Introduction
Coinciding roughly in time with the controversy over the creation of a national intelligence structure, a heated and at times bitter struggle in the Department of State took place over whether the Department should have its own intelligence organization, and, if so, how it should be organized and to whom it should answer. Within the Department of State, the two issues of national and Department intelligence organization were in fact closely linked.
Under the pressures of wartime, the Department had developed a few rudimentary intelligence operations. A Division of Foreign Activity Correlation handled liaison with the FBI and OSS, provided such Departmental and Foreign Service support as they required, and disseminated their reports in the Department. For matters that required scrutiny at a higher level, beginning with Assistant Secretary Adolf A. Berle, Jr., an officer of that rank seems to have had, as part of his portfolio, a general watching brief over intelligence matters. The Department also had a representative on the Joint Intelligence Committee of the Joint Chiefs of Staff, an arrangement that at least gave it some voice in what was then the only part of the U.S. Government where something resembling national intelligence assessments were being prepared. But there appears to have been no real Departmental backup for this process other than whatever backstopping the Department’s representative might elicit from the policy offices.
By 1943–1944, there were stirrings about the future of the U.S. intelligence system in the postwar world. In 1944, an officer of the Division of World Trade Intelligence proposed the establishment of an office of intelligence analysis and for a period of a year or so a somewhat desultory discussion of the proposal went on in the Department. The Bureau of the Budget encouraged the plan and worked with its proponents, but the initiative seems to have produced nothing except exchanges of memoranda and draft Departmental orders.
Secretary of State Edward R. Stettinius, Jr., showed some interest in intelligence during his brief tenure. At the time of President Roosevelt’s death in April 1945, Stettinius was meeting with Secretary of the Navy Forrestal and Attorney General Francis Biddle on a plan for postwar intelligence organization, but it is not clear whether he was involved in efforts to create a Departmental intelligence capability. By June he had been replaced by Secretary Byrnes.
Thus at the end of World War II, the Department of State still had no intelligence organization in the modern sense of the term. The Foreign Service was, of course, the major source of foreign information and intelligence [Page 181] available to the U.S. Government, but the analysis and interpretation of the information was the responsibility of the Department’s policy offices. There was no independent mechanism within the Department for assessing and correlating information.
At the time of his appointment, Byrnes asked the Bureau of the Budget for a study on the State Department’s organization and management. Chapter VII of the report, entitled “Intelligence and Research,” described the Department’s operations, both foreign and domestic, devoted to the collection and analysis of information and concluded that “the Department has not fixed the responsibility in any office to determine what information is needed and to ensure that it is brought together in a form that is reliable, adequate and available for use.” (Report by the Director of the Bureau of the Budget, August 15, 1945; see the Supplement)
To meet this problem, the report recommended that “The Department should establish, as an adjunct to the Office of the Secretary, an Office of Intelligence and Research to plan a coordinated program of foreign intelligence and research and maintain a central intelligence service.”
Byrnes apparently discussed this report with Dean Acheson at the time he persuaded him to return to the Department as Under Secretary. Acheson, who believed that Byrnes was committed to the report, was convinced of the need for an intelligence system in the Department and was to become one of its strongest advocates.
By this point, events were beginning to move quickly. On September 5, 1945, Assistant Secretary for Administration Frank McCarthy recommended to Acheson that the Department act on the Budget Bureau’s proposal by appointing a special assistant to the Secretary to handle intelligence. (Document 72) But other developments overtook this recommendation and, indeed, the original Budget Bureau proposal. By late August, the decision had been made to break up OSS and transfer its Research and Analysis Branch to the Department of State.
When the President signed Executive Order 9621 on September 20, making the transfer effective on October 1, the Department of State thereby inherited its first intelligence organization. (Document 14) A week later, Acheson announced the appointment of Colonel Alfred McCormack, a New York lawyer with a distinguished wartime record in the Military Intelligence Division, as the first Special Assistant to the Secretary of State for Research and Intelligence.
The Executive order was somewhat more complicated than a simple transfer. It established within the Department of State an Interim Research and Intelligence Service (IRIS), which was to remain in existence until December 31, 1945. The Secretary of State was required to wind up the affairs of IRIS by that date, but in the process he was authorized [Page 182] to transfer any of its functions to any agency within the Department that he might designate. In other words, he was authorized to dismantle the Research and Analysis Branch but to reassemble it by the end of the year as a new organization within the Department.
The September 20 Executive order was accompanied by a Presidential letter of the same date which conferred on the Secretary of State the responsibility to “take the lead in developing a comprehensive and coordinated foreign intelligence program for all Federal agencies concerned with that type of activity.” (Document 15) The Department of State thus had acquired a second responsibility that was closely linked to the establishment of its own intelligence organization.
Inevitably, the two aspects of McCormack’s responsibilities became entangled. He was inclined to hold back on plans for an interagency structure until he launched the Department’s own organization, but the pressures from the other agencies involved compelled him to move on both fronts. The chief critic of McCormack’s plan for a Departmental intelligence organization, Assistant Secretary of State for Administration Donald Russell, also became a player in the interagency maneuvering over the future intelligence community by proposing his own plan and engaging in certain contacts with Secretary of the Navy Forrestal which, although sanctioned by Byrnes, worked against McCormack’s position. Finally, the strong resistance within the Department to a centralized “in-house” intelligence organization became a factor in the interagency debate, with Secretary of War Patterson in particular opposing any State Department leadership of the national intelligence structure if the Department’s own intelligence system were decentralized to the policy offices.
When McCormack took office, he already had a functioning intelligence research and analysis organization, although it had to be slimmed down and many of its staff were anxious to return to private pursuits. He had to find a way to integrate it into the Department’s structure.
McCormack’s plan was for a centralized intelligence component, a set of geographic and functional research offices that would roughly parallel the policy offices, working closely with them but remaining independent of their control and steering clear of involvement in policy. By late October 1945, this concept was running into opposition within the Department and in Congress where funding for the intelligence program was being drastically reduced.
On October 27, 1945, Under Secretary Acheson called a meeting of senior Department of State officers to discuss the future of the intelligence question. As Acheson recalled, it was his and McCormack’s intention to head off opposition to the intelligence plan, which would create a unit of slightly over 1,000 people from the former Research and Analysis Branch. (Present at the Creation: My Years in the State Department (New [Page 183] York: W.W. Norton, 1969), pages 159–160) Acheson regarded the geographic policy offices, conservative elements in the Department, and Assistant Secretary Russell as the main opponents. They were promoting a decentralized intelligence system in which responsibility for intelligence analysis (and some but not all of the 1,000 or so intelligence personnel) would be assigned to the policy offices. According to Acheson’s account, Spruille Braden and Loy Henderson, Chiefs of the Latin American and Near Eastern Divisions, respectively, combined with Russell and his deputy, Anthony Panuch, at the October 27 meeting to argue that the plan for creating an intelligence office was a duplication of effort and a waste of money. (Ibid., page 160)
Spruille Braden charges that Acheson and McCormack were clearly trying to force acceptance of an Office of Research and Intelligence without adequate consultation with the policy offices. (Diplomats and Demagogues: The Memoirs of Spruille Braden (New Rochelle, NY: Arlington House, 1971), pages 346–348) Braden also raised the question of the loyalty and political sympathies of some of the employees the Department had inherited from OSS (an allegation that was to reappear in the debate). In Braden’s view, the October 27 meeting was a successful revolt against Acheson and McCormack. Both Acheson and Braden agree that the immediate result of the meeting was to delay a final decision on issue.
Within a week of the October 27 meeting, Russell laid the issue before Secretary Byrnes. In a November 3 memorandum to Byrnes (Document 81), Russell quoted the following language from the Bureau of the Budget’s recent report on Intelligence and Security Activities in the government: “Extreme centralization of the intelligence operation is no more workable than would be the centralizing in one agency of the job of producing all statistics for the Government.” (Report, September 20, 1945; see the Supplement)
Russell went on to argue: “If a centralized over-all governmental intelligence unit is not workable—the Budget’s conclusion—then it follows that the proposed plan of extreme centralization in the Department itself is not workable.”
The President’s directive of November 7 (Document 44) required McCormack and Byrnes to consider plans for an interagency intelligence organization, but the internal controversy in the Department of State broke out again in December. On November 29, Russell appointed a working group to “conduct a survey within the Department of State to determine where in the Department there should be intelligence activities.” (Memorandum from Nelson to McKay and Simmes, November 29, 1945; see the Supplement) The working group proceeded to draw up a report and recommendations for an “Intelligence Advisory Board.” The Board (not to be confused with the Intelligence Advisory Board established under the Presidential Directive of January 22, 1946) apparently [Page 184] had been established before Acheson’s October 27 meeting on intelligence but had held only one previous meeting when it convened on December 19, 1945, to consider the working group’s report. The report contained a majority recommendation for a centralized intelligence organization and an alternative proposal, submitted on behalf of the geographic policy offices, under which intelligence analysis would be decentralized to them while other intelligence functions (e.g., acquisition and dissemination) would be centralized. No copy of the report has been found, but it is summarized in the draft minutes of the Advisory Board’s December 19 meeting. (See the Supplement)
The Advisory Board, on which the policy offices had substantial representation, split evenly. The chairman, Sherman Kent of IRIS, cast the deciding vote in favor of an interim, centralized Office of Research and Intelligence to meet the problem of the imminent disestablishment of IRIS while leaving open for further study “the method by which research and intelligence should be organized in the Department.” To this end, the Board also recommended that its own existence be extended. The only record found of the Board’s December 19 meeting is a set of draft minutes which was circulated for corrections under cover of a note dated December 26, 1945. (See the Supplement)
Although the Intelligence Advisory Board had thus answered the immediate question of what to do when the December 31 deadline arrived, it was obviously aware that it had not solved the problem. On the contrary, its proceedings had underscored the depth of the division over the issue and especially the strong resistance of the policy offices to an independent analysis function in the Department of State.
On December 29, 1945, Russell reminded Byrnes and Acheson by memorandum that the order establishing a successor to IRIS had to be issued by January 1, at the same time commenting that “an irreconcilable difference of opinion exists.” Russell went on to argue “that research at the geographic level must be under the immediate direction of those who are to use it. In my judgment, the divorce of research from the policy action taken after the evaluation of information will … breed confusion and disorganize the operations of the Department.” (Document 82) Assistant Secretaries Braden and Dunn followed up with memoranda of their own on December 31, in which they argued for immediate transfer of the intelligence analysis function. (See the Supplement)
McCormack, returning the fire on December 31, asserted that decentralization “would end all possibility of organized State Department intelligence, and the President’s idea of State Department leadership in government-wide intelligence could not be attained.” (Document 83)
At this point, Byrnes was planning to leave for London and was under pressure to reach a consensus on the national intelligence structure. On January 5, 1946, he informed Russell that he wanted McCormack’s [Page 185] plan for a centralized organization adopted temporarily but on the express understanding that a final decision on “the ultimate location” of the Office of Research and Intelligence would be made by March 1, 1946. (Document 84)
Russell prodded Byrnes again on February 25 and Byrnes again postponed a decision. (See Document 85 and the source note thereto.) By mid-April the issue was still unsettled. The dispute had spilled into the press. The Secretary had given the principals a deadline of April 16 to reach agreement; a “mediator” acting at McCormack’s behest was trying to resolve differences (and was criticized for acting as an advocate of McCormack’s views rather than as a negotiator); and Russell, or one of his associates, had polled the heads of the political offices, asking them to indicate their preference for McCormack’s or Russell’s plan (all of the signatures were in the Russell column). (Document 87) On April 18, Russell reported to Byrnes that “the discussions now in process hold no substantial promise of reaching a settlement.” (Document 88) Finally, after further fruitless “mediation” efforts and one more memorandum from Russell to Byrnes, the Secretary issued an order on April 22, 1946, adopting Russell’s plan for a decentralized organization. On April 23, McCormack resigned. (Document 90)
The operation of the Department’s intelligence analysis organization under the Russell plan was short-lived and it is not clear how wholeheartedly it was applied in practice. For one thing, the decentralized unit continued to be colocated rather than physically dispersed to their respective policy offices. Nor is it clear how enthusiastic McCormack’s successors were about the system they had inherited. His immediate successor, Professor William Langer, was an interim replacement who served only a few months, and Langer’s successor, Colonel William Eddy, was quick to recommend abandonment of the plan as soon as Russell was no longer on the scene.
In any case, Russell resigned as Assistant Secretary for Administration effective January 20, 1947, one day before Byrnes left office. According to Acheson’s account, immediately following General George C. Marshall’s swearing-in as Secretary of State, Marshall requested that Acheson stay on as Under Secretary for a time and then asked him if there were any urgent matters requiring decision. Acheson mentioned two, one of them the need “to reverse Mr. Byrnes’ deplorable decision to split up the intelligence work among the geographic divisions.” Marshall, Acheson wrote, “needed no long explanation of what should be done” and directed that the necessary paperwork be undertaken immediately. (Acheson, Present at the Creation, pages 213–214) The Departmental instructions recentralizing the intelligence offices were issued on February 6, 1947, and, in Acheson’s words, “Thus a year too late my recommendation to Secretary Byrnes was put into effect and his [Page 186] own unhappy action of the preceding April undone.” (Ibid.; and Department of State Bulletin, February 23, 1947, page 366, and March 23, 1947, pages 556–559)
After Marshall’s decision, the documentary materials on the internal history of the Department’s intelligence organization are far less focused than the 1945–1946 documents on the contest over the nature of the organization. In part, this suggests that with the end (or suppression) of the controversy, the Special Assistant for Intelligence and his staff were freer from distractions and able to concentrate on their job of analyzing intelligence.
Secretary Marshall’s decision did not obliterate either the controversy or the attendant uncertainty over the future of intelligence, but it did create a different context for these issues within the Department. The persistence of some of the old attitudes was reflected in the Dulles Report of January 1949 (Document 358) which suggested that the Special Assistant for Research and Intelligence and his organization did not enjoy the confidence of the Department’s policy offices. To deal with this problem the report proposed that a separate system be created within the Department for the handling of intelligence estimates, a proposal that was never acted on.
With the settlement of the internal controversy, a principal concern for the Department’s intelligence organization was the delimitation of Department of State and CIG/CIA responsibilities in the area of intelligence research and analysis. This issue and the related one of designing a national mechanism for intelligence estimates are treated in sections on the history of the Vandenberg and Hillenkoetter directorships of the Central Intelligence Agency.