Professor Samuel Rascoff’s Presidential Intelligence1 reflects both the conceptual and research strengths of the author, which are formidable, and the practical difficulties of intelligence reform, which are no less so. Rascoff is certainly right that to be effective — in the still-unfolding constitutional environment that must contend with terror groups armed with unprecedented weapons and communications technology — the intelligence community (IC) must act within the law and the rules governing that community must be reformed to make this possible. He is inclined to believe that the answer lies in heightened presidential management. I’m not so sure. The actual presidential control over the vast intelligence apparatus is much more extensive than Presidential Intelligence acknowledges,2 although it is so secreted in the daily flow of tasking, decisionmaking,3 and after-action reporting that few outside a very limited number of White House offices can actually know and appreciate its extent. I am very doubtful that the mistakes attributed to the zeal of the IC were committed without direction from the most senior officials in the Bush and Obama Administrations. Moreover, Rascoff’s proposals to enhance presidential control may well be eclipsed by the recent announcements made by Director John Brennan about the reorganization of the Central Intelligence Agency (CIA) because Brennan’s proposals go to the real problem with the intelligence agencies of the United States.
That problem4 is not an insubordinate IC, but one less effective than it might be, although I agree that organizational reform can make an important, positive contribution. The CIA was created in 1947; President Harry Truman explained that its purpose was to prevent another Pearl Harbor.5 Nevertheless, despite many successes in the Cold War, the U.S. intelligence community was unable to prevent the surprise attack on New York and Washington in 2001. Almost a year later, the National Intelligence Council produced a National Intelligence Estimate (NIE) concluding that Iraq had reconstituted its nuclear weapons program, among other weapons of mass destruction (WMD).6 A subsequent review by the Iraq Survey Group determined that this assessment was inaccurate and premature.7 The NIE also claimed that Iraq’s biological weapons capability had surpassed the arsenal it had amassed before the first Gulf War and that Iraq had developed mobile biological weapons labs.8 The NIE claimed that Iraq had obtained unmanned aerial vehicles to deliver biological weapons and that it had resumed its production of chemical weapons and had accumulated stockpiles up to 500 metric tons.9 All these conclusions were wrong.
The commissions convened to analyze the 9/11 and Iraqi WMD failures found little in common. The 9/11 Commission Report faulted the IC’s failure to share information within the community: “With each agency holding one or two pieces of the puzzle, none could see the whole picture.”10 There was a “failure [to] ‘connect[] the dots.’”11 The inattention, even uninterest, of the President regarding al Qaeda before the attacks was, if anything, a contributing factor, something a more intensive control of intelligence collection priorities by the White House could only have made worse. The WMD Commission Report, by contrast, found that rather than a failure to share information, it was the general conformity of the analysts in different agencies, working from a few pieces of widely shared intelligence, that got the picture wrong.12 These analysts had enthusiastically connected the dots, creating a myth much as drawing lines connecting the constellations created images of a hunter, a bear, a bull, and a boastful if beautiful queen that were in the end only stars. It seems that there was too much presidential (and vice-presidential) guidance.
These two historic fiascos did share something, however: the fundamental structure of problem-solving within CIA and throughout the IC.13 That structure is determined by the following dichotomies:
(1) the division between the public and private sectors; (2) the separation between the domestic and the international; (3) the different rules we apply to law enforcement and intelligence operations; (4) the different reliance we place on secret as opposed to open sources (newspapers, articles, and monographs, or interviews with their authors); (5) the distinction between intelligence collection and analysis; and finally, (6) the differing roles of intelligence producer and intelligence consumer.14
And although these structuring distinctions enabled the U.S. intelligence services to successfully navigate the challenges posed by agencies operating in secret in the Cold War, the disasters of 9/11 and Iraqi WMD can be directly attributed to them.15
CIA Director Brennan’s reforms are an effort to overcome this structure and the problems it imposes as the United States confronts a new intelligence reality — a reality that is more diffuse, where the collection of human intelligence is more difficult, and where America’s adversaries are more innovative.
The key reform is the creation of “mission centers” not linked to any particular CIA directorate, each center being led by an assistant director. Do not be misled by the fact that the mission centers, like the directorates from which they draw personnel (and like their counterparts at State or Defense), are organized regionally and functionally. The creation of these centers represents a profound change. Each center will have teams of analysts and operators working side-by-side who are drawn from the directorates. Each center will have responsibility for espionage, analysis, and covert action within its assigned mission area. Indeed, it is most significant that the current National Counterterrorism Center (NCTC) provides a precedent for these mission centers, because the changes in terrorism since al Qaeda emerged to declare war on the United States represent a threatening phenomenon that is most challenging to those six structuring dichotomies: al Qaeda outsources its activities, merging private and public spheres; it operates across borders, moving between the domestic and the international, unconfined by any particular state territory; it commits crimes to further its political goals, and in fact depends upon criminal activity to operate; it is difficult to penetrate but advertises itself relentlessly in the media, including social media; it requires the closest collaboration between producer and customer to confront it because the usual customers in the policymaking departments can’t be relied upon to ask for briefings or task analysts in such novel and unpredictable circumstances; and its defeat cannot be achieved if analysts are unsure of the sources of the information on which they rely and collectors are not tightly confined to information that is actually useful to analysts, a requirement that is far from obvious when we move away from state-on-state collection.16
The most important objective of this reorganization, however, is not simply to enhance collaboration by the colocation of analysts and collectors and by organizing around missions and objectives, or even to achieve a greater coherence than the structure we inherited from the Cold War. Rather, that objective is to create accountability through the assistant directors. At present, responsibility is diffused among the directorates and there is no one person the CIA director can call on to summarize future trends, current operations, and threats in any particular area outside counterterrorism and counterintelligence. The head of the directorate for operations is responsible for collection but not for the use to which the information that is collected is put; the director for intelligence is responsible for analysis but not for the reliability of the intelligence on which the analysis is based, and so on.17
The Snowden/Silicon Valley problems on which Rascoff focuses18 are grounded in objections to the IC’s collection practices. Subjecting these practices — the sensitive “sources and methods” that are invoked by the IC when it fears exposure — to greater scrutiny may well be necessary to regain the legitimacy of the IC in the public’s eyes, but such exposure is fraught with danger. As recently as October 10, 2015, the Editorial Board of the New York Times wrote an article entitled Why Is Money Still Flowing to ISIS?19 Apparently the editorial board of the Times is innocent of the knowledge that it was the Times’s own reporting — over the passionate objections of the White House — that exposed the penetration of the SWIFT banking transfers system by the US which had been so important in identifying and thwarting terrorist financing.20 To be fair, the Times’s public editor later expressed regret over this.21 Perhaps someday, as part of a plea bargain no doubt, Mr. Snowden can be brought to concede that whatever the benefits of his actions, some of those benefits have accrued to the very organizations and states who are our adversaries.
There are executive agencies that deem themselves apart from presidential control; in some respects the post-Watergate Department of Justice has evolved in the direction of Vestal Virgins rather than in response to the wishes of the incumbent in the White House. The ludicrous and embarrassing arrest of an Indian vice-consul in New York22 is a recent consequence of this hermetic separation of information sharing and decision making from supervision and direction by the President. But the IC is not at the top of this list, by a large margin.
I hope that I am not totally insensitive to the public’s concern about the chagrin and embarrassment a failed or faulty espionage operation can cause, though I imagine the news media is more acutely sensitive to such matters than the public at large, perhaps in an excess of schadenfreude toward a fellow institution engaged in not dissimilar activities. The New York Times Rule — “never do anything you wouldn’t want to see reported on the front page of the New York Times” — is, as Slick points out,23 a perfectly awful guide to govern intelligence collection not only because it puts embarrassment ahead of effectiveness but also because it would animate a self-censorship that would disappoint even the investigative journalists who profess to be most outraged on behalf of their readers by the conduct they relish in exposing.
Before we adopt Rascoff’s proposed reforms, we had better satisfy ourselves that (1) the description he gives of an IC insufficiently tethered to presidential oversight is accurate and (2) that the reforms he proposes would not only address this lapse of control — if it exists — but would make the work of the IC more effective by making it more responsive to the needs of the President and the country. That is why I think the Brennan reforms are to be much preferred. Effectiveness and oversight should not be opposing goals. When they are, the analysis and the prescriptions must be rethought.
* Herbert Wechsler Professor of Federal Jurisprudence, Columbia Law School; Distinguished Senior Lecturer, University of Texas Law School.