Judith Yaphe, who served 20 years as the senior analyst on Middle Eastern and Persian Gulf issues in the Office of Near Eastern and South Asian Analysis for the Central Intelligence Agency, and was the CIAs chief political officer for Iraq, discusses the problems of supplying intelligence information for policymakers who may, or may not, agree with the CIA conclusions.
Currently director of the Middle East Project at the National Defense University in Washington, D.C., Yaphe says, Intelligence, in many ways, is like reading the Bible. You find what you want to support almost any argument.
Yaphe was interviewed by Bernard Gwertzman, consulting editor for cfr.org, on June 4, 2003.
Much has been made of the fact that, in the lead-up to the American decision to go into the Iraq war, the Bush administration claimed Iraq had a dangerous weapons program and that Baghdad might give weapons of mass destruction to al-Qaeda. How do such seemingly inaccurate assertions emanate from the intelligence community?
An intelligence agency in general receives all sorts of reporting. The kind of intelligence that youre referring to would probably come primarily from what they call humint, which is a human source telling you something, and from sigint, which is acquiring information through technical means [like wiretaps]. The problem is determining how accurate such reports are.
How do you do that?
If you are an analyst, you ask: Why is the person telling me or saying what he or she is saying? Whats the motive? What positions does he or she represent? Whats to gain out of it? Whats the goal? Can I trust what I hear or am I being led to believe something that just isnt true? A professional would look at these things, this information, and should be able to balance, to weigh, to judge their veracity against other sources and against the analysts own instincts.
If youve got experience in covering a country, a leader, a group, or a region for a long time, as many people have on Iraq, you should be able to say, This doesnt taste right, this doesnt sound right, read right for a number of reasons that [cause you to make] that judgment call. In assessing intelligence, for example, from the [Iraqi] opposition groups, you have to say: What is their access? Why are they telling us this? What do they want from us? What kind of action or reaction are they trying to frame? Whats important in all of that would be an ability to verify. One source telling you something is insufficient. Its crucial, if you can, to get a second source, or even [compare what you are hearing with] what makes sense to you, what fits your sense of history.
Heres where we get into a more subjective area, but if it doesnt taste right, if it doesnt seem to fit in the context of what you understand, then you have to question it. I make this point because I think if you bring in people to look at past intelligence, you can find what you want. Intelligence in many ways is like reading the Bible. You find what you want to support almost any argument.
Certainly, the intelligence was distilled for Secretary of State Colin Powell, who went before the U.N. Security Council with a rather strong, lengthy statement saying that Iraq in fact had arsenals of chemical and biological weapons.
First of all, when an intelligence report is written, whether it is an estimate or a write-up of weve received the following information, an intelligence officer will always try to categorize the level of reliability that information is assessed to have: Does it come from a source with reliable access or not? My experience over more than 20 years is that [in most cases], once that paper or that information goes into the hands of consumers--policymakers— they dont read those warnings. Its like the label on your drug prescription— you should not take this if --how many people read those inserts?
Thats an important point, because you have to be aware of some of the warning notices. Intelligence is not a perfect system. It doesnt know all and see all. Even what some would consider the most reliable sources, the things that you can hear, or the things that you see, are not always reliable. We know very well that the Iraqis, like many others, were taught over the years how to fool [information] collection systems. We know that the Iraqis were very good at fooling, deceiving, and denying information from UNSCOM [United Nations Special Commission, which oversaw the elimination of Iraqi weapons of mass destruction from 1991 to 1998] inspectors and others. And its not just the Iraqis who [try to thwart information gathering]. Indians and Pakistanis are pretty good at fooling us in terms of our being able to monitor their nuclear-testing sites.
In the Vietnam War— Im going back a bit because I covered it in Washington— there were charges made at the time, and I think the archives have now made the case, that a lot of the information provided by the CIA was so politicized by administration officials that it distorted what was going on in Vietnam. Is this always a danger— when people want to take intelligence information and turn it into their own political ends?
Of course thats a danger, but I think the problem is this: the intelligence community works for the president of the United States. If intelligence is going to be used, the political leaders, whoever they are, have to decide theyre going to use it. Intelligence does not make policy. Policy can use or ignore intelligence as it chooses. The intelligence community is a tool; its an instrument of national policy. Ive heard the stories from Vietnam— that was before my time at the company--but theres always the danger of politicization. You can provide the best intelligence, the most accurate intelligence, or the most inaccurate, but it is up to the political leaders youre delivering it to, to decide if theyre going to use it.
Do you have any thoughts about the small group that was set up within the Defense Department because the Pentagon felt it wasnt getting good intelligence information on Iraq and on terrorism?
The problem goes to a broader definition of what politicization is. Politicization on the one hand is pressure on intelligence analysts to produce the right kind of intelligence. And no one is going to be naïve and say it never happens. Usually if and when that does happen, the intelligence is massaged at higher [levels than desk officers], when somebody wants to provide what will support policy. But the other form of politicization is when a policymaking arm of government sets up its own cabal, to look at intelligence reports. As I said, its like reading the Bible. You can go back and find things that will support your point of view, and you can select those out, and you can ignore the rest, which is the same thing theyre accusing the agency of doing. Who am I going to trust? I wont answer that question. I think you know where I stand.
Please do answer it.
As a professional intelligence officer, I think that judgments as to the veracity of reports and intelligence should lie with people who make those professional judgments, without being part of policy or looking to support a policy. But once policies are made, its up to administration [officials] to decide what they want to do. They certainly have the right to pull in people to go over these things and to look to see what is missing. Nobodys arguing with that. The question then is: who is to judge what is true or not? Is Monday morning quarterbacking a real way to run a football game?
But its done all the time.
Right, exactly, and thats true with intelligence as well.