Indignant about the larger bureaucratic turf wars, the Director of the Federal Bureau of Investigation (FBI) leaked a memo regarding centralized intelligence from the Director of the Office of Strategic Services (OSS) in 1944. Subsequent newspapers reported this plan to be creating an “American Gestapo” and a “super-spy system,” thereby forcing President Franklin D. Roosevelt (FDR) to table the issue and thus delaying any meaningful progress in centralized intelligence by an entire year. Such manifestations of numerous interagency rivalries resulted from an ad hoc (i.e., when necessary) approach to U.S. intelligence activities during the twentieth century.
The postwar debate on central intelligence most fully realized the preexisting turf wars and competitions of command present during World War II. The bloating of the wartime bureaucracy resulted in an OSS that duplicated many duties of the preexisting intelligence agencies. This bloating and duplication worsened interagency turf wars and resulted in a bitter jurisdictional battle. The sudden death of President Roosevelt propelled a woefully unprepared Harry S. Truman into the White House to address these bureaucratic faults. In considering postwar American intelligence, President Truman admitted: “the war taught us a lesson – that we had to collect intelligence in a manner that would make the information available where it was needed and when it was wanted.”1 Rather than personally leading the effort, Truman tasked the heads of the State, War, Navy, and Justice Departments to address these rivalries and produce plans for a peacetime central intelligence organization. However, relevant scholarship neglects the institutional development of intelligence organizations and the roles that bureaucratic rivalries and governmental factors played in the decision to develop the postwar national security state. In filling this gap, this research will demonstrate the significance of how interagency dynamics directly influenced the capability and efficiency of the national security apparatus.
This paper will discuss the extent to which bureaucratic issues and interagency rivalries influenced President Truman’s handling of foreign intelligence at the dawn of the Cold War. Starting with a consideration of the prewar condition, these sustained rivalries were identified throughout the postwar debate, eventually culminating in the National Security Act of 1947. This paper traces and examines the core themes of interagency conflicts among policymakers’ personalities, departmental jurisdiction, and prior institutional experience. Relevant scholarship accepts that Truman’s decision to reorganize intelligence purely aimed to counter the growing threat of Communism. By contrast, this paper argues that these interagency rivalries, in addition to the threat of Communism, heavily influenced Truman’s decision to reorganize American foreign intelligence.
The massive overhaul of the American intelligence apparatus in the late-1940s had deep roots in the nineteenth and twentieth centuries. Throughout its history, the United States government long invested in its intelligence capabilities, resulting in the fermentation of institutional rivalries long before the National Security Act of 1947. The country adopted a piecemeal approach to intelligence which for decades challenged the coordination and efficacy of its operations.
During the first eighty years of the United States’ existence, for example, there was only nominal progress in intelligence capability and organization, largely owing to public distrust of standing armies and a palpable desire to maintain its distance from European affairs. Even so, the Continental Congress created the Committee for Secret Correspondence during the Revolutionary War that appealed to European countries for their support in the American cause by covertly communicating with agents overseas.2 But with no sustained or centralized intelligence effort, Congress and General George Washington preferred to approach intelligence in an ad hoc manner, in which intelligence collection occurred in reaction and when deemed necessary. President Washington, however, succeeded in securing legislation – the Secret Service Fund of 1790 – that created an executive fund for intelligence activities without the requirement that the president must inform Congress of the monetary allocation.3 This law confirmed presidential control over intelligence. Widely recognized as the first official intelligence agency, beginning in 1863 during the Civil War, the Union’s Bureau of Military Information (BMI) collected reports from “human intelligence, observation balloons, flag signal intercepts, and cavalry reconnaissance” as the first attempt at “all-source intelligence.” However, in 1861 the State Department (DOS) employed unofficial counterintelligence operations because, since the establishment of the Federal Government, the DOS had become “somewhat of a catchall for duties no other executive agency was designed to handle.”4 In other words, the DOS expanded its operations and responsibilities to fill any void that existed within the duties of the government.
The Civil War helped the United States to appreciate the importance of intelligence collection and propaganda warfare. However, the early intelligence structure operated in a complex and stumbling manner. As the Navy and Army professionalized, the military established two permanent intelligence organizations: The Office of Naval Intelligence (ONI) in 1882 and the Army’s Military Information Division (MID) in 1885.5 These units focused their efforts on acquiring information about foreign military capabilities and technologies. Despite a severe lack of funding in the first few decades, government officials familiarized themselves with peacetime military intelligence.
As the United States engaged more deeply in world affairs in the early twentieth century, American intelligence fell into the jurisdiction of several competing groups. As the sole reserve for trained detectives, the Secret Service under the Treasury Department acted as the only civilian intelligence agency.6 The Justice Department (DOJ) established a Bureau of Investigation in 1908, partially under the jurisdiction of the DOS. Although initially the Bureau relied on agents of the Secret Service, it soon professionalized and expanded its domestic counterintelligence operations into South America, thereby coming at odds with the MID and ONI.7 President Woodrow Wilson authorized the State Department to coordinate intelligence, thereby introducing a new institution into the mix.8 As World War I began, the War Department revived military intelligence as General John J. Pershing established intelligence functions throughout American Expeditionary Forces, thereby linking intelligence more firmly to fundamental military functions. As a result, several entities such as the Secret Service, Bureau of Investigation, MID, ONI, and the State Department could lay a legitimate claim to coordinating the nation’s intelligence.
During the Interwar Period, allocated resources rapidly decreased, but the ONI, MID, State Department, and FBI remained committed to intelligence. Specifically, the FBI massively expanded its mission and workforce as director J. Edgar Hoover ensured a permanent role for the agency as global tensions rose once again in the 1930s.9 Under Hoover the FBI actively worked against Nazi efforts to exploit the Germany’s efforts in South America to expand Nazi interests throughout the hemisphere. Due to the premises of the Monroe Doctrine that promised to protect the Western Hemisphere from European entanglements, Hoover sought an ever-increasing role for the FBI in hemispheric counterintelligence.
The early balkanized approach to intelligence contributed to policymakers viewing overseas operations in departmental terms. These rivalries induced very little coordination, uniformity of procedures, and sharing of information. This failure of synchronization also prevented clear dissemination of information to flow to the appropriate policymakers on the eve of World War II. During this time, President Roosevelt relied heavily on trusted friends who shared his zeal to bureaucratize and centralize America’s international activities. William J. Donovan, for example, became a determined proponent of a centralized, civilian intelligence apparatus.10 In July 1941, as he drew the nation more deeply into the war against the Axis powers, FDR appointed Donovan to serve as the Coordinator of Information (COI), a non-military intelligence organization that collected and analyzed information as requested by the other agencies. The loose charter, however, specifically barred the COI from interfering with military advisers, with Hoover’s FBI guaranteeing that its influence in South America would not be jeopardized.11 With these restrictions, the COI could not function as Roosevelt and Donovan had planned. Despite its limitations, this small step toward the coordination of American intelligence at the beginning of World War II initiated years of intelligence reform.
Although Roosevelt recognized the importance of increased coordination of intelligence, he took a hands-off approach to the creation of the OSS. The massive and sudden need for warfighter intelligence prompted FDR to place the COI under the JCS, thereby creating the OSS on June 13, 1942.12 The order to create the OSS provided no clear direction regarding, from a bureaucratic perspective, what or how the OSS should carry out its purpose. As a result, the JCS determined the range of OSS activities and the procedures to regulate them. With the JCS being executive military leadership, the Chiefs were not friendly to the creation of the OSS and thus took eight months to execute the task.13 This decision owes largely to the loyalty the JCS had to their departments (e.g., MID and ONI), thereby exemplifying the cause for such issues as a matter of power/jurisdiction and the justification of previous experience. Procedural rules, unclear authority, and supervisory committees thereby ensnared the OSS from its inception.
The JCS remained skeptical of the OSS’ abilities, while the FBI, ONI, and MID actively competed for power and jurisdiction, which ensured criticism surrounding the OSS. The Navy’s decision to bar the OSS from operating in the South Pacific theater and the FBI’s decision to block the OSS from South America are examples of such limitations.14 Simultaneously, the MID formed an operating arm, the Military Intelligence Service, to complement the Army’s analytical wing. A key factor in the decision to reorganize intelligence after the war, this duplication of duties energized wartime interagency rivalries.
The OSS strategically expanded into any avenue possible as FDR’s unclear direction enabled the complicated regulations of the JCS and enlarged the impact of Donovan’s personality. The highly complex nature of the OSS structure and duties became a direct output. As the OSS assessment staff reflected on the selection of OSS recruits for various positions within the agency, they wrote, “OSS undertook and carried out more different types of enterprises calling for more varied skills than any other single organization of its size in the history of our country.”15 Despite the small size of the OSS, the organization handled such a vast array of responsibilities. At its height in late 1944, the organization totaled just 13,000 men and women.16 Personnel from the Army made up two-thirds of its strength, with the remainder consisting of Navy officers and civilian employees. Despite the wide array of responsibilities, the OSS’ $135 million in spending over its four-year life equaled just $1.1 billion in today’s dollars.17 Despite the slight membership and budget, the range of the OSS’ duties succeeded in developing United States intelligence capabilities while simultaneously worsening the interagency rivalries as its duties overlapped with that of the FBI, ONI, and MID.
However, Donovan’s dream proved to be far from perfect. The ironically complicated structure plagued an organization created to remedy the complexities of bureaucratic intelligence. William J. Donovan, director of the OSS, then headed the permanently changing structure of an intelligence agency whose structure is most effectively illustrated using a complex flowchart.18 The organization was composed of branches, sub-branches, divisions, sub-divisions, sections, and sub-sections, all of which were subordinate partly to the director and partly to one of the deputy directors. These two deputy directors represent the dual purpose of the OSS: “Intelligence Services” and “Strategic Services Operations.”19 The former purpose refers to the techniques and functions of intelligence collection, collation, and analysis, while the latter purpose refers to the clandestine and espionage operations both to feed the former raw information and to act on its intelligence output. With such complexity, the OSS duplicated many intelligence activities, which only worsened the jurisdictional turf wars plaguing United States intelligence.
Although most of the OSS’ subdivisions duplicated certain operational aspects of the other agencies, the core branches of the OSS relevant to duplication and rivalry consisted of the Counterespionage branch (X-2), Research and Analysis (R&A), and Foreign Nationalities Branch (FNB). The X-2 directly challenged Hoover’s FBI as the Counterespionage branch aimed to collect and disseminate intelligence to Allied agencies regarding infiltration by enemy spies, working to carry out diversionary measures and send double agents into the field. However, the FBI held core expertise in counterintelligence, thereby demonstrating a case of rivalries in both prior experience and jurisdiction. FDR’s decision in November 1938 recognized this expertise as he sought to provide the FBI with extensive funding for expanded counterespionage functions, which dwarfed the military’s appropriation.20 This move infuriated the MID and ONI as Hoover’s FBI encroached on their historical duty of foreign counterintelligence, which in turn ensured there would be turmoil concerning the OSS’ X-2. The OSS’ R&A exclusively utilized Ivy League academics and professionals, which became a trademark of the OSS.21 To various extents, the MID, ONI, FBI, and DOS already pursued research and analysis with military men, diplomats, or civilian law enforcement professionals. However, the R&A distinctively utilized Ivy League academics to analyze foreign conditions. Nevertheless, the DOS argued a breach of its foreign relations jurisdiction.22 Further, the OSS’ FNB conducted the analysis of foreign news through “foreign nationality groups” living in the United States. The chief of the COI’s propaganda division planning board, of which a portion of its duties was transferred to the OSS’ FNB, admitted, “The State Department woke up with a start to the realization that one of Donovan’s impertinent little offices was making policy for it.”23 With this, Donovan’s “little offices” had worsened the interagency rivalries that preceded the War and would unknowingly contribute to its own demise. These groups became the core transgressions of OSS operational jurisdiction as they conflicted with the FBI-ONI-MID Delimitations Agreement of 9 February 1942.
The effectual manifestation of these interagency rivalries can be understood in a microcosm. For example, in late January of 1942, Donovan’s officers secretly infiltrated the Spanish Embassy in Washington to collect code books and other documents from Spain’s pro-Axis government.24 Angered and believing this was an invasion of the bureau’s operational territory, in conjunction with his personal rivalry with Donovan, Hoover ordered his agents to turn on their sirens and reveal the COI operatives in the act, thereby sacrificing the benefit of their actions to protect departmental jurisdiction. Hoover even argued the FBI’s prior experience and capability meant that such operations should be conducted only by his agents. This case provides a microcosm to understand the manifestations of these rivalries as the war emerged and continued.
During the war, the continuation of these rivalries led to numerous failures by the immature United States intelligence community. A gleaming example of such lies in the case of 1944, in which the OSS had worked closely with the Russian NKVD (Moscow’s intelligence and secret police agency).25 Donovan had organized the first American penetration of the NKVD in an exchange program; however, FDR tabled the agreement after Hoover’s outcry.26 There are countless such examples of the deterioration of capability at the expense of departmental relevance, ranging from the DOS objection of covert OSS passports to the remarks by FBI officials (names redacted) that Roosevelt would be “very receptive to a suggestion that Mr. Hoover take over OSS.”27 The sacrifice of efficiency and capability in the name of increased agency jurisdiction and competition of command had slowly become a transparent fault that largely characterized wartime intelligence and highlighted the need for a transformation.
When he surveyed the voluminous scholarship on the national security state, historian John Prados concluded that “there is a need for study of the ‘Washington command post” of intelligence, adding that the topic exists as a poorly understood but important aspect of modern American foreign relations.28 Writing in 2013, many years after the leading relevant scholarship, Prados’ recognition of the void in “command post” scholarship suggests an opportunity relevant to modern issues. This paper aims to help fill the void by building on previous works that emphasized strategic and ideological threats and domestic politics as the driving forces in developing postwar intelligence.
Melvyn Leffler leads the broadly researched “strategy to counter communism” argument within the historiography of the creation of the national security state. Writing in 1992, Leffler argues that, after the defeat of Germany in 1945 and the Soviet occupation of Eastern Europe, the Truman administration acted on the threat of another totalitarian enemy expanding beyond its borders. Leffler believes this military threat, combined with the ideological threat of communism, greatly influenced the reorganization of the national security structure. Referring to the Truman administration’s fear of war with communism, Leffler writes, “this contingency was improbable but not impossible,” adding, “narrow strategic calculations, therefore, reinforced the geopolitical, economic, and ideological factors that dictated America’s global concerns.”29 Leffler insists that “the institutional rivalries should not be exaggerated.”30 Writing in 2002, Arnold Offner agrees but places more emphasis on Truman as a “parochial nationalist” too quick to transcend the Soviet threat with American power.31 Although convincing, Leffler and Offner give little weight to the importance of the simmering bureaucratic rivalries that boiled over in the postwar period and shaped the modern intelligence bureaucracy.
Michael Hogan, by contrast, posits that the creation of the national security state alone cannot be explained by the communist threat. Writing in 1998, Hogan gives more attention to the role of the institutions in this decision as the United States accepted new superpower responsibilities without becoming a “garrison state,” thereby referencing a popular Cold War concern that the American economy would become dominated by military spending, coinciding with an erosion of civil liberties. Hogan, therefore, argues that the creation of the national security state involved “more than the country’s chances for success in the battle against communism,” such as addressing the restrictions caused by the complexities of the wartime intelligence bureaucracy.32 He cites President Truman’s December 19, 1945 message to Congress, which urged that the time had come “to discard obsolete organizational forms and to provide for the future the soundest, the most effective and the most economical kind of structure.”33 Hogan argues that the Departments of the Army, Navy, and Justice had incentives to exploit public and Congressional fears of communist expansion to justify expanded budgets.34 Hogan, in this way, differs from Leffler’s emphasis on the Truman administration’s communist overreaction by weighing the influence of interagency rivalries. Yet, he gives greater credence to Red Scare politics rather than viewing that discourse as a symptom of the defense and expansion of bureaucratic bases of power. While accepting the importance of strategic and ideological factors, this work takes a deeper look at people and institutions as well as a longer view of the development of intelligence in the first half of the twentieth century.
Pre-War Rivalries and the Competition of Command
The explanations for the rivalries that persisted before, during, and after the war can be understood from a few perspectives. The symptoms of these rivalries should be understood as the overlap of duties coupled with a lack of coordination. With this, the root causes of such conflicts were based on multiple factors that will be traced throughout the paper: individual personality, departmental power and jurisdiction, and prior experience and capability. Individual personality can be understood as in reference to the personalities of the respective agencies’ leaders. This factor will be most clearly shown through the tumultuous dynamic of Donovan and Hoover. Secondly, departmental power and jurisdiction also caused such rivalries, as each department aspired to increase power and relevance through increased jurisdiction. Lastly, to rationalize their behavior, each department and respective leaders lobbied under the belief that their agency had prior experience and/or proved more capable of effectively carrying out intelligence duties. The themes of personality, power and jurisdiction, and experience and capability will thus be shown to be present throughout the reorganization process.
The rivalry of personality between Donovan and Hoover was perhaps the most important and influential interpersonal relationship that worsened interagency rivalries throughout the war. The roots of this rivalry can be traced back to 1924, when Donovan took charge of the criminal division of the DOJ. J. Edgar Hoover, the acting director of the FBI, became subordinate to Donovan. Public perception had been fixated on Donovan’s successes, rather than on Hoover’s FBI. Donovan’s reputation grew as the “driving force in the [DOJ]”, as mentioned in a Daily Mirror of Washington article published in April of 1925.35 Furthermore, throughout Donovan’s time at the DOJ, Donovan criticized Hoover for wiretapping of private citizens which resulted in his recommendation to fire Hoover.36 This led to a long-lasting quarrel between the two leaders, which got reignited with the conception of the COI and deepened with the creation of the OSS.
With their history of personality clashes, Donovan’s COI and OSS entered the realm of United States intelligence just as the Army, Navy, and FBI had ineffectively attempted to address their own rivalries. A June 1939 Roosevelt directive intended to extinguish their rivalries and provide some clarity, stating Roosevelt’s desire that, “all espionage, counter-espionage, and sabotage matters be controlled and handled by the [FBI], [MID], and the [ONI] … the Directors of these three agencies are to function as a committee to coordinate their activities.”37 As a result of this directive, the departments bore responsibility for resolving their own turf wars. Therefore, the heads of the FBI, MID, and ONI formed the Interdepartmental Intelligence Committee (IIC). Although the IIC successfully created the Delimitations Agreement of June 1940, which very simply gave the FBI a role in civilian intelligence while the ONI and MID received military intelligence, the extremely vague agreement contained catch-all phrases that plagued the document.38 As expected, the IIC failed as its members became wary of sharing information, which resulted in the neglect of the IIC meetings in 1940. The IIC barely improved the Agreement in 1942,39 in that it addressed the jurisdiction of the collection; however, the document failed to address the coordination and dissemination of such information.40 Ambiguity and rivalry, therefore, plagued United States intelligence simultaneous to its vital importance to the war effort.
Established in 1940, the FBI’s Special Intelligence Service (SIS) significantly heightened these competitions. In the years prior to World War II, Hoover’s FBI expanded its funding and addressed the threats posed by Hitler’s agents in Latin America, further legitimizing the FBI’s control over counterespionage. In June 1939, FDR directed his cabinet members that the FBI would “take charge of investigative work in matters relating to espionage, sabotage, and violations of neutrality regulations.”41 On June 24, 1940, FDR issued a directive that formally gave the FBI responsibility for foreign intelligence in Latin America.42 As a result, Hoover quickly established the SIS, the new counterintelligence unit of the FBI. The SIS marks the first civilian foreign-intelligence organization in United States history as it worked in conjunction with the State Department.43 Thereafter, Hoover’s FBI gained the ability to lobby for increased departmental power and jurisdiction in the postwar debate based on the justification of prior experience and capability through the SIS.
However, the ONI and MID had long controlled and withheld the belief that the military should retain sole jurisdiction to lead in the collection of American foreign intelligence. In a memorandum to the head of the MID in August of 1940, Hoover addressed these concerns by promising the SIS focused its efforts on Central and South America.44 The FBI operated in those regions for years prior and used this experience as an argument for FBI prevalence abroad. The SIS thereby legitimately operated undercover FBI agents and legal attachés in the Western Hemisphere. Despite this relatively successful delimitation of jurisdiction, these department executives and organizations here acknowledged the existing rivalries. This specific delimitation proved only the “tip of the iceberg” of the turf wars.
Soon thereafter, the energetic and determined future Director of COI and OSS, “Wild Bill” Donovan, blazed a path for centralized intelligence amid the turf wars. With the war in Europe serving as a catalyst for preparation, the increasingly emergent need for centralized intelligence quickly became clear to FDR’s administration. As a result, Donovan submitted to FDR a memorandum on June 10, 1941, urging FDR to establish centralized intelligence. Donovan wrote: “These [Military Intelligence] services cannot … obtain that accurate, comprehensive, long-range information … All departments should have the same information.”45 In reference to these detriments caused by rivalry, he argued that the central intelligence agency should be directly responsible to the president and works alongside an advisory panel consisting of representatives of the FBI, ONI, and MID.46 This agency, directly responsible to the president and alongside an advisory panel, increased the efficiency of United States intelligence in the upcoming war effort.
When Donovan became a vocal and adamant proponent of centralized intelligence just before the United States entered the war, Hoover postured immediately against it. A well-known feature of their rivalry, Hoover held animosity toward Donovan and his plan for intelligence. R. Harris Smith, one of the scholars that argues the institutional factors of intelligence were equally significant to that of the emerging threat of communism, writes that the animosity became clear to British intelligence that “Hoover keenly resented Donovan’s organization from the moment it was established.”47 Hoover not only wanted to best his personal rival but also believed centralized intelligence should be the task of the Bureau. The Hoover-Donovan conflict resulted not only because of their clashing personalities, but also because of their struggle over departmental power/jurisdiction with arguments to prior experience and capability, citing FBI experience operating the SIS.
Hoover’s adversarial position became so transparent that Donovan aimed to address possible jurisdictional conflicts between the COI and FBI in a call to Hoover’s assistant. Referencing their June 1941 conversation, Hoover’s assistant Edward A. Tamm wrote to Hoover regarding Donovan’s “extremely anxious” comments, which suggests an acknowledgment by Donovan of their preexisting personality conflict. According to Tamm, Donovan “stated that the new agency was not in any way to interfere with the functions of the Bureau, ONI and military intelligence,” and recognized that these agencies “are afraid their jurisdictions will be usurped.”48 This quotation helps to understand the FBI’s anxiety over lost power/jurisdiction. Although Donovan broke his promise here during the war, one can sense Donovan’s recognition of the preexisting turf wars that the COI would further unsettle. This memorandum thereby recounts a conversation in which Donovan attempted to address the turf wars resulting from personality as well as from power and jurisdiction.
Nevertheless, the FBI did not welcome Donovan’s COI. On behalf of the FBI, Tamm responded to Donovan’s above statement by claiming “complete coordination” between the bureau and the military at all levels.49 The intelligence agencies refused to see enough of an issue with their current processes of coordination to justify the creation of the COI, which further challenged any solution. Perhaps the most notable came in the form of one of Donovan’s last statements in the same conversation, in which he “reiterated that he did not seek this job [FBI Director] and did not want it”, which further shows the clash of personality and power/jurisdiction.50 In this way, Donovan’s relationship with the FBI can be seen as one of duality. That is to say that Donovan challenged not only the power and jurisdiction of the FBI but also the personality of Hoover as director and as Donovan’s former subordinate. No matter the cause, the result proved the same: bureaucratic rivalries worsened on the eve of the American entry into World War II.
With the creation of the COI, Donovan must now calm the tensions that his office had worsened. FDR created the COI just a couple of weeks after the above June 1941 conversation, which formally submitted the COI to the wrath of its rivals. Weeks after the attack on Pearl Harbor, in a January 1942 memorandum to Attorney General (AG) Biddle, Donovan addressed the AG’s and FBI’s worries over the COI’s cooperation with the State Department in “this letter of explanation,” to calm the tensions. He argues that the COI cooperation with the State Department as “for the better integration of our work in all countries.”51 Although not stated, the attorney general’s confusion over COI’s cooperation with DOS likely stems from an insecurity that fears the COI will replace the FBI’s recently accepted role as legal attaches in United States embassies abroad. In a similar vein, the Joint Chiefs of Staff (JCS) distrusted Donovan and his new COI. In a clear consolidation of departmental power and jurisdiction, the JCS quickly adopted the COI under its control just months after its creation.52 The COI and subsequent OSS were, of course, submitted to a host of regulations and a slim budget in its early years.53 This general nature of distrust and infighting continued to plague the wartime intelligence community throughout the war.
Rivalries Engulf the Debate on Post-War Intelligence
As the first to do so, Donovan seriously considered peacetime central intelligence in late 1944. The scope of Donovan’s plan for centralized intelligence transcended wartime activities as he lobbied for the creation of a centralized, peacetime intelligence apparatus. On November 18, 1944, at the request of FDR, Donovan sent a memorandum known later as the “Donovan Plan” to the White House outlining his plan for post-war intelligence. He argues that peacetime intelligence will require that “intelligence control be returned to the supervision of the President” and the “establishment of a central authority reporting directly to you” would be necessary.54 However, Donovan sent this memo not only to the president, but also to eleven senior officials in the State, War, Justice, and Navy Departments.55 At this point, Hoover became fully aware of Donovan’s post-war plans.
In a breach of national security, Hoover leaked the memo to the press, which subsequently and dramatically labeled Donovan’s plan as a “super-spy system” that would be an “all-powerful intelligence service to spy on the post-war world and to pry into the lives of citizens at home.”56 With fears of fascism omnipresent in the context of World War II, newspapers flooded February of 1945 with reports that this would be an “American Gestapo” and a “super-spy system” that faced opposition derived from “bureaucratic jealousies.”57 So vibrant were the rivalries that even newspapers recognized the turf wars that bogged down American intelligence. A congressional uproar followed this leak alongside much public suspicion, as seen in the newspapers, about the creation of a peacetime intelligence service. Within this context, Roosevelt decided to table the issue, despite his favor for centralized peacetime intelligence, thereby handing Hoover the victory he desired.
However, FDR attempted to restart the debate over central intelligence as his health greatly diminished. One week before his death, Roosevelt wrote to Donovan that he believed various executive departments held an interest in centralized intelligence and that, “They should all be asked to contribute their suggestions to the proposed centralized intelligence service.”58 Indeed, the contribution of countless suggestions occurred. In September of 1945, newspapers reported on the heightened rivalries, such as in the Sunday Star’s “America’s Postwar Espionage Control Stirs Savage Fight.” The author, Wallace R. Deuel, correctly stated that the departments have a “large measure of agreement here that America needs [centralized intelligence]” but that “the real contest is over who should control it.”59 Although the departments recognized the necessity of centralized intelligence, the rivalries led the charge as each department wanted to control the central service.
Following the death of FDR in April of 1945, President Harry Truman led this reorganization effort with a predisposition to cutting bureaucratic waste. In his time as a senator, Truman led the “Truman Committee” (i.e., Special Committee to Investigate the National Defense Program) from 1941 through 1944, in which the senators investigated bureaucratic waste during wartime bloating.60 In this way, within the context of a clearly concluding war, Truman tended to pay attention to suspected duplication. Recounting his view of intelligence in his memoirs, Truman wrote that United States intelligence “first struck me as being badly organized when I was in the Senate… Our Senate committees … were often struck by the fact that different agencies of the government came up with different and conflicting facts on similar subjects.”61 Truman had experienced a front-seat view of the detriments caused by these rivalries; however, the Truman Committee did not expressly focus on intelligence. Moreover, FDR’s administration failed to advise Truman on the issue during his brief stint as vice president, thereby rendering his understanding of intelligence to a basic and distant level. As a result, Truman entered the scene ready to reorganize and reduce waste, yet without an adequate understanding of the complexities and proposed needs of central intelligence.
With the sudden transfer of power to Truman, Donovan quickly realized that he must appeal to the new president but soon discovered Truman’s skepticism regarding an addition to the bureaucracy. In August of 1945, Donovan forwarded a memorandum to Truman containing a statement of principles upon which centralized foreign intelligence should be established.62 However, Truman proved neither receptive to this plan nor to Donovan himself. In his memoirs, Truman’s White House Counsel Clark Clifford wrote, “for reasons I never fully understood, President Truman never liked Donovan.”63 Perhaps, this discontent originated from Truman’s exposure to the bloating produced, in part, by Donovan’s OSS. He continued that in September of 1945, “Truman prematurely, abruptly, and unwisely disbanded the OSS. He had been persuaded by a bitter critique from Army intelligence that was inspired by jealousy,” in that the Army argued Donovan’s central organization would simply continue the greatly hindered wartime OSS, which played to Truman’s disposition.64 Clifford’s statements here exemplify the role that rivalries between the departments played in the manipulation of an ill-equipped president.
Despite this, Truman slowly recognized the ferocity of these rivalries and their illustration of the need for intelligence reorganization. In his memoirs, Truman references a memorandum that “pointed out that a tug of war was going on among the FBI, [OSS], the Army and Navy Intelligence, and the State Department.” With this knowledge, Truman argued that he “considered it very important … to have a sound, well-organized intelligence system.”65 He accepted that “there had never been much attention paid to any centralized intelligence organization in our government,” which resulted in the “uncoordinated methods of obtaining information.”66 Even as a senator, Truman recognized that the rivalries resulted in significant inefficiencies and obscured the implementation of American intelligence. Once Truman became president, the department leaders presented him with a few plans for reorganization. He soon became aware of Donovan’s proposal and of the Navy’s counterproposal. Truman thus asked the DOS to draft their proposal for consideration, possibly due to Secretary of State James Byrnes longtime friendship with Truman.67 While Truman recognized inefficiencies caused by such rivalries, he chose to approach the problem by tasking the various Department executives with the formulation of a plan for central intelligence. This method put the turf wars on full display as it spawned an interagency debate for the next year.
However, with U.S. intelligence existing in a realm of uncertainty and with turf wars beginning to drive the postwar debate, Truman abruptly disbanded the wartime OSS. Executive Order 9621, signed September 20, 1945, was adequately titled, “Termination of the Office of Strategic Services and Disposition of its Functions.”68 EO 9621 thereby transferred the R&A branch of the OSS to the DOS, despite the fact that its own Under Secretary of State Dean Acheson realized the inexperience of the State in intelligence management in relation to that of the OSS.69 Further, the duties of the FNB became spread among the FBI and DOS. The War Department absorbed the remaining divisions, such as with the Special Intelligence and X-2 branches.70 As noted previously, the R&A, FNB, and X-2 became the most contentious branches of the OSS in the interagency turf wars because they directly and indirectly overlapped jurisdiction with the other agencies. The decision is fitting that a few sources of such conflict resulted in them being given to the departments that most challenged those operations, thereby demonstrating the catalyst to the reorganization that became of these rivalries.
The swift timing of EO 9621 came as a surprise to everyone except Truman. The OSS, Donovan, Bureau of the Budget, and the rest of the intelligence agencies had realized the incoming dissolution of the OSS early in Truman’s tenure, or at the very least the OSS’ direct transformation into a peacetime service. Just one week before the order, Donovan pleaded with Truman that these functions not “be severed and transferred to separate agencies,” urging “whatever agency has the duty of intelligence should have it as a complete whole.”71 Donovan understood he would lose the OSS but hoped its functions would remain centralized in a cemented structure. Truman initially believed that a quick-fix solution may suffice in that the agencies form a central council among themselves, rather than Truman fully opposing centralized intelligence. The Budget Bureau itself had drafted such plans in September of 1945; however, the bureau assumed that the termination would stretch over months, thereby expecting a deadline of January 1, 1946.72 The sudden decision defined a surprise not in function, but in haste. Signed September 20, 1945, EO 9621 provided a deadline of October 1, 1945 for the complete dissolution of the OSS. The abrupt nature of this decision drew criticism from his own administration, such as with the aforementioned comments by Clark Clifford, and has continued in historical scholarship since. Truman ordered the dissolution of the OSS at the time the various agencies only began to debate post-war intelligence.
With no surprise, Hoover ferociously seized the opportunity to appeal to a new president with the FBI plan for reorganization. In a memorandum eventually destined for the secretary of state and then Truman, in late August of 1945, Hoover proposed that the successful SIS program in the Western Hemisphere be expanded. He wrote, “I am not seeking for the [FBI] the responsibility for the world-wide intelligence system.” However, Hoover continued, “I do believe … we are well qualified to operate such a service … upon the same general basis as these operations have been carried on in the Western Hemisphere.”73 In other words, Hoover proposed that the “coordinated” intelligence program that had been operated through the State Department,74 by the FBI, and in conjunction with the ONI and MID, should be expanded to cover the entire scope of United States foreign intelligence. However, DOS and military intelligence each proposed conflicting plans.
Before the FBI could truly join the post-war debate on peacetime intelligence, Truman shut the plan down. A few weeks after Truman received Hoover’s memorandum, Tamm advised Hoover that “the President allegedly had expressed the opinion that the FBI should act only as a domestic agency.”75 Nevertheless, AG Clark soon sent a memorandum to Truman officially proposing the FBI plan. In it, a military attaché (MID), naval attaché (ONI), and legal attaché (FBI) would work through DOS channels to collect foreign intelligence controlled by a “committee to control basic policy” and an “operational committee,” both composed of the respective heads of the agencies.76 Despite the illusion that these agencies worked together on this proposal, these agencies remained divided.
In fact, Hoover expressed concern when Truman gave the State the purpose of leading the formation of this worldwide intelligence organization. “I am not at all optimistic as to the sympathy which Colonel McCormack [of DOS] may have toward our program.” He urged AG Clark “to bring to President Truman’s attention in the immediate future our views” of post-war intelligence.77 In another interior memorandum to Hoover in November of 1945, Truman’s position against the worldwide expansion of the FBI became clear as he preferred that the Bureau would “confine its efforts to the domestic field.”78 A meeting of the secretaries of State, War, and Navy just a few days later clarified the reasoning for the decision. According to the meeting minutes regarding comments of Assistant Secretary of War Robert A. Lovett, “there was a general agreement that intelligence should be divorced from police powers,” as the combination may result in a “gestapo” amid the postwar concern for domestic fascism.79 The FBI’s status as a law enforcement agency essentially removed itself from the post-war debate on foreign intelligence, eventually leading to Hoover’s decision to immediately end the SIS program with no transfer of information. The post-war debate then quickly became a duel between the DOS and JCS, taking place within the Lovett Committee,80 yet elements of the Donovan proposal still held strong.
Despite being modeled largely upon Donovan’s proposal of November 1944, the proposal from the JCS should be recognized as a counterproposal. The JCS agreed with Donovan in that they desired the “further coordination of intelligence activities” that “can be more efficiently conducted by a common agency” in the “synthesis of departmental intelligence on the strategic and national policy level.”81 The JCS model is closely related to Donovan’s vision in that the proposal argued for the establishment of a national intelligence authority and a Central Intelligence Agency. However, a strong deviance remained: the JCS worried that Donovan’s proposal, “would over-centralize the national intelligence service and place it at such a high level that it would control the operations of departmental intelligence agencies without responsibility … to the heads of the departments concerned.”82 As shown in this statement, departmental autonomy remained key in the JCS plan for post-war intelligence. Such was the reason the JCS made departmental autonomy a core pillar of their proposal that the creation of an overseeing committee, “would include representatives of the Secretaries of War and Navy.”83 The agencies worried that departmental representation would not be concretely established in this centralized intelligence authority to lobby for their own agency’s relevance.
The State Department’s proposal agreed with the views of the Budget Bureau, thereby forming a coalition to counter the proposal of the JCS. The proposition that the modern intelligence community would be responsible to and under the direction of the secretary of state became a core pillar of the DOS proposal. In this way, the State interpreted Truman’s letter, of the same day as EO 9621, which directed the DOS to “take the lead in developing a comprehensive and coordinated foreign intelligence program … through the creation of an Inter-departmental group, heading up under the State Department.”84 The DOS interpreted this phrase to mean that the State should lead the program’s development and operation; whereas the War and Navy Departments interpreted the phrase as a directive to simply “formulate plans for the approval of the President.”85 Nevertheless, the DOS saw this as a justification to lead and direct centralized intelligence as derived from rival departments.
Throughout the last months of 1945, the core controversial issues discussed in the opposing plans, therefore, addressed two major factors, with the first regarding the character of a coordinating organization. As previously mentioned, the wartime JCS had proposed to FDR a plan for reorganization that mirrored Donovan’s design but diverged by arguing that this central intelligence agency should be directed under the authority of a council consisting of the secretaries, with the executive having a nominal role on that council. The Central Intelligence Agency (CIA) would therefore be an independent coordinating agency under the oversight of a council of the Secretaries. However, the State’s proposal urged the creation of a National Intelligence Authority (NIA) with no independent central intelligence agency. Rather, the NIA would lead this coordination under the umbrella of the DOS. As Clark Clifford put it, the DOS would be “designated the sole conduit of intelligence reports to the President.”86 The consolidation of centralized intelligence under the distanced authority of the State Department can be further seen in their position on the character of the executive.
The second core issue regarded the character of the executive of centralized intelligence. Both camps accepted that a coordinating body consisting of the secretaries would oversee the executive of the centralizing organization. Therefore, this discussion centered around whether the new agency’s executive would be derived from any one department or completely independent, and thus whether they should be civilian or military. The State Department’s plan favored that the agency’s executive, “be named by the Secretary of State and that he be an employee of the State Department,” even suggesting the NIA Chairman be the Secretary himself.87 DOS believed this would permit the “coordinating machinery” to “get going without much delay.”88 The State, of course, preferred to have an executive derived from the Department, as to ensure the State remained active in intelligence. The State’s insistence to be placed atop this intelligence bureaucracy on the eve of the Cold War stems from its experience in peacetime foreign relations. In contrast, the War and Navy Departments’ plan argued that the CIA executive should be “named by the President upon the nomination of and responsible to the Secretaries of State, War and Navy.”89 Thus the military plan suggested the executive be an individual not derived from any of those core agencies. In both core problems, the influence of turf wars stemmed from conflicts over departmental jurisdiction, and prior experience seem to drive the reorganization.
The post-war debate over the reorganization of United States intelligence should therefore be understood as driven by preexisting interagency rivalries and turf wars, in addition to the looming Cold War. The two months leading up to Truman’s directive to establish the NIA and Central Intelligence Group (CIG) further demonstrate this perspective. In the meeting notes of the DOS committee on November 27, 1945, the officials were clearly driven by such worries. The secretary stated that when “selling” the plan to the president, the State must “avoid unnecessary questions of jurisdictional conflict with the Army and Navy … . If the military services present a plan which is not complicated, while our plan is complicated, the military plan will have a much better chance of being adopted.”90 There is a strong likelihood that this rationale stems from the common knowledge that Truman wanted to reduce waste and complexity. The discussion also included the recommendation that the NIA Chairman be the secretary with the admittance that the “War and Navy Department had much more experience with intelligence problems than the State Department.”91 Nevertheless, the DOS remained committed to their prevalence.
These disagreements rooted in rivalry quickly became quite clear to the president, his advisors, and the secretaries themselves. The Director of the Budget Bureau, in alignment with DOS, advised the president that these conflicts continue to hamper the process.92 Secretary of State Byrnes explicitly stated the function of their plan as “to reduce competition and duplication” between the existing departments.93 In a December 27, 1945 memorandum from the Navy to Clark M. Clifford, the Navy argued that the JCS plan for an independent executive would mean that “no one department could influence unduly the type of intelligence produced… . no single department would be dominant.”94 Finally, after many months of jurisdiction-based debates, the secretaries produced a memorandum for Truman that synthesized their proposals while ensuring each department retained certain responsibilities.
On January 7, 1946, the three secretaries submitted that proposal to Truman, only for it to be undercut by the Budget Bureau. The memo recommended the creation of an NIA composed of the secretaries. The NIA would thereby establish a CIA and director of the CIA that are directly responsible to and appointed by the president.95 Two days later, the Budget Bureau produced a memo that stated, “when a subject is left to three departments to divide up among themselves, the worst possible compromise results,” rather urging Truman to decide how he wants intelligence to be organized.96 Director Smith of the Budget Bureau further showed such discontent to the president by stating his “gloomy opinion” because the subject of intelligence proved “one of the most far-reaching problems of interdepartmental coordination that we currently face” and that their efforts would continue for multiple stages, concluding with “God bless bureaucracy!”97 Truman deeply considered and respected the comments of the Budget Bureau, likely due to his concern for bureaucratic waste in relation to the budget. The Budget Bureau, having analyzed the subject simultaneously to the debate, felt that the Bureau itself should continue the analysis not from the budgetary standpoint, but rather through the “intelligence aspect.”98 Addressing the Director of the Budget Bureau, Truman pushed back: “I know you have expert intelligence men in your office, but I like [the JCS] plan.”99 Nevertheless, this challenge undermined Truman’s confidence in the proposal.
Truman submitted the indecisive directive to develop centralized intelligence ambiguously. On January 22, 1946, the directive established a Central Intelligence Group (CIG) under a Director of Central Intelligence (DCI) to assist in coordination by the NIA.100 The directive presented the most neutral choice, as Truman’s decision straddled the two leading proposals. Rather than creating a new agency through the CIA, according to Clifford, Truman “set up a loose coordinating group [CIG]. He was not ready to create a new intelligence agency.”101 The CIG was, in fact, loose; The CIG got its entire staff and money from the State, War, and Navy departments.102 Despite Time magazine’s reference to this decision as the end of the United States’ “historical innocence” in foreign intelligence, the CIG became little more than the wartime COI. Its purposes were the ambiguous functions of planning, development, and coordination of foreign intelligence.103 The CIG became quickly enveloped by rivalries, which mirrored the lifecycle of the COI and OSS.
Due to its ambiguous language and indecisive foundation, Truman’s January 22 directive failed to safeguard the now-established CIG from rivalry and impotence. In fewer than six months, the interdepartmental rivalries prevented the Group from performing its mission.104 A wholly dependent and loose organization with no independent budget, personnel, statutory mandate, nor concise mission, unsurprisingly resulted in futility.
Hoover’s FBI became the most glaring example of the CIG’s challenges outside of the turf wars among the three departments. An intelligence vacuum quickly formed in Latin America following Truman’s decision to limit the bureau’s scope to the domestic sphere. The creation of a new centralized intelligence agency aimed to fill that void; however, the previous year of interagency debate produced only the measly CIG. The NIA thus had to scramble to address this void by allowing the FBI to continue its work in South America alongside the ONI and MID.105 By this time, however, Hoover’s FBI felt deeply frustrated and wanted no participation. Assistant Director Ladd stated in May 1946, “the most I will agree to now is to stay in the Western Hemisphere for 1 year… this is a project we must get out of … we would be carrying most of the work with none of the full authority to get it done properly.”106 Hoover’s FBI wanted no part in these operations, as they knew they would be subordinate to an NIA with no permanent FBI representation. Nevertheless, national security needs required that the FBI remain in theater long enough for the CIG to materialize and be capable of receiving the transfer of information and responsibility from the Bureau.107 In response to Hoover’s decision to withdraw the FBI in one week, rather than in one year as previously agreed, Truman and the NIA wrote to the AG that Hoover’s plan to withdraw was unacceptable.108 Despite the requirement to retain FBI presence in Latin America, Hoover immediately slashed its presence to a fraction of their previous contingent. In the process, Hoover ordered his agents to destroy their intelligence files, rather than bestow them to their CIG rivals.109 Not powerful enough to withstand such pressures, the CIG thus drew uncertainty regarding its future.
These turf wars engulfed the group and quickly rendered the organization futile. When Clifford asked whether the concept of the CIG laid out in the executive order truly works in practice, top officials of the CIG showed their frustration. “We do not have the power to hire anybody, fire anybody, or assign anybody … We are nothing but a stepchild of the three departments we are supposed to coordinate.”110 Internal documents from the CIG corroborate this feeling. With the post-war budget cuts, the “parent” departments lost resources. As a result of this, with the insistence that the Departments’ intelligence remain sovereign, this ensured that their “stepchild” would be neglected. The “impending dissipation of departmental intelligence resources” would render the CIG and NIA “unable to effectively discharge its responsibilities.”111 This dire situation owes to the fact that the CIG remained completely dependent on those departments for funding. Referring to this same issue, Acheson admits, “no committee can govern and no man can administer without his own people, money, and authority.”112 Therefore, the impotence soon became clear to Clifford, Acheson, the Group itself, and many others (nominally Truman) that the CIG “had run its course, and something new was needed.”113 However, the conflict of jurisdiction would once again slow this process, ensuring that these faults would not be addressed for another year.
The National Security Act of 1947 aimed to finally address these rivalries with Congressional legislation. However, Truman prioritized the unification of the War and Navy Departments under the new National Military Establishment that would be formed to better equip the United States military for the country’s new role as a global superpower. Truman reportedly worried that introducing the reorganization of intelligence into the bill too early would complicate the unification and risk its passage.114 This prioritization for military unification over central intelligence can be sensed in Truman’s memoirs, where the intelligence issue seems like an afterthought. The intelligence clause thus became added to the National Security Act only once the Secretaries of War and Navy came to an agreement.
The National Security Act, passed in July of 1947, formed the CIA as an independent agency with its own resources, authority, and ability to conduct operations abroad. The Act, therefore, remedied the many rivalries that had plagued U.S. intelligence for decades. Section 102 of the act abolished the CIG and NIA in favor of the Central Intelligence Agency (CIA) and the National Security Council (NSC).115 In addition to overseeing the CIA, the DCI also exercised authority over the rest of foreign intelligence in the United States government.116 The DCI now had the authority over all foreign intelligence, meaning the central apparatus morphed away from its predecessor’s role as the “stepchild.” Section 104A included the catch-all phrase that the CIA shall “perform such other functions and duties related to intelligence affecting the national security as the President or the Director of National Intelligence may direct.”117 According to Clifford, these “other” functions became understood to include a wide range of covert intelligence collection activities, thereby purposely not being specified.118 The intelligence bureaucracy, therefore, stumbled through a World War to become one that successfully waged a Cold War known for its reliance on intelligence and espionage.
As a result of the years-long debate between the departments, the National Security Act transformed the balkanized, ad hoc, and bloated American intelligence structure into a cornerstone of United States national security: the modern American intelligence community. Therefore, with the interagency debates defining central intelligence, the National Security Act of 1947 made significant progress in addressing rivalries rooted in individual personality, departmental power and jurisdiction, and prior experience and capability.
However, the continued effect of interdepartmental turf wars on United States intelligence does not ultimately end with this act. Some scholars, such as Amy Zegart, argue that the early years of the CIA and the DCI show that U.S. intelligence still experienced the remnants of these rivalries. The NSA ’47 version of the CIA, she argued, satisfied the War and Navy Departments, thereby being weak by design.119 Similarly, Clifford admitted that in practice, “military intelligence resisted taking directions from the DCI” and that “to this day , the DCI’s oversight of the intelligence community falls far short of our original intent.”120 Nevertheless, there should be a recognition that the decision to reorganize the post-war United States intelligence community was, at the least in part, due to these interdepartmental rivalries.
The significance of this research on the “Washington Command Post” of intelligence reveals the degree to which bureaucratic overlap and interagency rivalries influenced Truman’s decision to reorganize the U.S. intelligence apparatus toward a more centralized operation. This paper, moreover, sheds light on contemporary issues of intelligence failures and government reorganization plans regarding the expansive national security apparatus. In this way, the analysis of these bureaucratic hurdles and factors provides a key case study that demonstrates, more broadly, how internal forces influence external organization, capabilities, and effectiveness. In this case, intelligence can therefore be significantly affected not only by United States national security interests, but also by petty bureaucratic interests, sometimes adversely. This case study of the most formative period in modern American national security, therefore, demonstrates that vital and strategic systems of government are simultaneously transformed and stunted through interdepartmental relations. As discussed throughout the research, these competing bureaucratic interests manifest through competing leadership personalities that vie for increased departmental relevance upon a foundation justified by prior experience and capability.
In the contemporary context, understanding the continued impact of historical interdepartmental issues on the National Security apparatus will better inform the development of solutions to address such complexities. While the National Security Act of 1947 significantly diminished these rivalries and the extent to which they affected intelligence capability, the U.S. intelligence apparatus, similar to bureaucratic processes in general, continues to be influenced by interdepartmental issues. This sustained dynamic has resulted in the intelligence reforms derived from the Dulles Report (1949), the Schlesinger Report (1971), the Pike (1975) and Church (1976) Committee Reports, and the 9/11 Commission Report (2004). These reports investigated and sought to reform the issues of internal interagency coordination, organization, and preparedness that defined the external effects of the modern intelligence community. The competition for intelligence command has continued to dictate the dynamics of the interdepartmental relationship as the individual personalities fight for departmental power and jurisdiction through a justification of prior experience and capability.