DCI Hillenkoetter: Soft Sell and Stick
CIA HISTORICAL REVIEW PROGRAM
RELEASE IN FULL
22 SEPT 93
No Foreign Dissem
Central intelligence becomes an agency, still struggling to establish its position and function.
DCI HILLENKOETTER: SOFT SELL AND STICK
Arthur B. Darling1
The man to succeed General Vandenberg at the head of the President's information service had been under consideration for some time. Though often credited with the choice, Admiral Souers took no part in selecting Admiral Roscoe H. Hillenkoetter. Another personal representative of the President, Admiral Leahy, did.
A New DCI
While Ambassador at Vichy from 1940 to 1942, Admiral Leahy had formed a high regard for his naval attach6's skill in working with the French underground. Hillenkoetter was expert at helping patriots escape into Africa and acquiring information from both French and German sources. "He never got caught." Earlier, he had been so successful with Ambassador Bullitt in Moscow and in Paris that the State Department wished to keep him; the Navy, according to Leahy, had had to recall him into service so that he might learn something about ships.
In 1942 Hillenkoetter organized an Intelligence Center at Pearl Harbor for Admiral Nimitz and won his commendation. Then General Donovan tried to get him to take charge of OSS operations in the Pacific, but the Navy would not release him. After the war he returned to Paris, where he was engaged in collecting intelligence when he was ordered, against his wishes, to Washington to become Director of Central Intelligence.
Admiral Leahy and Navy Secretary Forrestal recommended Hillenkoetter to their fellow members of the National Intelligence Authority2 when the Army asked to have Vandenberg returned for high command in the nascent USAF. It was on February 17, 1947, that the NIA and the President approved this assignment for the recently promoted Rear Admiral, effective when Vandenberg should leave. The date is to be noted. It was but five days after Vandenberg had been named, as DCI, executive agent for the Secretaries of State, War, and the Navy in intelligence matters, an event which was to have a decided effect upon Hillenkoetter's administration of the Central Intelligence Group and its successor Agency.
From the point of view of central intelligence, it may well be said that General Vandenberg should not have been called back into military service at that time. He had been in charge of the CIG for less than a year; its new offices were not in full working order. He had just established in the minds of the departmental secretaries that the DCI ought to be their executive agent. He had not convinced the departmental chiefs of intelligence that he was an officer above them, not under their control. They were not accepting the distinction between advice and consent. They were still, as Vandenberg left and Hillenkoetter came in the spring of 1947, endeavoring to have the business of the DCI come before them, constituting the Intelligence Advisory Board, for consent or dissent on its way to the National Intelligence Authority.
General Eisenhower, responsible in large part for the recall of Vandenberg to the Air Force, may have known little or nothing of these matters. But he readily agreed, testifying on the pending CIA legislation that May, that frequent change was wrong; there should be stability in the office of Director of Central Intelligence. Three years, Eisenhower then thought, should be the least term of service, subject of course to exigencies.
One may wonder why the third DCI was not drawn from the Department of State. It was, so to speak, the State Department's turn after one each from the Navy and the Army. There was in fact a plan fostered in the State Department to make Allen W. Dulles the first civilian DCI. Robert Lovett had mentioned him to the secretaries in November 1945; his effective work for OSS in Switzerland had attracted attention. He was publishing a book on his experiences. His views had been sought on scientific intelligence by the Technical Advisors of the Joint Research and Development Board. He was soon to give testimony before Congress on the need for central intelligence and its possibilities as a civilian career.
No answer to the question can be final. There were personalities involved. President Truman's own ideas and opinions of men had been at work when he abandoned the Office of Strategic Services, let Donovan return to his law practice, and established the Central Intelligence Group; then in the first days of CIG Secretary Byrnes raised objections which led Truman to emphasize that this was his personal information service and Admirals Leahy and Souers his personal representatives.3 Congressional antipathies toward the Department of State were endemic for many reasons, including the suspicion that it was infested by radicals. But the most important factor may have been the influence of the Army and the Navy, supported by Admiral Leahy, who were uneasy at the prospect of a preponderant central intelligence organization now moving from the basis of executive order to institution by law.
Legal Status and Practical Problems
The Central Intelligence Agency would be created by the National Security Act of July 26, 1947, whose main purposes were the establishment of an independent Air Force and the unification of the military departments under a Secretary of Defense. Much work had gone into drafting an enabling act for the CIA, but it was decided that the time was not ripe for so complete and detailed a measure. Some of its provisions were too controversial and subject to attack by other agencies; objections would at least delay the passage of the unification act.
In the draft National Security Act proposed by the President, the brief section devoted to the Central Intelligence Agency provided essentially only that such an agency, under a DCI, should take over from the CIG the functions it was performing under executive direction and should be responsible to the National Security Council, replacing the National Intelligence Authority. The Congress, however, because of fears that such a carte blanche to the Executive might some day be abused, added amendments in which some functions, prerogatives, and limitations of the new Agency and the DCI were spelled out.
Headed by a DCI who might be drawn from either military or civilian life, the Agency, according to the Act now before Congress in final form, was to advise the NSC on the policies and objectives of the national intelligence mission and make recommendations for the correlation of departmental intelligence activities. It would correlate, evaluate, and disseminate the national intelligence product. It would have no police powers and would not infringe on the internal security functions of the FBI. All intelligence relating to the national security should be open to the inspection of the DCI, and he was responsible for safeguarding the sources and methods of intelligence. He could terminate any CIA employee at his discretion. The Agency should perform such services of common concern as the NSC might judge to be most efficiently so centralized, and it should have other functions and duties related to intelligence that the NSC might assign.
These stipulations followed generally those of the President's Directive of January 22, 1946, which had set up the CIG. There were slight modifications and differences in emphasis. There was now less distinction made between the functions of the Agency and those of the DCI with respect to the departmental intelligence agencies, and the DCI was no longer empowered to inspect the activities of the latter. His right to fire employees at discretion was new; strictly, in CIG he had no employees of his own. A significant omission from the Act was any provision for an Intelligence Advisory Board composed of the departmental intelligence chiefs to work with the DCI. He was empowered to appoint an advisory committee if he wished.
But the departmental chiefs were determined that the IAB should be perpetuated and should have governing functions. They had been annoyed that Vandenberg insisted upon being in a sense their superior, certainly not their servant. Now Hillenkoetter, a newcomer among the admirals and generals, was made at once sharply aware of the animus toward Vandenberg for getting himself designated the executive agent of the Secretaries of State, War, and the Navy. The military men let Hillenkoetter know that the Army and Navy had been in existence a long time while he was merely head of a civilian agency but recently established.
Vandenberg had urged that the DCI be designated Advisor to the NSC in the National Security Act. This suggestion was too controversial, but the concept remained in the CIA function of advising the NSC on intelligence matters pertaining to the national security. Admiral Hillenkoetter thus had authority from Congress to advise the National Security Council if he chose without first consulting a board of departmental intelligence chiefs—unless of course the NSC should direct him so to consult. This direction the members of the expiring Intelligence Advisory Board were determined to obtain.
The new DCI furthermore inherited complicated relationships with the joint Chiefs of Staff and the Research and Development Board in the national military establishment, and with the Atomic Energy Commission quite apart from the latter.4 The production and delivery of scientific intelligence—vital to all three, the joint Chiefs, the RDB, and the AEC—would have been difficult enough to accomplish had there been complete cooperation among the departmental intelligence services and the central intelligence organization, as there was not.
Matters were still to be arranged in detail with the Federal Bureau of Investigation, particularly with regard to the coordination of counterespionage activities. The problem would trouble the CIA Office of Special Operations for some time to come.
Admiral Hillenkoetter had also to contend with internal turbulence and disagreement due to ceaseless rows among the ambitious or pertinacious or zealous men who are found in any young and growing enterprise. There was friction between the interdepartmental Coordinating and Planning Staff and the Office of Reports and Estimates. Within ORE, the conflict between its Branches and Intelligence Staff had brought about a reorganization in less than a year after its establishment to replace the Central Reports Staff.5 Boundaries between the Office of Operations and the Office of Special Operations were still to be marked at every point. Within OSO, just completing the absorption of the Strategic Services Unit which had preserved some wartime assets of the OSS, there were plans to shift secret operations from a functional to a geographic organization.
Something had to be done to stop the Intelligence Advisory Board's interminable bickering and delay over every issue. The Interdepartmental Coordinating and Planning Staff which General Vandenberg had established to work for him with the IAB had been effectively thwarted in that purpose by the interposition of successive ad hoc IAB committees. This battle continued as Hillenkoetter took up the task of revising the directives of the old NIA and CIG for the new National Security Council. In view of the great amount of work done on them through the previous year, this should have been a relatively simple task. It proved to be far from that.
Three Anti-Centralization Thrusts
When Admiral Hillenkoetter took responsibility as DCI and head of CIG on May 1, 1947, Admiral Inglis with close support from General Chamberlin was pressing measures in the Intelligence Advisory Board which General Vandenberg had opposed. One of these would disperse the production of intelligence among the departments according to their dominant interests, as had been done for collection.6 To complicate the issue, controversy over air intelligence was rising between the Navy and the Air Forces which were about to become a separate department under the National Security Act. The Navy wished to keep its own air intelligence.
Another measure called for redefining "strategic and national policy intelligence" notwithstanding the fact that an explicit formula had just been established by the National Intelligence Authority.7 The production of this final intelligence—coordinated national estimates for the makers of policy—was the responsibility of the Director of Central Intelligence. Admiral Inglis, however, would have it made clear that the control of "operational" intelligence was still reserved to the armed services. In other words, the DCI would have to produce his national estimates without access to items of military information, however pertinent to those estimates, if the service chose to withhold that knowledge from his estimating staff. Inglis maintained that "strategic" intelligence and "national policy" intelligence were separate and distinct, not the unitary concept of General Donovan, who had originated the phrase, and others who had followed him in central intelligence.
The third measure to greet Admiral Hillenkoetter as he came to his first meeting with the IAB on May 15, 1947, was Admiral Inglis' plan, first submitted the preceding February 20, to have all recommendations of the DCI to his superiors pass through the IAB. The agenda for a meeting of the NIA should be referred beforehand to the IAB together with copies of all papers to be considered. IAB members should either informally express concurrence or furnish comments to the DCI for transmission to the NIA. On important matters any member might request a formal IAB meeting to discuss proposals before they were submitted to the NIA.
The counterpart paper prepared in CIG on behalf of the DCI emphasized that the IAB was advisory to him. He was not responsible to it but to the NIA. He was not at liberty to reveal to the IAB all recommendations which the NIA requested of him. This of course was so true with respect to budgeting and expenditure that the departmental intelligence chiefs, though curious, never sought to interfere in these matters, thus prejudicing their right with regard to others.
On the first of the three measures, decentralization of production, the new DCI endeavored to maintain the position of his predecessor that he should have supervision over the production of intelligence by the several agencies. The IAB, however, agreed to successive phrasings of a directive which left out all specific reference to the DCI and stipulated merely that the production work should be done. The minutes of the meeting do not state that Admiral Hillenkoetter expressed disapproval, but neither do they record his assent. Although he let the action of the IAB pass for the moment without disapproval, he still had the right to recommend his own ideas to the NIA along with the IAB proposal.
The hope had been that there would be few split opinions in the IAB, that differences would be removed in its deliberations so that the NIA would receive from the DCI and his advisers considered and concerted judgments, the result of true coordination. It was ideal to talk of thus resolving problems and reconciling opposite views. But the right of decision had to belong either to the DCI or to a majority of the Board. In terms of political science, sovereignty must reside somewhere, either in the will of an individual or the tyranny of a majority.
The issue was clearly drawn. Admiral Hillenkoetter inherited from General Vandenberg the view that the position of the DCI was the official central intelligence position regardless of dissents. Admiral Inglis, General Chamberlin, and other chiefs of intelligence in the departments clung to the opinion that they had inherent right to make the IAB a working staff of the National Intelligence Authority which their secretaries constituted with Admiral Leahy, personal representative of the President. In accordance with this theory, the IAB was entitled to know the agenda of the NIA in advance. Navy Secretary Forrestal had come to this conclusion by June 26, 1947, when the NIA met for the last time.
Agent of the Secretaries
The theory had been given some support by Admiral Leahy. In the preceding July he had admonished Vandenberg that the President held the departmental secretaries primarily responsible for coordinating intelligence activities. He had advised Vandenberg to drop the word "agent" and put in its place the phrase "act for" the NIA.8 Since then, however, he had modified his position to endorse Vandenberg's designation as executive agent of the secretaries.9 Now in the NIA meeting of June 26, 1947, Leahy stood by this position, and when interviewed on the subject in 1952 he still favored the idea that the DCI should be individually responsible. There should be room for dissent, he said; the policy-makers had the right to accept the dissenting view. But—and he made no distinction between estimates and other questions in this respect—the DCI alone was responsible for the central intelligence opinion, in questions of coordination as well as other matters.
In the historic final meeting of the National Intelligence Authority on June 26, 1947, however, Admiral Hillenkoetter declared before Secretaries Marshall, Patterson, and Forrestal and Admiral Leahy that the DCI did not need the authority which they had given to General Vandenberg on February 12, to act as the executive agent of the Secretaries of State, War, and the Navy. Its revocation, Hillenkoetter said, would create better feeling with the agencies represented on the Intelligence Advisory Board. If he should need the authority in the future he would be the first to request its reinstatement. Both General Chamberlin and Admiral Inglis were present to hear this abnegation.
Secretary Patterson, who had favored the grant modified by the right of an aggrieved agency to appeal through its secretary, would offer no objection to the withdrawal if the authority were no longer needed. Admiral Leahy remarked that he saw no reason for withdrawing it though he would agree if Hillenkoetter wished to have it revoked. Secretary of State Marshall was concerned to know if the withdrawal would adversely affect the relationship between the central organization and the departmental agencies; Hillenkoetter replied that on the contrary he expected the relationship to improve, and Marshall appeared to be satisfied. Secretary Forrestal, whose assent in February had been fairly reticent, came out now with a definite stand; the DCI's authority to issue orders in the names of the secretaries, he said, made CIG look like a Gestapo and caused unnecessary friction. Further discussion was not recorded. It was agreed to withdraw the authority.
Last NIA Directive
Admiral Inglis followed up the advantage which Admiral Hillenkoetter had given him. He argued at length his plan for IAB control of DCI recommendations in Hillenkoetter's second meeting with the IAB on July 17, 1947. There were present the usual representatives of the intelligence services and the aides who served on their ad hoc committees. Mr. Eddy, for the State Department, took the side of the DCI, But General Chamberlin, along with General McDonald for the Air Forces, supported Admiral Inglis. The IAB was intended to be something more than an advisory council, they said: it had authority to commit the departments to action; it brought their intelligence services into cooperation with the central agency.
The success of intelligence in government, declared Chamberlin, was dependent entirely on cooperation. He called for an ad hoc committee to draft a new paper. General McDonald supported him, and so an ad hoc committee of the familiar persons took over once more the job of trying to reconcile the fixed views of the intelligence chiefs and the concepts of the DCI.
There was much discussion in the same meeting on the origin of the concept "strategic and national policy intelligence." Admiral Inglis made clear that he was willing to accept the concept so long as control over "operational" intelligence was not taken from the armed services. It was finally agreed that his view should prevail until the joint Chiefs of Staff had finished reorganizing their joint Intelligence Committee. At that time the definition of "national" as distinct from "departmental" and of "strategic" as distinguished from "operational" intelligence might be agreed upon among most if not all interested parties.
At the next meeting of the Intelligence Advisory Board—on July 31, five days after the President approved the National Security Act—it agreed to ask that the National Security Council, when formed, should continue all of the directives under which CIG and the IAB were functioning until it could make such changes as it saw fit. According to the Act, its provisions should go into effect one day after the Secretary of Defense took office or the sixtieth day after it became law, whichever came earlier. This gave time for the IAB and its ad hoc committee to finish up the business of gaining control of the DCI. The result was one more NIA directive under the old setup before Forrestal became Secretary of Defense—on September 17, 1947, in the midst of rising tension over Russian activity against the Marshall Plan.
The report of the ad hoc committee reconciled the views of the DCI and the IAB by finding for the most part in favor of the latter. It should have some governing power. All recommendations from the DCI to the National Intelligence Authority and its successor, the National Security Council, should be submitted to the IAB in writing, with the necessary attachments and with voting slips denoting concurrence, dissent, or the request for an IAB meeting. Its members should have seven working days to consider each subject. Any recommendations, proposals, or other papers which any IAB member might originate should similarly be sent to the others. A recommendation by two or more members would be submitted to the NIA or NSC along with the opinion of the DCI.
A suggestion of the ad hoc committee for incorporating that part of the Fifth NIA Directive which authorized the DCI to act for the NIA "in coordinating all Federal foreign intelligence activities related to the national security" met resistance from both the Army and the Air Forces. It involved control over matters of espionage and counterespionage which the Army was not yet willing to concede in any form to the Central Intelligence Agency. This part of the ad hoc committee's plan for control by the IAB was therefore deleted.
Admiral Hillenkoetter allowed the report of the ad hoc committee on Admiral Inglis' original proposal, thus modified, to become the Eleventh NIA Directive on September 11, 1947. It was but nine days before he took office under the provisions of the Act of Congress, when the National Intelligence Authority ceased to exist. Why he did not withhold his approval of a measure placing him under the restrictions of his Advisory Board is not to be explained by a desire to reverse Vandenberg's policy. He had under consideration at the time a program for continuing the essentials of that policy. He stood ready to accept advice and to safeguard the right of dissent but would not yield his independent right of making recommendations to his superiors in the National Security Council.
Perhaps it would have been to his ultimate advantage if Hillenkoetter had settled the issue then and there. But he found tempers so high and feelings so hard that, as he put it later, he preferred to indulge in a little "chicanery" and let the Eleventh Directive go through. After all, he said, both sides must have known that it would not remain determining. He hoped that in time everybody would cool. Whether or not the DCI was called the executive agent of the secretaries was not of great consequence.
On the same September 11, 1947, Admiral Hillenkoetter sent to the Secretaries of State, War, and the Navy and Admiral Leahy a memorandum of suggestions for the first meeting of the National Security Council which they were about to constitute, with a copy for Admiral Souers who had been named Executive Secretary of the NSC on August 17. Hillenkoetter suggested that he and his associates in CIA should prepare papers on a set of new NSC directives within sixty days following the establishment of the NSC. This administrative detail had not been discussed with the IAB in accordance with the provisions of the Eleventh NIA Directive. Moreover, Hillenkoetter went on to suggestions of policy which also had not been discussed with the IAB. At least we have yet to find evidence that they had.
One was that there should be a subcommittee of the National Security Council to act as the NIA had acted in control and supervision of the DCI and CIG. The idea had been discussed in the congressional hearings, where Allen Dulles especially had advocated a small governing authority over the DCI and CIA and where Donovan of course still insisted upon having only one responsible officer, the Secretary of Defense, between the DCI and the President. Hillenkoetter suggested that the subcommittee to furnish "the active direction" might be merely the Secretaries of State and Defense. His alternative was to add the Secretaries of the Army, Navy, and Air Force, but he preferred not to, so that the Department of State would not be overshadowed by the military establishment. And then he proposed that the Director of Central Intelligence should sit with the National Security Council as "observer, counsel, or advisor," to keep in touch with the thoughts of the NSC and to answer its direct questions.
In August there was some uncertainty whether Hillenkoetter would continue the Intelligence Advisory Board. By September 19, however, its members had been informed that he intended to use his authority under the National Security Act and have an advisory committee to help him carry out his functions and those of his Agency. He told them that he wished also to readjust the Interdepartmental Coordinating and Planning Staff so that it might work with a standing committee of the new Intelligence Advisory Committee. This one standing committee for the departmental intelligence chiefs would take the place of the successive ad hoc committees they had been using. Members of the Standing Committee would remain in their respective agencies but be ready on occasion to go over to CIA and confer with ICAPS. ICAPS too would be composed of representatives from the departments, but these men, as officers in CIA, would not always be able to vote according to the wishes of their departments. The hope was that the Standing Committee for the IAC and ICAPS for the DCI would somehow be able to reconcile differences and reach coordinated recommendations. But it did not work out as Hillenkoetter hoped. The Standing Committee was to behave like its predecessors, the ad hoc committees of the old IAB.
On September 19 Hillenkoetter sent formal recommendations to the National Security Council for its first meeting on September 26: all of the NIA and CIG directives should continue in full force until changed; CIA should have sixty days in which to submit revisions. He presented his plan for the new Intelligence Advisory Committee in a separate memorandum. Then on the following day, September 20, 1947, he took office as the Director of Central Intelligence under the National Security Act.
The 1AC Clash
Hillenkoetter's General Counsel advised him on July 29, 1947, that under the National Security Act, just approved by President Truman, the DCI as head of CIA was "solely responsible for the performance of the Agency's duties." He therefore could go to the NSC without waiting upon advice from a committee. Any committee which he chose to have would be his own. Its membership might be supplied from the respective intelligence services, but they would sit at his request. They might take adverse opinions to their own departmental heads, who constituted the NSC with other appointees of the President. But his advisory committee would in no sense be a governing board to control his thought or action. There was no idea that its members had first to give their consent before he could proceed.
Admiral Hillenkoetter, in his memorandum of September 19 to the NSC, accordingly pointed out that he was not obligated to continue the old Intelligence Advisory Board. He might have a committee which for all intents and purposes would continue the Board, but it would be more subject to his control. He requested therefore that the National Security Council should authorize participation by members from the several departments in what he named "the Director of Central Intelligence's Intelligence Advisory Committee." He would have representatives of the State Department, of the Army, Navy, and Air Force, and of the Atomic Energy Commission to serve as the permanent members of the committee. Others would come at his invitation.
The DCI would supply the secretariat for the committee. It would meet on his call as chairman, Any dissent by a member of the committee would be formally recorded so that it would accompany the DCI recommendation to the NSC. The DCI would avoid discussion of any matters on which members had not yet studied the related papers and obtained if possible the opinions held in their respective departments. He wanted a concerted opinion before making a recommendation to the NSC. In deference to the wishes of Admiral Inglis and General Chamberlin, Hillenkoetter's proposal provided also that recommendations from two or more members of the committee would be presented to the Council.
The first response to Hillenkoetter's suggestions of which we have record came on September 23 from Robert A. Lovett, Acting Secretary of State. (Secretary Marshall was attending the Assembly of the United Nations in New York.) The proposal for a subcommittee of the Secretaries of State and Defense to handle CIA affairs for the NSC found favor with Lovett, but he wished to add the personal representative of the President in order to make it an authority comparable to the old NIA. The DCI should attend as a non-voting member. It would be desirable also to have him present in the meetings of the full NSC.
Lovett wrote that the DCI should consult with an advisory board to insure "prior consideration by the chiefs of the intelligence services" of matters which should come before the NSC. This made clear that the Department of State wished there to be an advisory board for coordination at the so-called working level; the secretaries ought to have the benefit of its deliberations when they discussed intelligence matters in the NSC. There was no implication, however, that the DCI should be subject to a governing board of the departmental intelligence officers.
The second reaction to Admiral Hillenkoetter's program came in the National Security Council on September 26. It adopted his recommendations that the old directives remain in full force and that sixty days be allowed in which to submit any necessary revisions. The Intelligence Advisory Board, therefore, continued to have legal standing until it should be replaced by a new directive. It was decided that the DCI should attend all NSC meetings as observer and adviser. He was authorized to submit the CIA budget for 1949 to the Bureau of the Budget.
No action was taken at this meeting with regard to establishing a subcommittee to direct the Agency, nor is there record of opinion on Hillenkoetter's proposal for an advisory committee. But there is evidence elsewhere that there could have been considerable discussion of these DCI suggestions in the first meeting of the NSC. Secretary Royall of the Army wrote on October 6 that the subcommittee was incompatible in his view with the purpose of the National Security Council, which was supposed to operate as an entity on all matters within its cognizance, giving broad directives to the DCI.
This statement might have been construed as an invitation to Admiral Hillenkoetter to manage the CIA as he saw fit, looking to the NSC only for guidance in large matters of policy. But it was quite apparent that the Secretary of the Army did not so intend. Royall was taking exception to the suggestion of a small governing board consisting of the Secretaries of State and Defense to the exclusion of the Secretaries of the Army, Navy, and Air Force. His demurrer was closely related to other things to come. Admiral Souers knew this well, and prevailed upon Admiral Hillenkoetter to withdraw the suggestion on October 17.
Hillenkoetter's suggestion of a new advisory committee came under revision in the office of the NSC's Executive Secretary. Both Admiral Souers and his assistant, James S. Lay, had been in the Central Intelligence Group from its beginning. Together they changed the wording in several places with Admiral Hillenkoetter's consent. The DCI was made to recommend rather than request. His proprietary emphasis on the DCI's advisory committee was softened—"the Intelligence Advisory Committee proposed by the Director of Central Intelligence."
Souers eliminated the provision that recommendations from two or more members of the IAC were to be sent by the DCI to the National Security Council. He did this on his own responsibility as Executive Secretary, on the grounds that the chiefs of intelligence already had proper access to the NSC through their superiors, the secretaries of the departments. This provision, drawn from the Inglis plan, is not to be confused with the stipulation that the dissents of IAC members be submitted to the NSC along with the opinion of the DCI. The DCI might properly report disagreements with his own position that came out in the deliberations of his advisory committee; but there was no reason he should be used as a vehicle for the initiatives of the departmental chiefs of intelligence. As Hillenkoetter remembered it, after some discussion he flatly refused to relay to the NSC their proposals that he opposed.
Approvals of the plan for the advisory committee as modified by Souers and Lay came back to the Executive Secretary from Secretary of Defense Forrestal on October10, from Secretary of the Air Force Symington on October 17, and from Acting Secretary of State Lovett and Chairman of the National Security Resources Board Hill on October 20. No replies from the Secretaries of the Army and Navy are filed with these, but a memorandum from the Secretary of the Army for the Executive Secretary was sent on November 26, 1947, through the office of the Secretary of Defense. Forrestal's Special Assistant kept a copy in forwarding it before the Secretary had yet seen it.
Secretary Royall opposed Hillenkoetter's plan. He declared that the DCI had been required by the first NIA directive to refer all recommendations through the intelligence Advisory Board, which therefore not only performed the service of advising the DCI but also insured that there would be full departmental coordination of all matters before they were submitted to the NIA. He insisted that the new Intelligence Advisory Committee should have a mandatory review of the same nature. It was due notice that in the NSC the Army would support General Chamberlin and Admiral Inglis rather than Admiral Hillenkoetter.
On Wednesday, December 3, a formal communication from the National Security Council to the DCI enclosed Secretary Royall's memorandum and requested DCI comment on it for "concurrent consideration." Hillenkoetter did comment within the week, orally before the departmental secretaries and chiefs of intelligence, Souers, and Forrestal in the office of the Secretary of Defense. Hillenkoetter remembered this conference vividly, he said, as one of the dramatic moments in his life. He could recall the words almost as they were spoken. But let us bring to the same point the parallel story of the NSC directives before we enter into the historic occasion.
The NSCID Battleground
It was evident by this time in another quarter that the intelligence services of the armed forces were entrenching against Hillenkoetter as they had in the preceding fall against Vandenberg. Members of the Interdepartmental Coordinating and Planning Staff and representatives of CIA operating offices had gone systematically to work revising and consolidating the old NIA and CIG directives according to the September 26 instruction of the National Security Council. The NSC directives "NSCIDs," were to lay down the principles, and directives issued by the DCI, "DCIDs," would carry the relevant administrative orders.
Drafts of the new measures were ready by October 16 and circulated for discussion within the Agency on October 20. Three days later revisions had been completed and forwarded to the DCI's deputies, assistants, and legal counsel for further suggestion. The directives were practically in order for submission to the Intelligence Advisory Board, as Acting Secretary Lovett had advised be done. There had been no concealment of this activity within the Agency. The Director had sent a memorandum about the undertaking to the departments on October 9.
The only reply to the October 9 memorandum came from W. Park Armstrong, Jr., Acting Special Assistant to the Secretary of State and representative of the Department on the IAB, and it reflected indirectly the displeasure of the departments at this activity in the Agency. It gave also a direct view of the tension between the State Department and the members of the military establishment. It proposed that the new directives should define intelligence in conformity with the concept of national intelligence which Vandenberg had got approved on February 12, 1947 by the National Intelligence Authority.10 This would hardly please Admiral Inglis.
Armstrong urged moreover that the DCI's right of inspection over the operations of the intelligence agencies as well as their materials be restored, as provided in the President's Directive of January 22, 1946. But Armstrong would go farther and specify that the DCI should determine the causes of omissions, inadequacies, or duplication and propose corrective measures to the NSC. The suggestion may not have been just a broadside aimed at the chiefs of intelligence in the armed services. There were suspicions in CIA that Armstrong was thinking of his own Department, where the chiefs of geographical areas opposed his work in research and intelligence just as they had McCormack's.11
Armstrong's letter, however, was trained on yet another target. The State Department had in common with the armed services antipathies toward the Central Intelligence Agency. The proposal in regard to the DCI's right of inspection carried with it a repeal of the section in the Fifth Directive authorizing the CIG to undertake research and analyses.12 Armstrong would have "centralization of functions" only when, by agreement among the departments and CIA, such functions could be "most beneficially and effectively accomplished on a central basis." His contention was that the intelligence agencies of the departments should each produce finished intelligence in the fields of their dominant interests, and the DCI should perform the inspection to see to it that they did. CIA would not produce national intelligence from source materials which it had processed for itself but from the finished departmental intelligence.
Along the same line of reasoning, Armstrong proposed to abandon the directive of February 12, 1947, in which CIG had laid down the national requirements for the collection of intelligence on China.13 These, classified as economic, social, political, scientific, and military, were matters of primary interest to the respective departments and not the immediate concern of the central agency. With this suggestion, Armstrong could count upon entire agreement from General Chamberlin.
The CIA reply to Armstrong on November 3 was lacking somewhat in candor. It said that almost everything he desired had been incorporated in the drafts which would be complete and ready for delivery to the departmental chiefs of intelligence on November 10. Actually, there was no provision in the drafts for CIA's abandoning research and analysis on "source materials" and depending on finished intelligence from the departments for the construction of national estimates. The directive with regard to the national requirements for collection in China, too, was to be incorporated in a new NSCID. But then Armstrong was to have another chance in the IAB and, as it proved, in yet another ad hoc committee if he wished to press his case.
Telephone calls went out to the members of the IAB by November 13, inviting them to a meeting with the DCI on November 20, 1947, to discuss the proposed NSC and DCI directives. These were to go to the NSC on November 26, as it had directed in its first meeting two months before.
Stalling in the IAB
The conference of the intelligence chiefs with Admiral Hillenkoetter on November 20 was notable, but not for analysis and criticism of the proposed directives. The chiefs spent time arguing whether or not they were the Intelligence Advisory Board. It was idle discussion. The IAB continued to exist so long as the First and Eleventh NIA Directives remained in effect, and the NSC had decided on September 26 that the old directives should continue in full force at least sixty days. Admiral Hillenkoetter may have befogged the issue by pointing to the fact that there no longer was an NIA to which the Board might report, but he himself accepted the IAB as such when he accepted its request that the proposed NSCIDs and DCIDs be referred to an ad hoc committee for discussion with ICAPS.
The meeting of the IAB on November 20 was notable, then, for the demeanor of its members. The naive observer might have thought that they were relatively uninformed and so piqued at being taken by surprise, that being conscientious men who did not like to be unprepared for their duty, they were provoked because they were not ready. We have followed these men and their aides, however, through a year of meeting and maneuvering over the duties and responsibilities of the Director of Central Intelligence, the facilities and functions of the central intelligence organization, and the relationship which the DCI was supposed to have with the departmental intelligence agencies.
The thought is hard to resist that the IAB members were present on this occasion not to cooperate in an enterprise of common concern but to take exception to this, to be hesitant over that, in a word to stall and delay. They had had not just the ten days since receipt of the November 10 memorandum to consider the proposals. They had been engaged with almost all of these matters as members either of the IAB or of its ad hoc committees for more than ten months.
Instead of pressing his case for a quite different concept of central intelligence Armstrong announced that the State Department had not been able to arrive at a firm position on the directives. Secretary Marshall was leaving for London; Mr. Lovett had the matter in hand, but the Department could not be committed as yet. Hillenkoetter hoped that the members of the Board, as heads of the intelligence agencies, might be able at least to get through the first four directives, having to do respectively with the duties and responsibilities of the DCI, with collection, with production, and with the objectives of national intelligence. But Armstrong insisted that as these matters were coming before the National Security Council, the head of each intelligence agency must have the position of his department established before he could speak finally.
This was turning the function of the advisory committee upside down. If this were the true interpretation, the DCI could not seek the advice of his advisory committee with regard to his recommendations to the secretaries in the NSC until they themselves had made up their minds on what they wished their intelligence officers to advise the DCI to advise them. Admiral Inglis cut straight to the point. As chief of Naval Intelligence, he said, he was not the "mouthpiece of the Secretary." What Inglis had to say in the meetings was his own opinion.
But Admiral Inglis did not like the present situation. The Agency had produced these papers "on its own initiative and its own authority"; they should have been considered by the IAB while being formulated. Had they been new papers, there would have been more strength in his argument. But they were revisions of things with which Inglis himself had been conversant for much more than a year.
General Chamberlin concurred that the procedure being used by CIA here was difficult for them. Chamberlin had other things to do as chief of intelligence for the Army; he could not turn his responsibility off or on "at somebody else's command." As far as he himself was concerned, he said, he had come "unprepared to discuss these papers," but his subordinates had worked on them and found many things to challenge. There were "important differences of principle," he said, that had never been approached, nor included in past directives., What those principles were he did not specify.
Later on, however, he revealed a marked difference in principle between Admiral Inglis and himself over the propriety of expressing their own views in the advisory committee. General Chamberlin objected to a procedure in the committee which might "drive a wedge" between the chief of intelligence in a department and his superior the secretary; this completely destroyed "command principles." He would be inclined, he said, "to keep quiet at all times" because he would be afraid that an action in the committee would be appealed over his head.
General McDonald for the Air Force then unleashed his statement. Analysis of the proposed directives had revealed to him, he said, that it was going to be necessary for him to recommend many changes. It would be impossible to cover the directives that afternoon; no attempt therefore should be made to go into either their philosophy or their composition. There should be an ad hoc committee "for the purpose of reconciling views." And so there was another such committee, made up of the familiar aides who by now must have been expert on the philosophies, the details, and most assuredly the conflicting views. This was to become known as the Ad Hoc Committee.
Admiral Gingrich, who was relatively new, having come on the IAB to represent the Atomic Energy Commission, went to the heart of the situation. "One point I might mention, Hilly," he said, "there doesn't appear to be any provision in these first two directives here for an Intelligence Advisory Committee, or Board, such as is executed under our old setup." Hillenkoetter replied that the law gave him the power to appoint an advisory committee. All present must have known that he had submitted his plan to do so on September 19. The members of the IAB in any case were not to be diverted from their intention to have the NSC direct him to establish such a committee as they wished. The matter went to the Ad Hoc Committee.
No one could have been surprised at the revisions of the drafts by that group. The changes, often small in detail, were persistently designed to restrict the DCI, to make him defer to the Intelligence Advisory Committee, and to remove his supervisory authority over the departmental intelligence agencies. The new draft of NSCID 1 outlined the duties of the DCI and his relations with the IAC: The IAC, consisting of the intelligence chiefs from the Departments of State, Army, Navy, and Air Force, the Joint Chiefs of Staff, and the Atomic Energy Commission, was to advise the DCI on "all recommendations and advice" to the NSC and upon his own directives or administrative orders for carrying out the NSC directives. He should act for the National Security Council "through the Intelligence Advisory Committee." The coordination of intelligence activities should be accomplished "by recognizing primary departmental requirements and by supporting the intelligence agencies."
The DCI was to disseminate intelligence subject to the security regulations of the agency in which the information had originated. He was to perform services of common concern as determined with the IAC. He was to obtain personnel from the departmental agencies in agreement with their intelligence chiefs. He was to arrange with the latter for "surveys and inspections of departmental intelligence activities."
Here the Ad Hoc Committee had made a slip, but it was soon corrected. Though Armstrong for the Department of State recommended DCI inspection of departmental intelligence activities, the chiefs of intelligence for the armed services could not contemplate such interference with their operations. When the measure came from the meeting of the IAB on December 8, the words "intelligence activities" had given way to "intelligence material," and control of even such inspection by the department concerned was restored before NSCID 1 was issued by the NSC on December 12.
Admiral Hillenkoetter notified the Ad Hoc Committee on November 25 that he could not accept as a whole its revisions in the drafts of the NSCIDs. He called another meeting of the intelligence chiefs for December 8 to consider the changes which he would make in the committee's proposals, and he sent those changes to them on December 1 so that they might bring to the meeting whatever statements of nonconcurrence they chose to submit.
Records are not available for all of the determining events between the action of the Ad Hoc Committee on November 24 and the meeting of the IAB with Hillenkoetter on December 8. But memories of some who participated are still keen. There is recorded evidence to show why the chiefs of intelligence were in a different mood from that of November 20. And the minutes of the meeting on December 8, steno-typed but never circulated, have come to light.
When Hillenkoetter learned on November 26 of Secretary Royall's opposition to his plan for the new Intelligence Advisory Committee, he went to Forrestal. Forrestal had approved the plan in October. There followed some days of telephoning and conversation, and then Forrestal called a meeting of the armed service secretaries and their chiefs of intelligence, a representative of the State Department, and Souers. Hillenkoetter was there to explain his plan for the operation of the Central Intelligence Agency under the National Security Act. Forrestal had declared to Congress in the preceding spring that the Agency, after the Security Council, would be the most important institution in the forthcoming Act.
As Hillenkoetter recalled the scene in the office of the Secretary of Defense, Forrestal asked for no opinions when Hillenkoetter had finished his statement but turned to Admiral Inglis and General Chamberlin on one side of the table. Forrestal did not include the representative of the State Department in his glance. Nor, apparently, was the representative of the Air Force in his line of fire. He spoke directly to Inglis and Chamberlin: "You are not going to interfere with this thing," he said. "It is going to run as Hillenkoetter says. Do you both understand that now?" Hillenkoetter was quite sure of that last question and of the remark to him later by Admiral Inglis: "He talked to us like a couple of plebes. I guess that makes us your servants now." The record is not yet clear whether this meeting came before or after Forrestal received a note written on Friday, December 5, by Vannevar Bush, head of the Research and Development Board. It seems likely that Bush's letter arrived shortly after he had told the military and naval chiefs of intelligence what they were not to do. In any event, the statements by Bush, chief adviser to the Secretary of Defense on scientific matters, added weight to Hillenkoetter's authority when he met with the chiefs of intelligence again on Monday, December 8, to discuss their differences over the NSCIDs.
Bush declared with effectively restrained language that the Central Intelligence Agency was not in a good position to provide scientific intelligence to the Atomic Energy Commission, and Mr. Souers should be warned of the situation. To amplify his statement, Bush included memoranda from officers in his organization who were in direct touch with events. His chief of intelligence reported that under the leadership of the State Department's representative the Ad Hoc Committee was seeking more authority for the Intelligence Advisory Committee. The director of his program division reported that the intelligence chiefs wanted an executive order, apart from the NSC directive, to establish the IAC as the "governing committee" of the CIA. In this situation the officer responsible for scientific intelligence in the Agency was "completely stymied." Bush urged that someone "at the highest level" determine the relationships between CIA and the "operating" services so that the production of "information, detailed intelligence, and integrated strategic intelligence" could proceed.
Hillenkoetter read this statement in the meeting of December 8. The response of General Chamberlin was that he was not conscious there was any question whether the Intelligence Advisory Committee was to be a controlling or an advisory body. "I frankly admit," he said, "it is an advisory body"; others might have a different opinion. He felt that he had authority to commit his own department "on certain things" in agreement with other members of the IAC and carry out the decisions "loyally without any command to do so." Thus the IAC could eliminate a lot of "minutiae" but would still be an advisory committee.
The position was tenable, and it was logical after the admonition from the Secretary of Defense that CIA was going to run as the Director of Central Intelligence said. But there had been some evolution in the thinking of the Army representative since the meeting on July 17 when he had called the IAB "a little more than an advisory body," and that on November 20 in which he maintained that the DCI's right of appeal to the secretary of a department over the head of its chief of intelligence destroyed "command principles."
There was action in the meeting of December 8 which the stenotypist could not record. The memory of Hillenkoetter's General Counsel, who was present, is certain on that score. Although the record is one of friendly words in half-finished sentences, Hillenkoetter's demeanor was as strong as Lawrence Houston ever saw him use. Houston sat where he could observe faces and catch fleeting expressions. As General Chamberlin's overtones conveyed his acknowledgement that "Hilly" was the boss, Houston saw an aide of the General "turn white." Admiral Inglis sat shaking his head in an unmistakable "no."
The representative of the Navy had attended nearly every meeting of the IAB from its beginning under Souers in February, 1946. Admiral Inglis had insisted all along that in most respects CIG, and then CIA, should be a cooperative interdepartmental activity. To the statement by Vannevar Bush that the Agency had to be either "almost completely self-sufficient" or "a small coordinating body" surrounded by strong departmental agencies, his response was now that there could be a "middle ground" for the Agency. As an "integrated operating agency," he said, it should have as little interference from the Advisory Committee as possible; the IAC should be "purely advisory, and absolutely nothing more." But in the relationships between the central agency and the departmental intelligence agencies, it should go beyond advisory capacity; it had something to do with "liaison, coordination and implementation." That was his "philosophy," he said, "for whatever it was worth."
There was further discussion in general terms. But the remainder of the meeting on December 8 was given for the most part to examining in detail revisions which the Ad Hoc Committee had made in the NSCIDs. For the time being the Director of Central Intelligence had his way. With the exception of the change regarding inspection of "intelligence material" and the inclusion of a phrase regarding "national policy" intelligence that was later deleted, NSCID 1 went to the NSC practically as it had been recast in CIA on December 1. Hillenkoetter sent with it on December 9 the suggestion that the NSC might name the DO chairman of the IAC to emphasize at it was an advisory body to help him and not a "Board of Directors or Board of Management," but the NSC approved it on December 12 without this addition.
1 Adapted from a history of central intelligence to 1951 prepared by the author in 1953. For previous portions see Studies VIII 3 p. 55 ff, X 2 p. 1 ff, XII 1 p. 55 ff, XII 3 p. 79 ff, and XII 4 p. 73 ff.
13 Studies X11 4 p. 83 ff.
No Foreign Dissem