FRANKLIN D. ROOSEVELT AND “WILD BILL”
America’s First Central Intelligence Apparatus
The teams of OSS officers and enlisted men on the ground were trained to expect and anticipate almost anything. They were trained to operate well behind enemy lines, in this instance in Burma during World War II. But no one had prepared them for a visit by their own commander, William “Wild Bill” Donovan, brigadier general and head of the Office of Strategic Services (OSS). He held an incredible storehouse of American military secrets in his head; he really shouldn’t have exposed himself to the risk of being captured.
But Donovan was, in the words of his assistant, Ned Putzell (who was also on this trip), the sort of leader who “couldn’t ask somebody to do something he himself wouldn’t do.”1 He would appear unexpectedly in dangerous locations just to pay a visit to the operatives he’d sent there. In this case, Donovan arrived with Colonel Carl Eifler, head of OSS operations in Burma, in a “damn little biplane,” landing on an improvised airstrip at an OSS base well behind Japanese lines. After Donovan had completed his tour of inspection and was talking with both OSS personnel and the Kachin guerrillas they were training and supplying, someone came running. The enemy was closing in. Donovan had to leave. Donovan’s men pulled the camouflage off the very short airstrip, which ended in a river. On the other side was the face of a cliff. The men on the ground held the tail of the biplane until the pilot had the prop up to full speed. The plane carrying America’s only spymaster then shot down the runway, barely getting off the ground before reaching the river, and just gaining sufficient altitude to make the turn before crashing into the cliff. Donovan was thrilled. Putzell was “sure as the Lord made little apples” that Donovan had escaped certain death.2
Never since Donovan has one of America’s spymasters been so deeply involved in operations. None of his successors would have dreamed of making a clandestine visit behind enemy lines. To be honest, when I was director of central intelligence, I considered myself a manager, nothing more. But “Wild Bill” Donovan, who got the ball rolling, couldn’t shake the romance of the profession. He had to live it.
Now, Donovan didn’t come to the job with any intelligence experience to speak of. He was a lawyer. He was not a close friend of Franklin D. Roosevelt. In fact, he was a Republican. So how did such an unlikely character end up crafting America’s first centralized intelligence service?
Preparing for war, a war that would propel America permanently onto the world stage, President Franklin Roosevelt made some changes at home. With incredible political dexterity, he threw the Republican National Convention of 1940 into disarray by appointing two Republicans, Frank Knox and Henry Stimson, to his cabinet. Knox would become secretary of the navy and Stimson, who had served under President Hoover just a few years before, would be secretary of war. Both believed, as did the president, that the war that had broken out in Europe in 1939 would inevitably, and soon, involve the United States. Roosevelt did not know at the time that he had just paved the way for Donovan to enter his inner circle of advisors, and that doing so would lead to the creation of America’s first full-fledged intelligence service.
What Roosevelt did know was that the intelligence apparatus he had was quite inadequate. What the president had available to him were offices dedicated to intelligence in the Army, the Navy, and the State Department, plus some foreign intelligence gathering by the FBI. None of it, however, was very aggressively oriented. Today, with Pearl Harbor, the Cold War, and 9/11 behind us, it is difficult to appreciate how naïve we Americans were about collecting intelligence clandestinely prior to World War II. Spying simply did not fit with our notions of “fair play.” For instance, the president felt it necessary to emphasize in a press conference in 1938 that he would never “sanction espionage by American agents abroad.”3
The one exception, where our intelligence was reasonably aggressive, was in intercepting and deciphering radio and cable messages of other nations. Back in 1929, Secretary of State Henry Stimson had closed down Army code-breaking operations. Fortunately, the Army pursued code breaking nonetheless. By August 1940, we had broken the code known as PURPLE, in which the Japanese government communicated with its embassies around the world. Roosevelt, then, did have available to him deciphered Japanese diplomatic messages. These were of inestimable value as Pearl Harbor approached. Fewer than forty people—including the president; the secretaries of war, navy, and state and the top military brass; and those who worked on the codes—knew of this operation and had access to the decrypted messages. This system of intercepting and deciphering was named MAGIC, and it was the most productive intelligence operation we had as we headed for World War II.
Through all this, the way MAGIC was processed for delivery to the president was almost a comedy. While the Army had broken the Japanese code, the Navy had the best translators of Japanese. Processing the sheer volume of messages required both their efforts, and they worked together smoothly. But a certain amount of friction developed over the question of which military service would deliver the transcripts to the president. As will be seen throughout this book, intelligence services get their light and air from the chief executive. How much a president brings them into his plans and thinking determines how well they can target their efforts to serve him. How much a president, in turn, supports them in bureaucratic disputes is vital to their ability to function. How strongly a president backs their secret activities before Congress and the public can determine their long-term viability. So a fight that developed between the Army’s and Navy’s intelligence services over who would deliver MAGIC to the Oval Office was over more than just bragging rights.
The compromise arrangement was that the Army would deliver messages during odd-numbered months and the Navy during even ones. This was not so bad in itself, until July 1941, when there was a breach of security with MAGIC. It came from FDR’s Army aide. This led the Army’s chief of intelligence to begin sending summaries only, rather than complete messages, to the president. Thus, if there were another breach, it would not be so obvious that we had the actual texts of Japanese messages. The Navy, however, continued providing the complete texts. It took some time for Roosevelt to deal with this ridiculous anomaly. As diplomatic relations with Japan deteriorated, he demanded the actual transcripts from the Army, rather than summaries. It had to be crystal clear to Roosevelt that there was a serious lack of coordination and cooperation between at least two of his three key intelligence services: Army, Navy, and State. The fourth, the FBI, was collecting a certain amount of intelligence in Latin America and elsewhere through its personnel stationed at various American embassies. He could only guess how much his lack of adequate intelligence was due to such rivalry.
Roosevelt went about filling his intelligence gap in his own inimitable manner. Not only in intelligence, but also in almost every area of concern to the executive branch, he liked to have several sources of information. In the case of intelligence he simply deputized a number of individuals and groups to be unofficial intelligence services reporting to him directly. One was Joseph Grew, U.S. ambassador to Japan. Grew had graduated from Groton and from Harvard two years ahead of Roosevelt. They knew each other from days on the Harvard Crimson. Grew, whose view of Secretary of State Cordell Hull was less than flattering, visited FDR as well as Hull every time he came to Washington. He also felt free to write to the president directly from time to time. On December 14, 1940, he wrote: “It seems to me increasingly clear that we are bound to have a showdown [with the Japanese] some day, and the principal question at issue is whether it is to our advantage to have that showdown sooner or have it later.”4 Interestingly, because Army intelligence had so few observers, let alone spies, in Japan, the chief of Army intelligence considered Grew his best source on what was going on there.
There were also some private citizens FDR used for intelligence gathering. One of them was the wealthy and well-connected Vincent Astor. Astor organized a secret club called “the Room” that met in New York to discuss gossip in the guise of foreign intelligence, aided by heavy drinking. In 1938 the president sent Astor and Kermit Roosevelt, a cousin of his and a son of Theodore Roosevelt, into the Pacific on Astor’s yacht to collect information about Japanese installations. It appears that Astor had a thrilling adventure but did not return with any groundbreaking intelligence. Astor, though, was a director of the Western Union Telegraph Company, which allowed him to provide FDR with the text of sensitive telegrams and cables. And he had a number of bankers in his “Room” who allowed him to gather intelligence on transfers of funds. Roosevelt’s directions to Astor are not on record, but Astor’s messages to Roosevelt suggest that FDR fully approved of these questionable activities.
Still another unofficial intelligence operative was John Franklin Carter, a friend of the president’s, who obtained FDR’s approval for “a small special intelligence and fact-finding unit” that would get its funding from the State Department. Assistant Secretary of State Adolf Berle, a skeptic about centralized intelligence, grumbled, “I am not, of course, familiar with what the President has asked him to do, nor do I wish to be.”5 Carter himself later evaluated his contribution as minimal: “It was a picturesque and wildly funny affair at times. Very fantastically amusing things happened as they always do in off-beat operations and I think we all had fun.”6 One of Roosevelt’s first directives to Carter was to spy on and report about Vincent Astor’s operation.
Much more effective and even more mysterious was Wallace Banta Phillips, a rather quiet and colorless businessman with a London-based rubber company. With Roosevelt’s approval, Phillips entered the espionage field through the Office of Naval Intelligence in May 1940. The fact that Phillips’s rubber company already had an industrial espionage ring in place meant that the Navy could spy without having to endanger its own attachés. Astor quickly learned of this newcomer and complained to FDR that the businessman was just a dangerous “social climber.”
In each of these failed experiments, we see a president hungry for information direct from the source, delivered by people untainted by politics. In some cases, we see Roosevelt using a shroud of secrecy to bypass Congress. In spring 1938, he wanted to support the Spanish Loyalists, then fighting Francisco Franco’s Nationalist forces, but the 1936 Neutrality Act tied his hands. Roosevelt, though, was able to route supplies to the Loyalists in the Spanish Civil War in direct violation of the neutrality legislation. He built a tiny spy ring for this operation, consisting of his wife, Eleanor Roosevelt, who cabled messages to Ambassador William Bullitt in Paris; a journalist; and his brother-in-law G. Hall Roosevelt. A scheme to supply 150 airplanes to the Loyalists fell apart when France began to seal its porous border with Spain.7 Much later, when Roosevelt again felt tied down by the neutrality law, he was able to turn to Bill Donovan.
Donovan was a second-generation Irish Catholic born in Buffalo, New York, in 1883. His family, which had fought for labor rights, was decidedly not Republican. Donovan showed early promise in college (while also working in factories and on construction sites) and obtained a scholarship to attend Columbia University Law School. Athletic, articulate, and handsome (those who knew Donovan, male or female, always remarked specifically on his dazzling blue eyes), Bill Donovan came into his own at the law school where FDR was a student contemporary, though Roosevelt barely knew Donovan there. After graduating, Roosevelt would soon be involved in politics and public service, while Donovan would return to Buffalo to start a law firm.
During the First World War, the defining event of their generation, each served in a way appropriate to his station in life. Roosevelt was an assistant secretary of the navy, while Donovan was a lieutenant colonel serving in the army in France. Roosevelt cultivated the image of a rule-breaker, a rebel inside a staid bureaucracy who got things done for the good of the men and women in uniform. Donovan already had earned the nickname “Wild Bill” while chasing after Pancho Villa along the Mexican border in 1916. Then, serving in the “Fighting 69th” regiment in France in 1918, he personally led an assault on an extremely well fortified German position and remained engaged and exposed even after being shot by an enemy machine gun. He received medals from several foreign governments for this action, and eventually the U.S. Congressional Medal of Honor, the Distinguished Service Medal, and many other medals, becoming one of the most decorated men in American military history. But he often broke down crying when he spoke of the men who “died out of loyalty to me” and requested that his Medal of Honor be deposited with his regiment.8
After the war, Donovan’s law practice flourished, but he was also drawn into state politics on the Republican side as U.S. attorney for the western district of New York. He made an unsuccessful bid for lieutenant governor of New York. When one of his law school mentors, Harlan Fiske Stone, became U.S. attorney general in 1924, Donovan joined him as an assistant attorney general. Now Donovan started to get a glimpse of where his unbridled ambition might take him. Within a year, he thought of being in the president’s cabinet, maybe not as secretary of state right away, but eventually. By 1928, there was muted talk of a Hoover-Donovan ticket.
Donovan’s labors on behalf of the Hoover campaign of 1928 did not go unnoticed; it seemed almost certain that the administration would reward him with an appointment to a significant position in the government. He was offered the job of governor-general of the Philippines. He refused, citing family reasons, but the real reason was that he felt it meant being exiled from Washington. Donovan tried to reenter public life by running for governor of New York in 1932, the same year Governor Franklin Roosevelt first ran for president. The gubernatorial election was a disaster for Donovan. His defeat—he received only a third of the votes—by a banker in the midst of the Great Depression seemed to put a final cap on his political ambitions.
Still, he traveled to Europe frequently in the early 1930s, and each time he came back he was more convinced that it was time for America to take a more active role in the world. Because of his distinguished war record, Donovan had entrée to important people all over Europe. He met with Mussolini in Italy and a number of top officials in London. He saw impressive demonstrations of German military hardware. After these experiences, he told a group of American veterans: “In an age of bullies, we cannot afford to be a sissy.”9 By the end of the 1930s, more people were coming around to this view, but a majority of the American public (not to mention the Republican party) were isolationist, so Donovan was going against the grain. His attempts to draw the nation’s attention to the dangers posed by the rise of two aggressive powers, Germany in Europe and Japan in East Asia, brought him closer to Roosevelt.
Donovan had no experience in intelligence—either spying or analysis. He had limited experience in government, and practically none in managing large bureaucracies. So why did Roosevelt give him the job—initially with the intriguing title of Director of the Office of the Coordinator of Information (COI)—of creating and heading a new intelligence operation? Donovan had a wide and diverse network of connections that enabled him to recruit talented people ranging from businessmen and Wall Street lawyers to academics from Yale and other elite institutions. He essentially was in the right place at the right time. He also had enormous charm and charisma. “He could charm anybody,” recalled Ned Putzell. “He experienced idolatry everywhere. And people admired him so damn much.”10 Another former Donovan subordinate in the OSS, Fisher Howe, remembered, “When you were with him, you were the only person that counted, and that is a tremendous talent.”11
In 1940, Donovan’s daughter was tragically killed in an auto accident. Donovan was not yet involved in White House discussions and there was no indication that he would ever be a person of consequence in government, but Roosevelt wrote a personal note of condolence that meant a great deal to Donovan. (I have some insight into how he might have felt. When I was in an accident myself in early 2000, President Clinton, whose charm and empathetic power has often been compared to Roosevelt’s, took time to call my family and spent about a half hour on the phone with them. I had retired from government service, and I barely knew the president. The fact that he took this time out of his day made me believe it was much more than a gesture.) The letter from FDR to Donovan was their first contact in a number of years, a personal touch that helped create a bond between the two men.
Just a few months later, on June 5, 1940, German troops crossed the Somme and the Battle of France began. On the fourteenth, the Germans entered Paris. On the sixteenth, Marshal Henri-Philippe Petain became head of the French government, and the next day asked for an armistice. On July 9, Secretary of the Navy Frank Knox and Roosevelt discussed the fall of France. They had already come to the conclusion that such a rapid fall could never have been achieved by regular military operations alone. Very quickly, rumors of a German “fifth column” provided a convenient answer. Knox suggested to Roosevelt that the non-isolationist Republican Bill Donovan be sent to Europe on a fact-finding mission. If such a fifth column existed and had played a role in the amazingly rapid collapse of French forces, Donovan was tasked to find out how this operation of internal subversion had succeeded. Second, he was asked to address an even more important question: Could Britain survive? The British had such a vested interest in convincing Donovan they would survive that they unveiled their war plans to him. They claimed they would win not through “traditional” warfare but through strategic bombing, economic pressure, the creation of widespread revolt in Germany’s occupied territories, and the “making of chaos and revolution.”12 They were vastly outnumbered and alone. Initially at least, fighting such a non-traditional war with limited resources was their only hope. Donovan reported this to Knox on August 5. A meeting with Roosevelt followed on August 9 at Roosevelt’s home in Hyde Park.
Roosevelt needed a spymaster, since his intelligence resources consisted of uncoordinated agencies, each going its own way to meet the perceived needs of its parent organization and not necessarily cooperating with the others. Donovan was a candidate for heading an entirely new intelligence operation, someone who would, because of the evolving personal relationship, have a direct line to FDR. The president had, at least in the early days, a high opinion of Donovan because of his open interventionist streak, his unconventional thinking, and his enthusiasm and verve. He must have seen Donovan as an ideal coconspirator. “Wild Bill’s” scattershot enthusiasm fitted neatly with FDR’s willingness to try almost anything if it had a reasonable chance of working (and if the Supreme Court or Congress didn’t stop him). Donovan was a bull in a china shop and an empire builder who was quite naïve when it came to what other, more powerful players thought of him. The same qualities and behavior that would have caused other presidents to double-bolt their Oval Office doors caused Roosevelt to swing them wide open when Donovan came knocking. Both men have been described as the sort who believed in not letting the right hand know what the left hand was doing. Both enjoyed being the one holding the invisible strings. For instance, when Donovan later needed to forge some German passports, he went to FDR to obtain the services of seventeen counterfeiters in federal prisons. FDR agreed, with the condition that if the Secret Service ever found out, Donovan would be out of business. They never did.13
Neither FDR nor Donovan concerned himself with details. Both were idea men. Donovan spent so much time focusing on the big picture that he proved to be a disastrous administrator. Both ran with ideas long before they had the details or framework even barely sketched out. Remembering Donovan, Fisher Howe recalled something that Dean Acheson used to say. “Ah, but a man’s reach should exceed his grasp (except for diplomats and trapeze artists).”14 For domestic policy, Roosevelt had the “brain trust,” an exciting group of eager, young go-getters. On the other hand, he had a cautious and traditionalist secretary of state, Cordell Hull, a man FDR found “exceedingly vague.”15 With such a diverse mix of personalities and temperaments, FDR knew how to play each one, telling Hull one thing one day, and cabinet secretaries Harold Ickes or Henry Morgenthau something quite different on another day. By masking his intentions and playing one man off the other, FDR got where he wanted to go. But this method could work only to a limited degree. In the military, he was dealing with men like General George C. Marshall, the army chief of staff, and Admiral Harold R. Stark, the chief of naval operations, neither of whom were inclined to try out wildly innovative ideas. Nor were their civilian bosses, old-line Republicans Stimson and Knox. Donovan was quite another matter. He embarked on many a bold and innovative scheme but would find himself, over the next five years, dealing with and outmaneuvered by consummate political and bureaucratic operators, men like J. Edgar Hoover of the FBI.
But Donovan was a breath of fresh air for FDR, and an innovator. To an insightful OSS historian, Donovan was FDR’s “Oval Office playmate,” someone with whom he could bounce around ideas, someone who shared his enthusiasm.16 “The relationship,” recalled Donovan’s assistant, “was such an easy one”17—at first.
One CIA legend suggests that their relationship was so easy that Donovan could get away with firing a gun in the Oval Office. In order to demonstrate to FDR a new super-silent .22-caliber pistol, Donovan emptied the magazine, ten shots, into a bag of sand. Roosevelt was on the phone and never heard a thing. Donovan put the smoking gun on the president’s desk and told him what had just happened. We can presume that they never informed the Secret Service that Donovan had entered the Oval Office with a loaded weapon.18
In April 1941, Roosevelt expressed concern about a “twilight zone” between several of his departments. He had asked at least twice before that the FBI, Army, Navy, and State get together and “coordinate” intelligence. Now, with the help of the British, in particular Rear Admiral John H. Godfrey and the future creator of James Bond, Ian Fleming, Donovan had drafted a sweeping plan to take over intelligence by becoming the “Coordinator of Information” (COI). He submitted this directly to the president. It was written to please Roosevelt’s interventionist leanings and to send the message that Donovan was ready to take charge. Donovan promised to build an agency that would “constitute a means by which the President, as Commander-in-Chief, and his Strategic Board would have available accurate and complete enemy intelligence reports upon which military operational decisions could be based.”19 Roosevelt, quite tired of mediating Army-Navy disputes, approved everything in Donovan’s memo.
As could be expected, Donovan’s proposal invited attacks by the military services, presaging a struggle between centralized intelligence and the military that continues to this day. The last thing the military wanted was some civilian interloper telling them how to conduct intelligence operations. And, alarmingly to them, the directive stated that the COI was to provide the president and his “Strategic Board” with information “upon which military [emphasis added] operational decisions could be based.” Putting this out in front of military commanders at a time like that was like waving a red flag in front of a bull. The military and State opened fire on the plan, particularly the part Donovan valued the most—his direct access to the president.
General Marshall, already a towering presence, diagrammed a scheme whereby intelligence developed by the COI would be routed through the Army and Navy before going to the president. Donovan consented to this plan in principle, but in practice seldom filtered his information and evaluations through the Army and Navy while he was Coordinator of Information. Roosevelt did not object.
Next, the people in the president’s own budget office fretted over the fact that no one really knew what this emerging office with the evasive name of Coordinator of Information was actually going to do and cost. And there were any number of specific decisions on whether COI would encroach on other entities in order to do coordination. In a number of these cases Roosevelt did not give COI all the authority Donovan sought. The most glaring of these was that COI was not given access to MAGIC and was forbidden to set up its own code-breaking operation. Access to MAGIC was so restricted that it appears that no one, including Roosevelt, even considered that Donovan should ever have it. How useful, though, could a “coordinator of information” be if the best information on the increasing tensions in East Asia and the Pacific remained out of his reach? And what does that say about how Roosevelt viewed the role Donovan and this new agency should play? Such limits on our central authority for intelligence have persisted ever since.
Still another issue was how far COI’s writ carried into the field of propaganda, or “psychological warfare,” something about which Donovan felt strongly. New York mayor Fiorello LaGuardia had been appointed director of civilian defense and was assuming a role in domestic propaganda. FDR assured LaGuardia that COI would conduct psychological attacks only on foreign populations, whereas LaGuardia’s responsibility was to “sustain the morale of our people within the national boundaries.”20 Donovan, nonetheless, developed domestic propaganda plans, but this placed him in direct conflict with a powerful opponent.
Donovan also found himself in an internecine struggle with the wealthy and well-connected Nelson Rockefeller, who was serving as coordinator of inter-American affairs, a presidential appointment. Rockefeller saw himself as responsible for managing the propaganda message in South America and viewed Donovan as an amateur and interloper. Donovan overplayed his hand when he told Rockefeller that he would not “enter into a compromise” and insulted his adversary with a condescending conclusion to an uncompromising letter: “I think I don’t need to assure you, Nelson, that this is no mere jurisdictional question. It is a matter of major policy.”21 Rockefeller responded by going directly to FDR, bringing with him a draft executive order prohibiting COI from doing propaganda work in South America. Roosevelt made a few changes and sent the order out on the same day. Donovan made a feeble attempt to regain control of radio broadcasts in South America, to no avail.
Thus, the first presidential effort to give central direction to American intelligence still left the country with a mélange of virtually autonomous intelligence agencies and uncertainty as to just what the role of the new coordinator would be. It was less than five months before the country was jolted by what, until 9/11, was the greatest intelligence failure in our history—Pearl Harbor. This failure was, at least in some measure, the result of both a lack of exchange of data among intelligence agencies and a failure to coordinate and focus their efforts. Alarmingly, both issues still plague our intelligence today.
Since Donovan’s agency was quite new, and since he did not have access to MAGIC, hardly anyone has placed any blame on Donovan for failing to predict Pearl Harbor. But Donovan was part of the problem. As Coordinator of Information, he did very little coordinating. Instead, he added to the pile on FDR’s desk—more than 260 memos and phone calls during his first six months as COI.22 Very little in this blizzard could be considered either evaluated intelligence or the “accurate and complete enemy intelligence reports” that Donovan’s original proposal had promised. Instead, Roosevelt got volumes and volumes of unprocessed material. When it came to Japan, Donovan failed to offer an evaluated estimate. Instead, he offered two contradictory statements.
On November 13, 1941, Donovan relayed “the substance of statements” made by Hans Thomsen, German chargé d’affaires, to one of Donovan’s “assets.” According to Mr. Thomsen, Japan was going to attack: “If Japan waits, it will be comparatively easy for the United States to strangle Japan.” Time was not on Japan’s side. “Japan is therefore forced to strike now, whether she wants to or not.”23 Of course, this was just one man’s opinion. And it was buried in the transcript—neither highlighted nor analyzed.
This statement of Japan’s intentions, as useless as it was, was soon contradicted by another COI report. A journalist traveling in China on a COI mission asserted that as long as Russia was still a force, “Japan would do nothing to provoke a major war.” He concluded that “time is really on our side.”24 On the twenty-eighth of November, Donovan discussed the issue over breakfast with Roosevelt, and the president lamented that “it was difficult now to find a formula in dealing with Japan.”25
The stunning thing about this conversation is that Roosevelt had already concluded, based on what he had read in intercepted Japanese messages, that war with Japan was inevitable. But he continued to keep Donovan, his own intelligence chief, in the dark about MAGIC, as well as the conclusions he had drawn from reading it. To the military chiefs, who knew about MAGIC and what it foretold, Donovan must have looked like a rank amateur who did not, in fact, enjoy presidential support.
On November 25, 1941, days before that breakfast with Donovan, Roosevelt had met with the secretaries of war, navy, and state, as well as the chief of naval operations and the chief of staff of the army. The subject was budgets. Roosevelt, though, told them in no uncertain terms that the Japanese were planning a surprise attack, perhaps as soon as December 1. In short, rather than these officials informing the president of what their intelligence agencies told them, the president was drawing his own conclusions and warning his subordinates.26
As a result of the president’s warning, the military leaders transmitted an alert to the Pacific commanders with no ambiguity whatsoever: “This dispatch is to be considered a war warning. Negotiations with Japan looking toward stabilization of conditions in the Pacific have ceased and an aggressive move by Japan is expected within the next few days.”27 What was missing, of course, was where our commanders should expect the action. Tokyo was apparently being very close-mouthed with its diplomats on this. Concurrently, though, there had to be Japanese military message traffic that was more specific. It probably lay in Japanese naval transmissions, which were not being deciphered.
When the Japanese attack struck Pearl Harbor at dawn on December 7, it was about 1:00 P.M. on the East Coast of the United States. Donovan was surprised when loudspeakers called him out of a football game in New York. He reached Washington and was in the Oval Office late that evening after the president’s meetings with the cabinet, congressmen, and military advisors were over. Roosevelt still had a long-standing appointment with newscaster Edward R. Murrow, which he elected to keep, even though it was around midnight. It was the last meeting of the worst day of Roosevelt’s presidency. They—FDR, Murrow, and Donovan—ate sandwiches, and Roosevelt continued to lament the losses. “They caught our ships like lame ducks! Lame ducks, Bill,” and “They caught our planes on the ground, by God, on the ground!” Then he remarked to Donovan (as Donovan remembered it), “It’s a good thing you got me started on this . . . ,” meaning Donovan’s having pushed for centralization of intelligence in COI.28
When Donovan left the Oval Office very late that evening, he must have thought that he was in a uniquely promising position. Here he was, one of a handful of advisors the president had summoned when the news of Pearl Harbor broke. He was the last of them to be with the president that night, and his meeting closed with Roosevelt’s acknowledgment that Donovan’s ambitious plan for coordinating intelligence had merit. Beyond that, Donovan was the Coordinator of Information. That meant he had direct access to the president, even if he were required to coordinate some of his reporting with other agencies and departments. Donovan must have assumed he might well become the leader of U.S. intelligence.
It did not work out that way for three reasons: The first was that Donovan overplayed his hand; second, the Joint Chiefs of Staff, who became a very powerful force in wartime, chose to cut him down to size; and third, the president did not protect Donovan. In retrospect it is easy to suggest that Donovan was an ill-suited personality to be Coordinator of Information if that meant that providing information was his principal responsibility. He was a man of considerable ambition, likely aspiring to be president of the United States, certainly governor of New York. He was a man of action, as is any winner of the Congressional Medal of Honor. Simply providing information upon which others would make decisions was far too passive a role for someone with a nickname of “Wild Bill.”
Still, Donovan’s activist image was a big plus for COI. It helped attract highly qualified people who wanted to make a contribution to our role in a war that so many saw inevitably coming. Even college presidents, such as James Phinney Baxter of Williams, signed up for the Research and Analysis (R&A) branch of COI. R&A under Donovan played a very important role in the war and at the same time laid the foundation for the worldwide analytic ability the United States would need after World War II.
Donovan also started spying with people—gathering “human intelligence”—at a time when no one in the military or State Department would do it. During World War II, this was accomplished primarily by placing agents in neutral countries—the prime example being Allen Dulles, who served in Switzerland and later would become Eisenhower’s DCI. But Donovan’s heart lay in what we now call covert action. Covert action is making things happen without it being known the United States is the instigator; whereas intelligence is providing information. Donovan thought COI should focus not just on gathering intelligence but also on covert actions like propaganda, sabotage, assassination, political manipulation, psychological warfare, and guerrilla actions. His passion for action, as well as his attempts to influence policy-making and military strategy, carried him past where Roosevelt could support or even respect him.
One such instance concerned Baja California. Immediately following Pearl Harbor, rumors of Japanese infiltration into the West Coast flew to Washington on wings of near panic. Talk of five thousand Japanese moving from Central America and mainland Mexico to Baja California was among the rumors. Donovan’s R&A branch looked at these rumors and concluded that Mexico did not have the trains, boats, or roads to support such a large-scale move and that the barren desert of the region could not conceal a secret base. Furthermore, the Mexicans had been steadily pushing the few Japanese on the peninsula out of the fishing business, leaving any potential fifth column force landlocked.29 This was an early example of how work between geographers, sociologists, and area scholars could create a useful strategic evaluation. Donovan sent Roosevelt the report on December 15.
However, Donovan’s political objectives soon had him contradicting the conclusions of his own organization’s report, giving us one of the earliest examples of a chief of intelligence “cooking the books.” On December 21, Donovan wrote Roosevelt that there was a “strong possibility” that the Japanese had “hidden bases” for submarines and airplanes in Baja. The infiltrators perhaps numbered no more than one thousand, he averred, but he felt there were enough “to do considerable damage.” He asked the president for permission to recruit the local fishing unions “to ferret them out.”30 Why the change of tune? Donovan’s desire to undertake operations and get a piece of the action trumped the logic of the cooler heads who had written, just a week before, that there was not, and could not be, a Japanese threat from Baja. The very next day, Donovan sent his plans to create his own guerrilla warfare unit to the Oval Office. Then another memo from Donovan to FDR followed, recommending a no-holds-barred covert “intervention” in Baja.31
Luckily, the president did not approve Donovan’s plan. In fact, he was already moving in an entirely different direction—away from Donovan. The same day he received the recommendation to flood Baja with operatives, he met with J. Edgar Hoover to discuss wartime espionage operations in South and Central America. Hoover felt that Donovan was getting in his way. After meeting with Hoover, FDR sent out an unambiguous directive: All other organizations had to seek approval for any “intelligence work” north and south of the U.S. borders with Hoover’s FBI. Donovan supposedly didn’t learn of the order for over a week,32 but that didn’t stop him from sending the president a report a month later boasting that he had “arranged with the United Fruit Company officials to have all of their division heads cover the territory through Central America. . . .”33 Was Donovan assuming the president’s directive didn’t apply to him? Others in the government saw a man who had gone off the reservation. As Adolf Berle told the president, “Bill Donovan wants to take over the FBI work in South America. At least he does not say so, but he wants to put his own man in. He does not even say that, but you can never pin him down to saying what he really wants to do.”34
Hoover and Donovan were locked in combat for the entire war, with Hoover remaining on the offensive and Donovan meekly trying to chip away at the limits set by FDR. Hoover even went to the length of planting a woman in COI’s communications center. She supplied Hoover with verbatim communications regarding Donovan that could be embarrassing. Hoover, in turn, supplied these to the columnist Drew Pearson. Eventually the woman was caught and fired. Everyone, including FDR, knew who was behind this scheme.
A flurry of memos from Donovan—sometimes several in a day—were crossing FDR’s desk. Many of them contained questionable intelligence and recommendations. For example, Donovan advised the president to conduct a direct attack on Japan from the sea with ten to fifteen thousand troops. He passed on rumors that the Germans were preparing to invade Spain. In response to such advice, Roosevelt was always polite to Donovan on paper and in person, but within six months of Pearl Harbor, FDR’s enthusiasm for “Wild Bill’s” “innovative thinking” had evaporated. According to Adolf Berle, Roosevelt was considering making Donovan a brigadier general, “after which he was thinking of putting him on some nice, quiet, isolated island, where he could have a scrap with some Japs every morning before breakfast. That, he thought, would keep Donovan happy and out of trouble.”35 By this time, then, any possibility that a strong, centralized intelligence authority would evolve from COI was dead. In brief, it was a personal relationship with the president that brought Donovan onto the scene, but it was the deterioration of that relationship that severely limited Donovan’s effectiveness. As we will see, this relationship between presidents and their chiefs of intelligence can bear importantly on the effectiveness of our intelligence operations.
It is remarkable that COI survived in any form. J. Edgar Hoover, in addition to keeping COI out of Central and South America, had defeated Donovan on a number of other fronts. Moreover, the Joint Chiefs of Staff were ready to recommend doing away with COI as an intelligence agency when it was saved by one man, Lieutenant General Walter Bedell Smith (aka “Beetle” Smith). Smith, secretary of the JCS, persuaded the Joint Chiefs to absorb COI instead of dismembering it. His reasoning was that it would be better to have it under their control. Smith himself later became one of five of us who have served as director of central intelligence while on active military duty.
Donovan acceded to Smith’s proposal to operate under the Joint Chiefs because it was his only chance to survive. Fisher Howe says, “Selfd-efense—it was the least bad of the choices that he had available.”36 Shortly after Smith proposed this, FDR signed two executive orders. One changed the name of Donovan’s organization from COI to Office of Strategic Services (OSS). The other had OSS report to the Joint Chiefs of Staff. In addition, and much to Donovan’s discomfort, one of the orders stripped OSS of the propaganda, or “psychological warfare,” function. When told of this, according to one of his deputies, Donovan flew into “a state of fury such as I had never seen before, and was never to see again. It was controlled fury, but it was real fury, for he felt that the President had betrayed him.”37
Far more debilitating to the basic function of OSS was a decision by the president to continue excluding Donovan from the intelligence yielded by Allied code-breaking operations. Donovan complained, quite rightly, that exclusion would impair the ability of OSS to discharge its mission. His pleas fell on deaf ears. What was probably the best raw intelligence of the war remained out of his hands and thus eliminated any possibility of Donovan’s becoming a true coordinator of intelligence. Donovan personally, though, was probably much happier under the Joint Chiefs than under Roosevelt. Roosevelt’s directives were often ambiguous and vague, reflecting the push and pull of various subordinates. The military operated with tight definitions of who did what. That was good for Donovan, the mediocre manager, and good for his organization and the morale of his people. And Donovan rediscovered that he was at heart a military man.
It was in part because of this military ethos that Donovan managed to make his mark with OSS. His units in the field and behind enemy lines provided some real intelligence and, as the war wound down, some great war stories, which Donovan fed into the publicity machine. An army report concluded that the landing in Southern France was the “best briefed invasion in history” and “a single example of what can be done by an agency of this kind.”38 Operation Torch, the invasion of North Africa, began with OSS reconnaissance to facilitate the landing, and all manner of cloak-and-dagger derring-do involving insertions of agents carrying bars of gold from submarines via rubber rafts. The army, it should be noted, had no special forces of its own at the time.
The least storied part of OSS, and perhaps the most effective, was R&A. This branch was “so respectable and respectful, so conscientious and helpful,” concluded one historian, “that it inevitably acquired the aura of virtue.”39 R&A was the department that Donovan had focused on first while setting up COI, and he had turned not to the military for staff, but to academia. “One of the basic ideas behind it,” wrote Roger Hilsman, an OSS operative in Burma and later head of the State Department’s Bureau of Intelligence and Research, “was the novel and almost impish idea that scholars could in some respects take the place of spies.”40
During the war, however, Donovan paid much less attention to R&A, and it developed its own culture and mores. The analysts realized something Donovan did not: OSS did not have any real power or standing of its own. If R&A people went around making policy recommendations, inevitably someone would disagree with them. If that someone was in a position of power—say, a political appointee or, worse, a cabinet secretary—then that person would make it a personal mission to quash OSS. The only way to survive was to stay neutral. “There is . . . no future for R&A as a pressure group, no matter how strongly we believe we are ‘right,’ ” wrote an R&A man, Richard Hartshorne, to his colleagues. “On the contrary, to the extent that we do special pleading, we will very soon lose our entrée to all policy-makers other than those already committed to the same special causes.”41 So R&A decided that it would “not suggest, recommend, or in any way determine the strategy or the tactical decisions of the war.”42 Its realm was the facts. This was the genesis of an ethic of American intelligence that has been official policy ever since.
As early as September 1943, Donovan began looking to the longer run and the postwar role for himself and the OSS. He first discussed this with Beetle Smith. Donovan proposed that the Coordinator of Intelligence be a full and equal member of the Joint Chiefs of Staff. This idea was a nonstarter. There was no way the four-star members of the Joint Chiefs were going to accept as an equal this now one-star officer and former lawyer whose writ was limited to intelligence. In addition, back then the chiefs of the army and navy reported to the president through the secretaries of the army and navy. A chief of intelligence would presumably report directly to the president and might well become the most influential of the group, especially in peacetime. It is not surprising that this plan was quickly shelved, never to resurface. But it was evidence that Donovan was a very ambitious man, and that he was very interested in enhancing his own position. He took winning the war seriously, he cared about the OSS and the men and women in it, but, as one of his former subordinates points out, OSS “was a justification also for personal ambition.”43
Just under a year later, on August 4, 1944, Donovan made another attempt to ensure OSS would have a role after the war was over. He sent the JCS a very credible report developed by R&A on the looming difficulty of establishing spheres of influence between the Allies and the Soviets in postwar Europe. Against the advice of his analysts in R&A, he appended to it an appeal to the JCS to recognize the importance of having a permanent organization such as R&A to do long-range studies. In short, R&A should become a permanent fixture within the JCS. One month after they received Donovan’s suggestion, the Joint Chiefs slapped Donovan down, hard, with the following letter from General Marshall:
Activities of the Office of Strategic Services in Europe after the collapse of Germany should remain under the control of the United States Theater Commander, who can best determine in what manner and to what extent such activities can be employed most usefully in furtherance of our national policies. It is, therefore, deemed unnecessary for the Joint Chiefs of Staff to give consideration at this time to your memorandum of 4 August.44
Life under theater commanders would spell an end to the strategic, big-picture aspirations of Donovan for R&A. He realized that there could be no future under the military and began to long for a revival of his former relationship with Roosevelt.
Donovan’s next proposal for continuance came quickly, and it took him full circle—once again he wanted to be appointed by and report directly to the president of the United States. He developed a bold plan that removed the proposed intelligence organization from military constraints altogether and put it “on the policy or strategy level.” As such, the director was to be “appointed by the President, and be administered under Presidential direction.” The OSS was to be the foundation: “It is not necessary to create a new agency. The nucleus of such an organization already exists in the Office of Strategic Services.”45
In stark contrast to his approach in 1941, Donovan did not march into Roosevelt’s office with his plan. Instead, he engaged in weeks of quiet lobbying around the White House to win the approval of those close to the president. Donovan’s marketing of his plan at the White House, however, failed to sufficiently impress anyone who had significant access to FDR. The keeper of the gate, Harry Hopkins, was not impressed. The one true champion of the plan was Dr. Isadore Lubin, an economist in the White House Map Room, who on October 25, 1944, wrote to Roosevelt about it and attached a copy of the report. No one who has written on this period of intelligence history can resist quoting from this memo of Lubin’s to Roosevelt, because it starts with the line: “As you no doubt know, Bill Donovan’s Office of Strategic Services has been doing some swell work.”46
Lubin continued: “It occurred to me that there will be more room after the war for a service in the United States Government which would carry on some of the work now being done under Donovan’s auspices.” To think that Roosevelt was unaware of Donovan’s efforts and would actually believe that this idea had simply “occurred” to Lubin is amusing. Lubin even “borrowed” Donovan’s language verbatim in his letter to FDR: “The nucleus of such an organization already exists in the Office of Strategic Services. It has the trained personnel, the foreign contacts, the administrative organization and the operating experience.”47 Attached to the letter was Donovan’s plan. Roosevelt’s only recorded reaction was to send both Lubin’s letter and Donovan’s plan to Acting Secretary of State Edward Stettinius Jr.
By November 7, no action had been taken on the plan placed in Roosevelt’s hands by Lubin on October 25, but Donovan’s staff was hard at work on a new executive order for the president’s signature (even though he hadn’t asked for one). This arrived at the White House on November 18. Donovan now wanted to focus on “the problems of peace.” This meant that “intelligence control should be returned to the supervision of the President,” or, more specifically, it should be a “central authority reporting directly to you.”48 This proposal essentially preserved the functions of OSS, including espionage, but transferred them to the executive office of the president. In fact, the proposed agency was really just a much better defined version of COI, with spying and “subversive operations abroad” actually listed as functions in the executive order.49 The COI had actually not been given any directive in writing regarding such subversive activities, which both Donovan and Roosevelt thought best. If Roosevelt had put his pen to Donovan’s proposed executive order, he would have openly endorsed “subversive operations abroad” and made himself and future presidents more publicly accountable for the agency’s activities. It’s not surprising, then, that this forward-thinking president, with an uncanny sense of where the political minefields lay, did not sign this order.
Donovan simultaneously sent copies of his plan to General Marshall; Admiral Ernest J. King; the secretaries of state, war, and navy; and six other military and civilian leaders, consistently stating (falsely) that the president had asked for this plan.50 But instead of gathering opinions from these people at the top, Roosevelt did exactly what Donovan had been trying to avoid—on November 22, he kicked it down to the Joint Chiefs of Staff. From there, the sky darkened considerably for Donovan. The plan was passed down the line until it reached the Joint Intelligence Staff for evaluation and comment. Here, it met a group of military and naval intelligence officers with their knives sharpened.
A group of men met on December 22, 1944, to determine the fate of U.S. secret intelligence. They didn’t know it at the time, but these individuals, holding jobs several notches down from the real “decision-makers”—the cabinet secretaries and chiefs of staff—would provide the framework for the Central Intelligence Agency. They would also determine the organizational relationship between future presidents and future directors of central intelligence.
That relationship was the main point of contention. Donovan wanted a direct line, but by the time of the December 22 meeting, Donovan’s proposal had been pushed aside entirely in favor of two counterproposals—one backed by the military agencies, the other by the civilian members of the Joint Intelligence Staff. Neither of these tied the DCI directly to the president. The military couldn’t even accept the DCI’s being appointed by the president, as was proposed by the civilian staff members. One military participant, Major General Clayton Bissell of Army Intelligence, suggested that if the DCI were appointed by the president, the DCI “could go any time to the President and give the Secretaries a run-around, so that their control over him would be next to nil.” The result would be a man who could direct the intelligence sections of other agencies and departments. “Such power in one man is not in the best interest of a democratic government. I think it is in the best interests of a dictatorship. I think it would be excellent for Germany; but I don’t think it fits in with the democratic set-up we have in this country where you run things by checks and balances.”51 The issue Bissell identified, that of a chief of intelligence directing the intelligence sections of other agencies and departments, is still a key one being debated today.
Donovan’s representative on this group barely spoke to Bissell, who complained that an important man like himself shouldn’t even be wasting time in meetings like this while there was a war going on. But Max Ways, a member of the Joint Intelligence Staff and a self-described “mere civilian timid bureaucrat,”52 became the voice of moderation and reason. What he said is important because it was heard by three men who were silent during the meeting but who would, each in his own way, contribute to the shaping of the CIA—Captain Sidney Souers, USN, who would become Rear Admiral Souers and America’s first DCI; Ludwell Montague, who served with Max Ways on the Joint Intelligence Staff, and who would later become a CIA analyst and legend; and finally, James Lay, who would become executive secretary of the National Security Council (NSC), which dominated the CIA during its early years.
What these pioneers heard during this meeting, and what they would later apply to the CIA, was that the DCI had to be politically weak. “I think this man [the DCI] is a pretty feeble animal,” Max Ways argued. He compared the government to a jungle, “in which the smaller an animal is, in one sense, the more chance he has to survive. . . . We are trying to get [the DCI] down as far as we can get him, and give him as much protective coloration as we can.”53 The protection was to come in the form of presidential appointment and the director’s consequent prerogative of going directly to the chief executive when his or her intelligence evaluations ran counter to those of the cabinet secretaries. This setup, however, presumed a president who was interested in and responsive to intelligence. Mr. Ways’s message did not carry the day but deserves to be highlighted: The DCI is weak by design, and the only protection the DCI has comes from the president. And, yes, government is a jungle.
Donovan, discouraged and frustrated, wrote to Roosevelt, trying to put the decision back in his lap. He wistfully hoped for a return to 1941, when the president could create an intelligence agency with a stroke of the pen. The situation appeared to him to be lost unless the president intervened. In fact, the plans debated on December 22 were reconciled a few days later without presidential intervention. In the resulting compromise the DCI would report to a National Intelligence Authority (NIA)—a board made up of the secretaries of war, navy, and state plus a representative of the Joint Chiefs. Below the DCI, monitoring him on a working level, would be the heads of the military and civilian departmental intelligence agencies. So the DCI and his agency were to be stuck between policy-makers on the cabinet level and the heads of all the other intelligence agencies on the working level. This arrangement seemed designed to guarantee that the DCI would have his every move watched by a member of the military or the State Department.
What was even more abnormal was that the DCI would be hired and fired by the president only on recommendation of the National Intelligence Authority.54 The president couldn’t do it without the approval of the NIA. Just as he had to get the advice and consent of the Senate for his political and judicial appointments, the president would have had to get the permission of several of his own subordinates to hire and fire a DCI! So not only was the DCI placed in a weak position, but the plan reflected an attempt to weaken presidential power. Further, the idea of the NIA led directly to a postwar attempt to significantly limit presidential power in foreign affairs. This was how we got the National Security Council, which was originally envisioned as a body that would be a check and balance on the president. That is, while only the president could make decisions, the existence of the NSC as an advisory body made it necessary for presidents to tread cautiously if they decided to go contrary to the advice of the NSC.
The compromise plan for intelligence organization was sent up to the Joint Chiefs, where it gathered dust for months. Roosevelt, hearing about it from a displeased Donovan, didn’t express a strong preference for either Donovan’s plan or the compromise plan. He also didn’t seem to be in a hurry. He felt that “at the end of the war there simply must be a consolidation of Foreign Intelligence between State and War and Navy.”55 Roosevelt did not say it had to be achieved through the creation of an agency like the one proposed by Donovan. In fact, by saying that it had to happen “between” the three national security departments, he seemed to be asking for mere coordination between departments—effectively reverting back to the way things were before Donovan and COI came on the scene. Further, he emphasized, “I think it should be limited to military and related subjects.”56 Donovan didn’t want to be limited to military subjects—his aspirations were to have a hand in foreign policy, but he had failed to make his case either for immediate action or for direct responsibility to the president. Roosevelt left for the Yalta Conference without making a decision on Donovan’s plans and pleas.
On February 9, 1945, the secret compromise plan that had been sent to the Joint Chiefs of Staff was printed in the Chicago Tribune, thanks to a leak to a reporter named Walter Trohan, who was hostile to Roosevelt. The article made it look like preparations for an American Gestapo. Two prominent congressmen denounced the plan. Donovan was furious and helpless. He started an investigation into the source of the leak, but the list of potential suspects was far too long. His own people determined that the reporter had connections to the Department of the Navy, but Donovan was inclined to believe that it had been J. Edgar Hoover, who wanted to see his own South American spy operations expand globally after the war.57 Some later historians suggested that it could have come from the Map Room in the White House.58 The Tribune reporter, interviewed years later, stated that the president’s secretary contacted him, claiming that Roosevelt “wanted the story out.”59
While the source of the leak remains uncertain, what became obvious was Donovan’s isolation. No one in the executive branch stepped up to defend the compromise plan. Donovan’s appeals to the president claimed that “the disclosure was no mere leak but a deliberate plan to sabotage any attempt at reorganization of this government’s intelligence services.” The sabotage was focused, he felt, on his relationship to the president: “You will note that the strong effort in the revised plan is to avoid the direct reporting of the Director of the Intelligence Agency to you.”60 On his return from Yalta on March 1, at a press conference Roosevelt dodged questions about the plan, failing to either condemn or endorse it. To some, he appeared to have given the plan only cursory consideration.
At this point, Lubin pleaded with Roosevelt to reopen the issue. He prepared a memo, from Roosevelt to Donovan, for Roosevelt’s signature that authorized the OSS director to “call together the chiefs of the foreign intelligence and internal security units in the various executive agencies, so that a consensus of opinion can be reached.” The memo mentioned that all ten executive departments had “a direct interest in the proposed venture.”61 When Roosevelt signed this, it was the last known word from him on the subject, and they weren’t his words—they had been prepared for him by Lubin (who had already shown a willingness to feed FDR words written by Donovan).
On April 12, Secretary of the Navy James V. Forrestal, Attorney General Francis Biddle, and Undersecretary of War Robert P. Patterson met with Secretary of State Edward Stettinius in his office to decide what to do with Donovan’s latest approach. Stettinius left the group to respond to an urgent summons to the White House next door. He returned with the news that Roosevelt had died at 3:35 P.M.62
Donovan was in Paris when a messenger arrived at his hotel room with the news. His surprised aides watched him rush out of the bathroom in his shorts and undershirt, shaving cream still on his face, demanding a line to Washington. It may have been an indication of how Donovan felt about FDR’s failure to support him that Donovan did not return for the funeral. He said there was too much to do in Europe.
Donovan would soon begin to realize how much his relationship with FDR, in its better days, had mattered. He started putting in requests to meet with the new president, but they met in person only once, on May 14, 1945. Truman had his own very different plans for organizing intelligence, and Donovan was not part of them. Truman’s jaundiced view of Donovan was reinforced by the negative opinion of Donovan held by Budget Director Harold Smith and a devastating report from an army officer citing a wide range of alleged OSS misbehavior.
When all is said and done, centralized intelligence was truly Roosevelt’s creation. Only an extremely strong president could have overcome the passionate objections to any centralized intelligence agency, as did Roosevelt in creating the Coordinator of Information. At the same time, once COI existed, Roosevelt was, at best, halfhearted about supporting his creation and protecting it from its rivals and detractors. There are a number of possible explanations for Roosevelt’s unwillingness to have a strong director of intelligence operations:
• Roosevelt’s personal style was to rely on multiple sources, and to keep each source largely in the dark about what the others were doing (as he did with Donovan and the breaking of Japan’s codes).
• Roosevelt saw the American public as being more isolationist than interventionist, and still not prepared to sanction espionage and shady dealings.
• Roosevelt did not foresee the need for an espionage service during peacetime (because he did not foresee the Cold War).
• Roosevelt may, indeed, have worried that a secret intelligence organization could become an American Gestapo, one that could even become a rival power base to the presidency.
• There was strong opposition from the military (something that has never abated).
• There was strong opposition from J. Edgar Hoover—a man who had real power over several presidents.
Or, finally, Roosevelt just didn’t like the idea of giving up control. “There’s no accounting for presidents,” reflected James Dunn, assistant secretary of state under Roosevelt. “They all become a little mad and because of the office’s constraints and frustration, they tend to look on intelligence, or whatever, as their own private model railroad which they can run as they personally and petulantly see fit.”63 Whether he intended it or not, though, Roosevelt had started this train rolling and had given it enough momentum to ensure that some form of centralized intelligence would emerge.
Responsibility for deciding just what shape that would take had passed to Harry S Truman. Roosevelt had notoriously kept his vice president uninformed and uninvolved. It was logical to think that Truman would want some time to get up to speed before making major decisions on the structure of U.S. intelligence. That was not to be the case.