The terrorist attacks in 2001 brought renewed calls for intelligence reform, with some of the most persistent advocates arguing, “If not now, when?” Even so, the purposes of reform have not been entirely clear. Several different purposes, not all of which are mutually exclusive, can be discerned.
• To improve the intelligence community’s ability to deal with terrorism overall
• To prevent further terrorist attacks against the United States
• To determine if the attacks occurred because of specific intelligence lapses, and, if so, who was responsible for them
• To use the attacks as an opportunity to push intelligence reform concepts, whether or not related to the attacks or the war on terrorism
However, the issue that provided the ultimate impetus for the intelligence legislation of 2004 was not the investigations into the September 11 attacks but the issue of Iraq weapons of mass destruction (WMD). The gap between prewar estimates and what was found (or not found) in Iraq since military action commenced in 2003 helped push many who had been undecided into the camp of intelligence reform. This factor also helps explain why the legislation focused so heavily on the management and oversight of analysis, with less attention given to the perceived problems that led to the September 11 attacks.
One final factor that must be taken into account is the misperception that the advent of multiple round-the-clock news media makes the intelligence community redundant. Those who hold this view believe that the community must transform itself to be more competitive.
journalism and intelligence have some interesting similarities: the need for reliable sources, the need to make complex stories comprehensible, the tyranny of deadlines. But there are also important differences. Deadlines may be even more tyrannical for the news media—both print and broadcast, but especially the latter—than they are for the intelligence community. News broadcasts must go on the air as scheduled, regardless of the day’s events. Journalists accept this operating necessity and use updates, corrections, or retractions as necessary. Whenever possible, intelligence managers and analysts seek to delay reporting (sometimes too long) until they have the story correct, or as correct as collection will allow. Indeed, Gen. Michael Hayden. when he was the director of the National Security Agency (NSA), urged his staff “to get the intelligence out” as soon as it was useful to someone—in other words, to publish intelligence as soon as it informs someone, even if it can be refined further, at which point it still can be published for yet another audience. Also. the average noncrisis news broadcast contains a great deal of filler and repetition over a twenty-four-hour period. The intelligence community also needs to report, but not around the clock. This is a saving grace. Also, the intelligence community seeks to do more than report: value-added analysis is an essential part of what it produces. Such analysis happens much less frequently in the news media, particularly the broadcast media. When it does occur, analysis can spill over into opinion. Squabbles among the twenty-four-hour news networks about which of them has a liberal or a conservative bias underscore the problem.
Still, the misperception persists, even among some policy makers, that round-the-clock news sources upstage the work of the intelligence community. The misperception that the two are in competition may reveal a less than firm understanding of their fundamental differences, although the news media may employ concepts, technologies, and approaches that would be of use to intelligence.
ISSUES IN INTELLIGENCE REFORM
Discussions about intelligence reform tend to fall into two broad areas: structure—or reorganization—and process. Both approaches have their advocates. Ideally, the issues should be approached together. Altered structure and unaltered process can become little more than moving boxes on the bureaucratic organization chart. Changing the process without changing structure would likely end in few, if any, meaningful results, as the old structure would probably resist the new processes. The following are some of the more frequently discussed issues in intelligence reform, some of which have been mentioned in preceding chapters. Some issues have been settled by the 2004 legislation, but they are likely to be touch-stones of future debate as the new intelligence structure begins to work and remains under scrutiny.
THE ROLF OF THE DCI AND THE DNI. The most central issue in the management and functioning of the intelligence community was the gap between the responsibilities (extensive) and the authority (limited) of the DCI. Under Executive Order 12333 (1981), the DCI was “the primary adviser to the President and the NSC [National Security Council] on national foreign intelligence.” The designation included “full responsibility for [the] production and dissemination of national foreign intelligence,” which included the authority to task agencies beyond the CIA. These responsibilities have passed to the director of national intelligence (DNI), although it is not clear that the problem of the DNI’s authorities has been resolved.
The DNI’s authority remains limited and may be subject to even more stress than was the case for the DCI. Some 75-80 percent of intelligence agencies and their budgets remain under the direct control of the secretary of defense. Any additional power granted to the DNI can come only from the secretary of defense. Although Secretary of Defense Robert Gates (2006- ) and his undersecretary for intelligence, James Clapper (2007- ), have taken a much more cooperative approach to their relationship with the DNI than was the case with their predecessors, it still remains unlikely that there will be major shifts of actual power or control over the two collection agencies [National Geospatial-Intelligence Agency (NGA) and NSA], which are also defense intelligence agencies and combat support agencies. Moreover, it is not at all certain that the agreements struck by the current senior leadership will survive the 2009 transition to new officials who may have less of long-term personal relationship on which to rely. Congress also is a factor in any redistribution of power, with the House and Senate Armed Services Committees jealously guarding the turf of the Department of Defense (DOD) that they oversee. The argument over power is a zero-sum game. Although few, if any, secretaries of defense have believed that the DCI threatened their authority, DOD was clear about preserving all of its authority during the congressional debate in 2004. Defense officials worry about two fronts: the authority of the secretary (often referred to as “Title 10 prerogatives,” as spelled out in the U.S. Code) and intelligence support for military operations. The latter became the main point pushed by DOD supporters in the 2004 debate. In addition to DOD, the DNI will likely be engaged at some point in a struggle with the director of the Central Intelligence Agency (DCIA) over control of covert action and perhaps human intelligence (HUMINT), and perhaps the director of the National Counterterrorism Center (NCTC) when the NCTC director is engaged in strategic operational planning, a function for which the NCTC director is allowed direct access to the president.
Much of the problem with the DCI’s authority stemmed from the origins of the office and how the intelligence community developed. The designation DCI predates the creation of the CIA. The first DCIs ran the Central Intelligence Group (CIG), which became the CIA in the National Security Act of 1947. President Harry S. Truman’s goal in creating the CIA under the DCI was to have a central organization that could coordinate the disparate analyses coming from the State Department and the military. No one envisioned the CIA’s producing finished intelligence in its own right or conducting operations. Thus, the limited authority granted to the DCI was consistent with the role as coordinator. The CIA was seen as the agency that supported this coordinative role.
As the CIA moved to fill both analytical and operational voids, the DCI’s power base grew, but it also diverted the DCI’s attention from the community-wide role. This is the issue that the 2004 legislation sought to correct, freeing the DNI from running any agency and thus allowing the DNI to concentrate on the larger role. What remains at issue is the degree to which this larger role can be accomplished without the strong institutional base that the CIA afforded the DCI. As former acting DCI John McLaughlin (2004) noted in his testimony on the proposed legislation, the reason DCIs have relied on the CIA was that it was the only agency they could command. The DNI does not have this base on which to fall back.
The DCI’s role could have been significantly enhanced if the office had been given budget execution authority over the National Foreign Intelligence Program (NFIP: now the National Intelligence Program). The ability to direct the allocation and spending of money is a major source of power and control, which is why DOD fought to keep this power from the DNI. But a political issue also was involved in choosing this course: It was too bureaucratic and not dramatic enough to suit those seeking major changes. Furthermore, some of the September 11 families proved an effective and difficult-to-refute lobbying force in favor of the DNI I legislation. Although some have questioned the propriety of allowing the families to dictate national security structure, they had tremendous political clout.
DOD’s arguments against ceding spending control to either the DCI or DNI are based on the view that such a change runs the risk of limiting intelligence support to military operations and that, without direct DOD control, the intelligence support may be wanting. The likelihood of this happening seems small. It is difficult to believe that any DCI or DNI would run the political risk inherent in not giving full support to the military in peacetime or in war, if for no other reason than self-protection, to avoid being blamed for military setbacks or casualties. Nothing in past practice over the more than sixty years of the modern intelligence community’s existence would suggest that any substance can be found behind this argument.
The ultimate success or failure of the DNI position remains uncertain. The first DNI, John Negroponte (2005-2007), was criticized by his congressional overseers for not being firm enough in establishing and using his authority. Negroponte deserves credit for getting what many perceived to be an unwieldy structure up and functioning, but he did not exercise much guiding authority over the intelligence community. His successor, Mike McConnell (2007- ), has put more emphasis on the problem areas that he sees as being the main inhibitors to a more collaborative and integrated intelligence community. His
100 Day Plan
and
500 Day Plan
also emphasize how little time he thought he had to get these changes made. Finally, several of the points in the two plans still depend on the DNI having sufficient authority to force compliance or to mete out consequences for obstruction. As DCI Richard Helms (1966-1973) observed about the DCI’s position, much of the DNI’s authority stems from both the real and the perceived support the DCI has from the president. The initial signals from President George W. Bush were ambiguous: granting authority to the DNI, as in the case of the morning briefing, but assuring CIA that its role was not diminished. As was the case when the secretary of defense position was created in 1947, the legislation may have to be revised after a few years, once the flaws have been revealed. The DNI will be under great pressure and scrutiny from the public and from Congress to improve information sharing and the overall coherence of the intelligence community. Whether the legislation provides the levers necessary to do these jobs is not clear. The DNI will also come under tremendous political pressure should another major terrorist attack occur in the United States. Supporters of the new law argue that the creation of a DNI freed from any agency and the emphasis on information sharing will lessen the likelihood of an attack. Many intelligence officers and some outside observers fail to see any connection between the new structures and what is needed to prevent a terrorist attack, as the law seems largely to have created another layer without making any marked changes in how terrorism is addressed.
How will we know if the DNI is working? This is one of the great uncertainties in the role of the DNI, any agreed set of hallmarks that will help determine whether or not the new structure is working, and whether it is making a positive contribution to the management of U.S. intelligence. Most of the “doing” functions of intelligence—collection, analysis and operations—take place within the agencies and not at the DNI level. It will likely be very difficult to prove that improvements in any of these functions can be tied directly to initiatives from the DNI. Some of DNI McConnell’s initiatives—such as improving the security clearance process, improving first generation and immigrant recruitment, improving foreign language capabilities—can be tracked and assessed and would make valuable contributions to the functioning of intelligence. But these are still management functions that would likely fall short of the somewhat hyperbolic aims set for the DNI in the debates over the 2004 legislation. Thus, we are left with either few useful hallmarks by which to judge the DNI function or one very stark hallmark—another terrorist attack—which most observers judge to be likely regardless of which structure is chosen.
STOVEPIPES. As discussed in earlier chapters, the term “stovepipes” refers to agencies in similar or analogous lines of work (collection or analysis) that tend to compete with one another, sometimes to a wasteful and perhaps harmful extent.
The stovepipe issue is most often discussed in reference to the big three collection disciplines (INTs)—signals intelligence (SIGINT), geospatial intelligence (GEOINT), and human intelligence (HUMINT)—and particularly the technical INTs (especially SIGINT and GEOINT). Some have proposed putting at least the technical INTs (SIGINT, GEOINT, and measures and signatures intelligence or MASINT) under a single agency with the authority to decide which INTs should respond to which requirements, thus limiting some collection that may not be optimal or necessary. This solution raises questions of its own.