Chapter Sixteen


Final Observations

Today, intelligence remains the only area of highly complex government
activity where overall management across departmental and agency lines is seriously attempted.

In the wake of the terrorist attacks in the United States on 11 September 2001, many observers believed that the biggest intelligence problem involved was a failure to “connect the dots.” Available but disparate pieces of information were not properly correlated and evaluated. The same judgment, of course, had been reached 60 years earlier in assessing why America was surprised by the attack on Pearl Harbor.[2] In both cases, correcting the problem about the dots—held by various agencies—was deemed best accomplished by making changes in organizations and process. In the 1940s, the solution was to create a DCI and a CIA to “centralize” the process of connecting the dots, and a principal response so far to 9/11 has been the creation of a new Department of Homeland Security and a new director of national intelligence.

But is organizational change the answer? After all, although created to provide strategic warning, the DCI and the CIA had not been able to predict the more recent deadly attacks. Some thoughtful students of intelligence have expressed skepticism about the value of organizational change in improving intelligence performance. Walter Laqueur, in looking at intelligence reorganization efforts in the 1970s, judged them to be unnecessary and even harmful, reflecting less sensible solutions for well-identified problems than “the growing bureaucratization of modern intelligence and the unhealthy preoccupation with managerial problems—the tendency, to paraphrase Karl Mannheim, to turn substantive problems into problems of administration.”[3] Earnest May, having examined the role of intelligence in several West European countries, concluded that the “type of organization appears to have had little effect on the quality of assessment. The examples of Edwardian Britain, tsarist Russia, and imperial Germany…showed collegial and centralized structures each to have flaws.” He went on to judge, like Laqueur, that reorganizations can be harmful, asserting that the examples he studied suggest that “it may be a mistake to change organization—that it is better to live with the ills one has than fly to others against which one has no built-up immunity.”[4]


Initial Conception of DCI’s Community Role

President Truman’s main motivation in creating a DCI and a CIA was to connect the dots better. He wanted an officer subordinate to himself, someone other than his chief policy advisers, to pull together and make sense out of disparate items of information about the world. He declared himself satisfied, at least initially, with how that task was performed at the simplest level, the provision of daily intelligence to the White House. Such a task seemed suitable for a presidential staff aide, and RAdm. Sidney Souers, the DCI who established the daily feed of information to Truman, stayed only long enough to set up the CIG (he later returned to the White House as an aide to Truman, serving as the first executive secretary of the NSC).

But the DCI’s job was broader than just correlating and evaluating daily reports. He was supposed to lead a coordinated effort to produce better strategic, or “national,” intelligence for the president and other senior policymakers on the major issues affecting US national security. This meant drawing on the full resources of the federal government and gave the DCI his first “community” role. In the early months of the new CIA, DCI Hillenkoetter drew criticism for his performance in this role by failing to coordinate the production of adequate substantive national “estimative” intelligence. Thus, a complaint about the quality of finished intelligence products was defined so as to point a finger at a managerial, or leadership, problem.[5] Connecting the dots was not just a matter of better thinking or better data, it also was a question of the process by which organizations cooperated to achieve a desired outcome, and the acknowledged leader or manager of that process was the DCI.


Community Role Expands

DCI Smith’s solution of a better estimates process addressed the complaint about inadequate output in the form of national, strategic intelligence. Over time, however, the DCI became increasingly drawn into paying attention to the inputs that underlay and in the end made possible the finished intelligence products. This came about because of persisting dissatisfaction, expressed by President Eisenhower, with the gaps in knowledge that intelligence was not covering. Human source operations were not revealing key secrets about Soviet military capabilities and policy. Not only were dots not being connected, they also were missing. Collection had to be improved. As long as the president looked to the DCI to achieve the needed progress, there was no logical end to the areas of interest to the DCI short of looking into the full range of intelligence activities.

The task of penetrating the USSR’s secrecy barriers, however, required more than better cooperation or coordination between the existing members of the US intelligence community. It demanded the building of major new national intelligence capabilities, and Eisenhower’s insistence that the new strategic reconnaissance programs be placed under Dulles’s aegis pushed the DCI toward a new and somewhat different community role.

In attempting to accomplish the new tasks laid on his doorstep, the DCI did not at first seek to intrude into how other intelligence organizations were managed or led, nor did he threaten their traditional roles. As had been the case with respect to the expansion of covert action capabilities, he looked to build and manage these expanded capabilities within CIA. He did have to become involved with technology-rich projects, however, and he did need the help of other organizations such as the US Air Force. With the advent of the space age, the strategic reconnaissance programs became increasingly expensive and complex, giving rise to an imperative that they be managed for maximum efficiency and effectiveness. The DCI thus found himself deepening his working relationship not with his traditional counterparts in Army or Navy intelligence but with the secretary of defense and his key civilian subordinates.[6] His community role was growing because the community was growing, adding elements not anticipated in 1947 by anyone other than technical specialists.

CIA was not the only intelligence organization growing as a result of the Cold War. Pentagon intelligence activities were also expanding, partly in cooperation with CIA and partly in support of military requirements. This growth prompted President Eisenhower in his second term to press Dulles to do more to improve the management of the entire federal foreign intelligence enterprise. By Dulles’s lights, he already was managing the entire Intelligence Community via his committee structure for coordination of activities of the various intelligence agencies and via his delivery of coordinated national intelligence to the president and the NSC. But what President Eisenhower wanted went beyond that. He wanted more efficiency in the use of resources, more “bang for the buck” for intelligence dollars as well as for defense dollars. As we have seen, faced with Dulles’s inability to satisfy this new expectation, Eisenhower in the end accepted the status quo and thus remained dissatisfied with the DCI’s community leadership role.


New Kind of DCI

Whereas “analysis” had been at the heart of the original coordination duty assigned to the DCI in the 1940s, a growth spurt in “collection” capabilities came to the fore in the 1950s and 1960s as expensive national reconnaissance and signals intelligence programs burgeoned. The new technical programs responded both to the continuing need to pierce Soviet secrecy as the USSR became militarily stronger and globally more active, and to the new need to fight a growing war in Vietnam. Existing organizations such as NSA adapted to accommodate some of the growth, but new organizations such as the NRO also sprang up. As these programs grew and became more important, “management” became the focus of studies about the DCI’s proper role in the community.

DCI John McCone was determined to take an activist, leadership role in all important aspects of these developments, and he deepened his involvement in DOD organizations. The earlier focus on correlating the products of intelligence entities other than CIA gave way to increasing interest in their internal operations. This was particularly the case with the NRO, in which McCone took an intense interest, and with imagery programs, where CIA was intimately involved in the processing and final presentation of data. NSA’s signals work—apart from technical data related to Soviet weapons testing, in which CIA took a keen interest—remained more distant from close DCI attention, partly because much of it was military or tactical in nature and partly because it had never been a part of what the DCI had been directed to oversee.

McCone followed up on President Kennedy’s formal endorsement of an enhanced DCI community leadership role by developing staff support for that role that had not previously existed. McCone’s establishment of the NIPE Staff in 1963 marked a shift from support appropriate for a chairman of a coordination process geared to achieving consensus in committees to support fitting for a chief executive officer intent on pressing his own guidance in leading a large-scale enterprise. This occurred none too soon as the explosive growth in the nation’s intelligence capabilities prompted cost-cutting impulses to find efficiencies in managing intelligence resources. Within DOD, reform led to the creation of a succession of senior civilian positions in OSD charged with overseeing DOD intelligence. Within the Intelligence Community, DCI Helms sought to work with the new OSD leadership in managing the now larger aggregation of programs for which he assumed responsibility.

With the advent of the Nixon administration in 1969, White House attention to intelligence increased. Senior administration officials connected their concern with a perceived lack of payoff at the product end of the business with a belief that the management of community resources, including the expensive collection systems that had come to dominate the budget, was inadequate. These twin concerns about effectiveness and efficiency drove the Nixon White House to formally charge the DCI in 1971 with enhancing his community resource manager role as well as continuing to provide national intelligence.

Nixon’s guidance went into surprising detail in designating guidance and resources mechanisms the DCI was to employ in exercising his newly expanded role. The mechanisms did not live up to their promise, however, and Helms was as reluctant as Dulles to move outside the realm of consensus leadership at which he excelled. As late as the mid-1970s, CIA veterans were resisting the wider community role being pressed upon the DCI. “The DCI,” one senior retiree said to DDCI Hank Knoche in 1976, “in looking outward to the community, should control it only via substantive aspects and not with respect to money, people and resources.”

The Nixon, Ford, and Carter administrations charged all five DCIs in the 1970s—Helms, Schlesinger, Colby, Bush, and Turner—with exercising more direct and effective leadership over the intelligence community. The more direct DCI hand on the reins of intelligence community leadership and management affected largely the topmost level of the community, but it drew the DCI into more dialogue with the secretary of defense and his principal deputy. This working relationship became an unsettled arena of constant discussion, mutual reminders of basic authorities, and occasional friction. Turner’s tenure marked a high point of striving and strain; no DCI tried harder to gain direct control over community resources outside CIA.


Political Dimension

The Watergate scandal, which broke into the open in 1973, and then startling public allegations of intelligence wrongdoing spawned a public and congressional debate that broke the back of the old patterns of intelligence oversight. It forced the president—always the most important factor in supporting or protecting the DCI’s position—from office and weakened the executive branch as a whole. President Ford and DCIs Colby and Bush found themselves having at every turn to defend their positions against newly unleashed pressures of congressional interest and public attacks. Studies and proposals suggested new rules and mechanisms aimed at keeping intelligence in bounds. President Ford’s 1976 executive order on intelligence demonstrated a new level of presidential oversight and an effort to preempt feared actions by the Congress. Intelligence became a presidential election campaign issue in 1976, and in 1977 a new president appointed a new DCI for the first time since 1953.

Congress took up a new formal oversight role. What began as a probe of illegal activities widened during 1976–77 into a regular process of oversight as both the Senate and the House established permanent committees devoted to intelligence programs. The DCI was the target of their interest in personal accountability. Ford’s executive order explicitly charged the DCI with being “the principal spokesman to the Congress for the Intelligence Community.” Although that designation did not mean that the executive branch surrendered its primary role in managing the federal government’s intelligence business, it brought the DCI into a new relationship with those who controlled intelligence budgets.

These congressional actions reinforced the executive branch steps taken early in the 1970s to give greater attention to, and create greater expectations regarding, the DCI’s community leadership. The armed services committees of the two chambers retained their oversight of DOD elements of the community and focused on the secretary of defense as an appropriate leader for those programs. But the new select committees for intelligence became competing forums for discussion and decision on a variety of resource and policy issues, and their “client” was unambiguously the DCI.

The development of the more “modern” DCI as community leader and the more “modern” congressional oversight of intelligence meant that the DCI was now beholden to a broader constituency than simply the president and the NSC. The arena in which DCIs had to operate had become wider and more political. DCIs were forced to deal with more political players in shaping decisions about the nation’s intelligence structure and policies, and they both faced more varied pressures for change and had available more outside avenues of support. William Colby was right to judge that his profession no longer occupied a secret cubbyhole adjacent to the president’s office.

As the Casey era in the 1980s showed, this development did not necessarily lead to a better working relationship between the DCI and Congress. It did mean, however, that the relationship could not be disregarded without costs to the DCI and the executive branch. During the 1970s and 1980s, some CIA veterans went to work on Capitol Hill, and during the 1980s and 1990s, a flow of congressional staffers migrated to the executive branch. In 1997, two former senior SSCI staffers, George Tenet and Keith Hall, became, respectively, DCI and director of the NRO, and the appointment of HPSCI Chairman Porter Goss in 2004 to be DCI created a new highwater mark of congressional influence.


Fat Years and Lean

After the turmoil of the 1970s, the 1980s provided a period of relative calm under DCIs Casey and Webster with respect to the DCI’s community role. Beyond President Reagan’s 1981 executive order, no new charters demanded adjustments in that role, and Casey’s close relationship with the president gave him a solid base of respect and support within the Intelligence Community. Casey’s staff contended with OSD on some intelligence program and policy issues, but basic agreement between the principals kept those issues in check.

After Adm. Inman’s departure in 1982, senior CIA officers filled the DDCI position for the next 10 years (John McMahon, Robert Gates, and Richard Kerr). They helped stabilize and support a “presiding” kind of DCI community leadership. Casey and Webster met frequently with community principals, and with their IC Staff seniors. From 1982 to 1992, senior military officers headed that staff, following the Schlesinger-Colby-Bush formula of the mid-1970s. The staff took up a large number of issues and rearranged itself in response to management initiatives, such as creating in 1986 a new collection discipline called “measurement and signature intelligence” and briefly trying yet again to improve the linkage between “requirements” and programs. But it also came to be viewed as more routinized in its activities and less connected to DCI leadership than to a consensus-style collective structure congenial to community members.

During Webster’s tenure, adjustments to a post-Cold War era and smaller budgets for national security measures caused intelligence leaders to rally around the DCI in a search for common solutions suitable for an era of “downsizing.” Dollars and personnel shrank for all, and integration in the name of efficiency seemed to offer an opportunity for closer community collaboration. CIA reached out to partner agencies via a new set of “centers” for subjects such as counternarcotics work and arms control, and NSA and CIA seniors held their first-ever joint offsite meetings to plan new levels of cooperation for signals intelligence operations. Looming behind all these activities was an increasingly activist Congress, restless with initiatives for structural change and “downsizing” suggestions.

Gates’s strategy of pressing limited change succeeded in keeping executive branch control over intelligence reform, but his delegation of responsibility for major community committees to other senior intelligence officers transformed those bodies from potential instruments of DCI influence into more technical staffing and community consultation adjuncts to their new bosses. In effect, Gates bet on his ability to promote change via his working relationships with the heads of the major intelligence agencies rather than use staffs with community representation as channels of DCI influence. Gates’s attitude paralleled Colby’s in having no patience with complex requirements mechanisms or community consensus committee products that perpetuated rather than challenged the inevitable inertia of the past.


Change as a Constant

The Bush and Clinton administrations in the 1990s gave the five DCIs of the decade—Webster, Gates, Woolsey, Deutch, and Tenet—newly prioritized lists of intelligence targets suitable for the post-Cold War period, but they allowed them to manage their own strategies of adjustment to the less clearly defined national policies that replaced the old containment strategy. The 1991 Gulf War put new pressures on DCIs to enhance intelligence support to military operations, and, thanks to the 1986 Goldwater-Nichols reforms, the rising power of regional combatant commanders within the US military structure amplified this demand on national intelligence. The DCIs of the 1990s sought to respond positively to military needs while at the same time addressing the increasingly important topics of international terrorism and weapons proliferation.

The orderly Bush-Gates relationship gave way to the distant Clinton-Woolsey non-relationship, and the Intelligence Community hardly had time to get used to one DCI before it had another to assess and look to for assistance in coping with downsized budgets and upsized demands. Gates, Woolsey, and Deutch stressed community-wide approaches, and they all tried to work cooperatively with senior executive branch leaders in coping with challenges that went beyond the community itself. Working with the White House and senior DOD leaders, they sought to keep intelligence as a preferentially protected element of national strength as defense budgets declined. Working with senior Department of Justice leaders, they struggled with the increasing involvement of foreign intelligence with law enforcement, mainly to enhance counterintelligence and security but also to cooperate against the new threats of information warfare and homeland defense against terrorism attacks.

In the 1990s, Congress reenergized its interest in intelligence, creating new senior positions for community management and demanding more detailed reporting from the DCI on community activities. Congress also successfully pressed for new commissions that studied the intelligence business broadly and made recommendations regarding community organization and management, including adding incrementally to DCI authorities. CMS gradually grew in size, as did the number of community committees dealing with various aspects of the more complex set of issues with which DCIs found themselves having to deal. By 2004, the DCI enjoyed more robust community staff support than at any previous time although the size of that support did not approach that of the chairman of the JCS, let alone that of a cabinet officer heading a department.


Look to the Future

In the early 2000s, the DCI, in his community role, looked like the kind of DCI who had emerged in the 1970s, the product of an evolution frozen in time. The basic community structure was the same, and decades-old issues of authority and accountability remained unresolved. The DCI was viewed mainly as a resource manager who should somehow be able to do a better job of pulling the Intelligence Community together. The tendency of most observers wishing to address the tensions inherent in the structure of authorities was toward centralizing solutions: give the DCI full budgetary authority or full line management authority over the “national” pieces of the community, often with the accompanying caveat that he should give up day-to-day management of CIA.

The new pressures on intelligence related to the threat of terrorism in the wake of the attacks of 11 September 2001, however, redrew the picture. Although the failure of intelligence to warn of surprise attack in 2001 reminded us of 1941, the 9/11 attacks altered the fundamental nature of the national intelligence mission, and thus, derivatively, the meaning of Intelligence Community leadership. Until they occurred, the focus of intelligence reform was on improving processes and connecting dots better within the foreign intelligence community, where the DCI at least provided a clear-cut, logical focus for improved leadership or management. Now, although improving foreign intelligence remains an important task, the challenge has been augmented by a new need to integrate foreign and domestic intelligence information better. Not all the dots useful for predicting 9/11 were items of foreign intelligence information, and the FBI has come in for its full share of scrutiny by the various post-9/11 review panels.

By 2004, the 9/11 commission, then the president, and finally the Congress decided that the DCI—as the position had been defined for more than 50 years—would no longer suffice as the singular leader charged with overseeing all activities needed to fulfill the new, broadened national intelligence mission. By creating a DNI and a National Counterterrorism Center, they intended to bring a new level of coordination and effectiveness to America’s Intelligence Community. But whether these reforms will actually improve foreign intelligence activities or adequately encompass the new emphasis on homeland security remains to be seen:

  • Will having a hybrid organization devoted to countering as well as analyzing international terrorism lead to better understanding of terrorism than that achieved by the DCI Counterterrorist Center in the years leading up to 2001? For that matter, will “centers” provide a boost to intelligence performance at all?[7]

  • Will the new DNI be able to work in more effective ways than the DCI with the heads of the Department of Justice and the FBI, already expanded both at home and abroad in response to 9/11? Or with the chief of the Department of Homeland Security, which is still defining its own intelligence role?

  • How much better will the new DNI be able to work with the DCI’s longtime partner in managing the foreign intelligence community, the secretary of defense? After all, he is busy grappling with how best to provide his own leadership to DOD’s intelligence activities and to generate intelligence that will improve the military’s effectiveness in fighting terrorism.

Experience tells us that redefining the “community role” of the nation’s intelligence chief has not ended with the passage of a new law and the appointment of the first DNI. Examination of problems has begotten reform. But reform can take a long time to achieve improved performance. DOD reform, decades old, is today still a work in progress. No less should be expected of the transformation of American intelligence.



[1]Commission on the Roles and Missions of the United States Intelligence Community, Preparing for the 21st Century, 47.

[2]Part of the failure lay in dots that were missing, that is, in failures of collection. This was true also for the Pearl Harbor case. The dominant impression so far, however, has remained that making sense out of information already at hand in one form or another is a primary area needing improvement. In the 1990s, corporations emphasized “knowledge management” as a way to draw comprehensively on information already held within various, sometimes widely dispersed, divisions, and some reformers see this as a key to remaking today’s Intelligence Community.

[3]Laqueur, A World of Secrets, 312–13. Laqueur cited the transition of the NIPE Staff to the IC Staff, the replacement of ONE by the NIOs, the supplanting of USIB by NFIB, and some internal CIA office changes in connection with this point.

[4]Ernest R. May, ed., Knowing One’s Enemies: Intelligence Assessment before the Two World Wars, 532–33.

[5]The focus during this early period on national intelligence estimates—then believed to be the most important finished intelligence product for the topmost customers—addressed what many view as the most essential intelligence mission. William Colby recalled that William Donovan, in addressing OSS employees for the last time, gave pride of place to analysis. “During the final ceremony, he referred first to his scholars and research experts in describing the OSS ‘team’ and only secondly mentioned the ‘active units in operations and intelligence who had engaged the enemy in direct encounter.’ In this he reflected his unique contribution to American intelligence, that scholarship was its primary discipline, that the acquisition of information was to serve it, and that its paramilitary adventures were an adjunct to its authority and expertise in secret machinery.” Colby, Honorable Men, 55. Richard Helms, in his foreword to a memoir by a former deputy director for intelligence at CIA in 1989, and in a memorial address at CIA on the occasion of CIA’s 50th anniversary in September 1997, also stressed that the supreme objective of national intelligence was to provide the president with the best information possible. Even if one agrees with this point of view, of course, achieving success rests on a base of intelligence activities other than analysis. Russell Jack Smith, The Unknown CIA: My Three Decades with the Agency, ix.

[6]In considering this important relationship, it should be remembered that intelligence is only a subordinate part of the duties of the secretary of defense, whereas the topmost federal intelligence chief—although a less highly ranked officer in the executive branch—is viewed as holding a singular role as the nation’s most senior intelligence official, responsible directly to the NSC and the president for the overall performance of US intelligence.

[7]The WMD Commission in March 2005, while recommending a small center for coordinating intelligence on weapons proliferation, stated that it was “skeptical more generally about the increasingly popular idea of creating a network of ‘centers’ organized around priority national intelligence problems.” The commission recognized the value of better coordination, but specified as potentially unappreciated costs that they might tend to crowd out competitive analysis, create new substantive “stovepipes,” and engender turf wars. The commission recommended instead the appointment of “mission managers” to help the DNI oversee intelligence on priority issues. Report of the Commission on the Intelligence Capabilities of the United States Regarding Weapons of Mass Destruction, 328.

Historical Document
Posted: Mar 16, 2007 08:48 AM
Last Updated: Jun 27, 2008 12:33 PM