Intelligence Liaison between the FBI and State, 1940–44

Effective Interagency Collaboration

G. Gregg Webb

Berle and Hoover’s collaboration reversed years of dysfunction between the FBI and State over intelligence.

The post-9/11 debate over intelligence reform has been framed as a response to the intelligence “failures” that led to that infamous day. Many commentators and policymakers have compared America’s current intelligence shortcomings to past disasters, such as Pearl Harbor in 1941 or the Bay of Pigs in 1961. The impulse to identify common errors in individual judgment and interagency action between the terrorist attacks in New York and Washington in September 2001 and previous tragedies cannot be ignored. Yet, dissecting mistakes should be only part of this nation’s strategy to retool its Intelligence Community for the fight against international terrorism.

Another important perspective for planning intelligence reforms comes from past instances of effective cooperation among agencies. Just as America stands to benefit from coolly analyzing intelligence missteps, careful consideration of intelligence successes also can be constructive. This article surveys one of the earliest, most extensive, and most successful examples of interdepartmental intelligence collaboration in American history. In a community famous for its deep fissures and debilitating rivalries, the working relationship forged between the Department of State and the Special Intelligence Service of the Federal Bureau of Investigation in Latin America during World War II is both unique and instructive.

What limited scholarly attention the FBI’s Special Intelligence Service (SIS) has received over the past 60 years has, quite deservedly, been focused on the agency’s successes in the field. These ranged from high-level penetrations of foreign governments to dogged hunts for smugglers and spies throughout the Western Hemisphere. SIS agents, in concert with State Department and armed services personnel, quashed virtually all Axis intelligence operations in Central and South America during World War II. The numbers are impressive. Between 1 July 1940 and 31 December 1945, the SIS identified 832 Axis “espionage agents,” apprehended 336 of these, and ultimately gained convictions against 105 individuals for a total of more than 1,340 years in prison. The SIS further identified 222 “smugglers of strategic materials” in the Western Hemisphere and captured 75 of them. SIS employees conducted 641 separate investigations at the request of other US government agencies and shut down 24 clandestine radio stations used by Axis agents to communicate with their handlers and each other.[1]

This work explores the largely unexamined bond between policymakers at the FBI and the Department of State that acted as the foundation for the SIS’s impressive accomplishments. Central to this spirit of interdepartmental cooperation was the cordial relationship between FBI Director J. Edgar Hoover and Assistant Secretary of State Adolf A. Berle, the wartime intelligence liaison at the Department of State. Through patience and mutual conciliation, these two bureaucratic heavyweights ensured the effectiveness of US intelligence and counterintelligence efforts in Latin America.

[Top of page]


Interagency Rivalry

That members of the FBI and Department of State were capable of forming a successful foreign-intelligence union during World War II is remarkable given the competition and ill will that plagued their pre-war interactions over domestic counterintelligence work. During the late 1930s, the FBI and State were important players in the US government’s counterintelligence program, competing with each other for presidential favor and scarce funding. This odd bureaucratic division, in which the de facto national police force had to battle the government’s foreign-policy arm for control over domestic counterintelligence operations, originated in a legislative quirk.

The FBI’s original authority emanated from an 1871 appropriations statute that limited Department of Justice investigations to “the detection and prosecution of crimes against the United States.”[2] In 1916, German espionage agents and saboteurs threatened both America’s national security and her highly valued neutrality in World War I. To counter this threat, Attorney General Thomas Gregory obtained an obscure amendment to the Department of Justice appropriations statute authorizing the Bureau to pursue “such other investigations regarding official matters under the control of the Department of Justice or the Department of State as may be directed by the Attorney General.”[3] Thus, without public fanfare or debate, the Bureau of Investigation gained legal authority to conduct non-criminal inquiries—such as those involving suspected intelligence breaches—with the catch being the addition of the Secretary of State’s permission.[4]

This potentially awkward arrangement remained benign through the gauntlets of World War I and the subsequent “Red Scare.” By the mid-1930s, federal counterintelligence activity had been temporarily stopped due to the embarrassing excesses of the Palmer Raids, which had featured unlawful detentions of individuals based on their nationality and political affiliation and instilled fear and skepticism of federal law enforcement agencies among large portions of America’s immigrant community. The FBI, in particular, was under strict orders to observe its original, narrow mandate and avoid investigations of “subversive” organizations “inasmuch as it does not appear that there is any violation of a Federal Penal Statute involved.”[5]

In August 1936, President Franklin Roosevelt ended this counterintelligence calm by requesting that FBI Director J. Edgar Hoover provide him with “‘a broad picture’ of the effects of Communism and Fascism on ‘the economic and political life of the country as a whole ….’”[6] Given this clear order for an intelligence report and his fear of further civil liberties complications, Hoover informed Roosevelt that, pursuant to the appropriations statute, he would need the Secretary of State’s authorization to proceed.[7] Secretary of State Cordell Hull gave his blessing on 1 September 1936; thereafter, State-FBI relations deteriorated rapidly.[8]

The rivalry between the Bureau and the Department of State over domestic security work between 1936 and 1939 flowed from two sources: stark policy differences and hopelessly blurred lines of authority. On the policy front, neither J. Edgar Hoover nor his chief adversary at State, Assistant Secretary George S. Messersmith, could agree on how the United States should organize its response to German, Russian, and other foreign infiltrations. For Hoover, the answer lay in consolidation of all civilian counterintelligence responsibility, investigations, and funding in his FBI.[9] Hoover recognized the rights of the two service intelligence agencies—the Military Intelligence Division (MID) and the Office of Naval Intelligence (ONI)—to conduct independent investigations where their respective personnel and installations were concerned. In return, he received steady support from both MID and ONI in his fight against Messersmith.[10]

On the other side, Messersmith and State were bent on preserving the more decentralized status quo, in which responsibility for counterintelligence investigations was divided among several agencies and the Department of State served as the chief facilitator for interdepartmental activity.[11] Additionally, the 1916 amendment to the Justice appropriations statute had given State a virtual veto over FBI counterintelligence activities.

Not only did the FBI and State deadlock on policy grounds, but also their relationship, as delineated by the 1916 amendment, was violated repeatedly by the president and other high officials. For example, in October 1938, President Roosevelt had become so alarmed by the threat from Axis agents and the dysfunction in America’s counterintelligence community that he created a “Committee to inquire into the so-called espionage situation” and to identify needed reforms and funding requirements. Roosevelt named Hoover’s immediate superior, Attorney General Homer Cummings, chairman of this new committee, but limited its participants to a fraction of the agencies then conducting counterintelligence investigations.[12] Both the Department of State and the Treasury Department’s Secret Service were left off the Cummings Committee. Not surprisingly, the Committee found that its restricted membership could manage the security burden of the US government without outside help.

The President soon added more confusion to the jurisdictional melee. In the spring of 1939, he directed Assistant Secretary Messersmith to lead a second counterintelligence panel made up of representatives from the War, Navy, Treasury, Post Office and Justice Departments. Messersmith’s job was to coordinate the individual and joint efforts of these agencies against foreign forces inside the United States.[13] According to one account, no FBI representative was included in this group, although a memberfrom the FBI’s parent Department of Justice was.[14]

Convoluted and conflicting divisions of authority were a hallmark of Roosevelt’s executive leadership style.[15] Yet, on 26 June 1939, the president finally settled the pre-war power struggle between the FBI and Department of State. In a secret directive, he expressed his “desire that the investigation of all espionage, counter-espionage, and sabotage matters” be centered in the FBI, MID, and ONI alone.[16]

Roosevelt’s decision was a clear bureaucratic defeat for the Department of State in the short-term, but the long-term impact of this move far outweighed any immediate loss of face for the Department. The president’s June 1939 Directive freed State from its nominal leadership role in domestic counterintelligence and helped clarify the overall intelligence relationship between the FBI and State. Thus, when war broke out in earnest the next year, the Department of State was able to focus on building a foreign-intelligence alliance with the FBI’s Special Intelligence Service.

[Top of page]


From Improvisation to Organization

Messersmith’s extended competition with Hoover for control over America’s internal security provided clear proof that the Department of State had no qualms about conducting domestic counterintelligence work. During the inter-war years, the Department’s attitude towards clandestine foreign-intelligence collection, or espionage, was very different. Many American diplomats did not regard espionage work as an appropriate method for fulfilling their duty to keep the American government informed about regimes and developments abroad. Even State’s intelligence czar, Messersmith, called the espionage work of German agents “un-American” in a 1938 letter to a friend about America’s tenuous domestic security situation.[17] In addition to moralistic arguments against secret intelligence work, the high cost in money and manpower required to obtain such information, as well as the potential for geopolitical embarrassment should such activity be discovered, added to the general distaste for espionage in the pre-war Department of State. Consequently, though State acted as the official eyes, ears, and voice of the US government around the world, the Department did not possess any covert intelligence organizations or responsibilities during the late 1930s.[18]

As the Department of State was taking a back seat to the FBI and others on the domestic counterintelligence front in the fall of 1939, a few US diplomats and service attachés around Latin America attempted to organize clandestine collection of intelligence in their host countries. The reports these pioneers sent back to their superiors in Washington played a vital role in alerting policymakers at State to the need for a foreign-intelligence capability in Latin America and the inadequacy of existing personnel and resources for undertaking such a task. Awareness of both these issues was central to the Department of State’s acquiescence in a wartime relationship with the FBI’s Special Intelligence Service.

By far, the most sophisticated of these improvised intelligence shops operated in Mexico under the guidance of Pierre de Lagarde Boal, counselor of the US embassy in Mexico City.[19] In late 1939, working closely with the naval attaché for Mexico, Lt. Cdr. William Dillon, Boal established a three-man intelligence “coordinating committee” composed of representatives from the embassy, the military attaché’s office, and the naval attaché’s office.[20] This committee met for one hour each day and maintained index-card files on a range of topics, including anti-US foreign nationals in Mexico, local confidential informants used by the embassy, and “reliable” Americans who could provide useful information to the legation.[21] Most of the information processed by this committee arrived through the operational exertions of Naval Attaché Dillon.

In December 1939, Boal wrote to Messersmith in Washington summarizing the lessons he was learning about interdepartmental intelligence cooperation and asking for money and personnel to enlarge his and Dillon’s activities. Messersmith refused Boal’s request. In an internal memorandum to other Department managers, Messersmith laid out his opposition, citing diplomacy (“we should not in any case engage in such work on such a scale without the knowledge and consent of the government concerned”) and finances (“[e]ven if it were desirable to go ahead … we do not have the money and could not do it”) as reasons not to devote more resources to foreign-intelligence work in America’s next door neighbor.[22] The official reply Messersmith sent Boal explicitly stated that leaders at State did not then believe the ends of investigating German and other anti-American activities in Mexico justified the clandestine collection means that Boal sought to expand.[23] However, Messersmith did authorize Boal to continue the activities of his intelligence coordinating committee.[24]

In the six months after Messersmith’s reining in of Boal and Dillon, both world affairs and the Department of State’s foreign-intelligence landscape changed dramatically. On the international stage, Germany’s invasions of Denmark, Norway, Belgium, the Netherlands, and France during the first half of 1940 bolstered the threat from an increasing number of Axis agents in Central and South America.[25] At the Department of State, Messersmith was dispatched to Cuba as US ambassador in February 1940. His replacement as the assistant secretary of state responsible for intelligence affairs was Columbia Law School professor and Roosevelt braintruster Adolf A. Berle. Berle quickly recognized the burgeoning threat to American political and financial interests from Axis intrigue in Latin America and, during the spring of 1940, began to press for a comprehensive interdepartmental response.[26]

[Top of page]


The Berle-Hoover Connection

On 24 June 1940, President Roosevelt issued a directive by telephone making the Federal Bureau of Investigation “responsible for foreign intelligence work in the Western Hemisphere, on the request of the State Department.” By 1 July 1940, FBI Director J. Edgar Hoover had established a “Special Intelligence Service” within his Bureau and had embarked on the colossal task of creating from scratch a foreign-intelligence capability in the FBI.[27] The president’s stipulation that this new agency—the first foreign-intelligence bureaucracy in US history—should conduct its activities in Latin America at the behest of the Department of State forced two longtime adversaries into common cause.

Fortunately, the two men tasked to direct this State-FBI intelligence union proved anything but adversarial towards one another. Adolf Berle, as the assistant secretary of state with intelligence liaison duties, and J. Edgar Hoover, as FBI director, were involved in every phase of the SIS project, from cultivating its roots in the pre-war Interdepartmental Intelligence Committee—established as a result of President Roosevelt’s 1939 counterintelligence delimitation directive—to resolving delicate administrative challenges in the Service’s wartime work.[28] Operating in tandem, the two men instituted several of the bedrock measures on which the SIS’s indispensable intelligence network in Latin America rested.

Among their most important collaborative successes was securing the assignment of SIS agents as “legal attachés” in US missions throughout Central and South America.[29] By October 1942, 77 FBI legal attachés, with diplomatic status, worked out of US embassies in 18 nations in the region.[30] These officers coordinated secret intelligence operations in their assigned countries, collecting information and investigative leads from indigenous contacts and undercover SIS agents.[31] The legal attachés passed these data on to FBI headquarters in Washington and sometimes used them to formulate local actions with embassy diplomats and armed services attachés. To implement this centerpiece of SIS organization, Berle played the role of intermediary between the FBI Director and skeptical ambassadors and bureaucrats at State. The pair also teamed up against opposition to a FBI-proposed courier system for SIS communications.[32] Such a system never developed, but SIS personnel did gain the ability to send correspondence back to Bureau headquarters through the Department of State’s official diplomatic pouches.[33]

[Top of page]


Diverse Backgrounds

Little in the backgrounds of Berle or Hoover suggested that they would become such close partners on the SIS. Though the two were born less than a month apart in January 1895, their professional lives followed very different paths until their intersection on intelligence in early 1940. Raised in Boston, Berle was the youngest graduate in Harvard Law School history when he received his J.D. in 1916 at the age of 21.[34] After a stint in the Army’s Military Intelligence Division during World War I, he became a professor at the Harvard Business School in 1924 and then at Columbia Law School in 1927.[35] He penned groundbreaking work in the fields of corporate law and economics during the 1930s. A member of Roosevelt’s “brain trust,” he also worked on New York City affairs with Mayor Fiorello La Guardia.[36] Appointed assistant secretary of state in 1938, Berle was assigned a wide-ranging portfolio, including Latin America policy. In February 1940, he gained intelligence-liaison duties.[37]

Hoover was born and raised in Washington, DC. He worked his way through George Washington University Law School, earning a LL.B. in 1916 and a LL.M. in 1917.[38] During the summer of 1917, he started in an entry-level position at the Department of Justice and rose rapidly, becoming the acting director of the Department’s Bureau of Investigation (later renamed FBI) in 1924 and director soon thereafter.[39] Hoover’s early career was dominated by his work in the various counterintelligence divisions of the Bureau during World War I and the subsequent “Red Scare.” His ascension was a product of his reputation in the Department of Justice as “an honest and efficient administrator” and occurred despite his close association with the Department’s contemporary civil liberties abuses.[40] By 1940, Hoover was firmly entrenched in his directorship of the FBI and confident in the organization that he had constructed over the previous 16 years.

Although both Berle and Hoover had spent considerable time in public service, they were opposites in many ways. Berle was a leading liberal in the Roosevelt administration, whereas, the outwardly apolitical Hoover held strong conservative convictions, particularly on social matters. Berle was a passionate internationalist; Hoover was wary of all things foreign. In personality, Berle was an intellectual, though not unskilled in policy administration; Hoover, as a master bureaucrat, was deeply practical, though highly intelligent. Notwithstanding these differences, their cooperation on the SIS proved exceptionally functional.

[Top of page]


Personal Dynamics

Several personal and institutional circumstances contributed to the general harmony between Adolf Berle and J. Edgar Hoover. On a personal level, Berle did not try to battle the director for administrative control over the Special Intelligence Service. Instead, he willingly left day-to-day management to Hoover and his subordinates, only intervening at the request of the Bureau or when FBI personnel and actions aroused the Department of State’s ire. Concerning the “big picture” policies of US intelligence in Latin America, Berle consistently sought, and for the most part obtained, frank communication with J. Edgar Hoover.

Berle’s approach contrasted sharply with that of his predecessor, George Messersmith. Hoover and Messersmith’s acrimonious relationship during the late 1930s distracted both the FBI and the Department of State from their shared responsibility to track down German and communist spies and potential saboteurs. Indeed, according to one account, Messersmith’s chief complaints about Hoover included the fact that he “was difficult to work with except on his own terms.”[41]

Berle could afford to be conciliatory with Hoover for two reasons. First, Berle, unlike Messersmith, was not a career diplomat. He did not feel obliged to obstruct Hoover out of interdepartmental jealousy or spite.[42] In fact, Berle repeatedly defended the SIS and Hoover from the machinations of others, including MID, OSS, and several factions within State itself. Second, Berle had far too many other duties within the Department of State to micromanage SIS affairs. He effected his liaison with Hoover through State’s Division of Foreign Activity Correlation (FC).[43] In addition to managing the relatively small FC staff, by 1944 Berle was directing State’s Passport Division, Visa Division, and
Special War Problems Division (which dealt with American prisoners-of-war), plus State’s Office of Transportation and Communication, which included Divisions for Aviation, Shipping, and
Telecommunications.[44] Berle also frequently drafted speeches for Secretary of State Cordell Hull and played a key role in the Department’s Latin America

The best evidence that Berle’s laissez faire handling of Hoover facilitated constructive cooperation between State and the SIS comes from Hoover himself.
The famously sensitive FBI Director was prone to curtailing communication with any government agency or individual that he considered a threat to his own or his Bureau’s authority. Yet, Hoover conscientiously kept Berle updated on SIS activities throughout his tenure as State’s intelligence liaison from 1940 to 1944. During this period, Hoover sent Berle reams of documents concerning the SIS, ranging from elaborate color maps with the disposition of secret, Bureau-run radio stations in Latin America to requests to send FBI agents on special assignments abroad.[46] Most of the information provided to Berle and his FC staff by Hoover were reports on intelligence operations and counterintelligence investigations.

Berle’s unobtrusive attitude proved useful as Hoover molded the nascent SIS. Atop the director’s list of professional pet peeves were indistinct lines of administrative authority and nebulous or conflicting agency mandates.[47] Hoover faced several threats to his power as sole collector of secret intelligence in the Western Hemisphere.[48] MID launched one such an assault in late 1940 and early 1941. During this period, Hoover and MID chief Gen. Sherman Miles fought an increasingly bitter battle over intelligence collection authority in the New York area.[49] On 12 February 1941, the Interdepartmental Intelligence Committee—with Hoover, Miles, and Berle all present—held a lengthy meeting to resolve the dispute. The FBI Director expressed his position on the New York conundrum as an ultimatum: either the FBI was in charge, or he would hand over SIS coverage to the military. This all-or-nothing approach to resolving a bureaucratic tangle was classic Hoover. Throughout the meeting, Berle worked to pacify the warring parties.[50]

Hoover’s obsession with strict divisions of authority extended into his relations with the Department of State. Although Berle labored to keep ambassadors and State personnel from meddling in the administrative affairs of the SIS, a certain amount of unsolicited input and criticism slipped into the FBI chain of command. For example, in November 1943, Ambassador to Peru R. Henry Norweb’s contacts at the State Department made inquiries at FBI Headquarters in Washington to see if the wife of a particular SIS agent could join him abroad.[51] Hoover’s reaction to this feeler was immediate and severe. He fired off a memorandum to his lieutenants declaring that he thought “it [was] rather presumptuous for the State Department or an Ambassador to inject himself into an administrative policy of this Bureau.”[52] Hoover concluded “I very definitely resent the intrusion into this aspect of our administrative policy ….”[53]

Just as Adolf Berle’s impressive discretion in dealing with Hoover can only be appreciated in light of Hoover’s strict bureaucratic principles, the FBI director’s respect for Berle must be measured relative to Hoover’s other contacts in Washington. Hoover had a close personal relationship with President Roosevelt and many of his successors. He often used this ace to circumvent his immediate superiors in the attorney general’s office when he had something to communicate to the president, whether it was a FBI operational success or a jurisdictional complaint. With such high access, Hoover rarely acknowledged FBI inadequacies raised by anyone outside his agency and not inside the Oval Office. One indication of his personal respect for Adolf Berle was manifest in his promptly addressing criticism of the SIS delivered by the assistant secretary in a conversation with FBI Special Agent Jerome Doyle on 3 January 1942.

According to a summary of this meeting prepared for the director, Berle had expressed concern that intelligence collection was not adequately covering the lower classes in South America. In the margin next to this account, Hoover scribbled, “Take steps at once to cover this aspect. H.”[54] Within a month, Hoover had a letter out to Berle describing relevant steps the Bureau had taken in Chile and Mexico to collect information on the working classes in those countries, along with an assurance that more such operations were in the works.[55] Berle’s willingness to raise the point constructively and Hoover’s rapid response illustrate the quality of their personal relationship.

[Top of page]


Institutional Dynamics

The way Berle and Hoover treated each other as individuals was only one reason their relationship succeeded. A second dynamic governing their collaboration was the unique institutional status of the Special Intelligence Service. Unlike Hoover’s jealous guarding of his domestic duties, he did not initially draw a line in the bureaucratic sand and fight off other agencies interested in the SIS. He rejected sharing the foreign-intelligence sphere under blurred authority, but he encouraged overtures to have the Service transferred out of the Bureau.[56]

Two considerations dominated Hoover’s calculation. First, the director had not pursued a major foreign-intelligence assignment; rather, the FBI’s presence in the field was a product of President Roosevelt’s organizational creativity. The Interdepartmental Intelligence Committee had considered placing the FBI in charge of clandestine work in Latin America, but no decision had been reached by the time the president divided the intelligence pie himself in June 1940.[57] Consequently, while Hoover and Berle built the SIS, the director checked his ambition and regarded it as a genuine “service agency,” collecting secret intelligence and conducting counterintelligence investigations for the benefit of others.[58]

Hoover’s initial lack of interest in aggrandizing his and the Bureau’s foreign intelligence role was expressed at the same IIC meeting in which intelligence jurisdictions in New York City were so hotly debated. He continued to see the Special Intelligence Service as an unsolicited and unwieldy burden until word of the unit’s spy-hunting success spread late in World War II, at which point he began seeking to expand the FBI’s post-war intelligence powers.[59]

A second reason why Hoover willingly cooperated with Berle and the Department of State, even as he tried to shed the SIS responsibility, was his fear of bad publicity. After devoting much of the previous decade to blowing hot air into public perceptions of himself, his “G-men,” and their crime fighting abilities, Hoover was intensely reluctant to see his or the Bureau’s reputation sullied by embarrassing intrigues abroad.[60] To avoid being disowned in a pinch, Hoover welcomed a record of close association between his FBI and the Department of State over SIS affairs. He expressed this anxiety over the FBI’s image openly during the IIC’s long debate on intelligence contacts in New York City.[61]

With Hoover skittish about his image and eager to cast off the SIS, one of Berle’s most consistent challenges was keeping the Special Intelligence Service in the FBI, and thus under the nominal control of the Department of State. No aspect of the Berle-Hoover liaison provides clearer proof of its exceptional nature than this one. Incredibly, between 1940 and 1942, Berle found himself several times either defending J. Edgar Hoover’s institutional interests for him or reminding the famously competitive bureaucrat why his presence in the foreign-intelligence field was necessary.

Berle’s most desperate defense of the SIS came in September 1941 during a push by Coordinator of Information William Donovan to assume responsibility for covert intelligence work in Latin America. Upon learning of Donovan’s ambition, Hoover feared the FBI would end up working under or alongside a Donovan-led unit. The director ordered his subordinate, Edward Tamm, to visit Berle at State and inform him that, as Tamm put it, “the Bureau … had no feeling one way or the other as to whether this transfer should be made.”[62] Berle dismissed this sentiment outright to Tamm, declaring that the Bureau had done “an excellent job” with the SIS and “that he would be opposed to having it transferred into untried hands.”[63] Berle’s forceful opposition to Donovan’s proposal at the next IIC meeting ensured that the Service stayed within the FBI.

When Donovan made a second stab at the SIS in December 1941, Berle once again was instrumental in protecting the FBI-State status quo from Donovan’s sticky fingers. In a January 1942 strategy session with Tamm, the assistant secretary professed that “he was more than ever convinced of the absolute necessity for so handling this situation as to insure the continuation of … SIS operations in the Western Hemisphere solely and exclusively by the Bureau.”[64]

The relationship between Adolf Berle and J. Edgar Hoover was not the only reason for the SIS’s resounding success in Latin America. The organization also benefited from the help of friendly governments and populations throughout the region. As the conflict progressed, the failing fortunes of the Axis states and the gathering strength of the Allied war effort also helped the SIS outmaneuver enemy agents and harvest political, economic, financial, and industrial intelligence.[65] Even so, the Berle-Hoover connection was an indispensable part of SIS dominance. The potent bond between the two—growing out of their respectful treatment of one another and complementary institutional interests—shielded the Special Intelligence Service from the debilitating discord that plagued other wartime intelligence organizations, such as Donovan’s Office of the Coordinator of Information (later OSS) and the US Joint Intelligence Committee. Nevertheless, the State-FBI relationship was far from perfect.

[Top of page]


Limits to State-FBI Cooperation

Whenever possible, Berle indulged the FBI director’s penchant for administrative control. Likewise, Hoover made a sincere effort to keep the assistant secretary abreast of the activities and requirements of the SIS. Unfortunately, this spirit of accommodation did not trickle down to all levels of the State-FBI partnership. Considerable tension between the two agencies arose at State from individuals who either mistrusted or were jealous of the FBI and its foreign-intelligence mandate.

US ambassadors stationed in Latin America were the most frequent antagonists of the SIS. Ambassadors existed outside the bureaucratic hierarchy at State. They served, instead, as the president’s personal envoys to foreign governments; as a group, they constituted a third party in the relationship between State and the FBI.[66] The most widespread friction between ambassadors and the Bureau during World War II erupted with the dispatch of undercover SIS agents to Latin America. In the first months of the project, all agents sent abroad by the FBI went in a clandestine capacity. The Bureau believed that keeping its agents’ identities secret from local American legations was essential for both the safety of SIS personnel and the security of their mission. Not surprisingly, US ambassadors rejected having an indeterminate number of FBI agents conducting investigations and running networks of informants in their zones of responsibility.[67] They feverishly set about discerning the identity and location of every SIS agent in the field.[68] British intelligence, likely feeling threatened by the Service’s presence in the region and perhaps looking for some fun, enjoyed identifying fresh American agents and reporting their arrival to unsuspecting ambassadors.[69] The anger and distraction this undercover policy created subsided with the assignment of legal attachés to most Latin American embassies by late 1942.

Among the critics of the Special Intelligence Service was Hoover’s former nemesis and wartime ambassador to Mexico, George Messersmith. While serving in Mexico from 1941 to 1946, Messersmith assumed two distinct attitudes towards the Service. On one hand, he supported SIS counterintelligence investigations and engendered close ties with the FBI’s legal attachés—called civil attachés in Mexico. The ambassador sent several messages back to Washington praising the FBI men under his jurisdiction.[70] In a December 1942 letter to Berle, Messersmith gushed, “I am very much pleased with the work which Mr. Jones (Civil Attache in Mexico) is doing here with his associates. They are showing good judgment and discretion and zeal ….”[71]

On the other hand, even as he fostered friendships with individual SIS agents, Ambassador Messersmith lobbied hard to limit the Bureau’s intelligence footprint. For example, Secretary of State Cordell Hull asked Messersmith’s opinion in August 1942 about having a short-wave radio set installed at the US embassy in Mexico City, with a FBI operator acting under Messersmith’s control to man it, saying that such radio units had already been set up in several other embassies around Latin America.[72] Messersmith argued against a radio in Mexico City, citing the delicacy of obtaining permission from the Mexican government.[73] Messersmith’s argument reflected his divided feelings. He stated, “The F.B.I. representative in this Embassy is a very good man … but I do not like the idea of communications between the [State] Department and this Embassy on all sorts of matters passing through the F.B.I. representative.”[74]

Back in Washington, interest in the SIS and opposition to its FBI management increased steadily during World War II. Before 1940, most diplomats at Foggy Bottom, like their peers in the field, cared little for intelligence work. As the SIS grew in size and stature, bureaucrats at State took covetous notice. Berle recorded this transition in his diary on 7 November 1940. Referencing his efforts to construct an Intelligence Division for the Department, Berle noted how “Intelligence is beginning to be interesting in the [State] Department now, so everybody wants to be in on it.”[75] By September 1944, a determined opposition had coalesced at State against FBI involvement in the foreign-intelligence field.[76] These anti-FBI forces helped shut J. Edgar Hoover out of the post-war intelligence picture, but they failed in their bid to become the Bureau’s sole replacement. Instead, State shared worldwide intelligence authority with several other agencies, including the new Central Intelligence Group, MID, and ONI.

[Top of page]



On 28 October 1943, Department of State administrator Rowena B. Rommel produced a long memorandum entitled “Relations of the Department of State to Other Federal Agencies.”[77] Rommel’s piece laid down conceptual guidelines for the Department’s wartime “role in the administration of government programs in the international field.” In one section, she considered the best technique for administering “those areas of activity where other agencies have operating responsibilities and the State Department a coordinating, advisory responsibility.”[78] This described perfectly the relative positions of the FBI and State in their SIS liaison. According to Rommel, “a conscious differentiation should be made between the kind of administrative methods and staff needed” in such collaborative arrangements “in contrast to those used in … direct operations.” Rommel’s subsequent list of qualities for Department representatives pursuing “advisory” relationships with other agencies fit Adolf Berle to a tee.

Her list included:

“Breadth of intellectual grasp”—Professor Berle was considered among the brightest minds of his generation;

“Willingness to understand another point of view”—Berle’s efforts to empathize with Hoover and defer to his administrative judgment were extensive and sustained;

“Planning ahead to give guidance and keep abreast of emerging problems”—Hoover’s efforts to keep Berle informed about SIS troubles and triumphs kept the assistant secretary on the organization’s administrative front lines;

“Decisiveness so all officials know where they stand and business moves along”—Berle maintained frequent and substantive contact with Hoover and several of his FBI subordinates and US ambassadors throughout Latin America;

“Delegation of authority to lower officials and backing of those officials so there is not a continuous appealing to higher courts”—most of the day-to-day contact between State and the FBI ran between Berle’s Assistant Fletcher Warren and the rest of the FC staff and several of Hoover’s lieutenants. Berle and Hoover never allowed a SIS-related dispute between them to travel up the chain of command. In contrast, fights between Hoover and Miles at MID and Donovan at COI shot to the Cabinet level, and even into the Oval Office.

The relationship that Adolf Berle and J. Edgar Hoover constructed was as close to Rommel’s theoretical ideal as the stresses of war and reality could be expected to allow.

The Berle-Hoover partnership was not the only instance in World War II where close personal relations among intelligence chieftains generated interdepartmental, and even international, cooperation.
J. Edgar Hoover fostered a productive relationship with at least two wartime heads of ONI, Rear Adm. Walter S. Anderson and Capt. (later Vice Adm.) Alan G. Kirk, as well as their organization. Anderson worked with Hoover and Berle in establishing the SIS during the summer of 1940 and witnessed the Service’s initial progress.[79] Kirk, as ONI commander in 1941, refused Hoover’s offer to transfer all SIS responsibilities in Mexico to Naval Intelligence, insisting that such a move would be counter to the government’s “best interests.”[80]

However, Berle and Hoover’s collaboration was unique, and all the more impressive because it reversed years of dysfunction between State and the FBI over intelligence. The two men accomplished this feat through patience, deference, open communication, and by pursuing common interests. Berle himself provided the best summary of his relationship with Hoover when he wrote in his diary on 28 February 1942, “This [SIS] is one case where cooperation between State and FBI is working out beautifully.”[81]

Before creating new intelligence agencies or overhauling old ones, contemporary intelligence reformers should consider the deeply personal dynamics that made the Berle-Hoover connection so formidable. Effective liaisons like theirs serve as compelling reminders that intelligence cooperation is, at its most basic level, a matter of individual, and not institutional, interaction.

[Top of page]



[1]Statistics from Tables 1 & 2, SIS Statistics; Section 10, File 64-4104, Administrative Records of the SIS, Record Group 65 (RG 65), National Archives at College Park, MD (NACP).

[2]US Statutes at Large, 65th Cong., April 1917–March 1919, vol. XL, 155. This sentence extends from one in US Senate, Select Committee to Study Governmental Operations with Respect to Intelligence Activities, Final Report, Supplementary Detailed Staff Reports on Intelligence Activities and the Rights of Americans, Book III, 94th Cong., 2nd sess. (23 April 1976), Report No. 94-755, Serial 13133-5, 379 [hereinafter: Church Committee Report, Book III].

[3]Ibid. Emphasis added. This sentence and the next draw from related passages in the Church Committee Report, Book III, 378–79.

[4]Ibid., 379.

[5]Memorandum from Hoover to Ridgeley, 14 May 1925, as cited in the Church Committee Report, Book III, 390.

[6]Church Committee Report, Book III, 395.


[8]Memorandum from Hoover to Edward Tamm, 10 September 1936, as cited in the Church Committee Report, Book III, 396.

[9]Raymond J. Batvinis, “In the Beginning: An Examination of the Development of the Federal Bureau of Investigation’s Counterintelligence Program, 1936–1941,” (Ph.D. dissertation, Catholic University, 2001), 37–40, and Thomas F. Troy, Donovan and the CIA: A History of the Establishment of the Central Intelligence Agency (Washington, DC: Center for the Study of Intelligence, Central Intelligence Agency, 1981), 12.

[10]Batvinis, 37–38, and Church Committee Report, Book III, 397–98.

[11]Batvinis, 45, and Troy, 12.

[12]The Cummings Committee was limited to representatives from the FBI, MID, and ONI. This and the previous sentence extend from Batvinis, 37, and the Church Committee Report, Book III, 397–98.

[13]This sentence draws from similar statements in Don Whitehead, The FBI Story, 165, as cited in Troy, 12, and the Church Committee Report, Book III, 402.

[14]Don Whitehead, The FBI Story, 165, as cited in Troy, 12. The circumstances of Messersmith’s intervention remain unclear. After World War II, Messersmith maintained that President Roosevelt had compelled him to undertake coordination of the counterintelligence field. However, some historical accounts have portrayed Messersmith as assuming this role for himself. Likewise, Messersmith argued that Hoover refused to participate in his initiative until forced by the president; whereas, Whitehead’s Hoover-sanctioned account asserts that Messersmith shut the FBI out. What is clear is that Messer­smith, likely with Roosevelt’s knowledge, sought to coordinate the counterintelligence field soon after the Cummings Committee had set out on the same mission. Neither effort succeeded, and both heightened the general confusion and interde­partmental rancor.

[15]Mediating rivalries was one mechanism Roosevelt used to control subordinates, hence the president’s apparent duplicity in assigning both Justice and State coordinating roles in the counterintelligence field. These statements are based on the author’s unpublished senior thesis at Princeton University entitled “Conflict and Creation: A Comparative Study of the US and British Joint Intelligence Committees in the Second World War” (2003).

[16]Presidential Directive, 26 June 1939, Section 2, File 64-4104, Administrative Records of the SIS, RG 65, NACP.

[17]This account relies on a description of a letter from Messersmith to Geist dated 19 May 1938 as cited in Batvinis, 32. Though Messersmith may have considered espionage work unsavory, he was not as reluctant to use aggressive foreign intelligence means as other diplomats at State. Before World War II, he sought to limit the use of local informants by members of the US legation in Mexico, but as wartime Ambassador to Mexico, he presided over an active and successful State-SIS intelligence collaboration that maintained extensive secret contacts throughout Mexico.

[18]This statement is based on the author’s research and writing for an unpublished senior thesis at Princeton University (supra note 15).

[19]Other ad hoc intelligence arrangements were operated, or at least proposed, before World War II by Ambassadors Jefferson Caffery in Brazil, E. C. Wilson in Uruguay, and Spruille Braden in Colombia.

[20]Letter from Boal to Messersmith, 22 December 1939, 5, and attached “Memorandum,” as reprinted in John Mendelsohn, ed., Covert Warfare: Covert Warfare in Latin America (New York: Garland, 1989), vol. 10.


[22]Memorandum from Messersmith to Warren, Duggan, and Chapin, 28 December 1939, as reprinted in Mendelsohn.

[23]Letter from Messersmith to Boal, 24 January 1940, as reprinted in Mendelsohn.


[25]Leslie B. Rout, Jr., and John F. Bratzel, “Origins: US Intelligence in Latin America,” Studies in Intelligence (Winter 1985): 50, in Folder 108, Box 9, Studies in Intelligence, Center for the Study of Intelligence, RG 263, NACP.

[26]Memorandum for the Files from W. M. Crane, 3 June 1940, 810.20 Defense/20, Box 3375, State Department General Decimal File, 1940–1944, RG 59, NACP.

[27]For details on the creation of the FBI’s Special Intelligence Service, see Troy, 16–17, and G. Gregg Webb, “The FBI and Foreign Intelligence: New Insights into J. Edgar Hoover’s Role,” Studies in Intelli­gence 48, no. 1 (2004): 46–49.

[28]The IIC was the first interdepartmental body in the United States for sustained intelligence-policy coordination. In May and June 1940, the IIC served as incubator for the idea that America needed a distinct foreign-intelligence organization. President Roosevelt’s 24 June 1940 directive was in response to the IIC’s proposal of such an agency.

[29]Draft Letter from Berle to Fletcher Warren, 19 May 1941, Folder 1, File 64-4104, Administrative Records of the SIS, General Records of the FBI, RG 65, NACP; and Memorandum for the Director from Tamm, 20 May 1941, Folder 1, File 64-4104, Administrative Records of the SIS, General Records of the FBI, RG 65, NACP (Berle’s advocacy within State of legal attaché concept). See also Memorandum for the Director from Ladd, 23 April 1942, Folder 4, File 64-4104, Administrative Records of the SIS, General Records of the FBI, RG 65, NACP (Berle’s pride on this point, update on legal attaché coverage, and FBI pressure for more legal attachés), 4.

[30]Memorandum for the Director from Ladd, 26 October 1942, Folder 6, File 64-4104, Administrative Records of the SIS, General Records of the FBI, RG 65, NACP, 2.

[31]This description of legal attaché responsibilities extends from a similar discussion in a FBI memorandum on the SIS, 12 February 1946, Folder 11, File 64-4104, Administrative Records of the SIS, General Records of the FBI, RG 65, NACP, 2.

[32]Memorandum for the Director from Tamm, 13 May 1941, Folder 1, File 64-4104, Administrative Records of the SIS, General Records of the FBI, RG 65, NACP.

[33]Ambassadors and other high embassy officials retained the right to read SIS correspondence sent via this method. Memorandum for the Director from Ladd, 23 April 1942, Folder 4, File 64-4104, Administrative Records of the SIS, General Records of the FBI, RG 65, NACP, 4–6.

[34]Jordan A. Schwarz, Liberal: Adolf A. Berle and the Vision of an American Era (New York: Free Press, 1987), 16.

[35]Batvinis, 63, and Beatrice B. Berle and Travis B. Jacobs, eds., Navigating the Rapids, 1918–1971: From the Papers of Adolf Berle (New York: Harcourt, 1973), xviii.

[36]Berle and Jacobs, xx–xxi, and Schwarz, 75, 91.

[37]Berle and Jacobs, xxi, xxiv, and Schwarz, 118–19, 169–70.

[38]Richard Gid Powers, Secrecy and Power: The Life of J. Edgar Hoover (New York: Macmillan, 1987), 40.

[39]Church Committee Report, Book III, 388.

[40]This sentence draws from a similar statement in the Church Committee Report, Book III, 388.

[41]In their pioneering book on the SIS, Rout and Bratzel write that “As early as the fall of 1939, Assistant Secretary of State George Messersmith … had denounced Hoover as a glory hound and difficult to work with except on his own terms” (37). George S. Messersmith Papers, University of Delaware Library, Newark, Delaware, file 2018/5, n.d., 3–5, as cited in Leslie B. Rout, Jr., and John F. Bratzel, The Shadow War: German Espionage and United States Counterespionage in Latin America during World War II (Frederick, MD: University Publications of America, 1986), 37.

[42]Batvinis, 63.

[43]Undated notes from Box 32; Entry 718; Working Papers and Source Materials for Histories of Organizational Units, 1938– 1949, Records of the War History Branch, General Records of the Department of State, RG 59, NACP.

[44] Ibid., and Dept. of State Organizational Chart and Table, 15 January 1944, Folder “Chapter II Office Departmental Administration,” Box 2, Entry 714, Drafts of Chapters for an Overall History of the Department of State during World War II, Records of the War History Branch, General Records of the Department of State, RG 59, NACP.

[45]One indication of the assistant secretary’s overflowing administrative platter appeared in 1943. Dismissing a critical letter sent to the Bureau under his signature, Berle admitted to Hoover’s lieutenant, Edward Tamm, that he had “to handle as much as 250 pieces of mail a day” and “naturally [could not] devote as much time to each individual piece as might be desirable.” Berle comments quoted by Tamm in Memorandum for the Director, 17 November 1943, Folder 4, File 64-4104, Administrative Records of the SIS, General Records of the FBI, RG 65, NACP.

[46]Letter from Hoover to Berle, 10 March 1943, 102.31/3-1043, Box 52, Department of State Central Decimal File, RG 59, NACP (radio “order of battle”); letter from Hoover to Berle, 27 September 1940, 811.20237/9-2740, Box 3728, Department of State Central Decimal File, 1940–1944, RG 59, NACP; and letter from Berle to Hoover, 4 October 1940, 811.20237/9-2740, Box 3728, Department of State Central Decimal File, 1940–1944, RG 59, NACP (investigation authorization).

[47]Batvinis makes a similar point, 50. For more on Hoover’s disgust for unclear mandates and commitment to direct divisions of responsibility, see Webb, “The FBI and Foreign Intelligence,” 51–56, 58.

[48]Webb, “The FBI and Foreign Intelli­gence,” 49–56.

[49]Army intelligence had established a New York City office to obtain information on Latin America from firms based in the city that conducted business in the region. Friction developed because the SIS also operated an undercover outlet in New York whose staff sought information from the same international businesses. See Troy, 46–49.

[50]Minutes of the Interdepartmental Intelligence Committee, 12 February 1941, 811.20200/3-2741, Box 3728, Department of State Central Decimal File, 1940–1944, RG 59, NACP, 7, 9. Despite Berle’s exertions and the Committee’s extended discussion, the IIC failed to adjudicate this jurisdictional dispute, and the matter wound up in President Roosevelt’s lap. The president attempted to solve the problem by bypassing Hoover and Miles and appointing his wealthy friend and confidante Vincent Astor as “Area Controller for the New York Area,” with authority to mediate jurisdictional disputes (Troy, 47, 49). For more on Astor’s amateur intelligence adventures, see Christopher Andrew, For the President’s Eyes Only: Secret Intelligence and the American Presidency from Washington to Bush (New York: HarperPerennial, 1995), Ch. 3.

[51]Memorandum for Tamm, Ladd, Carson, and Tolson, from Hoover, 13 November 1942, Folder 5, File 64-4104, Administrative Records of the SIS, General Records of the FBI, RG 65, NACP, 1.



[54]Memorandum for the Director from Ladd, 4 February 1942, Folder 2, File 64-4104, Administrative Records of the SIS, General Records of the FBI, RG 65, NACP, 1.

[55]Letter from Hoover to Berle, 26 February 1942, Folder 2, File 64-4104, Administrative Records of the SIS, General Records of the FBI, RG 65, NACP.

[56]SIS suitors included the Office of Naval Intelligence in early 1941, “Wild Bill” Donovan’s Office of the Coordinator of Information (COI) later in 1941, and the Military Intelligence Division in mid-1942.

[57]Troy, 17.

[58]Minutes of the Interdepartmental Intelligence Committee, 12 February 1941, 811.20200/3-2741, Box 3728, Department of State Central Decimal File, 1940–1944, RG 59, NACP, 8.

[59]In December 1944, Hoover submitted a proposal for a “world-wide intelligence system” to be run after the war by the FBI and modeled on the SIS. Webb, “The FBI and Foreign Intelligence,” 57–58.

[60]For more on Hoover’s public relations endeavors during the 1930s, see Kenneth O’Reilly, “A New Deal for the FBI: The Roosevelt Administration, Crime Control, and National Security,” The Journal of American History 69 (December 1982): 3. For references to Hoover’s publicity concerns regarding the SIS, see Batvinis, 63, 317.

[61]Minutes of the Interdepartmental Intelligence Committee, 12 February 1941, 811.20200/3-2741, Box 3728, Department of State Central Decimal File, 1940–1944, RG 59, NACP, 5.

[62]Memorandum for the Director from Tamm, 2 September 1941, Folder 1, File 64-4104, Administrative Records of the SIS, General Records of the FBI, RG 65, NACP, 1. For a fuller description of this and other takeover bids involving the SIS, see Webb, “The FBI and Foreign Intelligence,” 51–56.

[63]Memorandum for the Director from Tamm, 2 September 1941, 2.

[64]Memorandum for the Director from Tamm, 12 January 1942, Folder 2, File 64-4104, Administrative Records of the SIS, General Records of the FBI, RG 65, NACP.

[65]Letter from Hoover to Miles, 3 August 1940, Section 1, File 64-4104, Administrative Records of the SIS, General Records of the FBI, RG 65, NACP.

[66]This description of ambassadorial status draws on similar accounts in Memorandum for the Director from Ladd, 23 April 1942, Folder 4, File 64-4104, Administrative Records of the SIS, General Records of the FBI, RG 65, NACP, 2, and Memorandum for Ladd from F. C. Holloman, 14 January 1942, Folder 2, File 64-4104, Administrative Records of the SIS, General Records of the FBI, RG 65, NACP, 2.

[67]Memorandum for Ladd from F. C. Holloman, 14 January 1942, 2–3.

[68]Ibid., and Memorandum for the Director from Ladd, 23 April 1942, Folder 4, File 64-4104, Administrative Records of the SIS, General Records of the FBI, RG 65, NACP, 8–9.

[69]Memorandum for the Director from Ladd, 23 April 1942, 9.

[70]List of praise for the SIS from various US government officials, Folder 11, File 64-4104, Administrative Records of the SIS, General Records of the FBI, RG 65, NACP.

[71]Ibid., 8.

[72]Airgram from Hull to Messersmith, 12 August 1942, Box 3728, Department of State Central Decimal File, 1940–1944, RG 59, NACP.

[73] Letter from Messersmith to Undersecretary of State Sumner Welles, 18 August 1942, 811.20200(R)/8-1842, Box 3728, Department of State General Decimal File, 1940–1944, RG 59, NACP, 1.

[74]Ibid., 2.

[75]Berle and Jacobs, 351.

[76]According to Berle’s assistant Fletcher Warren, individuals at State compiled a list of “mistakes” in FBI political reports distributed to Berle and US ambassadors in Latin America. In retaliation, Hoover ordered the SIS to “Stop sending Political Inf[ormation] to State and Ambassadors and retrench in SIS coverage.” Though Hoover dug in and fought for a post-war role in foreign intelligence, he failed to gain President Harry Truman’s support and was elbowed out of the field. For quotation and details on State’s late-war opposition to the SIS, see Memorandum to Ladd from R. R. Roach, 20 September 1944, Folder 9, File 64-4104, Administrative Records of the SIS, General Records of the FBI, RG 65, NACP. For more on interdepartmental struggles over the post-war intelligence field, see C. Thomas Thorne, Jr., and David S. Patterson, eds., “Emergence of the Intelligence Establishment,” Foreign Relations of the United States Series, 1945–1950, (Washington, DC: Government Printing Office, 1996).

[77]Memorandum entitled “Relations of the Department of State to Other Federal Agencies,” from Rowena B. Rommel, 28 October 1943, Box 33, Entry 718, Working Papers and Source Materials for Histories of Organizational Units, 1938–1949, Records of the War History Branch, General Records of the Department of State, RG 59, NACP.

[78]Ibid., 5.

[79]Minutes of IIC Meeting, 3 June 1940, 811.20200/6-1040, Box 3728, Department of State Central Decimal File, 1940–1944, RG 59, NACP, and Troy, 17. See also Batvinis, 62.

[80]Memorandum for the Director from Tamm, 11 August 1941, Folder 1, File 64-4104, Administrative Records of the SIS, General Records of FBI, RG 65, NACP. See also G. Webb, “The FBI and Foreign Intelligence,” 52–53.

[81]Berle and Jacobs, 404.


G. Gregg Webb is currently pursuing a law degree at Stanford University.


[Top of page]

Historical Document
Posted: Apr 15, 2007 08:55 AM
Last Updated: Jun 26, 2008 12:31 PM