See: Directors of Central Intelligence, 1946-2005 [PDF 33KB*]

This office will probably be the greatest cemetery for dead cats in history.

James Forrestal used the above words in a private letter in 1947 to describe his new position as the first US secretary of defense.[1] As secretary of the navy, he had played a major role in designing the new office so it would not be able to wield much power over the military services. Now, thanks to President Harry Truman, he found himself occupying the position and facing the challenge of leading the nation’s federal defense establishment with deliberately limited authority.

Forrestal might just as well have been describing another new position then being created as part of a revised national security structure, that of the director of central intelligence (DCI). This post, originally created early in 1946 by President Truman within his own office, was given statutory basis in 1947 by the same National Security Act that established the office of the secretary of defense. Like the secretary of defense, the director of central intelligence was associated with a collection of already existing organizations. How well either official could make disparate elements work together was in question.

The similarity between the two jobs did not last. Forrestal’s original limited conception of the secretary of defense’s office—“it will be a coordinating, a planning, and an integrating rather than an operating office”—gave way soon to the view that he needed more direct authority and control.[2] In 1949, the Truman administration supported legislation that converted the 1947 act’s “National Military Establishment” into a single executive department, the Department of Defense (DOD), headed unambiguously by the secretary of defense and incorporating all the organizations for which he was responsible.[3] In later years, as DOD incorporated elements responsible for new national defense capabilities (including intelligence organizations), its chief automatically acquired authority over them.

No such strengthening of DCI authority with respect to the various federal foreign intelligence organizations took place. Whereas organizational “unification” of the military services was a major postwar presidential interest and congressional priority, consolidating all federal intelligence units in one department or agency was not.[4] The organizations associated with the DCI acquired no collective name analogous to “national military establishment,” the term “intelligence community” appearing only in the 1950s. The legislative and executive charters that shaped postwar intelligence put as much emphasis on not changing existing efforts as it did on creating new ones.

The DCI commanded the new Central Intelligence Agency (CIA), and therefore he exercised decisive control over some aspects of the nation’s intelligence capabilities, most notably the activities of the clandestine service in conducting espionage and covert actions abroad. Over the years, DCIs added to CIA’s capabilities—especially in the areas of all-source analysis and technical collection—and thus expanded their arena of direct control. But other major additions to America’s growing intelligence enterprise during the Cold War grew up outside the DCI’s domain. Because these capabilities—in satellite reconnaissance, signals intelligence, and other fields—dealt heavily with defense matters and contained many military personnel, they were placed in DOD and hence fell naturally more subject to direction from the secretary of defense than from the DCI.

These facts notwithstanding, the DCI from the outset has been associated with expectations that he would be able to integrate the nation’s foreign intelligence efforts. How he has exercised this “community role,” is the story told in this study. The questions that defined the research undertaken for the study were the following: How have the various DCIs through the years viewed and carried out their community role? What expectations regarding that role did they face? What priority did they give it? What specifically did they try to do? And how did their efforts fare?

The issue of the DCI’s community role is not, of course, a new one. But systematic treatment of how that role has evolved over time is surprisingly absent from the now quite large body of literature about intelligence. In doing the research for this study, the author encountered only one specific recommendation that a study of this sort be conducted. Walter Laqueur suggested in a footnote in a book published in 1985 that “a special monograph ought to be written about the attempts made by successive DCIs ‘to provide effective guidance and coordination’ to the entire intelligence community, to quote an internal directive issued by [President John F.] Kennedy to John McCone.”[5]

The story will take up first the roots of the DCI’s community role. It will then proceed chronologically, describing the various approaches that successive DCIs have taken toward fulfilling their responsibilities in this regard. At the end, it will pull together some themes and sum up circumstances as of 2005, when a new official—the director of national intelligence—replaced the DCI.[6] It will not propose recommendations for resolving the mismatch between responsibility and authority that has bedeviled all DCIs. Rather, it will attempt to clarify through historical research some of the issues involved and to provide future commissions and officials with a fuller knowledge base upon which to build recommendations for change.

This study is very much a first effort to sketch an outline of major developments over a lengthy period of time based primarily on CIA files. There no doubt are many sources of information not adequately reflected in it that can add useful new facts and insights to those presented here. Most useful would be perspectives from the vantage points of intelligence agencies other than CIA, various presidents and other senior executive branch officials, and Congress. The author’s hope is that this study will spur additional research into how the Intelligence Community has functioned, including exploration of how it can best operate and be led.


[1]Walter Millis, ed., with the collaboration of E. S. Duffield, The Forrestal Diaries, 299. The comment appears in a letter written to Robert Sherwood on 27 August 1947, three weeks before Forrestal was sworn in to the new position.

[2]Clark Clifford, with Richard Holbrooke, Counsel to the President: A Memoir, 156–62. The citation in the text is from page 159. Clifford’s account includes testimony of Forrestal’s belief that he had been “wrong” during 1946–47 to have helped water down the original definition of the secretary of defense position (he had plenty of help from Congress) and his determination to strengthen it during 1948–49. It also records President Truman’s satisfaction with the strengthening achieved in 1949. He had told Clifford in 1947 that he recognized the weakness in the secretary’s original authority and that “maybe we can strengthen it as time goes on” (157).

[3]There continued to be three sub-departments for the military services, but they were now “military” rather than “executive” departments and their heads were removed from membership in the National Security Council. Also, the secretary of defense now had the powers traditionally vested in an executive department head and exercised full rather than “general” direction, authority, and control. See Alice C. Cole, Alfred Goldberg, Samuel A. Tucker, and Rudolph A. Winnacker, eds., The Department of Defense: Documents on Establishment and Organization, 1944–1978, 108–111.

[4]This was the case despite the creation in the law of a director of central intelligence and a Central Intelligence Agency. Both had singular and important roles, but the emphasis on “centralization” did not lead to creation of a seat of comprehensive authority. For a useful exploration of the concept and how it has been incorporated in key documents defining the evolution of the DCI’s scope of authority, see Michael Warner, ed., Central Intelligence: Origin and Evolution.

[5]Walter Laqueur, A World of Secrets: The Uses and Limits of Intelligence, 19. The footnote was to a sentence noting the lack of budgetary authority exercised by the DCI outside of CIA, “which lessens his ability to fulfill his responsibility as the supreme controller of all intelligence.” The ease with which observers refer to the DCI’s community role as something implying he should have strong powers (here, “supreme controller”) helps feed a bias toward stronger centralization and personal authority as “solutions” to the community role “problem.”

[6]Another “community role” issue arose afresh with the terrorist attacks of 11 September 2001 in the United States. Former Pennsylvania Governor Thomas J. Ridge became a senior White House director for homeland security, but he operated as a coordinator of efforts without the authority of an executive department head until the establishment of the Department of Homeland Security in 2003.


* Adobe® Reader® is needed to view Adobe PDF files. If you don't already have Adobe Reader installed, you may download the current version at (opens in a new window). [external link disclaimer]

Historical Document
Posted: Mar 16, 2007 08:48 AM
Last Updated: Jun 27, 2008 12:26 PM