Eighteenth DCI, George John Tenet
George Tenet: Deputizing Integration
This is both an historic and a happy event for our Intelligence Community.
It’s our first coronation.
George Tenet, who had been DDCI since July 1995, acted as DCI from the time of Deutch’s departure in December 1996 until he was sworn in as DCI on 11 July 1997. He was not President Clinton’s first choice as Deutch’s successor. Clinton initially had turned to Anthony Lake, his first-term national security adviser, who began preparations to move to Langley. In a replay of Gen. Carns’s experience in 1995, however, Lake fell victim to confirmation problems and withdrew his name from consideration.
By 1997, Tenet’s resume included four years of senior executive branch experience with intelligence—he had been the NSC staff officer in charge of intelligence in the first two years of Clinton’s administration before serving two years as DDCI—in addition to his years of congressional staff work in the field. He sailed through confirmation and took command of an Intelligence Community still defining paths of change through a thicket of challenges. He retained the same honorary cabinet status Deutch had enjoyed, declaring that it would be useful to have the same status as his predecessor if that was what the White House wished.
Tenet wasted little time moving on a wide front once he was DCI. Within days, he named new deputy directors for intelligence and operations in CIA and a new chairman of the NIC. In October 1997, with White House approval, he released a public figure for the nation’s intelligence spending for FY 1997, resolving for the moment a nettlesome issue. In November 1997, he dealt with the long-standing issue of organizing to deal with weapons proliferation, naming senior CIA officer John Lauder as the new head of a strengthened DCI center that now contained CIA’s technical analysts on key proliferation issues and was more closely linked with the main DI and DO units dealing with the topic. He also worked with the deputy secretary of defense to allocate additional resources to the study of biological and chemical warfare in future years.
Tenet visited all the major intelligence agencies in his initial weeks as DCI. He stressed to each that he would fulfill his obligation to lead “the entire Intelligence Community,” and that he would focus his energies on “promoting a closer-knit Intelligence Community that shares a common vision for the future.” He followed up in subsequent months to deal with a range of community issues. NIMA posed a special challenge. Some at CIA believed it was straying from a national focus as it strengthened its ability to support military operations, and some at DIA believed that NIMA’s NPIC-style analysis was too much like the “all-source” analysis for which DIA was responsible. Tenet oversaw the negotiation that in February 2000 produced a DCI/secretary of defense memorandum of agreement resolving the status of CIA career personnel working in NIMA and the transfer of a large number of CIA positions to NIMA. He also had to devote attention to overseeing the future imagery architecture (FIA) program for which NIMA and the NRO were responsible.
Tenet announced in June 1998 a decision on a structure for measurement and signature intelligence (MASINT). A Central MASINT Organization (CMO) was to be maintained within DIA, and various initiatives to bolster support for its activities were approved. Tenet also decided in consultation with his fellow intelligence agency leaders to merge the Community Open Source Program Office (COSPO) into the Foreign Broadcast Information Service (FBIS). A range of other collaborative processes advanced during Tenet’s tenure. Soon after Tenet became DCI, for example, the NRO set up a deputy director for national support to show its interest in balancing its responsibilities between military and “national” coverage, and a new Overhead National User Exchange Group began meeting. To give an element of oversight to NSA’s work, a Cryptologic Senior Oversight Group was set up under the auspices of the DCI, the chairman of the JCS, and the deputy secretary of defense to assist cryptologic systems development.
Tenet was determined to show his colleagues and his overseers that he intended to play an active community role and that he could provide leadership conducive to decisions on even knotty problems. His community staff necessarily bore most of the brunt of the bureaucratic friction from these efforts, but he gave voice to the value of their and his efforts. Near the end of 1998, in addressing the Association of Former Intelligence Officers, Tenet endorsed the importance of using information technology to create a “common operating environment” for the Intelligence Community and of changing its culture and patterns of collaboration, of trying to “build a greater sense of Community within the Intelligence Community.” He also kept his eye on the substantive issues that drove the collaborative organizational and bureaucratic initiatives. In a December 1998 memorandum to his senior subordinates at Langley, he stressed the need to “redouble” efforts against international terrorist Usama Bin Ladin, whose al-Qa’ida organization had attacked two US embassies in August 1998. “We are at war,” Tenet declared, and “I want no resources or people spared in this effort, either inside CIA or the Community.”
New DDCI/CM and ADCIs
Tenet inherited the thorny issue of working out with Congress the addition of five new presidentially appointed, Senate-confirmed positions: a statutory general counsel at CIA, a new deputy DCI for community management (DDCI/CM), and three new assistant DCIs (ADCIs) to focus on major areas of community activity. DCI Deutch had notified the SSCI in the fall of 1996 that he was adamant in not wanting to accept the three ADCI positions called for in the recently passed FY 1997 Intelligence Authorization Act and that he would recommend to the president that he seek repeal or substantial revision of that provision. As acting DCI, Tenet chose to defer dealing with the issue.
Once sworn in as DCI, Tenet engaged a former SSCI general counsel with whom he had worked closely as SSCI staff director, Britt Snider, to try to work out a compromise with Congress. On 23 October 1997, the president sent a letter to Senator Specter (R-PA), the original sponsor of the legislation, noting that he had already announced his intent to appoint lawyer Robert McNamara to the general counsel position and senior DOD intelligence officer Joan Dempsey to the DDCI/CM position and asking that they be confirmed while Tenet worked with the SSCI to reach a mutually agreeable solution to the difference of views regarding the ADCI positions. Snider worked through the fall trying to achieve agreement with the SSCI on the issue, but he concluded by January 1998 that his efforts would not bear fruit. He was frustrated by the SSCI staff’s unwillingness to discuss alternatives and its lack of rationales for the new positions beyond a general expectation that the new ADCIs would help “manage the community.” Taking into account that the HPSCI was sympathetic to the DCI’s position and would support repeal, he recommended to Tenet that he cease trying to achieve a compromise in the near term, simply state his views, and let the issue play out over the next legislative session.
Accordingly, on 3 February 1998 Tenet sent a letter to SSCI chairman Senator Richard Shelby (R-AL) conveying a lengthy rationale for his opposition to the three-ADCI provision, asking that the dispute not hold up the confirmation of the new DDCI/CM and promising to address the concerns that had led the committee to advocate the ADCI positions. He argued that the legislation added “nothing to the authority of the DCI to manage the Intelligence Community,” that it created “an unnecessary bureaucratic layer which could hamper, rather than facilitate, the Community’s day-to-day work,” and that authority lines might become blurred and staffing unnecessarily unwieldy. He also objected to the enlargement of the number of intelligence officials subject to presidential appointment and Senate confirmation, arguing that it worked to the disadvantage of career professionals and “would inevitably lead to greater politicizing of the intelligence function.” This straightforward approach invited discussion, but it also put the DCI on record opposing an idea already in law that he might in the end be forced to accept.
In an exchange of letters in March 1998, the two sides continued their negotiation. Senator Shelby and Senator J. Robert Kerrey (D-NE), SSCI’s vice chairman, offered to forgo the third, more general ADCI, now called the ADCI for administration, but insisted on the need for the other two. Tenet responded to them promptly, turning the senators’ suggestion around and offering to accept a single ADCI for community management to assist the new DDCI/CM rather than the other two ADCIs. He was anxious to dislodge the SSCI from its delay of the confirmation of the DDCI/CM, whom all agreed should be installed but who was being held hostage to the ADCI issue.
By May, the deal had been cut, and the Senate moved ahead with the confirmation of Joan Dempsey as the new DDCI/CM. The Senate agreed to confirm just the ADCI for administration in addition to the DDCI/CM and to allow the DCI to proceed with appointing on his own authority, without Senate confirmation, the other two ADCIs, one for analysis and production and the other for collection. On 21 May 1998, Tenet sent a letter to Senator Specter (who had left the SSCI), explaining the deal and making it clear he considered the confirmed ADCI the principal deputy of the DDCI/CM. He noted that the 1996 law remained in place and that he and the SSCI would review in the future “whether or not the approach I have outlined above adequately strengthens the DCI’s ability to manage the Intelligence Community or whether, in fact, confirmation of the additional two positions is required.”
As soon as Joan Dempsey had cleared the Senate hurdle, Tenet moved to invigorate the mechanisms of his community leadership role. He gave Dempsey a charter memorandum describing her job as the third ranking position in the community and directing her “to undertake the actions necessary to carry out all of the DCI’s responsibilities as head of the Intelligence Community.” She was to reconfigure all DCI directives to take into account the new structure, starting with an “overarching” directive establishing “the DCID system as the principal means by which the DCI provides guidance and direction to the Intelligence Community.” She was to charge the new non-confirmed ADCIs with their responsibilities and see that they carried them out. She was also charged with a set of tasks aimed at improving the community’s approach to strategic planning, prioritizing requirements, and evaluating performance. She was always present when Tenet met with the heads of the various intelligence agencies, whether in group meetings or bilateral sessions, and he allowed her considerable autonomy in acting as his community deputy.
Three aspects of the new system deserve particular attention. First, the DDCI/CM position filled in 1998 hearkened back to the community position set up by Allen Dulles in 1957 in response to President Eisenhower’s pressure that the DCI do more to manage the community, and to the community deputy appointed in 1976 by DCI Bush (and commended for revival in the early 1990s). Like Dulles, Tenet was forced to accede in some ways to the pressures for a more activist community leadership role. However, in 1998, the pressures on Tenet were in line with his preference—also like Dulles’s—to assign much of his community role to a senior deputy rather than take them on himself.
Second, the overarching DCI directive, DCID 1/1, which became effective on 19 November 1998, was in a sense a self-drafted job description of the DCI in his community leadership role. From 1947 until 1998, the basic charters one could use to learn about the DCI’s community responsibilities had been laws, executive orders, and presidential or NSC directives. These sources of authority continued to be the basis for the DCI’s duties, including his community role. Rather than have Congress, the president, or the NSC review them and issue a refreshed charter for his community role, however, Tenet simply had Dempsey pull together in DCID 1/1, as the title clearly stated, an updated list of “The Authorities and Responsibilities of the Director of Central Intelligence as Head of the US Intelligence Community.” Since he had delegated to her the job of dealing with all his community responsibilities, it described her job as well.
Third, the installation of the two non-confirmed ADCIs did indeed increase the complexity of the bureaucracy. The heads of the collection discipline committees, for instance, who at the beginning of the 1990s reported to the DCI via his community staff chief, now reported via the ADCI for collection to the DDCI/CM—a new layer and level of separation from the DCI. Also, the ADCIs took over major community processes, including chairing the National Intelligence Production Board (NIPB) and the National Intelligence Collection Board (NICB). Gates had already based community-oriented analytic, or production, processes in the NIC, and John Gannon, the first ADCI for analysis and production (ADCI/A&P), was already the chairman of the NIC. But those processes became more separated from existing structures after Gannon left in 2001 (even though Mark Lowenthal, Gannon’s successor as ADCI, served simultaneously as NIC vice chairman for evaluation). Charles Allen, the ADCI for collection (ADCI/C), had been CMS’s senior requirements officer. But he moved his ADCI operations out of CMS into a separate office, thus creating some organizational distance from the program and resources work that remained in CMS. Indeed, CMS itself, while responsive to the DDCI/CM, continued to be headed by the Executive Director for Intelligence Community Affairs (EXDIR/ICA), thus creating a three-person layer cake (DDCI/CM, ADCI for administration, and EXDIR/ICA) in place of the CMS structure created by Gates in 1992, in which there had been only a single senior community staff chief reporting to the DCI and DDCI.
The formal chartering in DCI directives of the ADCI/A&P and the ADCI/C came in mid-2000, two years after the initial incumbents were appointed. DOD concerns regarding the authority of these two officers centered on assuring that military needs for intelligence collection would not be impaired by the new DCI structure of community management. The DASD/I had expressed concern about the absence of the military services from membership on the NICB and about the ability of collection entities to commit resources in response to NICB demands. The DDCI/CM believed that the representation of the JCS and DIA on the board ensured that military needs would be adequately addressed, and the directives acknowledged that board members would commit collection resources only with the agreement of appropriate principals.
Once DCI, George Tenet emphasized getting on with accomplishing the central mission of acquiring and providing intelligence to high-level executive branch customers. As DDCI, he had been an unabashed partisan of CIA, and especially of its DO, and his enthused hosting of CIA’s gala 50th anniversary celebration at Langley in September 1997 shortly after he was sworn in as DCI suggested he would continue to emphasize his agency leadership role. At the same time, however, he also paid attention to how the community operated. He relied on Joan Dempsey, whom he had selected to serve as his chief of staff, for advice and counsel, while also using personal meetings with other intelligence leaders to show continuing interest in community matters.
Tenet redesignated some community mechanisms early in his tenure. In July 1997, immediately after he was sworn in, a new directive established an Intelligence Community Principals Committee and a matching Deputies Committee, replacing the Intelligence Community Executive Committee structure Gates had set up in 1992. The principals group included the vice chairman of the JCS, signaling a desire to work closely with the military operators, and the deputies group was to be headed by the new DDCI/CM once she had been duly confirmed and sworn in. The new bodies, whose names mirrored NSC-related policy bodies, were chartered to provide advice on community-wide policies, planning, and processes. Biweekly meetings of NFIP program managers continued to be held, and other Intelligence Community entities such as NFIB and the NIC continued to exist. The resource bodies where DOD equities were considered along with the DCI’s—the Expanded Defense Resources Board (EDRB) and the Intelligence Program Review Group (IPRG)—were also continued.
Other steps reflected Tenet’s readiness to foster increased collaboration and integration across the community. In 1997 he began implementation of the Intelligence Community Assignment Program (ICAP), which encouraged officers to think in terms of community-oriented careers. The program was modest, but it represented a tangible step in the direction of turning the confederative community into a single organization. Although officers knew that careers were still made within individual agencies or services, the program aimed at influencing mindsets with the hope that more solid inroads on bureaucratic reality—if they were possible at all—might come later. Tenet also directed collaborative efforts on many subjects, most notably perhaps the serious and successful effort to anticipate computer program problems that might arise with the turnover of computer systems from the year 1999 to 2000 (called the “Y2K” issue).
In 1998, Tenet established an Intelligence Community chief information officer (CIO). Again, the step was arguably a modest one, but it did engage the principals of the various agencies in working together more closely to achieve common or collaborative information technology policies and programs. The new office replaced CMS’s intelligence systems secretariat, which had fostered the creation of Intelink, a common computerized system for sharing intelligence reports and data across agencies. The new community CIO initially took up issues such as e-mail policy and received from the principals a promise of more centralized authority in some areas.
Another important position Tenet established was that of Intelligence Community senior acquisition executive (SAE). This officer, first appointed in November 1999, was to ensure that NFIP systems acquisitions were accomplished prudently and with due attention to seeing that their capabilities were applicable to the highest priority intelligence targets and, where possible, were integrated across the community. This step was also seen as giving a potentially powerful tool to the new DDCI/CM. After all, the procurement of new systems addressed the largest resource decisions within the NFIP.
Tenet continued community participation in the executive branch-wide program for improving governmental efficiency and “customer” satisfaction sponsored by the vice president’s office. By undertaking projects under the mantle of the 1993 Government Performance and Results Act, the DCI’s small “quality” staff was able to portray various DCI or community programs, e.g., the hard targets process, as aimed at improving performance across the individual agencies and thus at coordinating or integrating the community as a whole.
These changes in community mechanisms were not novel, reflecting in many cases continuity of ideas and projects that had been fostered in various forms going back to the early 1990s, but under Tenet they achieved new milestones in the name of integration and cooperation. Although Tenet struck many traditional notes in how he approached his job, he seemed determined to act with respect to his community role, both in terms of guiding internal community progress on various fronts and in terms of representing the community downtown and in Congress.
Like some of his predecessors, DCI Tenet sought to reenergize centralized planning as a means of influencing community activities. He had devised a Strategic Direction document for CIA in May 1998 and had sent it to other intelligence leaders. The director of NSA complimented the effort and penned a quick reply: “Time to put one together for the IC.” Tenet asked his new DDCI/CM to prepare a similar document for the Intelligence Community, and in March 1999 she published his Strategic Intent for the US Intelligence Community after suitable coordination. Underscoring the importance of the new DDCI/CM’s role in acting fully for the DCI in community matters, Dempsey signed the letters transmitting the new strategy document to the principals within the Intelligence Community, whereas the DCI signed those going outside the community to principals within the executive and legislative branches of the federal government.
Its vision statement declared a simple overall goal: “A unified Intelligence Community optimized to provide a decisive information advantage to the President, the military, diplomats, the law enforcement community and the Congress.” The word “unified,” while arguably an overreach, captured the hope that technology could help US intelligence leaders achieve higher levels of cooperation even without fundamental change in organization. The use of “optimized” reflected a continuation of the decades-old hope that somehow the DCI could find efficiencies that saved money.
The document held out practices in the private sector as models worth emulating. “A cornerstone of our strategy will be to move toward establishing a collaborative community enterprise similar to those now found in industry,” the report declared, adding that “we propose more centralized corporate management of the intelligence enterprise.” The very next sentence expressed both the hope and the limits of the vision: “This will be a cultural shift from the loose confederation of independent agencies that we are today, although we will continue to share fiscal ‘ownership’ of intelligence activities with cabinet-level organizations like the Department of Defense.”
The DCI’s principal subordinates for community affairs took on the job of centrally managing the staff activities aimed at advancing the vision’s goals. In the fall of 1999 Joan Dempsey announced that the ADCIs and the EXDIR/ICA would be the “champions” for the formal DCI objectives in the plan, and that “expectation coordinators” would be appointed for each of the specific goals listed under each objective. For example, the ADCI/A&P and ADCI/C would champion DCI objective number one, “unify the community through collaborative processes,” and the EXDIR/ICA would lead the way on number five, “improve corporate management of resources.”
The DDCI/CM dutifully reported on early progress to Congress, and in 2000 the DCI directed that she “maintain momentum” in community strategic planning activities by writing detailed action plans and otherwise energizing agencies to advance the agenda of integration he had set. Inevitably, progress in many areas was accompanied by frustration in others as various projects moved forward or stalled. The limitations of “management” when one does not control incentives and resources were all too apparent despite her staff officers’ energetic activities.
The DDCI/CM oversaw the process of ensuring that strategic plans for particular issues were drawn up as needed. For example, in response to PFIAB’s concern about what it saw as a decline in science and technology capabilities and expertise in the community, the DCI in February 1999 tasked her with developing a strategic plan for S&T in the community. By June 1999, Tenet was able to sign off on a strategy for advanced research and development, advertised as the community’s “first comprehensive plan” for advanced research and development investments in the NFIP.
Setting Community Priorities
Tenet had overseen the “hard targets” effort as DDCI, and soon after Deutch departed Tenet directed the adoption of a more institutionalized approach. Formal executive boards were established for major countries of strategic interest to the United States, and a principals forum of Intelligence Community agency heads was formed to oversee the “hard targets process.” After the appointment of the DDCI/CM and the ADCIs in 1998, they took over the leadership of this community enterprise. They enlarged and regularized the numerous meetings and reports involved, and they attempted to connect these activities to the allocation of resources within the main intelligence programs.
In 1999, CIA’s inspector general conducted an inspection of the hard targets process that revealed criticisms of the process. The officers they interviewed complained that there was too much bureaucracy, that effectiveness of the process was uneven across the several main targets, and that the modest gains registered against the targets would have been achieved in the absence of the process. The stated purpose of the report was to assess “the value and impact of the Hard Targets Process for senior CIA officers,” and the CIA-centric views it expressed reflected the particular burden it placed on CIA, which supplied the chairpersons of the executive boards for all the hard targets.
The report drew a deft rebuttal from the DDCI/CM’s deputy, ADCI for Administration (ADCI/A) James Simon. Some of his points highlight well the perspective of greatest interest for our study, that of a DCI attempting to find methods of wielding leadership over the disparate parts of the community he is charged with heading. Tenet, who had contributed personal attention to the hard targets process in order to give it status and impact within the community, penned in the margin of his copy of the rebuttal: “This is a terrific response,” and read portions of it aloud at an internal staff meeting with the inspector general in attendance.
Simon complimented rather than criticized the inspectors, but he argued that what they had uncovered in their survey was “the inherent difficulty of the task.” To him, the task was “to force a common approach across the various inter-agency and intra-agency parochialisms.” That the inspectors had no suggestions about how the DCI could better advance community work on the top intelligence targets was, Simon asserted, “a sobering testament to how limited are the tools at the DCI’s disposal and how far we have yet to go to make our use of the word ‘community’ more than a convenience.”
Simon argued that “all of the complaints about ‘process’ are familiar,” and that “complaints of this sort abound whenever authority tries to institutionalize anything that affects the freedom of action, or inaction, that is so important to so many components of the Intelligence Community.” “Alas,” Simon intoned, “process is a poor thing but all we seem to have.” He went on to argue that the complaints voiced to the inspectors reflected more than anything the seriousness of the DCI’s objective to effect change, including “this DCI’s attempt, for the first time ever, to redirect Community resources toward his highest priority regional problems.” At the same time, he acknowledged some confusion in the establishment of the new DDCI/CM and ADCI structure, declaring that the lack of an accepted management structure for community collection and production management led intelligence agency leaders to be reluctant to accept DCI-directed recommendations for reallocating intelligence resources.
In the end, Simon stated, “we are dismayed, although not surprised, by the breadth and depth of opposition to the DCI’s efforts to impose his priorities over those of his subordinates, many of whom seem unclear on the role of the DCI as substantive leader of the Intelligence Community.” Although he recognized that some who were interviewed did not believe that all the countries highlighted in the hard targets process were worth their attention, he observed tartly that “our sympathy is constrained by the knowledge that the DCI believes that they are and that the United States government is not a participatory democracy.”
Simon’s diatribe reflected the frustration that many a DCI staff officer has felt about the difficulty of moving beyond the general acceptance of DCI leadership in the “guidance” stage of strategic planning to more specific decisions regarding how resources should be applied. The latter, of course, is more a “management” function in which the acceptance of direction throughout an organization depends upon a willingness to recognize superior authority and decisions passed down from above. The inspection report gave no help to the DCI on this critical dimension, instead suggesting that the DDCI/CM “decouple the Hard Target Executive Boards from the resource allocation process.” Simon’s anguished cry about process also reflects a senior staff officer’s frustration with the difficulties of moving beyond levels of cooperation limited by the willingness of other organizations to support DCI initiatives.
Mission Requirements Board
Simon was a central player in another initiative undertaken by Tenet, the establishment of a Mission Requirements Board (MRB) to validate customer needs and technical requirements for the acquisition of expensive intelligence systems. Tenet had been the author of PDD-35, the Clinton-era top-level statement of the substantive needs of the principal users of national intelligence, and he and his senior community deputies wished to move beyond the national intelligence “needs process” begun by Gates and Woolsey and the “hard targets process” begun under Deutch to something that would enable the DCI to speak with greater authority and to deal with DOD on intelligence systems that clearly had to meet non-military as well as military needs. There was much brainstorming by various DCI staff elements in the 1990s on how to improve collection management and the so-called requirements process, and it coalesced under the new DDCI/CM and ADCI organizational structure in 1998 and 1999.
Joan Dempsey, the new DDCI/CM, took as a departure point that the DCI had “no integrated process to articulate, prioritize and defend future needs of national intelligence consumers” in place in 1998. Her June 1998 charter memorandum from the DCI tasked her with creating a process in such a way that it informed the allocation of current resources and also dealt with future needs, and she set forth the goal of having a new system in place by mid-1999. She criticized the “issue coordinator” and “hard targets” approaches as uneven in quality and—echoing Lew Allen’s 1973 criticism of the USIB committee system—not useful in making choices between intelligence targets, hence the emphasis on an integrated method of prioritizing across the entire range of intelligence needs. Also, she wanted to focus on future needs within mission areas, thus supporting decisions about major downstream resource commitments. This was an area where the DCI might hope to have greater influence since his community-wide role as program-builder was recognized more than his role as program-executor. Tenet, in explaining in 1998 why he accepted recommendations for reform in the wake of the failure of US intelligence to predict Indian nuclear testing, stressed that reform should “give me the management tools I need to help us make the right trades [among resource allocation options]…[and] make sure that our coverage and our resources and our people are deployed in the right way.”
The DCI announced the creation of the new national requirements process and the new MRB in August 1999 in letters to principals such as the secretary of state, whose department he declared to be the “lead agent for defining future diplomacy and diplomatic operations intelligence needs.” The MRB, which held its first meeting in August 1999, was modeled on DOD’s Joint Requirements Oversight Council (JROC), a high-ranking body used by the chairman of the JCS and the secretary of defense to vet major system requirements. Its chairman was James Simon, Joan Dempsey’s deputy as the ADCI/A, and the vice chairmen were the other two ADCIs.
Its utility was questioned both by those within the Intelligence Community used to other means of validating elements of their programs and by DOD, which felt it could validate its own intelligence needs. But it persevered, mainly because its advocates realized that a non-comprehensive approach allowed major NFIP program managers to cherry-pick justifications for programs, and they wished to give the DCI a mechanism that would end the existing undisciplined method of setting and validating requirements. After a year Simon wrote to the chairman of SSCI claiming that it had had “a significant effect on the Intelligence Community’s business practices.” In particular, he asserted, “programs are being examined with a rigor not experienced in the past.”
Probably of even greater importance was Simon’s claim that the new board had “built an important bridge to the military requirements process of the Department of Defense.” “It has brought the DOD and IC together at the earliest stages in the requirements development process,” Simon declared, “which should heavily influence PPB [plans, programs, and budgets] and eventually acquisition.” He undertook efforts to make the projects of the MRB mesh as much as possible with the JROC so that the DCI could have an effective seat at DOD tables of discussion and decision. In fact, at one point, Simon wrote to the DDCI/CM’s legal adviser asking whether the MRB needed new legislation in order to vest authority in the MRB analogous to that of the JROC. “Just as NFIP funding supports JROC-validated military requirements,” he asked, “how can DOD be compelled to provide the required resources” for MRB-validated needs? The creation of the board was an innovative and ambitious effort to enhance DCI leadership within the community and the DCI’s dialogue with DOD. In one case, the MRB brought about a decision that reversed a prior JROC position on an intelligence program, using evaluation data to convince the JCS that the MRB could help reach decisions that served military customers better than existing processes could, a signal accomplishment.
CMS and the ADCIs Chime In
CMS had been since 1997 funding a staff activity called the Community Operational Definition of the Agile Intelligence Enterprise (CODA) that looked intently at ways the community could be conceived of and even operated as an integrated whole. This project, originally a modeling exercise sponsored within CIA, necessarily focused on processes since new authorities were not forthcoming, and avid staffers worked hard to devise ways in which the DCI’s staff could assist him in stitching the intelligence agencies together more effectively. The newly appointed ADCIs naturally took over or invented processes aimed at integrating the community better in their respective areas of responsibility, increasing communication among intelligence agencies and cooperation in areas where parties saw mutual interest. The ADCI/C oversaw a large-scale task force effort to look at ways collection management might be improved to overcome “stovepipe” behavior. Many of the ideas the task force considered, of course, were the same ones explored in the hard targets or other exercises being run by CMS, and the papers generated justified more activities by the new DDCI/CM and ADCI structure.
In 1999, Tenet’s top community deputies pressed an idea to “revolutionize” community-wide interaction on requirements and tasking. This initiative, called the Intelligence Community Multi-intelligence Acquisition Program, or IC MAP, envisaged a one-stop, cross-INT “shopping network.” It was an extraordinary initiative in that a self-appointed senior steering group took on leadership of the initiative and agreed to devote funds to advance it. This ambitious effort showed an unusual degree of commitment to change and to the DCI’s oft-stated goal of greater collaboration across the community. Even the endorsements of the agency chiefs, however, could not overcome the bureaucratic difficulties in implementing this ambitious project, which struggled to move toward its ambitious self-advertised goals.
By 2000, a Collection Concepts Development Center (CCDC) had been established under the aegis of the ADCI/C to find advanced, innovative ways to increase the amount and quality of information about the hardest intelligence targets. It aimed to add a more strategic element to collection planning and to add to the level of expertise about the full range of collection methods available to the community leadership, thus defeating the limitations of compartmentation applied to various “special” collection programs. The ADCI/C expressed the hope that the center might develop a “community-wide doctrine for collaborative collection management” and might “spin off new strategies and initiatives” to others for exploitation. Also, the NICB became a frequently used forum for voicing all kinds of ideas about current, and to a lesser extent, future collection problems and how to solve them.
Department of Defense
The DCI and the secretary of defense orchestrated policies and planning both within their respective empires and between them on several key issues. With respect to information operations, earlier called information warfare, one memorandum likened their cooperation to “combined” military operations involving more than one country, in which “the resources are governed by authorities (SECDEF and DCI) which have distinctly different legal bases and reporting chains.” Various committees worked on the issue within the community and between the community and DOD entities, and in 2000 the DDCI/CM sent a community-wide strategic plan on the issue to community principals, calling it “our latest strategic planning achievement.” The importance of issues such as this one guaranteed that the DCI would have to deal personally with them, usually with the deputy secretary of defense and the national security adviser. Even when staffers on both sides reached agreement on specific plans or actions, the legal and national policy dimensions required thoughtful top-level attention and authorization.
On the program and budget front, Tenet was acting DCI as DOD’s Quadrennial Defense Review was being finalized in 1997, and he kept careful watch on how that process affected the NFIP. He had help from Congress as the HPSCI urged intelligence to be more strategic, to go beyond simple “support” functions, and to be more robustly represented in DOD programs in light of the “dominant battlefield awareness” vision articulated by DOD. As Deputy Secretary of Defense John White departed office at the end of June 1997, Tenet warmly thanked him for the personal attention he had accorded “to maintaining the close cooperation between the Defense Department and the Office of the DCI,” citing especially his improvement of the “analytic basis for intelligence resource decisions” and his “commitment to the Joint Defense and the Intelligence Community process” of making decisions on intelligence programs. (A cartoon slide prepared during this period for Tenet—subtitled as being about things they “never told you when you were on the Hill”—depicts a superman-sized DCI at the guidance stage of program preparation and a boy-sized DCI at the operational stage of program execution.)
By the following spring, Tenet was telling Congress optimistically that “the days of stapling together the budget submissions of the program managers and calling it the NFIP budget are gone for good.” The hard work by his community staff to sustain his role in the budgetary process, however, had to continually deal with DOD initiatives that did not adequately take the DCI’s equities into account. A particular problem was lack of attention to intelligence at the top level and a tendency to handle community resource issues at a lower level. In the fall of 1999, Tenet wrote to Secretary of Defense William Cohen pleading “I need your help” in altering the draft of a DOD directive on the role of the ASD/C3I in order to remove a provision that sought to grant that official “authority, direction, and control” over the major intelligence agencies within DOD such as NSA. He argued that such delegation of authority from the secretary’s level was inconsistent with existing and past agreements and practices between his office and DOD and in fact appeared “to diminish the Secretary’s responsibilities for national intelligence.” He was certainly on sound ground in defending his need for a partner at the topmost level of DOD if he was to avoid having his community become subject to the bureaucratic whims of lower-level DOD officials.
Once again, when a new administration came into office in 2001, Tenet needed to remind senior DOD leaders, preoccupied with significant internal reforms and a new quadrennial review, of the value of joint processes of program consideration. In August, DDCI/CM Joan Dempsey wrote to Deputy Secretary of Defense Paul Wolfowitz complaining about a lack of coordination in developing DOD’s FY 2003–2007 fiscal guidance that departed from longstanding practices for developing the Intelligence Community budget. Dempsey pointed out that the uncoordinated DOD changes to the NFIP, resulting from applying inflation rates without regard for program impact, altered unilaterally previous joint DCI/DOD program decisions. Besides requesting negotiation to avoid this problem in the future, Dempsey warned that the Defense Planning Guidance should not treat the NFIP simply as a DOD component program when the DCI had to respond to non-DOD as well as DOD needs. Tenet also took the step of raising this issue with the deputy national security adviser, Stephen Hadley, warning him that in addition to discussing the issue with Secretary of Defense Donald Rumsfeld, he might have to raise it with the president.
Crime, Security, and Counterintelligence
Tenet continued his predecessors’ dialogue on the topics of crime, security, and counterintelligence with Attorney General Janet Reno, her deputy Jamie Gorelick, and FBI Director Louis Freeh. The arrest of Aldrich Ames in 1994 had spurred one set of decisions for organizing government-wide for counterintelligence, and suspicions of Chinese penetration of the Department of Energy’s nuclear weapons laboratories in the late 1990s brought new pressures of improvement from Congress as well as within the executive branch. Tenet and Freeh formed a “gang of eight” group that began meeting in early 1998 to spur progress in CIA-FBI cooperation. A senior former CIA official, former ADDO John McGaffin, had been hired as a senior adviser to the deputy attorney general, and he worked with Robert M. “Bear” Bryant, the head of the FBI’s National Security Division, to improve the dialogue with the DDCI and senior CIA officials, including DDO Jack Downing. They focused on practical steps such as joint training, improving coordination in the field, and cooperative operations on specific cases.
DCI Tenet turned to his other natural partner on security and counterintelligence, the deputy secretary of defense, who co-chaired with the DCI the Security Policy Board (SPB) set up in 1994, and in 1998 they constituted a second Joint Security Commission to review progress since 1994 and highlight emerging security issues. With the two principals chairing a meeting of the full SPB, Gen. Larry Welch, USAF (ret.), head of the Institute for Defense Analyses, briefed the results of the “JSC II” review in September 1999.
The panel praised the SPB structure set up in 1994 but cited a lack of follow-through on policy implementation. It urged the creation of a new SPB executive committee to correct this flaw and recommended that the National Counterintelligence Center become the central source of threat information for both government and industry in the United States. It recommended clearing reinvestigation backlogs and making information security authorities more coherent. It also drew attention to security threats in the information technology area, endorsing more stringent attention to training and system administrator certification and the creation of a formal Information Technology Service to enhance professionalism in the corps of such employees in sensitive positions. Recognizing the community-wide nature of all these steps, the DCI appointed the DDCI/CM as a co-chair of the new executive committee and gave her the responsibility for further consideration of the panel’s recommendations.
Also in 1999, the DCI worked with both FBI and DOD senior officials to create initiatives to strengthen counterintelligence at the national level. This effort, known as “CI-21,” envisaged a new top-level organizational structure to guide the nation’s counterintelligence programs. The deputy attorney general, the deputy secretary of defense, and the DDCI were to co-chair a National CI Board of Directors, which in turn was to appoint a new national counterintelligence executive who would develop a national counterintelligence strategy and oversee a combined governmental and private sector set of activities. Attorney General Janet Reno was skeptical of the value of what she saw as a new and large “superstructure” and preferred that the FBI fix whatever problems it had in counterintelligence before embarking on this overarching scheme. DCI Tenet worked hard to convince her of the value of the plan, keeping national security adviser Samuel Berger informed of the progress on this initiative and seeking his help in bringing Reno on board. This initiative stands as a useful instance when a DCI played what he saw as a necessary role representing the Intelligence Community on an issue of extraordinary scope, involving not only other departments and agencies, but also the private sector.
Department of State
Tenet accommodated the State Department’s new level of interest in community involvement. Undersecretary of State for Political Affairs Thomas R. Pickering wished to have the DCI’s program guidance take into account the State Department’s new Strategic Plan for International Affairs, a document analogous to the DOD’s Defense Planning Guidance, and Tenet was happy to oblige. He also welcomed State Department’s establishment of a new internal Intelligence Policy and Resources Coordinating Committee and continuing emphasis on “support to diplomatic operations” as a discrete intelligence mission analogous to the more often heard “support to military operations.” On the latter point, the DCI’s staff supported enhancing mention of it as a priority in revising PDD-35 and urged the State Department to draft a paper articulating its needs in that regard to fit with the DCI’s requirements process.
As manufacturers of non-government overhead imagery systems began to make products commercially available on the open market, Pickering raised the issue of who “leads” in approving foreign remote sensing arrangements. On Tenet’s behalf, DDCI John Gordon suggested to Pickering that they continue to follow a policy agreed upon in 1992 between DCI Gates and Deputy Secretary of State Lawrence Eagleburger concerning the licensing of satellite reconnaissance systems. At the same time as cooperation continued apace on a variety of issues, reminders to State of DCI equities were occasionally needed. DDCI/CM Joan Dempsey found it necessary in 2000 to write a letter to the undersecretary of state for management reminding her that a planned internal transfer of funds from State’s Bureau of Intelligence and Research to its Bureau of Diplomatic Security required DCI approval and congressional notification to the intelligence committees.
Commissions Urge Reforms
Early in his tenure, DCI Tenet endured a spate of outside critiques of the performance of the Intelligence Community he headed. In 1998, the Commission To Assess the Ballistic Missile Threat to the United States, headed by former White House Chief of Staff Donald Rumsfeld, took issue with both the conclusions and the approach the community had taken in the 1990s with respect to missile threats. In addition to its report, issued in both unclassified and classified versions, the commission forwarded on 15 October 1998 a classified “Intelligence Side Letter” that lectured the DCI regarding how the community should do its business. It took particular note of “the harmful effect of continued stovepiping of functions and information within the IC” and offered other observations generally critical of both intelligence and the Clinton administration’s stewardship of it. Tenet responded in a measured way to the sweeping criticism of the panel. He acknowledged the value of some of the commission’s general points by promising to undertake steps such as “red-teaming” problems, but he also restated continuing differences of view, making it clear he would defend the community’s work where warranted.
Also in 1998, India’s surprise nuclear weapons tests occasioned another commission, this time headed by Adm. David Jeremiah, USN (ret.), a former vice chairman of the JCS. His panel devoted a section of its report to “organization and integrating the IC,” charging that the lack of organized follow-through related to DCI tasking and of a community collection management process hampered community performance. In its recommendations, the panel urged the DCI to “vest a community manager with the authority to demand accountability from across the IC,” and to “install an overarching management structure to integrate collection systems and ensure better interagency allocation of resources.” “A ‘cross-INT’ collection mechanism is needed,” the report declared, “to address this need to task collection as a ‘system of systems.’” Tenet was primed to take advantage of the recommendation that a senior subordinate be charged with improving community management, telling the SSCI: “I accept all of Adm. Jeremiah’s recommendations, and I have asked the recently confirmed Deputy Director of Central Intelligence for Community Management to oversee development of action plans for each of them.”
Tenet in mid-1999 had to deal with the intelligence aspects of a national commission on proliferation headed by his predecessor, John Deutch. Deutch’s panel recommended the appointment of a kind of proliferation czar (“National Director for Combating Proliferation”), and Tenet had to make sure that administration or congressional actions in response to the panel did not undercut his authorities. Tenet also met with the National Commission on Terrorism, chaired by L. Paul Bremer, early in 2000 as it examined ways in which the Intelligence Community might improve its performance in countering international terrorism. Commission members were interested in how intelligence was working with law enforcement and whether the DCI needed new authorities, and Bremer was concerned that CIA’s human rights policies regarding assets might be constraining the agency’s effectiveness on this issue.
The findings of both the Rumsfeld and the Jeremiah panels were discussed at a senior intelligence leadership conference hosted by the DCI on 11 September 1998. One conclusion the participants reached was that “failure to improve operations management, resource allocation, and other key issues within the community, including making substantial and sweeping changes in the way the nation collects, analyzes, and produces intelligence, ‘will likely result in a catastrophic systemic intelligence failure.’” This stunning prediction—three years to the day before the 9/11 attacks in the United States in 2001 and remindful of the hauntingly accurate portrayals made in 1941 by US military intelligence officers of how the Japanese navy might attack Pearl Harbor—shows that the intelligence professionals involved were quite aware that “management” issues can affect substantive intelligence products dealing with world events. Actually, 10 years earlier a historian of US intelligence had said much the same thing, arguing that the “fundamental structural weakness” of a community pulled apart by centrifugal forces “may well invite another Pearl Harbor.”
There were also reports and suggestions made by groups sponsored by Congress or within the Intelligence Community itself. In 1998 and 1999, the SSCI’s Technical Advisory Group issued reports on the impact of changing technologies on the major intelligence collection disciplines. NIMA was the topic of a commission formed by the secretary of defense and the DCI in 2000 at the insistence of Congress. It took up major issues such as how much tactical intelligence to expect from national intelligence assets, technical competence, and how to deal with increasingly available commercial imagery products. Also, DCI Tenet created his own National Security Advisory Panel, chaired by Adm. Jeremiah, which helped him prepare in 2000 for the advent of a new administration. Talking points prepared for Tenet’s use in meeting with them provide a telling story of the lack of real substance in his relationship with top DOD leaders at that juncture: “Not at war with DOD, but no one there at home. No meetings. DOD needs to step up. SecDef needs ‘top cop’ to ensure the Services do the right thing.” The panel encouraged Tenet to follow his instincts to emphasize the building of personal and institutional relationships in engaging a new administration, but it also told him that he lacked the statutory and regulatory authorities necessary to meet his responsibilities.
The Intelligence Community came under new scrutiny after the inauguration of President George W. Bush in 2001. The United States Commission on National Security/21st Century, chaired by former senators Gary Hart (D-CO) and Warren B. Rudman (R-NH), issued a final report on 31 January 2001 that emphasized threats to US homeland security. Its recommendations for the Intelligence Community called for no major structural changes and commended the community management changes already under way. It did point to warning intelligence failures, however, and urged various improvements in intelligence policies and operations as well as presidential involvement in setting updated intelligence priorities.
Also, the administration established a commission headed by former national security adviser Brent Scowcroft to examine the intelligence business broadly and recommend reforms. This group came up with a draft report that suggested giving the DCI direct authority over the major national intelligence agencies (the NRO, NSA, and NIMA). At first, it seemed that the 9/11 attacks, which came just as the commission was finalizing its findings, would serve as a spur to adopting this most radical resolution of the DCI’s community role. Perhaps inevitably, however, the administration’s focus on retaliating against the murderous attacks and planning for a long-term “war on terrorism” left reform waiting at the altar during 2001–2002. Also, Secretary of Defense Rumsfeld could scarcely have been expected to be enthused about losing control over major intelligence assets, especially at a time when military user demands had come to the fore. The commission’s recommendations were never formally transmitted to the president.
Post 9/11 Pressure for Change
The terrorist attacks in the United States on 11 September 2001 marked a watershed for American intelligence as well as a turning point for US national security policy. The DCI faced three major challenges, each of which involved his ability to speak for the Intelligence Community and harness its efforts to fulfill new strategic tasks. Immediately, Tenet found himself a central figure among the top administration policymakers in identifying who had masterminded the attacks and outlining plans for moving quickly to retaliate against them. For this task, he relied especially on CIA and its ability to work with enemies of the terrorist leaders responsible for the attacks, and the agency moved front and center in the initial US actions taken to unseat the Taliban in Afghanistan and to pursue Usama bin Ladin and his key aides.
Secondly, Tenet had to deal with the issue of “intelligence failure” on his watch. The 9/11 attacks conjured up both the 60-year-old warning failure of Pearl Harbor and the impression that the Intelligence Community built to fight the Cold War had not fulfilled its core mission of remedying the earlier failure. This issue was mainly a Washington-area political battle, and again, it centered on the CIA (because most observers believed HUMINT was the best way to penetrate terrorist organizations) even though obviously signals intelligence and other non-CIA intelligence capabilities were involved. The popular characterization of the intelligence failing in this case as one of not “connecting the dots” pointed to the difficulties of getting disparate organizations to work together, always the bane of DCIs attempting to assert leadership within the community.
But the greatest impact of the attacks in the long run may be the third challenge that Tenet faced: the “dots” involved domestic as well as foreign intelligence information and organizations and were not, therefore, all within the DCI’s jurisdiction. 9/11 had recast the mission of American intelligence. It was still highly important, but no longer sufficient, to gather and analyze foreign intelligence. Needed now was a new structure and method to meld foreign and domestic information in order to protect against attacks that might be launched from within as well as outside of the United States.
The establishment of the Department of Homeland Security in 2003 dealt with part of the problem. It gathered under one organizational roof domestic security functions associated with a wide range of organizations that had been in various executive departments in the hope that their integration would achieve greater security within the United States. This step did not, however, resolve the role of intelligence. Indeed, some in Congress had hoped for greater intelligence integration within the new department than was in fact achieved. Instead, the administration in 2003 created a new Terrorist Threat Integration Center (TTIC) under the DCI in a move to connect “dots” better from both domestic and foreign sources of information and to help protect against terrorist attacks both at home and abroad.
A joint inquiry in 2002 by the Senate and House intelligence committees into intelligence issues connected with 9/11 concluded that it would be helpful to establish a cabinet-level DNI separate from the position of the head of CIA. In February 2003, DCI Tenet was asked at a Senate hearing if he was interested in the joint inquiry’s recommendation. Tenet said he was not: “The DCI’s direct relationship with, and control over, the CIA is essential to the DCI’s ability to carry out his mission and functions as the head of the IC. Dissolving the existing links between the head of the IC and the CIA would weaken both the Community and the Agency.”
Long Tenure Ends
In February 2004, George Tenet surpassed Richard Helms as the second longest serving DCI. Tenet had hung Helms’s official painted portrait in his office, and he made no secret of his admiration for his predecessor. Like Helms, he focused on delivering intelligence to the president and other top customers and on overseeing CIA’s activities, especially those of the DO. Also like Helms, he responded to pressures for more community coordination and management by instituting and overseeing new mechanisms and processes. In some ways, however, they differed. Helms saw a gaping mismatch between his limited authority and the ambitious community responsibilities assigned to him by President Nixon, and he eschewed initiatives to play an expanded community role. Tenet, on the other hand, acted as though his authorities were sufficient for him to play a leading community role, and prior to 9/11 he said much in the name of improving community-wide processes and practices.
In the late 1990s, Tenet used critiques of the community’s performance such as the Jeremiah and Rumsfeld panel reports as reasons for both insiders and outsiders to support the community initiatives of his new DDCI/CM and ADCIs. He had objected to aspects of the new community positions, but he used and praised the new officers for developing the community-wide activities that became the sum and substance of his deputized community role. By 1999, Tenet reported to Congress that “although we have a long way to go, I am optimistic that US Intelligence is already moving away from its traditional ‘stovepipes’ towards becoming a more collaborative and agile enterprise.”
In the wake of 9/11, he continued using the same levers of community management he had employed earlier. By 2004, there were more than six dozen community committees overseen by the DDCI/CM’s office busy promoting coordination among the various intelligence agencies. Work on a new requirements process, led this time by the ADCI/A&P and called the National Intelligence Priorities Framework (NIPF), followed the adoption in February 2003 by the White House of a new list of presidentially approved intelligence priorities. Like its predecessors, it envisaged resource allocations tied to substantive topics and the use of evaluation to improve management of the community.
Tenet also sorted through two issues with DOD that arose after 9/11. One was working out with the secretary of defense their respective roles in conducting covert actions abroad. Executive Order 12333 assigns normal primacy to the CIA, but Secretary of Defense Rumsfeld emphasized the need for military special operations forces to conduct various activities in the world-wide war on terrorism. The other was ensuring the DCI’s continuing community role in the context of DOD reform that created in 2003 a new undersecretary of defense for intelligence charged with exercising leadership over intelligence agencies housed in DOD. No surface friction initially accompanied this move, and Stephen Cambone, the first holder of the new DOD position, expressed his and the secretary’s desire to support the DCI’s community role. How these episodes and developments will alter the way intelligence chiefs fulfill their community role remains to be seen.
In June 2004, Tenet tendered his resignation to President Bush, noting that his last day, 11 July, would be his seventh anniversary as DCI. Under attack for providing faulty intelligence before the 2003 Iraq war (both HPSCI and SSCI weighed in with critical reports) and facing a forthcoming commission report that certainly would be critical of the community’s performance related to 9/11, he cited family reasons for his departure. Tenet was finally free to stand on the sideline and watch as his former colleagues and his successor wrestled with inevitable calls for intelligence reform.
An early postscript to Tenet’s tenure that appeared less than a month after his departure directed pointed criticism at the way he had carried out his community role. An article in a news magazine charged him with not doing enough to assist his topmost community deputies in their efforts to implement various community initiatives, citing former ADCI/AP John Gannon as saying: “There was no top leadership support for us.” This venting of frustration, which included criticism of the same sort of bureaucratic behavior by individual agencies as the 9/11 Commission had voiced, suggested that pressures for reform existed within as well as outside the community, and at senior as well as lower levels. Combined with public comments that Tenet’s call for “war” in 1998 had gone unheeded within the community, the criticism drew attention once again to the large potential for gaps between DCI statements of goals and actual changes.