Introduction

This volume takes as its starting point the surrender of Japan in August 1945, but it does so with the realization that the contest over the postwar U.S. intelligence structure had already begun during the latter stages of World War II. On November 18, 1944, Major General William J. Donovan, the head of the Office of Strategic Services, sent President Roosevelt a proposal—subsequently known as the “Donovan Plan”—to place peacetime intelligence directly under the President’s supervision rather than under the military, as it was during the war. Under this proposal, the director of the central intelligence service would have full authority in intelligence matters, subject only to Presidential control. Although the plan provided for an advisory board consisting of the Secretaries of State, War, and Navy, and others the President might choose to appoint, its function was to be consultative only. Donovan’s paper signaled the beginning of the controversy over the national intelligence system.

During the debate over Donovan’s proposals in 1944 and early 1945, the Department of State was at best a peripheral participant. The protagonists were Donovan and the OSS on one side and the armed forces on the other. The Army and Navy opposed Donovan’s plan and by late 1944, the Joint Intelligence Committee of the Joint Chiefs of Staff had developed two schemes for a permanent intelligence organization, the so-called “military” (or “services”) and “civilian” plans. The former provided for a “federal” system of intelligence coordination among the State, War, and Navy Departments, with a supervising body composed of the Secretaries of State, War, and Navy; an “intelligence directorate” headed by a Department of State official (with Army and Navy deputies); and a joint intelligence service to conduct whatever activities of common concern were agreed upon by the three Departments. The “services” plan explicitly rejected a “national intelligence service.”

The “civilian” plan took a number of elements from Donovan’s proposal including a central intelligence agency with an independent budget, responsible for clandestine intelligence collection and for intelligence coordination. The Director would be appointed by the President but would be under the “direction and control” of a board consisting of the Secretaries of State, War, and Navy.

By late 1944, the Joint Intelligence Committee had instructed its Joint Intelligence Staff to refine the two plans and by the end of the year [Page 2] they had become a single proposal—designated as JIC 239/5—which provided for a national intelligence authority composed of the three Secretaries plus the Chief of Staff to the Commander in Chief; a central intelligence agency headed by a director appointed by the President on the recommendation of the national intelligence authority; and an advisory board to the director composed of the intelligence chiefs of the State, War, and Navy Departments.

In February 1945, the Chicago Tribune and the Washington Times-Herald published the texts of Donovan’s proposal and of JIC 239/5. The resulting public controversy over the propriety of a permanent intelligence system led the Joint Chiefs to “withdraw” JIC 239/5. Thereafter, the official Army and Navy position was that the subject was one to be dealt with after the war.

President Roosevelt took no action on the intelligence issue. When he died in April 1945, the future of intelligence was still uncertain. His successor, Harry S. Truman, was inexperienced in foreign affairs, mistrustful of secret government agencies, and opposed to the unorthodox operating procedures that characterized the Roosevelt–Donovan relationship during the war. The evidence of his thinking about postwar intelligence suggests that foremost in his mind was liquidation of the wartime intelligence apparatus as soon as possible.

With the end of the war in August 1945, the future intelligence organization became a pressing issue. The decision to abolish the OSS was taken before the end of August by a White House committee charged with the liquidation of wartime agencies. The committee consisted of Presidential Special Counsel Samuel I. Rosenman, Office of War Mobilization and Reconversion Director John W. Snyder, and Bureau of the Budget Director Harold Smith (although Smith apparently waited until September 13 to get the President’s final approval). (Memorandum of conference with the President by Harold Smith, September 13, 1945; Roosevelt Library, Papers of Harold Smith, Box 4, Conferences with President Truman 1945; see the Supplement)

Donovan wrote to the President on August 25, indicating he was moving swiftly on liquidating OSS and enclosing a “Statement of Principles” to govern a future intelligence agency. (Document 3) He wrote Harold Smith on the same date, urging the need for a successor organization. (Attachment to Document 3) In his letter to Truman, Donovan noted that the President had earlier indicated a desire to talk to him about the subject, but there is only second-hand evidence that the two ever had a substantive conversation about intelligence. (Document 5)

In the meantime, as it became clear that the debate over the postwar system was about to resume, the armed forces revived JIC 239/5. Under the redesignation JCS 1181/5 and with only a few changes, it became the [Page 3] Joint Chiefs of Staff proposal for the national intelligence system and the model that was ultimately adopted.

The Bureau of the Budget was also advanced in its plans and was about to become a major contender in the battle over the intelligence system. During the war, the Bureau had provided organizational and managerial consultation to various armed services intelligence programs and had worked during 1944–1945 to establish an intelligence office in the Department of State. By the time of Truman’s succession, the Bureau was considering the full range of postwar intelligence problems. It was beginning to draft guidelines for national intelligence organization, and in an organizational study of the Department of State prepared for Secretary James Byrnes had urged greater emphasis on intelligence.

By September 20, 1945, the Budget Bureau had its package of proposals ready and presented the President with two documents, both of which he signed immediately. The first was an order that abolished OSS and transferred its Research and Analysis Branch to the Department of State and the remainder of the organization (primarily its clandestine intelligence and covert action elements) to the War Department. As signed by the President, it became Executive Order 9621. (Document 14) The second document was Truman’s letter to Secretary of State Byrnes, making him responsible for taking the lead in establishing an interdepartmental group, headed by the Department of State, to develop “a comprehensive and coordinated foreign intelligence program” for his approval. (Document 15)

The third element in the Bureau of the Budget’s package was the study on “Intelligence and Security Activities in the Government,” also dated September 20 (see the Supplement) but not formally sent to Truman until October 31. (Document 38) It presented a detailed set of guidelines for a national intelligence organization. In general, the approach was to emphasize the coordination of existing information resources; to urge the use of existing intelligence components of cabinet departments as building blocks for a new intelligence system; and to downgrade the importance of separate, centralized agencies (although the Bureau recognized that some subjects of common concern might require a separate organization). The Bureau had a strong conviction that a peacetime intelligence organization should be centered in the Department of State and to this end encouraged the Department to create an intelligence component and to establish itself as the “lead agency” in forming the national intelligence community. In these last two endeavors, the Bureau had to contend with resistance in the Department of State where, with some notable exceptions (e.g., Under Secretary of State Dean Acheson), there was little enthusiasm for either enterprise.

Within hours of the signing of the Executive Order, the Joint Chiefs told the Bureau of the Budget that their own paper and recommendations [Page 4] on intelligence had been forwarded to the Secretaries of War and Navy for transmission to the President. They were informed that the order had already been signed.

The Federal Bureau of Investigation was also involved in the intelligence debate. In 1941, at President Roosevelt’s direction, the FBI became responsible for clandestine intelligence and counterintelligence in the Western Hemisphere and for security liaison with Western Hemisphere governments. This responsibility was reconfirmed by Roosevelt in 1942 and the FBI’s jurisdiction in the Western Hemisphere made exclusive (except for military and naval intelligence collection by service attaches). Thus, the OSS was debarred from clandestine intelligence operations in Latin America.

In December 1944, FBI Director J. Edgar Hoover proposed a future “world-wide intelligence system” to be run by the FBI along the lines of its Latin American operations. By September 1945, with major decisions on intelligence impending, Hoover sought to maintain the FBI’s Western Hemisphere activity and, if possible, to win support for the FBI plan. The attitudes of the Department of State were of particular importance to the FBI, since the Bureau’s intelligence work in Latin America was generally well regarded in the Department. (Document 5)

President Truman and Harold Smith did not agree with this view. Smith discussed the subject with Truman, who reacted negatively to the wide-ranging U.S. intelligence activities in Latin America, “and commented that if we continue our present attitude toward Latin American countries in this respect we will not be in a position to complain very much when they send their intelligence people into the United States. He then said with considerable vigor that he was ‘very much against building up a gestapo.’” (Roosevelt Library, Papers of Harold Smith, Box 4, Conferences with President Truman 1945, May 4, 1945)

Two months later, the subject arose again, and Truman, while approving continued funding for 6 months for the use of the FBI in South America, added “that he, himself, had some question, from the standpoint of good neighbor relations, about our having the FBI in South America.” (Ibid., July 6, 1945)

In September, Truman told Smith “that he thought the FBI should be cut back as soon as possible to at least the prewar level; that he proposed to confine the FBI to the United States; and that he had in mind a quite different plan for intelligence.” (Ibid., September 5, 1945; see the Supplement)

Nonetheless, the FBI continued to press its case for the “World-Wide Intelligence Service.” (Documents 7, 8, 10, 19, and 43) Because Truman had such strong views on the subject, Hoover’s efforts were ultimately unsuccessful. By the time the struggle between the Department of State and the military services over the new system had begun in earnest, the [Page 5] FBI apparently was no longer a serious contender. There continued, however, to be strong sentiment in parts of the Department of State—especially the Office of American Republic Affairs—for a continuation of the FBI’s Western Hemisphere role and this issue was revived the following year when the Director of Central Intelligence sought to establish exclusive control over all clandestine intelligence.

When Truman told Smith in their September 5 conversation that “he had in mind a quite different plan for intelligence” he implied that he had his own blueprint. On September 13, in another conversation with Smith, “The President again commented that he has in mind a broad intelligence service attached to the President’s office.” (See the Supplement) There is no record, however, that Truman ever conveyed to his advisers his ideas for an intelligence system, and the debate in the fall of 1945 became a contest between the Department of State and the Bureau of the Budget on one hand and the Army and Navy on the other.

By early October 1945, Secretary of the Navy James Forrestal was the government’s most energetic advocate of a national intelligence structure. At least two factors seem to have precipitated his active involvement. One was the looming issue of armed forces unification; although national intelligence organization was a distinct problem, many discussions of military unification raised the question of the future intelligence system. For example, the study that Forrestal himself commissioned on unification, the Eberstadt report, had argued for centralization of intelligence.

The second factor that influenced Forrestal was the speed with which decisions on intelligence were being taken and in particular the dominant role of the Bureau of the Budget and the Department of State. The impact of these developments was magnified by the failure of the Joint Chiefs’ proposals to reach the White House before the President signed Executive Order 9621 and the letter to the Secretary of State.

At the end of September, Forrestal and Secretary of War Robert B. Patterson had sent the JCS proposal to Byrnes, as coordinator of intelligence planning, saying they assumed he would send it on to the President. By October 10, Forrestal was trying to set up a dinner with the State, War and Navy intelligence chiefs and J. Edgar Hoover, to discuss a national intelligence agency. On October 13, Forrestal proposed to Patterson that they should push the JCS plan “vigorously at the White House.” (Document 26) On the same day, he wrote to Byrnes to underscore the importance of the intelligence problem, and to suggest that it be taken up in the periodic meetings of the Secretaries of State, War, and Navy. (Document 27)

The Army was also active. Late in October, on instructions from Patterson, Assistant Secretary of War for Air Robert A. Lovett organized a study group to consider the future intelligence structure. The Lovett [Page 6] Board report, November 3, 1945, marked an important step in the decision-making process on national intelligence. (Document 42) Although the committee prepared its report in little more than a week, it heard testimony and received written submissions from a number of persons who had had significant experience in the intelligence field, which may have given added weight and authority to its recommendations.

The Lovett Board acknowledged that it inclined toward the JCS proposals, but modified them in certain important respects. In part due to General John Magruder’s influence, the group departed from the Joint Chiefs’ plan by recommending that a future central intelligence agency should have an independent budget rather than be financed by allocations from the State, War, and Navy Departments, and that it should be the sole agency for clandestine intelligence collection. Lovett’s group also recommended a strengthened role for the Departmental intelligence agencies on the intelligence advisory board proposed in the JCS plan.

Thomas F. Troy, in his study entitled Donovan and the CIA: A History of the Establishment of the Central Intelligence Agency, concludes that the “Lovett report did for Secretary Patterson what the Eberstadt report had done for Forrestal: it provided an independent justification for the Secretary’s endorsement of the JCS plan.” (page 319) Meanwhile, the Department of State had not taken the lead on intelligence, as directed by Truman. Indeed, Byrnes’ two key subordinates in this field, Alfred McCormack, the Secretary’s Special Assistant for Research and Intelligence, and Donald S. Russell, Assistant Secretary of State for Administration, “had clearly come to a parting of the ways on what should be done internally; without internal cohesion and growth, State could hardly play a strong leader role.” (Ibid.)

Another problem was that the military thought that Colonel McCormack was simply too slow. McCormack’s style was deliberate, and he made a conscious choice early on to establish the State Department’s own intelligence organization before turning his full attention to the interagency theater. In part, the decision was dictated by practicality; it would be difficult for the Department of State to preside over intelligence planning if it had no intelligence component of its own. Moreover, McCormack faced the deadline set by Executive Order 9621—the interim arrangements had to be wound up and a successor unit established in the Department of State by the last day of 1945, a problem compounded by Congressional difficulties over appropriations.

But by the end of October there was already a rising tide of opposition to McCormack’s plan for a centralized Department of State intelligence component, with Russell organizing the opposition of the geographic and policy offices. Moreover, Russell and McCormack were at odds on the interagency organization. Russell had proposed to Byrnes an interagency structure that was considerably different from what [Page 7] McCormack was slowly designing and close enough to the JCS plan to make Forrestal say they were basically the same. Russell had complained to Byrnes when McCormack did not respond to a memorandum he had sent McCormack enclosing a copy of his plan. At the October 30 meeting of the Three Secretaries, Forrestal was authorized to discuss the JCS plan in detail with Russell, suggesting that he had become a channel between Byrnes and Forrestal.

On October 31, 1945, McCormack wrote to Admiral Leahy, Chief of Staff to the Commander in Chief, expressing his view “that this Department should formulate its own plans before going ahead with the interdepartmental group. That position has been acceptable to the Army and, I think, also to the Navy, and the Army has had a Board functioning for the purpose of determining its position on the post-war intelligence problem.” (Document 39)

Leahy and McCormack conferred on October 31 and the letter presumably followed up on their conversation. Leahy’s diary for that date contains the following entry: “By direction of the President, discussed with Colonel McCormack of the State Department the formation of a Central Intelligence Service. Colonel McCormack has a very different plan from that proposed by the J.C.S.” (Library of Congress, Manuscript Division, Papers of William D. Leahy, Diaries 1945, page 182)

Beyond this brief reference, no record of the discussion between Leahy and McCormack has been found, nor any record of Truman’s instruction to Leahy. Moreover, there is no specific evidence of what prompted Truman’s inquiry about McCormack’s efforts. In any case, the Leahy-McCormack exchange indicates that Truman had become concerned by the pace of developments. An even stronger White House intervention occurred a week later, when a memorandum was put before the President that said that “The only apparently promising prospect of getting useful action” was for “The President to call a conference with the Secretaries of State, War and Navy, and direct them to work together in the preparation of a plan for the establishment of a Central Intelligence Service,” to be completed not later than December 31. (Document 44)

The memorandum carries no indication of addressee or authorship (although Leahy is clearly a strong candidate for the latter), but it became a Presidential directive when Truman sent it to his secretary with the annotation “Set this up.” The minutes of the Three Secretaries’ meeting on November 14 (Document 45) show that Forrestal and Byrnes were aware of the memorandum and understood it as a directive. Byrnes’ comments at his November 27 staff committee meeting (Document 50) suggest that the President had told him some action should be taken on intelligence and that Leahy had been given responsibility.

Although there was still some distance to go (and the contemplated meeting of the President with the Three Secretaries was never held), the [Page 8] November 7 directive broke the impasse over intelligence planning. In effect, it overrode the President’s September 20 letter with a new instruction ordering the Secretaries to prepare a plan jointly. Admiral Leahy’s role also weighed in favor of the Army and Navy, especially the latter. Moreover, by specifying a plan for a “Central Intelligence Service” the directive was weighed against the Budget Bureau/State Department side of the debate.

By mid-November, McCormack had a Department of State plan ready for consideration. (Document 46) Following closely the Budget Bureau proposals of 2 months earlier, the plan proposed a dual structure for coordinating foreign intelligence and security intelligence, based on a complex system of committees. The plan did not rule out the establishment of centralized agencies to perform common functions; in fact, it envisaged that certain common services would have to be performed centrally. But McCormack’s plan tended to regard centralization as a last resort, with the burden of proof on those arguing for it. This underlying attitude, which saw coordination of existing activity as the essential problem, separated it fundamentally from the Lovett Board/JCS blueprints. In one other respect, too, they were fundamentally at odds. McCormack took seriously the injunction in the President’s September 20 letter to Byrnes to “take the lead in developing a comprehensive and coordinated intelligence program for all Federal agencies” and proposed arrangements that would put the Department of State in firm control of the foreign intelligence structure.

At this point, Byrnes was beginning to have second thoughts about the Budget Bureau guidelines. In his meeting with the Secretaries of War and Navy on November 14, the day before McCormack formally presented his plan, Byrnes had already expressed doubts about the Budget Bureau approach and appeared to endorse the idea of a central intelligence agency.

By early December, McCormack’s plan had undergone two revisions in an effort to meet objections. On December 3, Byrnes sent Patterson and Forrestal the second revision of McCormack’s “Annex I” which set forth a proposed basic structure for the national intelligence system. (Supplement) In many respects the thrust of McCormack’s original proposals was not altered, but there were concessions to Army and Navy concerns. Even in the first (November 28) revision (see the Supplement) McCormack had replaced the original two coordinating authorities (for foreign and security intelligence) with a single National Intelligence Authority (NIA) consisting of the Secretary of State as chairman and the Secretaries of War and Navy.

Other revisions provided that while the Executive Secretary (the chief operating officer of the Authority) would be named by the Secretary of State, the appointment would be subject to the approval of the [Page 9] Secretaries of War and Navy. In addition, a new provision spelled out that the Executive Secretariat would “Act as the executive for the Authority in carrying out such program and all operating plans approved by the Authority,” phrasing that emphasized the executive and operational as distinct from the coordination and deliberative aspects of the plan. There were other changes as well that presumably were intended to stress the same point—the right to inspect departmental intelligence operations was added to the NIA’s powers and the language about the NIA’s power to determine which intelligence functions should be conducted centrally was strengthened.

More important, perhaps, than these changes was Byrnes’ covering memorandum, which defended the basic structural proposal while suggesting that in practice there would be possibilities for development along lines presumably attractive to the Army and Navy. Thus, Byrnes noted: “The Plan does not preclude any centralized intelligence operations…which may prove feasible and desirable as the program of the Authority is developed.” And he added: “With respect to clandestine activities (‘secret intelligence’ and ‘counter-espionage’) I understand the prevailing opinion to be that such operations, if they are to be conducted, might well be under a central agency; and the Plan sets up machinery for study of that problem in detail and for development of specific operating plans in those fields.” (Document 54)

However significant these modifications may have been from the Department of State perspective, it is doubtful that they even began to meet the armed services’ concerns. The elaborate committee structure still remained, as did the explanatory comment that said that “Such Committees will be the primary means by which the Authority will carry out its missions.” Moreover, despite the positive references to centralized operations, Byrnes’ memorandum made a distinction between a single central intelligence agency for the performance of agreed common functions and the centralization of agreed intelligence activities. The former was rejected because it was too conspicuous and would compete and overlap with existing intelligence agencies. The language used in connection with the latter suggests that the preferred approach was “either (1) by vesting responsibility for a particular field in a single existing agency or (2) by bringing together the working units of several agencies on a subject into a joint staff under direction of the Authority.” Both approaches differed from the ideas underpinning the Joint Chiefs/Lovett Board proposals.

Finally, the revised McCormack proposals and the covering memorandum still provided for Department of State predominance in the intelligence field. For the armed services, this remained a major stumbling block.

[Page 10]

There was more fine-tuning of the McCormack proposals and by mid-December the indicators were somewhat mixed. The War Department seemed to be wavering on some points and to be exploring possibilities of compromise. In so doing, the Army was perhaps reflecting a greater awareness than the other participants that the Strategic Services Unit (SSU)—the name under which the OSS apparatus for foreign secret intelligence had been lodged in the War Department—was a diminishing asset as time went by and uncertainty about the future led to a loss of experienced personnel. (See paragraph 2, Document 59.)

There were no indications that the Navy had softened its position. Of the three Cabinet Secretaries involved, Forrestal seems to have had the strongest convictions about intelligence. Moreover, the Navy had an ally at the White House in the person of Admiral Leahy. Byrnes was aware of this. At his November 27 Staff Committee meeting, he had said “it was necessary to bear in mind that whatever we propose will have to be ‘sold’ to the President. It will be turned over by the President to Admiral Leahy, who has strong views on the matter and… we must, therefore, seek to avoid unnecessary questions of jurisdictional conflict with the Army and the Navy.” (Document 50)

As 1945 drew to a close, the debate was plainly deadlocked. After the December 26 meeting of the Three Secretaries, McCormack had spoken to Harold Smith of “the apparent hopelessness of proceeding without further direction from the President.” (Document 65) McCormack said as much again after his December 27 meeting with the War and Navy Department representatives, telling them at the end of the session that “matters were at an impasse and that he felt the question would have to be decided by the three Secretaries and by the President.” (Document 65)

The White House in fact had become involved again. The documentation is sketchy and neither Truman’s Memoirs nor Admiral Leahy’s diaries provides any details. But toward the end of December, Truman apparently set in motion the developments that would break the deadlock, by seeking the advice of Rear Admiral Sidney Souers, Deputy Director of Naval Intelligence. (Document 64) Additional details are in Troy, Donovan and the CIA, pages 339–340; and a letter from Souers to Montague, August 13, 1970, in Central Intelligence Agency Records, No. 206102, Job 84–T00286R, Box 2, Folder 12.

During this period, one element of the problem was Secretary Byrnes’ availability in Washington. Byrnes had been away for meetings in Moscow from mid-December until almost the end of the month and he was due to leave for London on January 7, 1946. This time constraint was partly responsible for forcing the pace of decisionmaking on the intelligence issue at the end of 1945 and in the first week of the new year. On Sunday, January 6, 1946, Byrnes met with Forrestal and Under Secretary [Page 11] of War Kenneth Royall at his apartment in the Shoreham Hotel. According to Arthur B. Darling’s recounting of “the tale still going the rounds,” Forrestal told Byrnes: “Jimmy, we like you, but we don’t like your plan. Just think what might happen if another William Jennings Bryan were to succeed you in the State Department.” (Darling, The Central Intelligence Agency: An Instrument of Government, to 1950 (University Park: The Pennsylvania State University Press, 1990), page 70)

In an oral history interview almost 9 years after the event, Admiral Souers was less colorful. He recalled simply that Forrestal and Royall went to Byrnes’ apartment on Sunday afternoon and told him that they were prepared to submit their own plan for intelligence to the President, whether or not it had Department of State endorsement. According to Souers’ account, Byrnes approved the plan with one or two minor word changes and it was submitted to the President the next day. (Truman Library, Papers of Harry S. Truman, Post-Presidential Memoirs, Sidney W. Souers interview with William Hillman and David M. Noyes, December 15, 1954)

No contemporary record of the January 6 meeting has been found and none of the participants put down his recollections at a later date. What seems to have been presented to Byrnes was a letter, dated January 6 and signed by Royall and Forrestal. (Document 66) The letter rejected the Department of State plan and proposed two alternatives. The first was a modification of the Lovett Board proposal, with the central intelligence agency “at the outset” to “be housed for administrative purposes in the State Department,” headed by an official who would be or become a State Department official “unless the President otherwise determines”, and staffed with personnel detailed from the three Departments, i.e., it would not be an independent agency with its own budget.

If this were not acceptable, the letter proposed as a “possible alternative” acceptance of the Department of State plan as modified up to that point, with certain further modifications: the Executive Secretary would be appointed by the National Intelligence Authority and would be or become a Department of State official unless the President decided otherwise; the Executive Secretary would have no other Department of State duties; a specific provision that “evaluation and synthesis on a national level, direction of foreign espionage and counterespionage, and appropriate dissemination” were central functions; elimination of the provision that committees would be the primary means by which the Authority would operate; and provision for a single advisory group, to have the membership and functions recommended by the Lovett Board.

The letter’s sparse marginalia suggest that Byrnes worked through the first alternative, making only minor changes. The proposal thus approved was basically the JCS plan of September 19, 1945 (Document 13), and thus was closer to the Navy position, which had always rejected [Page 12] the independent budget concept, than the Lovett Board proposal. It was also close to the draft directive that Souers had sent Clark Clifford on December 27. The only genuinely new elements in the January 6 letter were the ties between the Department of State and the new intelligence agency: the provision that it would be “housed for administrative purposes in the State Department” and that its chief executive would be, or become, a State Department official. But these provisions had disappeared by the time the proposal was signed by Byrnes, Royall, and Forrestal and sent to the White House the next day. (Document 68) Presumably Byrnes dropped them during the Sunday afternoon discussion.

The letter from the three Secretaries to the President was received at the White House on January 8. Admiral Souers, and probably Clark Clifford, immediately started reworking the draft directive. (Library of Congress, Manuscript Division, Papers of William D. Leahy, Leahy Diaries, January 8, 1946; see also footnote 2, Document 68)

Budget Director Harold Smith seems to have been unaware of the Sunday meeting. But on Monday, January 7, he saw Byrnes briefly before the latter’s departure for London, probably about the State Department’s internal battle over how to organize its intelligence component. At their meeting, Smith learned “about a proposed Executive Order disposing of the matter of the organization of intelligence activities in the Government.” (Document 69)

On Tuesday, January 8, Smith telephoned the President’s secretary, Matthew Connelly, and asked that Truman not sign the order. The following morning, Connelly telephoned to tell Smith that the President had called a meeting on intelligence which was about to convene. Smith hurried across the street to join the meeting. Besides the President and Smith, the other participants were the President’s advisers, “and several persons from the Navy Department.” (Document 69; see also Library of Congress, Manuscript Division, Papers of William D. Leahy, Diaries, January 8, 1946)

Smith’s account (written in the third person) added that “the War Department and the State Department were not represented. The implication of most of the statements made at the meeting was that intelligence could not be handled in the State Department because that Department was too weak.”

Smith by his own account “took the part of the devil’s advocate,” arguing that “when a subject is left to three departments to divide up among themselves, the worst possible compromise results” and that “in all that he had heard on the subject there did not seem to be even a clear understanding of what kind of intelligence was being discussed.” Leahy commented curtly in his diary that “The Director of the Budget offered [Page 13] many objections that evidently were instigated by the Department of State.” (Ibid.)

Truman’s account of what is probably the same meeting (although it could possibly be a January 12 White House meeting on the same subject) is in the second volume of his memoirs. (Memoirs: Years of Trial and Hope, page 57)

On January 12, 1946, there was another White House meeting and a revision of the January 8 version of the draft directive. Leahy records that “The President conferred with his advisers, including Judge Rosenman and the Director of the Budget on the establishment of a Central Intelligence Agency.” (Library of Congress, Manuscript Division, Papers of William D. Leahy, Diaries 1946, page 4) A copy of the draft directive in the Truman Library has revisions that are apparently in Souers’ hand. The January 12 version simply established the Central Intelligence Agency rather than leaving the National Intelligence Authority to do so, as the January 8 draft provided. Consequently, paragraph 4 of the January 8 version, which made the submission of an organizational plan the first responsibility of the Authority and the Intelligence Advisory Board, was eliminated and the subsequent paragraphs renumbered. In old paragraph 5(a), an additional sentence was added at the end: “Full use shall be made of the staff and facilities of the established departmental intelligence agencies.” Finally, in old paragraph 7, the last sentence was changed to read: In the conduct of its activities the National Intelligence Authority will be responsible for fully protecting intelligence sources and methods.” A note on the revised draft indicated it should be cleared with the Bureau of the Budget and Attorney General Clark. (See the Supplement)

On January 18, Assistant Solicitor General Harold Judson had a 3-hour session at the White House with the drafters of the intelligence directive. Whether this was the only such session is not known but the version considered at the January 18 meeting differs in several respects from the January 12 text, suggesting that the White House may have worked with the Justice Department (and others) to develop the new draft. The January 18 version, with handwritten changes made at the meeting that day, is attached to a memorandum from “M.E.H.” to Clark Clifford, January 18, 1946. (See the Supplement)

The January 18 text was no longer an executive order but a Presidential letter of instruction to the Secretaries of State, War, and the Navy. The language concerning the NIA’s powers seems slightly weaker; the draft no longer establishes a Central Intelligence Agency but instead authorizes the three Secretaries to “designate persons from your respective Departments who shall collectively form a Central Intelligence Group and who shall, under the supervision of a Director of Central Intelligence, assist the National Intelligence Authority.” (Ibid.)

[Page 14]

Judson returned to the Justice Department with a marked-up version of the January 18 draft which included further changes made during the meeting. He met immediately with FBI Assistant Director D.M. Ladd, who was unhappy with the provisions demarcating responsibility between the FBI and the foreign intelligence machinery. (Memorandum from Ladd to Hoover, January 18, 1946; National Archives and Records Administration, RG 263, Records of the Central Intelligence Agency, Troy Papers, FBI Documents; see the Supplement)

Among other problems, Ladd was concerned about paragraph 9. In the January 18 version this had originally read:

The White House meeting had dropped the first sentence, leaving the last as the complete paragraph. (Ibid.)

Ladd protested that this and other changes “would leave no protection against this new national intelligence authority expanding into full investigations in the domestic field.” Judson replied that “the President did not desire to replace or duplicate the FBI and that those drawing up this directive had no intention of doing so.” Ladd countered that “these men may be gone and others who take their place are the ones who will be interpreting the phraseology here.” (Ibid.)

Finally, Judson drafted a new paragraph 9: “Nothing herein shall be construed to authorize the making of investigations, otherwise than as provided by existing law and Presidential directives, inside the continental limits of the United States and its possessions.” (Ibid.)

Ladd was satisfied with this formulation, but when Judson telephoned Clark to clear it “the Attorney General suggested milder language in that the President had just this morning advised members of the Cabinet that because of the strikes, he was so burdened with work he desired them to avoid conflicts which would necessitate his becoming involved for decisions. The Attorney General, therefore, thought that this was not a good time to interject such blunt language as here suggested and he suggested the phraseology, ‘Nothing herein shall be construed as affecting the present jurisdiction of any United States agencies inside the continental limits of the United States or its possessions.’” (Ibid.)

Ladd thought that this was “a negative statement which merely meant that although the Bureau would not be precluded from conducting the same investigative work in which it is now engaged, it would not [Page 15] prevent this new outfit from duplicating and coming into the same field.” (Ibid.)

Ladd’s account does not mention any further exchange with the Attorney General, but notes that Judson concurred in the FBI view. That afternoon, Judson telephoned the text of his proposed paragraph 9 to Clark Clifford’s office and with a minor variation in word order, it remained in the directive as finally issued. (Ibid.; Memorandum from M.E.H. to Clark Clifford, January 18, 1946; see the Supplement)

At this point, the directive was close to its final form. There were more adjustments of language over the next few days, and on January 22 the Presidential letter of instruction to the Secretaries of State, War, and the Navy was signed by Truman, bringing a national intelligence system into being. (Document 71)