When Intelligence Reports Become Political Tools . . .
By Jay Taylor
The Washington Post
Sunday, June 29, 2003
Last week, Chairman of the Joint Chiefs of Staff Gen. Richard B. Myers, commenting on the U.S. failure to find Iraqi weapons of mass destruction, stated, "Intelligence doesn't necessarily mean something is true. . . . It doesn't mean it's a fact. . . . you make judgments." A week earlier, Deputy Defense Secretary Paul D. Wolfowitz had explained that "intelligence is an art and not a science," a formulation suggesting the need to employ skill and imagination to produce a product pleasing to the senses -- at least to some people's senses.
Experience, imagination and an ability to think beyond the box are all elements of good intelligence, as they are in all professions. But intelligence, like science, must also adhere to rigorous standards of evidence and reasoning, and present facts in as objective a way as possible.
Unfortunately, national intelligence reports are not always fashioned this way. The current flap over whether the Bush administration has misrepresented national intelligence estimates (NIE) to suit its political goals reminds me of a National Intelligence Assessment drafting meeting that I attended at the CIA in 1984. The paper was about an expected crisis in an Asian country. There were two conflicting points of view: The State Department, which I represented, and several other agencies agreed on one set of judgments. The CIA alone had a distinctly different analysis. The group decided that the written assessment would reflect the majority view and that the CIA would explain its disagreement in a footnote. By the time I got back to my office across the Potomac, however, the CIA officer who chaired the meeting had left a message. The assessment would reflect the CIA's conclusions; the views of State and the other agencies would be relegated to the footnote.
William Casey, then the CIA director, had imposed his view on an intelligence analysis. This also happened on assessments regarding the Soviet threat in Central America, Soviet military power, the prospects for revolution in Mexico and the existence or non-existence of "moderates" in the Iranian government. In principle, Casey was within his rights. He was the analyst-in-chief. But the way he asserted his role distorted the assessment process. While all the elements of the intelligence community could get their views incorporated in a footnote, policymakers might never know the extent of disagreement with the Casey view. President Reagan would frequently only read the brief "judgments" section of an NIE or at most the "executive summary."
The making of intelligence assessments is the same now as it was then. There are two aspects: collection and analysis. Thousands of spies and technical people involved in communications intercepts and reconnaissance photography collect facts -- or try to. An important element of judgment enters into the collection process, but there is a procedure for assessing the reliability of sources and methods. This grading of reliability is critical. Drawing on these clandestine reports, but also on open material and historical experience, intelligence analysts construct a cogent narrative and draw conclusions. Here, an ability to synthesize facts and weigh conflicting ones is essential. A good intelligence analyst seeks to combine Mark Twain's description of a newsman colleague and himself: "He knows all that can be known. I know all the rest." But professionalism also requires strict standards of logic and a balanced presentation of conclusions.
During the Casey era, I supervised intelligence analysts in the State Department. On rare occasions, an analyst and I could not work out to our mutual satisfaction a disagreement over the conclusions of a paper we were preparing for the secretary or another "Seventh Floor principal." (The offices of top State officials are mostly on the seventh floor.) In this case, my view would prevail, subject to my own boss's concurrence. The summaries of all the papers we did, however, were required to highlight any striking piece of information or line of argument that contradicted an important judgment and explain why it was thought not to be compelling. The body of the report also had to describe the dissenting argument. It was and is especially important to be balanced and accurate in the executive summaries and the even shorter "key judgments," which provide tempting opportunities for distortion.
George Tenet, the current director of central intelligence, came into office in 1997 giving high priority to maintaining the integrity of the CIA. But over the past year, it appears that he has not served Congress and the American people well on the question of Iraqi weapons of mass destruction and alleged Iraqi ties to al Qaeda. He seems to have engaged in over- and under-statement; highly selective release of facts and assessments, including the clever use of "key judgments" and executive summaries; failure to correct exaggerated statements by the president and others; and failure to stop a maverick Pentagon operation producing intelligence as art.
It may not have been necessary to pressure individual analysts to distort public and congressional perceptions of what the administration knew and did not know. Analysts, like their chiefs, are human and to varying degrees are inclined to go along if the spin on the top of a report is done subtly. Nevertheless, during the buildup to the war, a number of CIA and Defense Intelligence Agency analysts risked their jobs by complaining to journalists about misperceptions that the administration was creating on major issues regarding Iraq. Throughout this period, the CIA director probably -- and this is a subjective judgment -- understood that the evidence of the Iraqi threat overall was flimsy, but he went along with this exercise or at least did nothing to stem the tide of misrepresentations.
Why would a man who has dedicated most of his life to professionalism in U.S. intelligence fail to warn his leaders that they were exaggerating the situation and, in at least one case , even relying on information that was probably fraudulent? One can only speculate. A major factor was probably Tenet's strong identity with the Bush team after Sept. 11, 2001. Following the traumas in New York and Washington, there was heavy criticism in Congress and elsewhere of the CIA's and Tenet's "intelligence failure" in not having had a clue about the coming catastrophe. Some members of Congress demanded he be fired.
But President Bush embraced Tenet, extended his protective umbrella over him, approved a major role for the CIA in the war in Afghanistan and allowed Tenet to publicize this supposedly covert activity. At the same time, Bush authorized a huge expansion of the CIA and its operations around the world. Within months, Tenet went from being a bitterly criticized Clinton appointee with an uncertain future to being the most powerful CIA director since Allen Dulles. For anyone, all this, plus war-room strategizing and daily briefings with the president and vice president, would amount to a heady experience. According to a profile in The Washington Post, Tenet and Bush "bonded." (It didn't hurt that Tenet had named the CIA headquarters after the president's father.) Perhaps Tenet, like Bill Casey, became too good a friend of the president and his political team to be a truly independent CIA director.
Eventually, America's first preemptive war and the resulting occupation of Iraq may end well. If so, the current Iraq flap may be but a historical footnote. If the Iraqi bucket of vipers proves to be bottomless, however, Tenet's accommodation with Bush may come to be seen as central to a major foreign policy failure, conceivably a debacle. Two Republican-dominated intelligence committees in Congress and a CIA in-house investigation are looking for evidence of whether what was known about Iraq was misrepresented.
Regardless of the outcome, several steps should be taken to prevent future distortions of the intelligence process. First, return to the old practice in which the CIA director is not a member of the Cabinet and thus not part of the president's political team. Second, let a senior career CIA officer other than the director brief the president and vice president every morning. Third, just as we now ask corporate chief executives to certify the accuracy of their companies' financial reports, Congress should add a specific pledge to the director's oath of office to assure that intelligence assessments and judgments are as objective as possible and not massaged or selectively released to lawmakers or the public to serve policy goals. Congress might also prohibit other Cabinet officers from setting up new so-called intelligence units without congressional approval (as did Defense Secretary Donald H. Rumsfeld and company).
Above all, the CIA director and other dons of U.S. intelligence must remain dedicated to providing information to policymakers and letting them judge the best course of action. That should be done by leaving craftiness out of the craft of intelligence assessments and giving the president the facts available without regard to anyone's political agenda.
Author's e-mail: jaytaylor888@sprintmail.com
Jay Taylor was the State Department's director of analysis for East Asia and the Pacific, and later deputy assistant secretary of State for intelligence and research, under President Ronald Reagan.
washingtonpost.com |
