearl Harbor
within sixty days. Stalin may well have passed this on to Roosevelt in
return for Sumner Welles’ helpful gesture in informing him of Hitler’s
plan to attack
Russia. One of the last things that Stalin would have wished to
have happen at this time in the
Far East
was the destruction of the American Pacific fleet. Most important of
all is the fact that a very prominent American Army Intelligence officer
in service in the Far East during 1941, whose name I am not yet free to
mention, had gained knowledge of the Yamamoto plan to send a task force
to attack Pearl Harbor and sent three separate messages to Washington
revealing this information, and at least two of these reached the Army
files well before the attack on Pearl Harbor. Moreover, as will be
clear later on when we deal with the Merle-Smith message, it is entirely
possible that Roosevelt could have read this on the evening of December
4th, Washington time, and known that the United States was already
involved in war because the Dutch had implemented ABCD and Rainbow 5
(A-2) on December 3rd, Washington time. The message must have been
available in
Washington by the 5th. Perhaps even more instructive and revealing is
the fact that some time before 5:30 P.M. on December 4th, Roosevelt had
discussed the Far Eastern situation with Stark and had approved Stark’s
informing London and the Dutch that Roosevelt was in favor of warning
Japan that if its forces crossed the magic line in the southwest Pacific
this would be regarded as a hostile act and Japan would be attacked by
the ABCD powers. Roosevelt was thus approving the ABCD (ABD) agreement
more than 24 hours before Halifax approached Hull, and he should have
been well prepared for the contents of the Merle-Smith message.
Another
unimpeachable item of information which indicates that Roosevelt was in
all probability informed by December 4th that the Japanese were planning
to attack
Pearl Harbor
on December 7th has not previously been presented, but, fortunately, it
has neither been destroyed nor suppressed. This is an entry in the
History of the
Sacramento Air Service Command for December 6, 1941. This History, declassified
in 1948, had been casually lying around for some time but had not been
carefully examined even by revisionist historians. A copy was noticed
by a revisionist student who was working for his master’s degree at
Indiana
University on the subject of logistic failures at
Pearl Harbor.
Having plenty of money, he had travelled about looking for sources.
When visiting the Wright-Patterson Air Force Base at
Dayton,
Ohio, he found the History of the Sacramento Air Service
Command available for inspection by interested parties admitted to
the Base. Reading the entry for December 6, 1941, he was immediately
impressed with its significance and sent it to Commander Hiles, who was
assisting him in locating source-material for his study. Hiles has been
the first revisionist expert to develop the full significance of this
material.
General
Henry H. Arnold was the chief of the Army Air Corps and one of
Marshall’s deputy chiefs-of-staff. Few men could have been more
vitally needed at this critical time in
Washington, the center of activities in getting ready for the war
with
Japan which had been regarded as imminent ever since
Hull sent his ultimatum on November 26th. Its approach was
amplified and confirmed by the codes destruction intercepts of December
1st and 2nd and by the Winds Execute intercept of December 4th, the
latter revealing that when war came Japan would attack the United States
and Britain, and not Russia. Against this background, it is obvious
that
Arnold
could have been spared from Washington only if he were to carry out an assignment of the utmost
confidential and strategic significance in the face of a Japanese attack
at any moment. On December 5th,
Marshall
ordered Arnold to make a transcontinental trip to
Hamilton
Airfield in California.
Arnold’s
mission was ostensibly to expedite the departure of a small squadron of
some twelve B-17 bombers from Hamilton Airfield to the Philippines via
Hawaii, and to repeat orders concerning the continuation of
reconnaissance while en route over Japanese mandated islands in the
mid-Pacific. This assignment surely did not justify a long trip by an
officer of
Arnold’s rank, experience and ability, even if there had been no
crisis in Japanese-American relations and he had all the time in the
world. It was something that could have been executed by any
experienced captain, major or colonel in the Air Force at
Washington. There was nothing complicated or unusual about it, since
this was by no means the first time that a squadron of B-17 bombers had
been sent to the
Philippines via
Hawaii and had photographed the Japanese islands. It does not
seem reasonable, or even credible, that such a lofty and capable
military figure as General Arnold would have been sent from Washington
to carry out so relatively trifling a mission as Watching a few bombing
planes depart from the Pacific Coast, especially when it was assumed
that the first Japanese moves in the approaching hostilities would be
made in the air and require Arnold’s full attention at Washington Hence,
we are compelled to look for the actual reason behind the Arnold
mission.
It
so happened that December 4th was the day on which the Chicago
Tribune published the implications of Rainbow 5, which fully proved
that Roosevelt had been planning war over many months, if necessary
without any attack on American forces, while at the same time he was
assuring the American people that all his actions were designed to keep
the United States out of war. Naturally, this sensational exposure
created great excitement in
Washington,
and Roosevelt ordered Marshall to try to locate the source of this embarrassing leak.
After
the war, it was revealed that it was an emissary from General Arnold’s
office who facilitated the leak of Rainbow 5 to Senator Burton K.
Wheeler, who, in turn, showed it to the Washington representative of the
Tribune, all three of them patriotically motivated by the hope of
forcing more adequate attention to the needs of the Army Air Corps if
the United States was to become engaged in a farflung Pacific war. Some
writers, working mainly on hindsight, have alleged that
Marshall
wished to get Arnold out of
Washington for the moment as soon as possible, lest his relation to
the “leak” be discovered. I personally doubt this explanation, although
Marshall was feverishly active in searching for the sources of the
leak, and Colonel Deane was working for him on this subject when he saw
Marshall
at his office in the Old
Munitions
Building about 10:00 on the morning of December 7th.
Whatever
the basis of Arnold’s mission, it had to be one of a secret, serious and
responsible nature, commensurate with
Arnold’s rank, distinction and ability. The account of what
Arnold actually did when he was on the coast provides the
soundest explanation of his mission and it rests on facts that cannot be
refuted. They are the following:
The
same message that had been sent to General Short on November 27th,
ordering action at
Hawaii
to prevent local sabotage had also been sent to the Army headquarters on
the
Pacific
Coast
at the Presidio in San Francisco. Accordingly, appropriate steps had
been taken at the McClellan airfield and the planes had been bunched
there to safeguard them against local sabotage. Presumably, they were
also bunched at the Hamilton airfield, but neither
Arnold
nor the Sacramento History mentions this matter. As the entry in the
History of the Sacramento Air Service Command for December 6th
puts it: “It looked like all the planes on the Pacific coast were at
McClellan field.” General Arnold “brought word of the imminence of war,
expressed stern disapproval of the planes being huddled together and
ordered them dispersed.” This was done at once and as rapidly as
possible, despite heavy rain and special local difficulties at the
moment. There were no revetments, so the planes had to be flown to
other airfields.
This
dispersal of the planes was an order that would not have been accepted
or obeyed if given by a junior officer however capable and well
informed. It superseded the Washington order of November 27th to the
Hawaii
air command in which Arnold had participated and had supplemented by later directions
on how to assure full protection against local sabotage.
The
action taken by
Arnold
can only be explained on the ground that Marshall and Arnold had learned
through December 4th that the Japanese were planning to attack
Pearl Harbor
on December 7th. Fearing an attack on the Pacific coast, as well, they
decided to order the dispersal of the planes that had been bunched there
in accordance with the orders of November 27th and 28th. Marshall and
Arnold did not dare to order the dispersal of Short’s planes at Hawaii,
although Hawaii is 2500 miles closer to Japan than California, and hence
far more vulnerable to a Japanese air attack, but they decided to take a
chance on alerting the Air Force on the Pacific coast. Both Marshall
and Arnold were well known for their fear of an attack there.
In
other words, Marshall and Arnold were greatly alarmed over the
information that the Japanese would attack at
Pearl Harbor
on the 7th. While their hands were tied with respect to alerting Short
and Martin at
Hawaii, they did have momentary freedom of action on the Pacific
coast and could surreptitiously alert McClellan airfield without
creating any great excitement or publicity. In any event, by the next
morning any possible adverse reaction to alerting the Air Command in
California would be rendered redundant by the news of the attack on
Pearl Harbor.
It is instructive to note that nowhere in his testimony about his trip
to California did Arnold mention actually visiting McClellan airfield,
which indicates that he wished to leave this visit in obscurity for
obvious reasons. Moreover, he made it a surprise visit, thus avoiding
the normal honors and publicity attending a visit by the head of the Air
Corps in
Washington.
This
would seem to be the only rational and valid explanation of Arnold’s mission to
California on the eve of Pearl Harbor; the expediting of planes to
Hawaii
and the Far East was only the excuse or coverup. Otherwise, we face the
double paradox of the century for Roosevelt’s defenders to explain: (1)
pulling
Arnold
out of
Washington
during the two most critical days of the whole crisis for a perfunctory
and routine operation, and (2) keeping Short’s planes bunched in
Hawaii,
while dispersing the planes in California. The Arnold mission and action is surely one of the best
proofs which we shall have that Roosevelt had advance knowledge that the
Japanese planned to attack Pearl Harbor on December 7th until the time
comes when we can produce absolute documentation of this fact.
One
can well imagine
Arnold’s
feelings as he sent off the B-17’s to
Hawaii, unwarned that they might in all probability be heading
for destruction the next morning. Their guns were unfit for use and
there was no ammunition for them, the latter having been dispensed with
to provide more room for fuel. Having been sent to
California
ostensibly to dispatch these planes, not even
Arnold dared to restrain them and cancel their flight. His
emotions must have been even deeper when he thought of Short’s huddled
planes, which would also be destroyed on the ground by Japanese bombers
the next morning, and of Kimmel’s battleships that would actually
provide sitting-ducks for the Japanese bombing and torpedo planes, but
he did not dare to alert Short, Martin, Bloch and Kimmel as to their
impending fate.
That
Arnold gave the officers at the Sacramento Air Service Command the
definite impression that war was right at hand is evident from the
statement in the History that: “When word came on December 7th,
1941, that the Japanese had attacked Pearl Harbor it did not cause any
surprise!”
I
shall only mention in passing a possibly significant slip of the tongue
on the part of Roosevelt at an important meeting of Roosevelt, Hull, Stimson, Knox,
Marshall and Stark at noon on November 25th, which has attracted
the interest of some revisionist scholars. Roosevelt observed that the
United States might be attacked “perhaps next Monday for the Japanese
are notorious for making an attack without warning.” One revisionist
critic has queried: “Why did he say ‘Monday’, which is Sunday in the
United states? A dispatch from the Orient from anyone except our
military or diplomatic services normally used the
Far East date rather than ours. Somehow, I just cannot believe
that Roosevelt would even have said Monday unless it slipped out
inadvertently as a result of his having read some warning message from
the
Far East.”
I am not inclined to overplay this item, and will leave it merely with
the suggestion that
Roosevelt’s
defenders give a better explanation for his mentioning “Monday” rather
than “Sunday.” This is something that had long aroused my curiosity.
Many
revisionist historians now regard the above material as adequate to
demonstrate that Roosevelt must have received impressive and precise
information by December 4th that
Japan
was planning to attack Pearl Harbor as the first act of war. Nevertheless, it is probably
best to recognize the plausibility and relevance of this assumption but
to depend more upon circumstantial evidence, based chiefly on the trend
of events from the 4th to the 7th which has now been presented in
detail. This is actually overwhelming, while the circumstantial
evidence—and there is no other evidence—supporting the contention that
even after December 4th Roosevelt still did not expect an attack on
Pearl Harbor is extremely fragile and unconvincing, as we shall now
indicate.
One
argument for
Roosevelt’s
ignorance of an impending attack is that, as a lover of ships and
especially our naval ships, he would never have sacrificed our Pacific
fleet to insure his needed attack. But he could have known or seen to
it by December 5th that the carriers, the heavy cruisers, and most of
the destroyers and pursuit planes had been sent out of Pearl Harbor,
leaving mainly the battleships, which were chiefly of sentimental
concern in the light of contemporary methods of naval warfare. This had
been done as a result of Stark’s order to Kimmel on November 27th. When
Roosevelt
was trying to “sell” his idea of a long patrol line, rather than a
double line, to the Orient, he did not seem disturbed about the prospect
of losing even a few cruisers. He wanted to see them “popping up here
and there” to fool the Japanese. He may have loved ships but he loved
politics and his own political ambitions far more.
Even
less plausible is the contention that Roosevelt would not have sacrificed the lives of thousands of
American sailors, soldiers and marines to obtain the attack. He was
then playing for high and crucial political stakes in which a few
dreadnaughts or a few thousand human lives were hardly a consideration
to override policy, however regrettable their loss. Roosevelt’s program was primarily political rather than military or
humanitarian. He surely knew that the war into which he was seeking to
put the
United States would cost millions of lives. Moreover, it is well
established that
Roosevelt
did not anticipate as great destruction of ships and life as the
Japanese bombers actually wrought. As Secretary Knox observed after he
visited Roosevelt in the White House immediately following the news of the
attack: “He expected to get hit but did not expect to get hurt.” There
can be little doubt that the Cockleship plan of December 1st was
designed to get the indispensable attack by a method which would precede
the
Pearl
Harbor attack, avert the latter, and save the Pacific fleet and American
lives.
It
is maintained that Roosevelt could have had his Japanese attack on Pearl
Harbor without its being a surprise and the forces of Short and Kimmel
could have been alerted as to the prospective attack and repelled it
with no serious losses. This fantastic suggestion runs counter to all
the well-known facts. Walter Lord and Gordon W. Prange, the main
writers on this subject, have shown with impressive evidence that
Admiral Nagumo would have been ordered to turn back if there was any
impressive evidence that Pearl Harbor had been fully alerted to the
prospect of an attack, even after the order of December 5th to “climb
Mount Niitaka.” Of course, we do not need the testimony of Lord and
Prange for these facts are fully supported by the available official
documents. There could not have been any Japanese task-force attack on
Pearl Harbor
unless it was a surprise attack.
Finally,
there is the fact that Roosevelt sent a message to the Japanese Emperor
on the night of December 6th, after he knew that the negative reply to
Hull was coming in, suggesting a peaceful settlement, but even Hull has
admitted that this was only sent “for the record” after it was too late.
Roosevelt had stressed this point of having a good formal record to
Harry Hopkins when Lieutenant Schulz brought to him, on the evening of
December 6th about 9:30, the detailed Japanese reply to Hull, which everybody in top
Washington
circles had assumed would be the moment when
Japan would attack this country. Moreover, as will be indicated
later, on the afternoon of the 6th Roosevelt had approved the
implementation of Rainbow 5 by the Dutch and British, which meant that
we were already at war with Japan, actually had been since December 3rd,
Washington time when the Dutch invoked Rainbow 5 (A-2).
There
is an alternative cogent, logical and completely factual explanation of
Roosevelt’s
decision on December 4th to concentrate on preventing any warnings from
being sent to Short and Kimmel. This does not rest upon circumstantial
evidence or any assumption that Roosevelt must have received precise
information by that time that Japan was about to attack Pearl Harbor.
Through
the “three small vessels” stratagem he had done all that he could to
secure his indispensable attack in the Far East. There was nothing left here except to wait and hope that
one of the “small vessels” would be fired on. This left
Pearl Harbor
as the only other remotely probable place that would invite and be
vulnerable to a surprise Japanese attack. Hence, nothing should be
allowed to obstruct or divert this final crucial necessity. There is no
doubt that he would have preferred a prior attack on one of the three
small vessels” and thus save the
Pearl Harbor
battleships. He hoped for this until the morning of December 7th.
It
is my firm personal opinion that this is the one unassailable and
impregnable explanation of Roosevelt’s action on December 4th for
revisionist historians to accept prior to published documentary evidence
that Roosevelt had been definitely and personally informed of an
imminent Japanese attack on Pearl Harbor. This is sound and true, even
though the circumstantial evidence of his having received such
information is overwhelmingly more convincing than the Blackout and
Blurout contention that he was surprised and shocked by the attack at
Pearl Harbor, at least beyond the shock over the actual extent of the
devastation produced by the attack. By December 4th he had brought the
country to the brink of war. Its outbreak had to come through an attack
on American forces if he was to have a united country behind the war
effort. The Far East, via the three “small vessels,” and Pearl Harbor
were the only places that remained at which he could reasonably expect a
surprise attack. The
Philippines, as has been indicated, had been so well informed of
Japanese intentions and operations through the Sadtler-Akin pipeline and
their own intercepts that MacArthur could hardly have been surprised by
hearing of immediate Japanese aggression. Moreover, Admiral Hart’s
Asiatic Fleet was so small that to destroy it would not have furnished
much protection for the extensive movements that the Japanese had
planned in the Pacific, once the war had started. Kimmel’s powerful
Pacific fleet would have remained intact.
Now
that it has been shown that apparently few top officials in Washington
except Roosevelt, Marshall and Arnold—and possibly Stark after the
4th—expected that the Japanese would first attack at Pearl Harbor, and
that Roosevelt may not have ordered all the top brass to refrain
from warning Short and Kimmel, we may indicate how he did prevent any
warning from being sent to Short and Kimmel between December 4th and
7th.
Roosevelt
first passed on his logical conclusions or specific information relative
to the impending attack on Pearl Harbor to Generals Marshall and Arnold on the 4th of December. Marshall had very special reasons for being subservient and
trustworthy to
Roosevelt. The latter, influenced by Mrs. Roosevelt and Harry
Hopkins, had rescued
Marshall
from obscurity after his conspicuous failure in the armed command of the
famous Eighth Regiment, and MacArthur had relegated him to the post of
an instructor of the National Guard in Illinois. Roosevelt promoted
Marshall to be full general over some thirty-four superior
officers, and even made him Chief-of-Staff of the Army. There is no
doubt that Marshall also greatly admired Roosevelt personally and, as
the events of December 4-7 demonstrated, put his loyalty to the
President above his loyalty to the military services and his country.
Nothing
else could account for
Marshall’s strange behavior from December 4th to 7th, right down to
his delayed sending of the “too-little-and-too-late” message to Short at
11:50 A.M. on the 7th, which we have already described. Neither
Marshall nor Stark personally wished the United States to go to war with
Japan in 1941 because they did not feel we were prepared to wage a
large-scale Pacific war, to say nothing of a two-front war in Europe and
the Pacific. They so reported on November 5th. They favored the
modus vivendi of late November which Roosevelt and
Hull
kicked over, followed by Hull’s sending an ultimatum to
Japan on the 26th. There is no reasonable doubt that if
Marshall had been left to his own convictions and impulses he would have
sent Short a real warning at least as early as November 27th, elaborated
it repeatedly, and been in his office on the afternoon and night of the
6th of December conferring with Short, if this had been needed.
Obviously it would not have been needed to deal with any immediate
attack on
Pearl Harbor
if Short had actually been warned on the 27th. Even the Army Pearl
Harbor Board stated that a clear and definite warning to Short on
November 27th, indicating the threat of an immediate Japanese movement
against
Pearl Harbor,
would have led to action by Short which would have averted the attack.
As
Admiral McCollum and others have revealed, Roosevelt quietly directed on
December 4th that no warning communications could be sent to Pearl
Harbor unless cleared by
Marshall, which bottled up Army Intelligence and the Signal Corps.
Marshall immediately informed Stark of this directive, thus preventing
any leak to
Pearl Harbor
through the Navy. This precluded sending Short or Kimmel the Winds
Execute message which was received on the 4th and was the most important
and decisive intercept that had been received indicating immediate war
with Japan, as well as all later evidence of an attack on Pearl Harbor.
Whether
Roosevelt personally emphasized to Stark this arrangement to black out
Pearl Harbor
before the night of the 6th is uncertain. When the news of the arrival
of the Japanese reply to Hull was brought to him about 9:30 on the
evening of the 6th, Roosevelt called Stark on the telephone, found that
he was out for the evening at the theater, and left word that Stark was
to call him on his return, which Stark did.
The
next morning, when Noyes, McCollum and Wilkinson showed Stark the “Time
of Delivery” message, and indicated to him that this probably meant a
Japanese attack on Pearl Harbor about 7:30 A.M. Pearl Harbor time.
Stark called Roosevelt, rather than Kimmel, and thereafter showed no
interest in contacting Kimmel, even ignoring the pleas of Noyes,
McCollum and Wilkinson for a separate warning message to Kimmel. After
discussing with Marshall the desirable content of the message to be sent
to Short—the “too-little-too-late” farce—Stark only suggested, as a sort
of afterthought, that this also be handed on to Kimmel by Short.
We
have already dealt with
Marshall’s strange behavior from December 4th to 7th and especially
on the 6th and 7th. How much influence he had on the frustration and
killing of McCollum’s clear message of precise warning to Kimmel on the
4th by Stark and Turner is not known. As reported in the officially
accepted legend, on the afternoon of the 6th, as soon as he learned that
the Japanese reply to Hull’s ultimatum would be coming in, which the
Pilot Message clearly indicated meant immediate war, Marshall abruptly
left his office and hid out somewhere, which he could not for a time
remember but later on reported to be his official quarters. On the
morning of December 7th, the
Washington-Times-Herald published an item stating that
Marshall attended a banquet of alumni of the Virginia Military
Institute on the evening of the 6th, but this has never been confirmed
or denied. Marshall did not appear again officially until Sunday morning,
whether some time around 9:00 at Stark’s office, which seems most
likely, or not until 11:25 at his own office. According to normal
military procedure he should have been in his office all of Saturday
afternoon and most of the night further informing Short and conferring
about protecting
Pearl
Harbor.
If
we accept the official legend of Marshall’s activities on December 6th and 7th,
Japan
might have attacked the Pearl Harbor fleet on the afternoon or night of
the 6th and
Marshall would have known nothing about it until he came out of
hiding late the next morning. When he did, he only sent Short at 11:50
the brief, vague, ambiguous and equivocal “too-little-and-too-late”
message, which was in no sense any warning that war was about to start
and, least of all, that the Japanese would probably attack Pearl Harbor
in about an hour. It gave Short little or no information that he did not
already have, except for the Time of Delivery message, and
Marshall deceived Short in withholding the significance he had
attached to this when reading it in his office. Finally, he refused to
use three rapid means that he had available to send his already
“too-late” message to Short, but let it be sent by Western Union to San
Francisco, and R. C. A. from San Francisco to Fort Shafter—not even
marked urgent—with the result that it did not reach Short until the
Japanese planes had returned to their carriers after the attack. The
delay in sending the message and
Marshall’s refusal to use a rapid method of transmitting it can
only be explained as due to a desire to have it arrive too late for
Short to take any action that might frighten off the Japanese attack. If
we accept the more probable version, earlier described, that the message
to be sent to Short had been agreed upon during Marshall’s conference in
Stark’s office before 10:00 on the morning of the 7th, then the delay in
sending it until 11:50 becomes all the more significant and
unpardonable, to indulge in understatement.
Marshall
saw to it that no warnings were sent to Pearl Harbor between the 4th and
the 7th. The only alleged attempt to do so came on the night of the
6th, when Knox has asserted that he made a serious effort to send a
clear and definite warning to Kimmel and to Admiral Hart, commander of
the Asiatic fleet at Manila. This never arrived at Pearl Harbor or
Manila
and Knox could not find any record of what happened to it in
Washington. Only
Marshall had the authority to kill it if Knox actually ordered such
a message to be sent of which there is some doubt.
As
conduct on the part of a trained soldier, assumedly dominated by the
ideals and professional stereotypes of those high in his profession, and
having the supreme military responsibility for the protection of his
country, it would seem both fair and reasonable to contend that
Marshall’s conduct can be explained on only three grounds: mental
defect, deliberately treasonable behavior, or carried out under orders
from President Roosevelt. The last seems the only plausible and
sensible interpretation. One thing is certain: however much
Marshall
was dominated and controlled by Roosevelt, his behavior during the brief period between December 4th
and 7th perfectly performed the function of keeping Short and Kimmel in
the dark about the danger of Japanese attack until the Japanese bombers
appeared over the Pacific fleet. And this was all achieved with a
minimum of risk and exertion on the part of
Roosevelt.
He only needed to give his blackout order directly to Marshall.
While
we are on or near that subject, it is desirable to point out that
altogether too much emphasis has been laid by both the defenders of
Roosevelt and his “Day of Infamy” rhetoric and the revisionist critics
on the alleged significance of possible “last minute” warnings late on
the night of the 6th or the morning of the 7th, whether sent or unsent.
Unless
the Japanese task force could have been frightened back more easily than
is likely, even in the light of its jittery and timid commander, Admiral
Nagumo, any warnings sent immediately after the first thirteen
points of the Japanese reply to Hull’s ultimatum had been received,
decoded and delivered before midnight of the 6th might not have made any
great difference with respect to Nagumo’s carrying through the attack.
The results could have been even more disastrous to the Pacific fleet
than it turned out to be. As Admiral Nimitz and others have suggested,
there might have been just time enough to get the ships out of port and
on the ocean, in which case they might have been sunk in deep water and
could not have been raised, salvaged and restored for action. There
would have been plenty of Japanese planes available for a supplementary
attack on the Army installations, machine shops, supplies and most
important of all, the oil supplies, still above ground, which would have
been far more of a disaster to the United States than the destruction of
the battleship fleet.
There
is little doubt that Short, Martin and Bellinger could have got many of
their planes distributed, fueled and ready for battle and some in the
air for reconnaissance, probably only to be shot down by the greatly
superior air force on the six Japanese carriers. The unarmed B-17
bombers that came in on the morning of the 7th, some of them only to be
immediately destroyed or damaged, might have been turned back. There is
little doubt that greater damage could have been inflicted on the
attacking Japanese bombers than took place in the actual attack, but it
is doubtful if the devastation wrought by them would have been greatly
lessened.
If
a warning had been sent to Short and Kimmel when the Pilot Message had
been decoded and read and the Kita message had been processed by
mid-afternoon of the 6th, it might have been a quite different story.
Defensive movements at Hawaii connected with an alert put in operation during the
afternoon of the 6th might have caused the Japanese task force to
abandon their bombing mission and turn back or to face an empty harbor.
That would have made a great difference in the fate of the ships at
Pearl Harbor. In this case, Kimmel could have put to sea with all his
available ships, joined Halsey who was returning from Wake, linked up as
soon as possible with Newton and Brown, and through a surprise attack
perhaps have inflicted a serious surprise blow on at least a part of the
returning Japanese task force whose location could have been rather
precisely determined by Commander Rochefort at Pearl Harbor. Nagumo
could have had no knowledge of the location of Kimmel’s reorganized
fleet. In any event, Nagumo’s bombers would have found an empty harbor
at
Pearl Harbor
and all of Kimmel’s warships out of sight beyond the horizon.
When
it comes to later warnings that could have been given, but were not,
such as Knox’s mysterious alleged message to Kimmel and Hart late on the
night of the 6th, a warning to Kimmel by Stark shortly after 9:00 on the
morning of the 7th, when Noyes, McCollum and Wilkinson explained to him
the significance of the Time of Delivery message, and Marshall’s
“too-little-and-too-late” message at 11:50, or even a clear and
forthright one by Marshall at least two hours earlier, there is only a
gambling chance that the disaster to the Pacific fleet would have been
greatly lessened. As pointed out above it would have been worse had the
ships been sunk in the deep Pacific beyond hope of salvage and repair.
The failure to get off these last minute warnings promptly, or not at
all, may have great significance for the historian and the moralist but
they are far less important strategically.
It
is only fair, however, to present here an informed critique by Commander
Hiles of my opinions on the probable results of a warning if sent by
Stark to Kimmel even as late as 9:30 A.M. on the morning of the 7th,
when Stark was made to realize that the Time of Delivery message implied
the probability of a Japanese attack on Pearl Harbor at around 7:30 A.M.
Pearl Harbor time. Actually, the bombers did not arrive until about
7:50. If, as McCollum maintains, Marshall was also in Stark’s office
between 9:00 and 10:00 on the morning of the 7th and they had there
decided to send a real warning to Short, this could have been sent by
10:00 A.M. instead of 11:50. Hiles admits that even a clear warning
sent at 11:50 by the most rapid method, would not have made possible an
effective averting of the Japanese attack, although both Short and
Kimmel would have had time to get some of their anti-aircraft armament
in shape for action and Short might have got more of his planes off the
ground by the time the Japanese bombers arrived, thus increasing the
damage to the Japanese. But let us stick to the approximately four
hours that Short and Kimmel would have had in which to take action if
Stark and/or Marshall had sent clear warnings to them by around 9:30 on
the morning of the 7th. According to Hiles:
It
would have required only four hours at the most for the Pacific fleet
that remained at Pearl Harbor on the morning of the 7th to sortie from
the harbor, and still less time for Short to have re-oriented his planes
and anti-aircraft defense against the attack. Kimmel could have sortied
and dispersed his entire force and perhaps rendezvoused with the three
main task forces later on, but most likely he would have kept his ships
dispersed until he learned the composition of the attacking force.
After this, it is problematic as to just what he would have done.
Admiral Kimmel has told me that he is not sure just what action he
would have taken. Two American carriers against six for the Japanese
might have been too great a risk to take, although the battle of Midway
was won in June, 1942, against a much superior Japanese naval force and
the four Japanese carriers were mainly destroyed by one American
carrier, the
Enterprise.
Now
let me set the stage for you, assuming a four-hour alert. It would have
been 3:30 A.M. at Hawaii. It would have been dark and would remain so for several
hours. There would have been no need to wait and recall the liberty
section which was ashore. The ships could always function with the duty
sections in an emergency. Long before daylight all the major units
would have been clear of the harbor and well scattered beyond visual
distance. A few of the smaller ships might have been visible by the
time the Japs arrived but even this is not likely because a large part
of the fleet was already away with Halsey, Newton and Brown, and four
hours would have been adequate for the depleted fleet to have sortied
and dispersed. It
is safe to say that the Japanese would have found both the horizon and
the harbor empty. This also presupposes that Nagumo would not have been
alerted by Japanese spies in Honolulu as to the sortie during these four
hours and withheld the attack, even have got ready to turn back to
Japan. Several scouting planes preceded the attack waves to report back
to Nagumo as to the state of the fleet and it is unlikely that Nagumo
would have ordered an attack on an empty harbor.
Let
us assume, however, that the bombing planes did proceed to Pearl Harbor
and found nothing there. The targets were gone and well scattered out
over the broad Pacific. The Japanese planes had no spare fuel to go
flying around completely blind, looking for targets they knew not where;
as it was, some of them ran out of fuel before they got back to their
carriers after the attack. With this unexpected denouement, Genda and
Fuchida (the Japanese bombing commanders) would have had no other choice
than to recall the planes or bomb the off-shore installations and the
shops, machinery and oil, which is not very likely under the
circumstances. And up to this point we have assumed perfect conditions
for Fuchida and have ignored the fact that Short and Martin would also
have had that same four-hour warning and that their planes would have
been in the air and the anti-aircraft guns ready to greet the bombers.
It
is well to have this authoritative and detailed portrayal of what might
have happened at Pearl Harbor and Fort
Shafter on the morning of December 7th if Stark and Marshall had
sent warnings to
Hawaii by or before 9:30-10:00
Washington time.
Of
course, Commander Hiles is assuming that all would have worked out
smoothly if the warning had been received about 3:30 A.M. on the morning
of the 7th, but how a situation looks on paper may be quite different
from how it will take shape in action. It would have been quite a shock
to officers, crews and soldiers to have been rudely awakened at 3:30
A.M. with the news that Oahu was about to be shattered by a Japanese
bomber attack when there had been no previous warning of any such move
and nearly every one in the armed forces there had been convinced that
Japan would not make war on the United States, a rich and powerful
country that no small island empire could hope to overcome.
Events
might have worked out as Commander Hiles has indicated. On the other
hand, there might have been much confusion, with the ships not all out
of sight when the Japanese bombers arrived. The channel leading out of
Pearl Harbor
was so shallow that the battleships had to move slowly, and in the haste
and confusion one of them might have run aground and made it impossible
for ships behind it to reach the open sea. But it is certainly true
that, if a clear warning had reached Fort Shafter and Pearl Harbor
between 3:00 and 4:00 on the morning of the 7th, the Army and Navy
forces and equipment on Oahu would have suffered smaller loss than
occurred, unless Kimmel’s battleships had been sunk in deep water as the
Repulse and Prince of Wales were shortly afterward in the
southwest Pacific.
It
is, of course, utterly abhorrent to have to conceive of Kimmel’s being
subjected, as a result of Washington neglect or treachery, to any such
shock and crisis as being warned of a Japanese attack at 3:30 on the
morning of the 7th, when he could and should have been effectively
warned days, weeks or months before, and there would not have been any
Japanese attack on Pearl Harbor.
The
main reason for deploring overemphasis on the failure to send last
minute warnings is that this obscures and confuses the real nature and
the extent of the guilt for failing to warn
Hawaii in plenty of time. There was every reason for sending a
clear warning there on November 27th, and any delay after December 4th
was nothing less than criminal neglect, if one wished to save the
American forces at
Pearl Harbor.
If one limits main consideration of the warning period to the late
night of the 6th and the morning of the 7th, the Blackout and Blurout
writers can conjure up all sorts of confusing alibis about timing and
the possible disastrous results of warnings sent at this late hour, but
there is no such way to counter or explain the failure to warn the
Hawaiian commanders at any time during the previous nine days, or even
as far back as when the first Bomb Plot message was decoded and read on
October 9th. Nor was there any excuse for having failed to provide
Pearl Harbor and
Fort
Shafter with a Purple Machine to intercept, decode and read the
Japanese diplomatic messages right there, thus learning of the danger on
the spot.
It
is also futile and misleading to exaggerate the minor acts of
incompetence or mis-judgment at Pearl Harbor between very early morning
on the 7th and the attack at 7:50, so much stressed by Roberta
Wohlstetter at the beginning of her book on
Pearl Harbor.
Such were the failure properly to interpret the discovery of a Japanese
scouting submarine right off Pearl Harbor on the early morning of the
7th, the apparent indifference shown by Lieutenant Kermit Tyler of the
Army Air Corps to the report from the Army radar station about some
strange approaching planes, which might have been thought to be those of
Admiral Halsey who was returning with his task force from Wake or the
approaching B-17’s, and the official closing down of this radar station
at 7:00 on the morning of the 7th, as had been ordered, but was not
actually closed. These have some curious interest as minor deficiencies
and mistakes of judgment, greatly bolstered by the impact of hindsight,
but they had little to do with the approach, diverting or repulse of the
Japanese bombing planes, which were already well on their way from their
carriers to attack Pearl Harbor.
Incidentally,
technically speaking and as a matter of curious interest, despite the
repeated orders of Roosevelt and his associates that Japan must be
allowed to fire the first shot, this was actually fired when the
commander of the American destroyer Ward fired upon and sank a
Japanese submarine off Pearl Harbor about an hour before the first wave
of Japanese bombers arrived there.
Defenders
of Roosevelt and Washington have sought to equate these trifling and
exaggerated errors, due to surprise, confusion and haste at Pearl Harbor
on the morning of the attack, with the failure of Washington to pass on
the Bomb Plot messages to Short and Kimmel, the refusal to give them the
Purple diplomatic messages, the denial of a Purple machine to Pearl
Harbor, and the killing of the efforts of McCollum, Noyes, Sadtler,
Bratton and others after November 26th to warn Short and Kimmel about
the approaching danger of a Japanese attack. These were the main
deliberate delinquencies of Washington which made possible the
successful Japanese attack—to say nothing of the fact that most of the
top Washington officialdom, civil and military, involved in such
matters, appear to have permitted the movement of Japanese task forces
down the southwest Pacific and the strategic assumptions of Rainbow 5 (WPL
46) to drive the whole traditional Pacific naval strategy relative to
the threat to Pearl Harbor from their minds.
The
time to have started warning Pearl Harbor was with the decoding,
translating, reading and evaluation of the first Bomb Plot message on
October 9, 1941, obviously by November 5th when it announced that Japan
had set a deadline for negotiations, and certainly by November 26th when
Hull rejected the Japanese diplomatic plans, kicked over the modus
vivendi, and sent his ultimatum to Japan. Literal warnings even if
sent by the time the codes destruction intercepts were at hand on
December 1st and 2nd, and when Winds Execute was received on the 4th,
would most certainly have averted the Pearl Harbor attack. And, surely,
a Purple machine should have been sent to Pearl Harbor by July, 1941,
where the competent operating force, headed by Commander Rochefort,
could have intercepted and read the revealing diplomatic messages from
Tokyo. If they had been ordered to do so, they could have read
the Bomb Plot messages in the J-19 and PA-K2 codes without a Purple
machine.
The
overwhelming responsibility for the war and the attack was, of course,
Roosevelt’s deliberate refusal to settle the relations between the
United States and Japan in a peaceful manner by honest diplomatic
negotiations, to achieve which Japan made unusually impressive gestures
and offered very reasonable terms that protected all legitimate vital
American interests in the Far East. As I pointed out in an article in
the Progressive, December 6, 1941, we were surely more thoroughly
and logically involved in continuing our profitable and peaceful
relations with Japan in 1941 than with supporting Chiang’s tottering and
corrupt regime, even though Roosevelt’s maternal grandfather may have
made no money out of trade with Japan. Japan
had no military designs against the
United States except for self-protection in the event of war. Japan
made two genuine but vain offers to withdraw from the Rome-Berlin-Tokyo
Axis in return for peaceful relations with the
United States; these were unceremoniously brushed off.
Roosevelt
was, however, so deeply involved in his anti-Japanese war plans and his
commitments to Churchill and others by the late summer of 1941 that it
is extremely doubtful if he would have accepted any Japanese diplomatic
proposals short of complete surrender. He would have been surprised,
shocked and annoyed if the Tojo government had been willing to humiliate
themselves enough to resume negotiations on the basis of
Hull’s ultimatum of the 26th. The desire to prevent this was a
major consideration of Roosevelt and Stimson in connection with
formulating the fake warnings to Short and Kimmel on the 27th.
Forward to Chapter VII
Back to Chapter I and Table of Contents
Back to Barnes Page