• Who would run such an agency? Would it matter if that person were civilian or military?
• Would this new agency be manageable?
• Given that DOD currently controls all the technical INTs, would this remain true? What are the implications either way?
The main goal is some modicum of collection efficiency and improved resource management. However, the suggested solution would create a large entity, one whose inherent power might rival that of the DNI. Cooperation has been growing between the National Geospatial-Intelligence Agency (NGA) and the National Security Agency (NSA), although still far from the point of a merger of any sort.
Some have suggested that the two HUMINT components—the CIA’s National Clandestine Service (formerly the Directorate of Operations, DO) and the Defense Humint Service of the Defense Intelligence Agency (DIA/DH—Defense Humint)—be unified, also to avoid duplication. Recognition of the need to improve coordination among the various HUMINT collectors—which include the Federal Bureau of Investigation (FBI), Drug Enforcement Administration (DEA), and the military services as well as the CIA and DIA—was the impetus behind DCI Porter Goss’s creation of the National Clandestine Service (replacing the old Directorate of Operations), which includes a deputy to coordinate HUMINT across the intelligence community. For the time being, the decision appears to be in favor of improved coordination but continuation of separate HUMINT efforts, which permits a broader and more diverse HUMINT collection effort. Along similar lines, some have proposed that the clandestine services (HUMINT and covert action) be a separate agency, either to improve management responsibility or to avoid contaminating analysis, or both. Sen. Pat Roberts, R., Kans., made such a proposal in 2004, at the outset of the debate over the new intelligence structure. judging by the reaction Roberts received, it is fair to say that very few support this concept, although it is not as radical as it is often portrayed, replicating, as it does, the British structure.
In the area of collection, open-source intelligence (OSINT) is a specific reform issue. OSINT was long underutilized and had no strong organizational locus. Reformers have advanced several ideas to improve the role of OSINT, including creating an OSINT agency or office or contracting out stronger OSINT services. The common goal is to elevate OSINT to a full-standing INT that is readily available to all analysts, as opposed to the more random situation that currently exists. One of the DNI’s first responsibilities was to report to Congress on the future of OSINT and the possibility of creating a separate OSINT agency. The WMD Commission (Commission on the Intelligence Capabilities of the United States Regarding Weapons of Mass Destruction) recommended creating an Open Source Directorate at CIA. DNI Negroponte created an Open Source Center (OSC), which was largely a renaming of the old Foreign Broadcast Information Service (FBIS), which had long been the leading OSINT producer. Negroponte gave management responsibility for the OSC to the CIA, which made sense given that FBIS had long been part of the CIA’s Directorate of Science and Technology (DS&T). However, this left the OSC open to criticism that it was old wine in new bottles. The DNI’s office and OSC have gone to great pains to show that there is now an increased reliance on and better utility of OSINT in intelligence products, including the President’s Daily Brief(PDB) and National Intelligence Estimates (NIEs), such as the Iran WMD NIE. Today OSINT probably has greater visibility than has been the case for many years, but it is not clear if it has the kind of bureaucratic backing within the DNI’s office or in Congress that many believe are necessary.
Two final collection issues that are part of the reform debate have already been discussed: the balance between HUMINT and the technical intelligence (TECHINT) and the need for improved TPEDs (tasking, processing, exploitation, and dissemination).
Turning to analysis, the main issues highlighted in the 2004 legislation were ways to improve the oversight of intelligence at the DNI level in terms of timeliness, objectivity, and quality of analysis as well as to foster more alternative analysis. Similar issues were discussed in the 2005 WMD Commission report. Although the goals are worthy, underlying the provisions may be a changing view about the acceptable tolerances within which intelligence analysis exists. Coming up with reliable standards for assessing the quality of analysis is difficult (see chap. 6). An obvious but stark one would be “right or wrong.” The problem with this standard is that most analytical issues play out over time, during which some analytical judgments are right and some are wrong. The analytical standards promulgated by the office of the deputy DNI for analysis recognize this time lag problem. In the end, creating a balance sheet could be possible, but doing so would be secondary in importance to whether policy goals were met over that period. The intelligence community’s experience with the Soviet Union is instructive. Over the forty-four years that the intelligence community spent analyzing the Soviet Union, it made numerous analytical judgments. Again, some were right and some were wrong. The wrong judgments, although problematic, never put U.S. security at risk. More important, however, is that U.S. policy, supported by intelligence, succeeded and the Soviet Union collapsed. Part of the problem is perceptual. By virtue of the issues they address and the highly charged political atmosphere in which they now exist, most attention goes to NIEs. These are high-value analytical products, although they have not tended to be influential in terms of policy making. A great deal of the intelligence community’s analytical effort takes place at a lower, more constant level, providing daily intelligence support to a broad range of policy makers.
But September 11 and Iraq WMD, for example, reflect a starker situation. Despite multiple warnings about al Qaeda hostility and intentions, no specific intelligence warning was possible about the terrorist attacks. Many have noted that the threads of intelligence make sense only in hindsight, but a body of opinion sees September 11 as a right-or-wrong issue. The discussion may be on firmer ground with Iraq WMD. The situation on the ground did not reflect prewar estimates. Although, as DCI Tenet pointed out in his 2004 speech at Georgetown University, parts of the estimate were borne out, both over- and underestimative judgments were made, and overall the analysis was not correct. But does this argue for the acceptability or the wider utility of a right-or-wrong standard? The fear among some in the intelligence community is that little tolerance now exists for anything other than absolutely correct intelligence judgments and that the new provisions regarding analysis reflect this view. If so, then the intelligence community is doomed to fail, as it will never achieve success in a right-or-wrong system. It is also difficult, although probably necessary, to have a discussion about how right analysis can or should be. The answer is “right as often as possible.” But the key to that answer is the word “possible.” A level of reasonable expectations of intelligence analysis may have been lost and will be difficult to regain.
The 2007 NIE on Iran’s WMD program is seen by many as the “antidote” to the Iraq NIE. Based on background briefings given by senior officials in the DNI’s office, there were “lessons learned” from the Iraq experience. These largely have to do with more intense vetting of intelligence sources, especially the new intelligence that led to a reversal of views from the 2005 NIE on the weaponization aspects only of Iran’s nuclear activities, and more rigorous uses of various competitive methodologies. These are all to the good. However, they do not necessarily mean that the Iran estimate is more likely to be correct than was the Iraq estimate. As noted earlier, the intense use of NIEs by partisans in both parties and in both the executive branch and Congress will make it increasingly difficult for analysts to make “tough” calls without fearing that not only their work but their motives for having produced analysis with a particular outcome will be suspect by one side or the other in a debate. Finally, the 2007 declassification of the national intelligence program total, $43.5 billion, will inevitably lead to questions as to why, given that sum of expenditure, the intelligence community cannot do better. Again, intelligence is among the most difficult government activities when it comes to trying to relate expenditures to results.
The reports of the 9/11 Commission (National Commission on Terrorist Attacks upon the United States) and the WMD Commission also focused attention on how best to organize analysts across the community. The 9/11 Commission recommended organizing all analysts by regional or functional national intelligence centers. In the commission’s concept, the centers would carry out all-source analysis and plan intelligence operations and “would be housed in whatever department or agency is best suited for them.” The center concept arose during the late 1980s and early 1990s as the community put increased emphasis on transnational issues that—by definition—crossed national borders. The centers allowed analysts from various agencies to be brought together to focus on an issue, although the centers have always been dominated by the CIA. The commission’s idea would do the same for other functional and, now, regional issues.
The 2004 reform law only mandated one center, the NCTC, but also stipulated that the DNI report to Congress on creating a center for nonproliferation, which has been done. A clear expectation is presented in the legislation that other centers will be created as well. The main advantage of the centers is, in theory, the ability to get cross-cutting analysis on an issue, assuming that they are true community centers and not dominated by one agency. But centers also have disadvantages.
• To date, centers have been primarily functional in nature. Although analysts in centers do consult with their regional colleagues, this does require some effort. Centers have an occasional tendency to focus on their functional topic and to be less able to bring in the regional or national context in which these issues occur. Organizing by centers—either regional or functional—could exacerbate this tendency. Also, some of the issues handled by centers have close relationships, such as terrorism and narcotics. Sharing analyses across these boundaries can also be difficult. In other words, centers may create analytical stovepipes of their own.
• Getting resources out of or away from centers has proven difficult once they are established. Although a DNI should be able to effect changes, past performance indicates that centers do run counter to the desire for greater analytic agility.
• Centers tend to focus on the most pressing issues. In a center-based community, devoting some level of resources to those issues or regions that have lower priorities may prove even more difficult than it has been.
The WMD Commission recommended the creation of one new center, the National Counterproliferation Center (NCPC), although it would not function like the other intelligence centers. As established by the DNI in 2005, the NCPC serves “to identify critical intelligence gaps or shortfalls in collection, analysis or exploitation, and develop solutions to ameliorate or close these gaps.” Thus, it is not an intelligence production center. Its ability to carry out its mandate depends on knowing exactly what the intelligence community is doing concerning proliferation, and the DNI’s ability to then make changes. The commission also recommended the creation of mission managers to oversee both collection and analysis for the more important topics or issues. The commission’s report was vague on who would decide among the mission managers when it came to resources. Presumably the DNI or someone on the DNI’s staff would be designated. But it is important to understand that the two major resource decisions—collection assets and analysts—tend not to occur within similar time frames. Decisions about analysts tend to be made for longer terms, except during periods when a particular area of higher interest emerges. Decisions about collection assets tend to be more tactical as more frequent calls to adjust collection priorities may be heard. Thus, the mission manager concept will not work without an adjudicator or adjudicators who are familiar with both collection and analysis. Also, existing structures within the intelligence community are similar to the mission manager concept, albeit without authority to move resources, which the new mission managers presumably will not have either.
Closely related to the center issue is the older issue of the flexibility and agility of the analytical corps. The analytical agencies have no reserve or surge capacity. Analysis is still organized around two basic structures: regional and topical offices. These are not mutually exclusive, but no intelligence service around the world has discovered a third organizing principle.
The problem stems, in part, from the fact that analysts have to be expert in something, which necessarily defines and limits the issues on which they can work. Creating a corps of intelligence generalists is impractical and dangerous. They will likely know a little about many issues but not much about any single issue. Successful intelligence analysis requires expertise, and long-term expertise is one of the major value-adds of the intelligence community. Thus, the problem is to maintain some level of flexibility or surge within this body of experts.
Surge is most important during crises, especially in areas that previously had a low priority. However, in giving a low priority to a particular issue or nation, the policy and intelligence communities have already decided not to allocate many resources to it. Short of either finding someone already on staff who has some working knowledge of the issue or dragooning others into working on it, not much can be done internally. A frequently suggested reform proposal is the creation and use of an intelligence reserve—a body of experts, either former intelligence analysts or outside experts, who can augment analytical ranks during a crisis.