Gjelten: How Things Have Changed At The CIA
NEAL CONAN, HOST:
This is TALK OF THE NATION. I'm Neal Conan.
Ten years ago, the CIA called it wrong on Iraq. The conclusion that Saddam Hussein had stockpiled weapons of mass destruction turned out to be flat wrong, and played no small part in the decision to invade. A decade later, the issue is Iran and its nuclear ambitions. And again, the stakes are very high. It could involve a decision on war. Last night on ALL THINGS CONSIDERED, NPR's Tom Gjelten reported on these questions after rare access to CIA analysts where he found out how things have changed in the agency since Iraq and how they're applying those lessons.
We'd like to hear from you. Has the CIA learned the lessons from WMD? 800-989-8255. Email: email@example.com. You can also join the conversation at our website. Go to npr.org, click on TALK OF THE NATION. Tom Gjelten's with us here in Studio 3A. Tom, always good to have you on the program.
TOM GJELTEN, BYLINE: Good to see you, Neal.
CONAN: And are the CIA analysts worried about getting it wrong on Iran?
GJELTEN: Oh, they're very worried about it. They're very worried about getting it wrong on any number of issues that they have to analyze, that they have to report on. But Iran carries - as you've just said in the intro, behind the decision to - of what to say about Iran, lies the question of whether to go to war in Iran. And any time a decision to go to war is based on intelligence, then yes, the intelligence analysts have a lot to worry about.
CONAN: In the case of Iraq, the U.S. intelligence got it wrong in 1991. They suspected Saddam Hussein had a nuclear weapons program. Turned out he had three and was much closer to a weapon than they knew at the time. Then in 2003, they assumed he had nuclear weapons or nuclear weapons program and, of course, other weapons of mass destruction, and they were wrong again.
GJELTEN: You know, it's interesting, Neal, you said the word assumed, because this is the - one of the top lessons that the CIA has learned from this, is you just can't make assumptions. And this is something that they now recognize, that their judgment in 2002, they prepared a national intelligence estimate, an official intelligence finding, that Saddam Hussein had weapons of mass destruction, they now acknowledge that it was based on assumptions, the assumption that because he'd had them before, and they now know that he'd had them before, because he had used - we're talking here not just about nuclear weapons, we're talking about biological and chemical weapons. He'd use chemical weapons against his own people. He had kicked out the United Nations and...
GJELTEN: Yes, exactly. And he had kicked out U.N. inspectors. So on the basis of all that, they assumed that he had those weapons. It was inferences, but it was not hard evidence.
CONAN: And interesting, one of the quotes in your piece that I thought was most fascinating: just because you intercept a telephone call and a friend tells somebody something, that doesn't mean it's true.
GJELTEN: Well, one of the things that I found interesting in talking to these analysts, Neal, is how open they were about the imperfections of intelligence gathering. That is with respect to signals intelligence. Signals intelligence or SIGINT, as they say in the intelligence business, is electronic: intercepted emails, intercepted telephone conversations and so forth. And yes, you're right. They say that just because you hear somebody say something to someone else, doesn't mean it's true.
And, you know, there is a certain mystique associated with intelligence and probably some analysts or intelligence gatherers, you know, excuse me, get carried away by what they've heard. The same thing goes for human intelligence. Just because you've got some human source who is in a prominent position, who says such and such a thing is happening, doesn't mean it really is.
CONAN: So what has changed since Iraq? What have they instituted that will conduct things differently?
GJELTEN: Well, I think the most interesting thing is in each department of the directorate of intelligence in the CIA, there is now what's called a tradecraft cell. There's a few specialists who are not, themselves, looking at the intelligence and making judgments. Their job is to show, to teach the analysts there various techniques for checking their own biases, for considering competing explanations for some of the developments that they're trying to follow. They are checks and balances within the agency built into the intelligence directorate, and their responsibility is to say to these another analysts, look, have you considered this possibility? Have you considered your own - the biases that you are bringing to this? And to help them with exercises, explore what their own biases are.
CONAN: This goes back to, well, about 35 years ago. There was - then the big dispute was over the capacities of the Soviet Union. And there was a team B that was instituted at the time to challenge the findings of the CIA and say, wait a minute. No, they're much more capable than you think.
GJELTEN: That's exactly right. And that team A, team B approach, Neal, is one that is still in use. In fact, a lot of these exercises that they have put in place at the CIA go back a number of decades. I mean, intelligence analysts have known about this, but they just were not using them. It's interesting. They now have this - I brought it with me - a trade craft primer. This is a primer that analysts are supposed to follow in making intelligence judgment. So it was only prepared in March of 2009, but a lot of the exercises in here, as you say, go way back.
It's interesting. On this page here, is strategic assumptions that were not challenged. Beginning in 1941, World War II, Japan would avoid an all-out war because it recognized U.S. military superiority. That was an assumption that was not challenged, and, of course, proved to be wrong. You mentioned - and then there's the 1962 Cuban Missile Crisis, a famous one. The Soviet Union would not introduce offensive nuclear weapons into Cuba. That was the assumption that proved wrong. And then, finally, of course, 2003, Iraq's WMD. Saddam failed to cooperate with U.N. inspectors because he was continuing to develop weapons of mass destruction. Another assumption proved wrong.
CONAN: So how do you build an institutional skepticism?
GJELTEN: You have specialists whose job it is to be skeptical, and then you have exercises that you force the analyst to go through before they make a judgment. The interesting one is this thing called the ACH exercise, analysis of competing hypothesis. When you, as an intelligence analyst, come up with a hypothesis, whether it's that Iran has developed a long-range missile and the idea is to give it the capability to deliver a warhead to somewhere else in the Middle East - that's your hypothesis.
But you are now forced by your own colleagues to look at all the evidence that is inconsistent with that hypothesis, not the evidence that's consistent with it. That's too easy. You are required now to tally up all the pieces of evidence that challenge that hypothesis. And if those pieces of evidence, you know, if after considering all the contrary indications, your hypothesis still stands up, then you can go ahead.
CONAN: But doesn't that raise another danger? And that is an analysis that says, well, maybe this, maybe that. We don't really know.
GJELTEN: Exactly right. And that is a big concern. You know, there is this notion in the intelligence community called making the call, and there has always been pressure. As the director of the - deputy director of intelligence for analytic programs Peter Clement told me, when he came into the agency in 1977, his bosses were saying to him, Peter, this is why you make the big bucks. Don't give me any nuance. Make the call. What's the answer?
That's kind of been a culture in the CIA. And now they have really put that aside. They've moved away from it. But you're absolutely right. There is a risk that - because presidents need to know. Presidents and policymakers that need to - they need to make the call. They can't afford to be nuanced. You know, they have to decide - President Obama had to decide: Do I send these special operations troops in to get Osama bin Laden, or do I not?
CONAN: So again, he can say, what's your best guess on whether he's there? And, obviously, we said, well, 55-45. But that doesn't really help you make the decision.
GJELTEN: It doesn't help the president.
CONAN: Let's see if we can get some callers in on the conservation. Our guest, of course, NPR's Tom Gjelten. You may have heard his piece last night on ALL THINGS CONSIDERED, A Peek Inside the CIA. We'll start with Bruce, and Bruce is with us from Captain Cook in Hawaii.
BRUCE: Yeah. Thanks for taking my call. I think there's too much emphasis - I appreciate the analysis, but I think there's much, too much emphasis on the intelligence gathering and the intelligence's conclusion. And the reference to the Cuban Missile Crisis is germane. But the most important lesson of that crisis is that the generals and the military all wanted to attack, and the intelligence supported that conclusion. It was Kennedy's sense of history, I think, that made him make the right call, which went against the intelligence, to not attack, because he had the bigger picture in his mind, which is this could start a very big war. And this could be the beginning of the end and...
CONAN: I think you're confusing intelligence with strategic advice. The intelligence was there were Soviet missiles in Cuba. They sent over U-2s and got pictures of them, and that intelligence turned out to be correct. You're correct, though, that the generals then said, we need to do something about it, and then President Kennedy made his decision. But, Tom, that's an instructive case.
GJELTEN: It is. And I think that that actually - that point, and it's a good one that Bruce made, goes to what we were just talking about, whether intelligence analysts should give the presidents and other policymakers that they serve nuanced analysis, saying that on the hand, on the other hand 50-50. Because in the end, the responsibility, the judgment has to be the president's. You - really, you elect presidents that you hope have that sense of history and have a sense of what's at stake for the nation. And, in a sense, you are, I think, doing them a favor by not saying to them, this is the way it is, and this is what we think you should do.
CONAN: Bruce is also right that the intelligence agencies do not make operational - or they're not suppose to make operational suggestions. Yes, the Soviet missiles are there, and you should do X or Y.
GJELTEN: No. But when George Tenet, the director of the CIA, says to the - to President Bush that it's a slam-dunk case, that certainly sort of takes President Bush off the hook for that decision - in a sense. You could make that argument.
CONAN: Tom Gjelten, NPR correspondent, with us here in Studio 3A. You're listening to TALK OF THE NATION, from NPR News. Rebecca's on the line with us from Hayward, California.
REBECCA: Hi. I don't think the CIA is that stupid. I think that they know full well what was in Iraq, what it had and what it didn't have. And I think we should admit why we're really there, was for oil, and also in Afghanistan for its mineral wealth.
CONAN: Well, it's hard to explain then, Rebecca, why the CIA would have been so - issued such an embarrassing report when it was flat wrong.
REBECCA: Well, maybe it was worth it.
GJELTEN: You know, whether we were in Iraq for oil or not, you know, we - I can't say. I can say, however, that we were not in Afghanistan for mineral resources, because we're about to get out of Afghanistan, and we have not gotten access to any of those mineral resources which are substantial. China is the country that's going to come in and benefit from all those mineral resources.
REBECCA: Yeah. But haven't we made it easier for China to get in?
CONAN: I suppose, but...
GJELTEN: No. I don't think that was the idea.
(SOUNDBITE OF LAUGHTER)
CONAN: May not have been the goal. Thank you, Rebecca. Let's see if we can go to - this is Jordan, Jordan with us from Charleston, in South Carolina.
JORDAN: Yes. Thank you for taking my call. I just wanted to say that I think the intelligence community has learned from the Iraq war. And I think two of the things that were the most important part was that, in that case, the intelligence was politicized, which is something that ideally isn't suppose to happen. For whatever reason, the Bush administration wanted to - wanted to go into Iraq. And...
CONAN: And, Tom, this is exactly something that you asked the analysts about in your story last night.
GJELTEN: Yeah. They, in a sense, you know, it would be easy for the analysts to say, look, you know, we were pushed into this. They twisted our findings. They politicized the intelligence. They actually don't say that. I mean, there are a few intelligence analysts who say that the Bush administration, in fact...
CONAN: Vice President Cheney.
GJELTEN: Vice President Cheney in particular, as Jordan just says, were determined to go to war, and I think that history has probably shown that. But the intelligence analysts that I spoke to sort of fault themselves for not having highlighted their own caveats, their own uncertainties, the gaps in their knowledge. They take full responsibility for not having made clear the level of uncertainty that was central to those findings.
CONAN: That - had they done so, those - even those people who had advocated the war might have looked at it and said, well, wait a minute. Maybe we ought to think again.
GJELTEN: Or they may have gone to war, anyway. But the point is they did not say that, and that that is the point that they're making, that if they had underlined those caveats, possibly people in the White House or in Congress would have had more second thoughts than they did.
CONAN: Jordan, thanks very much. And, Tom, before we let you go, I wanted to ask you about another story. And this was Secretary of State Hillary Clinton making sort of a remarkable statement that the United States has hacked into the computers of al-Qaida in the Arabian Peninsula and inserted, effectively, counterpropaganda.
GJELTEN: Well, we're actually going to have a report on this by my colleague Dina Temple-Raston, who follows counterterrorism issues for us. What's not clear, Neal, is whether they actually hacked into those websites, because the websites we're talking about are jihadi forum websites. These are websites where people can sort of weigh in with their own thoughts and observations.
CONAN: So there's access to them.
GJELTEN: So there is access to them, and it was - it's fairly easy for - and what the United States did in this case, I understand, is actually access those websites and put - posted comments on these forums. So it wasn't in the classic sense, necessarily, of hacking, but it was certainly going onto those websites using, cybertools to get onto those websites and post messages that countered the propaganda of the jihadis themselves.
CONAN: It pointed - these messages pointed out the number of civilians who have been killed in al-Qaida operations in Yemen.
GJELTEN: Exactly. Because they know that that has been a problem for (unintelligible).
GJELTEN: Mm-hmm. That that has taken away from the popularity and the draw of al-Qaida and other jihadi-type groups.
CONAN: Why would we publicize that?
GJELTEN: Well, that's a good question. You know, the truth is, Neal, that, I think, this is not the first time that the U.S. government has done something like this. It is the first time that the U.S. government has, in fact, as you say, advertised it. In the past, we've, you know, kept mum about it. I don't have an answer for that.
CONAN: And, of course, we've just seen a new call by al-Qaida in the Arabian Peninsula for e-jihad - that is, to attack U.S. sites or maybe post propaganda on our...
GJELTEN: Well, actually, it even went further than that, Neal. What you saw was a call by al-Qaida, who are looking for electronic jihadis, who could, in fact, hack into U.S. industrial control systems. Now, this is actually much more serious. This would allow them to carry out physical attacks on U.S. infrastructure...
CONAN: Shut down power plants and...
GJELTEN: ...et cetera, you know, through cyber means. That's what they're looking for, is some cyberterrorists who are willing to carry out attacks like that.
CONAN: Well, more on this story, as Tom mentioned, later today on ALL THINGS CONSIDERED from Dina Temple-Raston. You can find a link to Tom Gjelten's piece on the CIA's intelligence on Iran. Just go to npr.org, click on TALK OF THE NATION. He joined us here in Studio 3A. Tomorrow, TALK OF THE NATION: SCIENCE FRIDAY. Ira Flatow will be here. Have a good holiday weekend, everybody. We'll see you next week. I'm Neal Conan. It's the TALK OF THE NATION, from NPR News. Transcript provided by NPR, Copyright NPR.