National Security Intelligence

by Loch K. Johnson

  Readability

  Considerable effort goes into making intelligence reports inviting to read: the four-color quality of the PDB, for instance, fully equipped with charts, graphs, and photographs – all on high-quality paper. It is not enough to write a solid intelligence assessment about some event or condition in the world; the report must be marketed as well, to catch the attention of busy policy officials. The language must be straightforward, too, even if the subject is economics and the temptation to use econometrics and jargon is great. Policymakers have a low tolerance for the obscure – and few of them have doctorates in economics.

  Brevity

  Like many overworked government leaders, the great Secretary of State George Catlett Marshall placed a premium on succinct reporting. When George F. Kennan became the first director of the Policy Planning Staff in the State Department, he asked Marshall what his instructions were. “Avoid trivia,” the General replied. Prime Minister Winston S. Churchill of Great Britain felt the same way. In a compliment to one of his favorite policy advisers, Churchill referred to him as Lord Heart-of-the-Matter. Because of this emphasis on getting to the point quickly in our fast-paced society, praise descends upon those report writers and oral briefers who cut to the chase. It is also why the modern NIE will have to be trimmed down in length at least somewhat, although not to the degree of losing its in-depth research contribution to policy debates.

  Imagination

  The best analysts attempt to think imaginatively about how an adversary might try to harm the United States. Would a terrorist organization resort to aerial terrorism against the skyscrapers of New York City – a question insufficiently pondered before the 9/11 attacks would ultimately provide a tragic answer. What targets are most vulnerable inside the United States today? What are the likely strategic objectives of China in the South China Sea, or Russia on their border with Eastern Europe? What are the odds that a missile might strike the United States without anyone in Washington knowing who fired it? How likely is it that a pandemic might sweep from China or Africa toward the United States? To what extent is environmental degradation dangerous to U.S. security interests?

  Jointness

  The United States has seventeen intelligence agencies, plus a range of “ints,” for a reason: the world is large and several approaches to intelligence-gathering are necessary to ferret out the information Washington leaders desire. Yet a fragmented flow of information to decision councils from seventeen different “hoses” would be overwhelming and confusing. Instead, presidents and other leaders want “all-source fusion” – a thoughtful blending of all the int findings into comprehensive reports (“jointness,” in military intelligence jargon). This requires the sharing and integration of intelligence findings, which the Kean Commission inquiry found to be the weakest link in U.S. intelligence in the run-up to the 9/11 tragedy.

  Objectivity

  Intelligence reports must be free of political considerations and attempts to please decision-makers – a cardinal requirement. The Nazi Foreign Minister Joachim von Ribbentrop issued orders to his ministry that reports in contradiction to the views of Der Führer would be ill-advised, hinting further that anyone guilty of this offense would face an unpleasant outcome.105 With these sorts of edicts traveling out from Berlin to the various bureaucratic entities of the Third Reich, the end result was predictable: Hitler and his retinue descended deeper and deeper into self-delusion, cut off from accurate intelligence about the progress of the war against Britain, Russia, and the United States. A strong sense of professional ethics keeps most intelligence officers in the democracies on the straight and narrow path of honest reporting. Occasionally, though, some individuals in the intelligence agencies have set aside this sense of ethics, as when DCI Allen Dulles failed to stand up to inflated DoD estimates on Soviet bomber production rates; or when DCI Richard Helms deleted a paragraph from an Estimate on Soviet missilery, under pressure from the Nixon Administration to paint a more frightening portrait of America's Cold War enemy.106 This “politicization” of intelligence is much more likely to come, though, from the consumer side of the intelligence equation, as the less scrupulous of America's policymakers cherry-pick intelligence reports and otherwise bend and twist the findings to suit their political agendas.

  Specificity

  Finally, and most difficult of all, the best intelligence reports carry a sufficient degree of specificity to provide policymakers with an ability to take action. Intelligence officers are often able to monitor an increase in the flow of messages between members of an adversarial group or nation – say, the telephone conversations of ISIS lieutenants. This kind of “traffic analysis” that reveals an increase in enemy communications – more “chatter” – can be valuable to know, triggering a greater concentration of ints toward the target. Yet even more valuable would be a quick deciphering of the encoded messages and translation of them into English, which might point to an enemy's precise attack plans. In short, vague airport “orange alerts” about possible terrorist attacks provide little help; what is necessary are a terrorist organization's detailed and timely attack plans – actionable intelligence. The analyst must be explicit and honest about the limits of data in a report, conveying a sense of how confident he or she is about the findings.

  This list of reporting attributes adds up to a lofty set of standards, on top of which comes the necessity of easy access to top policymakers. Given a lack of rapport at the level of intelligence dissemination, the intelligence will go unread or will fall on deaf ears. That is why the relationship between top intelligence managers – the DNI and the various agency directors – and senior policy officials is so important. DCI James Woolsey was rarely able to sit down with President Clinton and talk about intelligence findings; William Casey, despite his many deficiencies as DCI, enjoyed open access to President Reagan and, as a result, could return to Langley with full knowledge of the main foreign policy problems confronting the Administration – an indispensable ingredient for the achievement of relevant intelligence tasking. Often the Intelligence Community will fall short of these demanding standards, because of inadequate collection (during the war in Vietnam, for example, the CIA was never able to penetrate the North Vietnamese government with a spy107); the slow translation of foreign language and coded materials; flawed analyses; the political misuse of reports by policymakers; or a lack of access to the President and other top officials. Despite such challenges, the secret agencies must aim for the highest possible quality in intelligence reporting; policymakers must constantly resist the temptation to twist these reports for political purposes; and the producers and consumers of intelligence must develop bonds of trust in their common quest for better information to inform decisions.

  As for NIEs, they often make a contribution in the mix of products for decision-makers, but improvements are necessary.108 Estimates have to be more nuanced, with dissents – sometimes known by analysts as “footnote wars” – boldly presented (“flagged” – including both within the body of the text and in the Key Judgments or “executive summary” portion of the report). Their production levels ought to rise as well (in one year under DCI George H. W. Bush, the community produced only five NIEs); they need to be shorter (thirty rather than one hundred pages long); and they should be completed in six months at the most – and much faster in emergencies. Moving the Office of the DNI to Langley would also make sense from the point-of-view of the NIE production process (and also the PDB). After all, the vast majority of the government's top strategic analysts are located in the CIA's Directorate of Analysis (DA) and the NIC, not in the Office of the DNI at Liberty Crossing, six miles away from Langley.

  Co-Location

  Some reformers have long believed that analysis could be improved by having a closer relationship between operatives and analysts at the CIA. The operatives in the DO (formerly the NCA) enjoy “ground truth” about countries overseas, since that is where they serve under official or non-official cover. This gives them a ce
rtain inside knowledge, from café life to the nuances of local slang. The analysts are experts about foreign countries, too, and travel abroad, but for shorter periods of time. Their primary knowledge comes from study; they typically have a PhD that reflects their advanced book-learning and research on international affairs. Though quite different in their career paths and daily experiences, both groups can bring something to the table when a specific nation or region is the focus of U.S. attention. Yet traditionally operatives and analysts have been located in separate places at the CIA, behind doors with combination locks that bar any outsiders from entering. This can have unfortunate consequences.

  For example, in the planning that went into the Bay of Pigs covert action in 1961, the DO operatives were enthusiastic and confident about the prospects of a relatively easy overthrow of Fidel Castro; the people of Cuba would supposedly rise up against the dictator once the Agency had landed its paramilitary force on the island beaches. In another part of the CIA, however, the analysts with expertise on Cuba understood that an uprising was highly unlikely; as they spelled out in a SNIE in December 1960, the people of Cuba adored their leader (Fidel Castro) and would fight a CIA invasion force door-to-door in Havana and across the island. The DO could have benefited significantly from rubbing shoulders with their colleagues in the DI (now called the DA); perhaps that interaction would have brought a stronger dose of realism to DO planning. Nor was President Kennedy made aware of the DI's views on the unlikely success of this scheme. The head of the Bay of Pigs planning, Richard Bissell, enthusiastic about the paramilitary operation and the advancement the successful overthrow of Castro might bring him personally in the CIA leadership hierarchy, had some corridor knowledge of the skeptical SNIE but elected never to bring it to the President's attention.109

  Aware of this physical and cultural distance between the DO and the DI, John Deutch took steps as DCI in 1995 to improve the cooperation between the two at Langley by physically moving together elements of both Directorates. This experience in “co-location” has been uneven. Sometimes the two types of intelligence officers – the doers and the thinkers – have displayed personality clashes that get in the way of sharing information; still, on other occasions, the experiment has led successfully to the achievement of its goal: a blending of in-country experience with library learning to provide intelligence reports with more richly textured insights into world affairs. In 2009, the Director of the CIA, Leon Panetta, said that there would be “more co-location of analysts and operators at home and abroad” in the coming years, adding that greater fusion of the two groups “has been key to victories in counterterrorism and counterproliferation.”110 In 2010, he announced the formation of a CIA Counterproliferation Center to combat the global spread of WMD. In the Center, which would report to Panetta and further upward to the DNI's National Counterproliferation Center, operatives and analysts would work cheek-by-jowl in the spirit of “co-location.”111 Panetta's successor John Brennan, who assumed office in 2013, has pushed the CIA even further in the direction of co-location, with a major integration of operatives and researchers through the Agency's Headquarters Building at Langley.

  Dissemination

  Once prepared by analysts, intelligence reports are distributed to those who make decisions (and their top aides). This may seem easy enough, but even this phase of the intelligence cycle is rife with possibilities for error. For one thing, policy officials are often too busy to read documents provided to them by the IC. “I rarely have more than five minutes each day to read intelligence reports,” an Assistant Secretary of Defense (a former Rhodes scholar and Harvard University professor) told the Aspin–Brown Commission in 1995.112 That a nation may spend a king's ransom on the collection and analysis of national security intelligence, only to have the findings ignored by busy decision-makers, is a disquieting paradox. No wonder Betts concludes that “the typical problem at the highest levels of government is less often the misuse of intelligence than the non-use.”113 Misuse occurs as well, though – all too often, as when leaders cherry-pick the portions of a report they like and dismiss the rest.

  Always of concern is the proper relationship between analysts and decision-makers. If a NIO becomes too cozy with those in power, the danger of politicization rises as the analyst may be tempted to bend intelligence in support of policy objectives – what is known as “intelligence to please.” Yet if the NIO is too detached and unaware of the policy issues faced by a decision-maker, an intelligence report risks being irrelevant. The skillful analyst will carefully navigate between this Scylla and Charybdis, developing rapport with policymakers to understand better their in-box pressures, while keeping a distance from the politics of an administration.

  An important debate on this topic revolves around whether or not DNIs, D/CIAs, and other intelligence managers and analysts should enter into discussions with decision-makers about policy recommendations. Or should they maintain a strict no-cross zone between the presentation of facts – a universally acknowledged intelligence duty – and commenting on ideal policy directions. Richard Helms, DCI from 1966 to 1973, argued for neutrality; so did Sherman Kent, who thought a high wall should exist between intelligence and policy officers. As Helms put it:

  My view was that the DCI should be the man who called things the way he saw them, the purpose of this being to give the president one man in his Administration who was not trying to formulate a policy, carry out a policy, or defend a policy. In other words, this was the man who attempted to keep the game honest. When you sit around the table and the Secretary of State is propounding this and defending this, and the Secretary of Defense is defending this and propounding that, the President has the right to hear somebody who says: “Now listen, this isn't my understanding of the facts” or “That isn't the way it worked.”114

  Other intelligence officials, though, such as DCIs John A. McCone (1961–65) and William J. Casey (1981–87), have taken an active – often aggressive – role in debates over policy options.115 McCone even strongly recommended a military invasion of Cuba in 1962. Whatever one's philosophy on this subject of intelligence activism versus neutrality, it might be difficult in practical terms to stay apart from the policy fray as an intelligence officer. As a senior analyst has noted, “When it's 8:30 at night and the Undersecretary of State says, ‘What do you think I should do?’, you can't say at that point: ‘That's not my job, Mr. Secretary.’ You just can't do that.”116 One solution followed by DCI Woolsey was to stay out of policy debates until after the formal meeting was over and the room cleared; then, if asked, he offered President Bill Clinton his personal views at that time. The most important obligation of an analyst or an intelligence manager is to resist political pressures from those in high office to twist intelligence in a manner that suits an administration's policy preferences, at the expense of the actual meaning of an intelligence report. Here is the soul-destroying sin of politicized intelligence – the gravest hazard of the intelligence cycle.

  With respect to the Iraqi WMD controversy in 2002, the DCI at the time, George Tenet, had an obligation to spell out for President George W. Bush the weaknesses in the intelligence reporting. While it is true that thirteen of the sixteen U.S. intelligence agencies at the time concluded that WMD probably did exist in Iraq, every analyst knew how soft the data were (as personified by Curve Ball's unsubstantiated speculations). Neither Tenet nor the 2002 NIE adequately communicated this softness to the President, however, or to Congress and the public. Had these weaknesses been highlighted, responsible officials could have argued for a delay in the invasion plans until firmer data were acquired. Intelligence reports, based on flimsy evidence, had helped pave the way to war – although the Bush Administration may well have taken up arms against Saddam Hussein regardless of intelligence findings, so intent were its leaders on regime change.

  Weaknesses in the Iraqi WMD data were improperly dealt with in Great Britain as well. The Prime Minister's communications director gave to the British people th
e impression that Iraq had WMD that could strike the British Isles, even though MI6's intelligence report on this subject noted only that Iraq probably had tactical WMD that might well be used against a British and American invasion force on the battlefield in Iraq. Neither the Prime Minister nor the Director of MI6 ever corrected the distorted record presented by the political leaders in the British government, which helped turn British public opinion more toward a pro-invasion stance out of fear that Saddam Hussein harbored nuclear (and perhaps chemical and biological) weapons that could hit London and other targets in the United Kingdom.117

  The ongoing quest for better collection and analysis

  Despite the well-intended efforts of many intelligence officers and policymakers to make the intelligence cycle function smoothly, serious questions remain about the usefulness of the information provided by the Intelligence Community. Even with the staggering amount of money spent gathering and analyzing national security intelligence each year, many consumers find its products lacking. “We never used the CIA stuff,” recalls a former U.S. ambassador and assistant secretary of state. “It was irrelevant.”118 A survey of intelligence consumers conducted by the Senate Select Committee on Intelligence during the 1980s revealed widespread disdain toward the value of the Community's analytic work. Most widely reported was Senator Daniel Patrick Moynihan's sweeping indictment of the CIA for failing to predict the fall of the Soviet Union.119