Loading summary
Transcript89 lines
- [00:02]
A
You're listening to the Cyberwire Network, powered by N2K. Welcome to Spycast, the official podcast of the international spy Museum. I'm your host, Sasha Ingber. And each week I take you into the shadows of espionage, intelligence and covert operations across the globe. In 1947, a new civilian intelligence agency was established, the CIA. But a series of intelligence failures undermined its credibility. The White House and Congress were up in arms and a new mission was formed to recruit Ivy League professors with uncanny skills. Leaving their so called ivory tower, the academics brought new ways of thinking about national security to the CIA, helping the US navigate the complexities of the Cold War. In one year, the academics analysis revolutionized the service, cementing the CIA as one of America's finest agencies. Political scientist Peter Gray's author of the Intelligence Intellectuals sits down with me to unfold how these bright minds shaped the agency. Good morning Peter. Thank you for joining.
- [01:32]
B
Thank you Sasha, for having me.
- [01:34]
A
So let's talk about some of the major intelligence failures that Americans saw. For one, the Soviets acquired an atomic bomb earlier than the CIA had predicted in the summer of 1949.
- [01:51]
B
Yes, that's correct. It's an interesting way of leading into this because in 1945, in July 1945, Truman meets Stalin at the Potsdam Conference and he tells Stalin that the Americans have come up with this unusual weapon of amazing power and destruction. And Stalin doesn't bat an eyelid. And there's a little bit of surprise at that because there's an expectation that Stalin would be shocked. But Stalin already knew about the development of the atomic bomb and he'd known about it since about September of 1941, so three and a half years earlier.
- [02:27]
A
So the CIA knew that it had gotten this prediction, this assessment wrong. When did that become apparent?
- [02:36]
B
At the testing of the atomic bomb by the Soviets. So the estimate was out that said that it was about 1953 that the Soviets would get the bomb. And here it is in 1949 that it's apparent the Soviets have already got it. So it undermines CIA's credibility.
- [02:54]
A
What did they say when it came out four years before? They had predicted that the Soviets would be testing this.
- [03:04]
B
Well, intelligence is not a definitive thing. You're always going to get things wrong or slightly wrong or you're going to miss something. So as far as CIA was concerned, they had done what they could do with the information that they had. And at that time there was no penetration of the Kremlin. So there was no way of getting inside the heads of what the Russians were thinking they could only go by what information they were getting. And it was pretty slim.
- [03:36]
A
So it was not just a failure of analysis, but it was a penetration failure, too, because they hadn't been able to access the Kremlin on that human level.
- [03:46]
B
Yes.
- [03:47]
A
So there were other failures as well. When the Korean War broke out In June of 1950, what had the agency been saying at the moments leading up to this?
- [03:59]
B
Well, this also demonstrates the fragility of intelligence analysis, because in the case of North Korea, the CIA had told Congress and had told the President that they had seen troops massing at the border. So as far as CIO was concerned, they had done 90% of their job by identifying that the North Koreans were mobilizing. What they hadn't done was get the exact time and date right. In fact, this is the point that Helen Ketter, who was the Director of Central Intelligence at the time, stressed when he was talking to Congress. You know, you can't expect us to get that kind of level of detail correct. We told you everything you needed to know to get prepared, but we didn't just give you the time in the morning that the tanks rolled over the border. This is the problem that Helen Ketter and CIA are having at the time, which is that they're not screwing things up. They're doing a pretty good job, but they're losing the narrative. So every time they have something that's called an intelligence failure, people are getting exasperated with them, and they're not doing a very good job of defending their position.
- [05:08]
A
So when these not failures, if we don't want to call them that. But when these mistakes or lapses happened, what did Truman, what did Congress say? How did they respond?
- [05:22]
B
There were two factions within Congress. There were the senior senators who were basically supportive of CIA and gave them an awful lot of rope. And then there were the other senators who did a bit of grandstanding. Eugene Milliken was one who. Who really used this as an opportunity to rub CIA's nose in it and said, you know, this is an enormous failure. We can't trust you. We don't think that we're getting the kind of quality product out of you that we're expecting, and this is just a disaster. And that was echoed in the national press. So the New York Times said that the CIA was the weakest link in the national security sector. They talked about it being staffed by chair warmers and empire builders. So at this point, CIA, this is talking about 1950. CIA has basically lost the ability to defend itself, and it becomes an Existential crisis.
- [06:17]
A
So whose idea is it to bring academics into the fray? And what is the thinking behind them being able to create a more methodical way of predicting national security matters.
- [06:33]
B
So what Truman does is he turns to a guy called General Walter Beedle Smith, who was Eisenhower's chief of staff during the war and then ambassador to Moscow. And Winston Churchill called Bedell Smith the American Bulldog. So he's a really tough guy. He's got a photographic memory. Everybody's afraid of him. Yeah.
- [06:53]
A
He was called a hatchet man under Eisenhower.
- [06:57]
B
He was a sort of person people were afraid of. And he had a stomach complaint, which made him enormously irritable as well. So you just didn't mess with Metalsmith, right?
- [07:06]
A
A stomach complaint?
- [07:07]
B
Yeah. He had to delay his move into CIA in order to have a stomach surgery.
- [07:12]
A
Okay.
- [07:13]
B
He had, like, a really chronic irritation.
- [07:17]
A
Huh.
- [07:18]
B
So Smith is the guy that goes in there and clears the pathway for the social scientists to arrive and Beetle,
- [07:26]
A
as he was called. He, as you said, had been the US Ambassador to the Soviet Union, and he was seeing the deterioration between US and Soviet relations firsthand. He wrote in his book, quote, we dare not allow ourselves any false sense of security. We must anticipate that the Soviet tactic will be to wear us down, to exasperate us, and to keep probing for weak spots. And we must cultivate firmness and patience to a degree we have never before required. Tell me more about his thinking in bringing academics from Ivy League universities into the agency.
- [08:06]
B
Okay, so there were two things that illustrated why CIA was going downhill. One was that they didn't have a tough guy at the helm. You know, they needed a benevolent dictator. Somebody that the military services, State Department, others were afraid of.
- [08:20]
A
Because he had been a senior officer in the army as well.
- [08:23]
B
Exactly. The other thing that they were lacking were these processes. It wasn't necessarily that their assessments were wrong. It was that they didn't have the processes that backed up and withstood scrutiny on that.
- [08:38]
A
The real question is, why would you bring in a bunch of university professors to a spy agency?
- [08:45]
B
Right. Beedle Smith was a chief of staff. And the thing about a chief of staff is that you have to have the information for the general. You have to have it on their desk when they need it. You have to be absolutely rigorous in the kind of information they provide, providing. So he would have understood as a background that CIA wasn't actually delivering on these things. And in fact, he arrives early. He's the first to get there, and he asked for an estimate on North Korean situation and CIA doesn't deliver it in time for him. So he would have been really frustrated by the kind of people he had around him. He was also looking at OSS Office of Strategic Services. And he would have seen that the kind of information he was getting during the war from the OSSS was, was particularly good, high quality stuff. And the guy that was running that was William Langer.
- [09:30]
A
And the Office of Strategic Services was the predecessor to the CIA. Let's talk about William Langer because he is one of the most influential academics that you followed. He came from Harvard. Tell us more about him.
- [09:45]
B
So William Langer was a European historian, European diplomacy at Harvard actually joined the Office of Coordination of Information much earlier, before it became the Office of Strategic Services. And so he was there right from the very beginning. And Langes was a very acerbic man. He had a sense of humor, but it tended to be quite cruel. Again, like Bea Smith, he was a leader as a sort of person that assumed responsibility for the work really early on. He was like a supervisor in an academic sense. He was a natural man to lead the project. He wasn't necessarily a particularly deep thinker, but he was extremely methodical in his ways and expected the best from his people. So he was the natural leader of what became the Office of National Estimates. So Lange had a very nasally whiny voice.
- [10:41]
A
And that's relevant because I think it
- [10:44]
B
was because he had been a consultant to the State Department after the war. And the argument was that he'd never got there because he had any charm, because he certainly didn't have charm. He was the sort of person that could reduce people, grown men, to tears and had extremely high expectations of the way people would deliver. He was a Bostonian, what is it, Brahmin? He was old school Harvard man.
- [11:13]
A
Now, one of the people that he was supervising was a man named Sherman Kent, who had been at Yale. He had also been at oss, and he would later be described as the father of intelligence analysis. Let's hear more about his background.
- [11:30]
B
So Kent also arrives at the Office of Coordination of Information under Langer. And he runs the Northwest European and African desk all the way through the war. He's a man who is enormously convivial. He was a dressy dresser. He used to wear red braces and seersucker suits and things like that. And Kent is also one of these people who really internalizes the thinking. He takes it enormously seriously. He's the sort of guy that if there's a problem, he sort of rushes off to his typewriter and starts writing a 30 page report. We're really lucky that he was this kind of person because he gave us so much information about the processes of the time and how they went about it. But also right up until his retirement, the successes and failures of CIA. He was a guy that was enormously humble, wore his heart on his sleeve, and if he got something wrong, if he screwed something up, he would tell you about it and tell you why. So he's fundamental to our understanding of CIA in that period.
- [12:37]
A
And he was the son of a congressman and a women's rights activist. So he came from a pretty prestigious family.
- [12:45]
B
That's true. And they owned quite a lot of land in Kentfield in Marin county in California. So they were quite a. I wouldn't say wealthy, but quite a well off family. And very, very connected, obviously through his father, the congressman. He knew everybody. And you look at Kent's archives and full of letters to the most unexpected people. Julia Child was a personal friend, former
- [13:06]
A
CIA officer turned chef, correct?
- [13:09]
B
Yes.
- [13:09]
A
I don't know if any other spy could cook a chicken as well.
- [13:14]
B
Yeah.
- [13:14]
A
And he also had imposter syndrome. You were telling me that he was worried that he was not going to be able to meet the task. This was a very high pressure position for him.
- [13:27]
B
So they arrive at CIA in 1950, here in Langer in late November 1950, and the Korean War is in full swing. And we know Kent was unnerved by this and really felt there was a 50, 50 chance of the US actually having any impact at all in the Korean War and was more worried about the level of deaths that were going to occur as a result. So he goes in with this rather impossible task, as he sees it, of trying to predict the future. This is not something he had any experience of at oss, neither had Langer. They were basically providing current and basic intelligence, providing maps. They were providing knowledge of things like railroad networks and ports and things. They didn't have any insight in how you would look 2 years, 5 years, 10, 15 years into the future and develop the kind of assessments that the US suddenly needed about where was the world heading. So he had imposter syndrome. He also felt a little bit daunted by what he called the rarefied intellectual atmosphere at CIA. The people that were there were all PhDs and they were all very clever people. What they lacked was a good leader and they were now going to get that in Bedell Smith. And what they lacked was a process, a way of going about things in A consistent way. And they were going to get that from now from Langer and Ken.
- [14:50]
A
Now both of these men had worked at the OSS under Wild Bill Donovan. So what made this endeavor different?
- [14:59]
B
Donovan is the person that recognizes you have to have the best brains in the business. And I think that's his biggest contribution to intelligence analysis. Don't have second rate people. So what Langer and Kent were doing at what was called the Research and Intelligence Unit was sort of a backstop to all of the operations that OSS were doing and a backstop to what Chief of Staff were doing, particularly things like the invasion of Northern Africa. When they get to CIA, that's not what they're being asked to do. Beadle Smith describes the new office that he creates called the Office of National Estimates. He describes this as an ivory tower in the academic sense. And he wants those guys to be put away where they don't have to deal with the day to day stuff of intelligence. And they're only working on long range estimates, what's going to be happening in the future. So you know, in a sense that area of comfort has been removed from them. The area that they know well of current intelligence and basic intelligence has been stripped away and shunted off to another department. And they are only being asked to concentrate on the long term predictions. So the impression redoubles in that sense.
- [16:20]
A
And there's another person who is now part of this effort and his name is Max Millicon. He had been at mit. Let's talk about him too.
- [16:29]
B
So Max Millikan had trained at Yale. He'd been at State Department after the war as chief Economist for Europe. Kent knew him and brought him into the organization. Milliken is the person that contributes probably the most of the early success that the Office of National Estimates has. He's able to think on a number of different levels. He understands that the problem of getting their heads around the Soviets is the key thing and that economics or social science can bring a process to understanding that better. So Millikan really lays the groundwork for a Cold War program of research and then he starts delivering on the fundamentals of that research really, really quickly. So he's only there for a year and in a year he turns that department around. He creates a department of Economic Intelligence. And it's feeding back into the work that Kent and Langer are doing and earning them an enormous amount of Brownie points.
- [17:30]
A
When we come back, the concepts the academics created and how they're still being used by the CIA today. In the world of Covert operations. Superior intel can be the difference between mission success and total compromise. Your enterprise is a high, high value target. Stop letting threat actors breach your digital perimeter. Palo Alto Networks provides the zero trust security platform you need to gain the upper hand. They turn noise into actionable intelligence. Automating your defenses across cloud, network and endpoints. Don't operate in the shadows. Gain total visibility and help secure your mission. Visit paloaltonetworks.com and fortify your digital defenses today. And what are some of the concepts that these three men, Langer, Kent and Millicon, come up with that ended up being important in shaping the agency?
- [18:47]
B
Okay, well, let's start perhaps with Milliken. So we're talking about him. Millikan comes up with this idea called the Inventory of Ignorance.
- [18:54]
A
Inventory of ignorance. This is a great term that I feel like should just be in everyday language.
- [19:02]
B
Yes.
- [19:03]
A
I might steal this for other concepts, but please, let's hear.
- [19:07]
B
Really is. It's expressing an enormous amount of vulnerability. There are things there that we. No, we don't know. His main job is to determine what the Soviet economy is because they don't have any idea. All of the experts up to that point are really struggling with trying to get their heads around how robust the Soviet economy is and then to drill down and work out what the military budget is. Again, this is a highly kept secret by the Soviets. So Milliken does this inventory of ignorance where he says, where are the gaps in our knowledge? Here are the things we do know. Here are the things that are missing. Let's start working on filling those gaps. And so everything that Millikan is doing is a filling the gap process.
- [19:52]
A
And when you're describing an agency that is filled with people who have PhDs who came from some of the most prestigious universities to say, these are the things we don't know, there is a certain amount of courage and vulnerability in even searching for that information, Admitting that information.
- [20:11]
B
Yes. And again, if you think of the highly pressurized environment they're in, to go out there at the very beginning and sort of admit that there's an awful lot we don't know. It's not human nature to do that. So Kent and Milliken are part of the legacy, I think, that they provide not only to CIA, but also to all intelligence agencies in the world is you. You can admit when you don't know something.
- [20:39]
A
But how. How do you figure out what you don't know?
- [20:43]
B
Well, Kent also had a model for thinking about that, which he called the pyramid. So Kent has this idea that, that the research project of an assessment is like a pyramid. You start what you know, at the very beginning, is that like the base of a pyramid, the amount of research that you're going to have to do and cover is going to be very, very wide. And he calls that an inductive process of having an understanding that you have to have no stone unturned. And then you will refine and refine and refine towards the apex of the pyramid, which is your conclusion. And he says that's a deductive process. So he says when you get to the very top, you've got a conclusion which is either supported by all of this evidence that you've sifted through, or sometimes that pyramid apex is broken off and it's not quite there, but it's as best you're going to achieve in the timeframe that you've got. So both Milliken and Kent have got what we'd say was a metaphysical understanding of the process, and they start with that rather than try to impose a process on it. Right from the beginning, they're saying, what don't we know? How do we reach good conclusions? What are the principles of good intelligence?
- [21:57]
A
You've also written about how Millican created this building brick method. Explain what that is.
- [22:05]
B
Yes. So the building brick method was a way of understanding what the Soviet economy and what the Soviet military was spending by comparing it with what it would cost for the Americans to build the same thing. So he would start with a brick, which would be a small unit, something like some tanks on the border. And he would say, okay, there are 10 tanks here. What would it cost for us to produce those 10 tanks? As he built up towards the building, the whole wall, he would roll that out and be able to say, well, the total sum, if it was in American dollars, would be this. Therefore, the Soviets are probably spending something quite close to that.
- [22:40]
A
So it was essentially comparisons, just kind of like block by block.
- [22:46]
B
Exactly.
- [22:47]
A
So they're sitting at their desks, they're creating these concepts, they're making these conclusions. Are their colleagues at the agency, Are they supportive? Are they skeptical? Do they think that these men are just absolutely nuts?
- [23:02]
B
They're a little bit ginger about saying, this is our new process that we're working on. So it doesn't appear in the assessments until they start to get their confidence that it works. I think there's a really a growing confidence in the robustness of their methods. And the way you can see that happening is because right up until 1940, 50, the military intelligence agencies and State were enormously critical of the assessments. And they have these things called dissents that they put at the end of an assessment where the other wings get to say that they don't agree with the conclusions in the assessment. And the kind of dissents that you're seeing right up until 1950 are absolutely damning. They say this is not an assessment that should be used for national security policy, things like that. After November 1950, those dissents die out. And there's a combination of reasons why they die out. One of them is because everybody's afraid of General Beetle Smith and they're not going to take them on convenient. And the other reason is because Kenton Langer and Milliken are providing really good evidence based assessments that are robust, that can't be challenged as easily. It's not to say that they're not going to get challenged or they're not going to have conclusions that the other service wings don't agree with, but they're finding it a lot harder to pick them apart.
- [24:18]
A
And all of this is happening inside the Office of National Estimates. This new office, it would ultimately make more than 1500 assessments to the President, to policymakers. What were they providing that proved incredibly valuable, which helped restore the agency's reputation.
- [24:40]
B
They managed a turnaround in the agency very quickly, and there's renewed faith that they're providing good product. The Office of National Estimates and Max Milliken's Office of Research and Reports both start to get big projects coming in as a result, which is a demonstration that they're being trusted again. In Millikan's case, he starts off looking at the Soviet economy, but then he's allowed to look at the whole Soviet bloc as a result of that. So what Millikin in particular is offering is a better understanding of the Soviets ability to fight the Cold War. And it's in an economical sense more than anything. You can start to plan for diplomatic ventures, you can start to plan for allied ventures, you can start to plan for defending Europe, things like that. If you have a better understanding of the economy.
- [25:28]
A
Okay. But not everything went well. I know that there were also moments that led to more accusations of failure. And I'm thinking of the Cuban Missile Crisis. So let's talk about some of the things that didn't go so well in terms of estimates and assessments.
- [25:47]
B
Well, the Cuban Missile Crisis is the case in point. And Kent actually writes about that in a 1968 article called Crucial Estimate Revised. And he says, why did we fluff it? Why did we get it wrong? He said, well, we had only a certain amount of information at our fingertips, and the problem was that we tended to go with the information that we were provided, and we ended up making conclusions based on incomplete evidence. And so, again, Kent is terrific at this. Just a willingness to accept that sometimes you screw things up. You have this difficult balance and intelligence between being trusted, people having faith in you, and the problem that you are going to never get it 100% right. And even when you get it 98% right, that 2% that you get wrong might be the bit that trips you up. It is a very fragile process. And the ability to keep your reputation, keep people coming back to you, asking you for details and things like that, people having faith in your judgment and your conclusions is always going to be on a knife edge. So the ability to have some sort of introspection about what you're doing, where you can succeed, where you are going to fail regularly, is, I think, enormous comfort to analysts today. Insights that people like Kent in particular provided is really something that probably keeps you on the straight and narrow when you're really feeling like everybody's against you and nobody likes you and nobody trusts a word you say.
- [27:25]
A
The stakes are high and it's a very, very stressful position. There's no doubt about it. So what became of these three men? Why was Kent the only one who stayed at the Agency?
- [27:38]
B
I think Kent really enjoyed the intellectual problem. And as I say, he was the sort of guy that liked to run off and type things out on his typewriter. So he saw himself as being sort of the glue that held everything together. He didn't have total control over the department. He lost that quite quickly. There were other people that stepped in, particularly Ray Klein, who got promoted above him, but Kent was always there as being the sort of the sounding board, the historian in the sense of the Agency's process. And so he's there for another 17 years. And as I say, he's probably the most influential man in intelligence analysis ever. And I think it's this ability to be able to look at every problem clearly and to be able to say this is not a perfect process that is part of his attraction.
- [28:30]
A
It created the high confidence, low confidence.
- [28:34]
B
Yes, yes, he did that.
- [28:35]
A
What about Langer and Milicon?
- [28:38]
B
So Langer goes back to Harvard and just continues to teach history. Langer is involved with the CIA, what is called the Princeton consultants, over a period of years, which is a sort of another sounding board that's external that where they can take the assessments to, but that doesn't really work out terribly well. Milliken is also involved with the Princeton's consultants for a much longer period of time. He also goes to MIT and sets up a think tank within MIT, which is enormously successful. And also with Rostow, the famous political scientist, sets up a whole idea about modernization theory, which is how American aid will help develop the third world countries. So they all have careers of sensors outside of CIA after that, but they still have quite a close connection with the agency?
- [29:24]
A
Yeah. So they're still in service of the country and they still have connection to CIA?
- [29:28]
B
Yes, they do.
- [29:29]
A
And you also have met repeatedly with Sherman Kent's son?
- [29:34]
B
Yes.
- [29:35]
A
Tell us about him. What is he up to? What does he have to say?
- [29:39]
B
Sherman Kent Jr. Said when I sent him my book, he said, this is a side of Dad I didn't even know existed. He has this view, which is about dad being a dad, about Dad's friends who worked for CIA. But he didn't have the view of some of the awfulness that was going on. And I think it surprised him a little bit. Kent writes to his brother Roger in 1951. He's only been at CIA a few months, and he says, I'm really struggling, he says, with the awfulness of the world and our ability to destroy the great works of man. He says, when I go to work every day, I can cope with it, but when I come home at night, my subconscious takes over and I'm really struggling to come to grips with it. So here's a man who very quietly has this family life, brings up a wonderful family, and yet he's got these demons, in a sense, haunting him that he brings home each night and can't talk about.
- [30:42]
A
I think that explains why so many people miss their colleagues at the agency, because there is a sense of understanding at the darkness that you have to deal with day after day.
- [30:53]
B
The darkness and also the enormity of that shared problem of trying to do good predictions.
- [31:01]
A
So how did Kent's theories play out or not play out in 9, 11, and onward?
- [31:11]
B
I think what Kent does is he gives us a foundation for thinking about intelligence analysis that's timeless. There are things that change over the next 50 years. You have computers and the ability to crunch numbers. You've got AI now that changes the nature of the process, but the fundamentals don't change.
- [31:31]
A
And this is totally speculative, but what would these three men say about where we are in terms of analysis and thinking in the age of artificial intelligence?
- [31:45]
B
I think they would be skeptical about artificial intelligence ability to deliver on a human level. What I mean by that is that human judgment may not be perfect, but things like gut instinct or moral values and things that also implicate on your assessment of what's going on. The computer can't do that. So I think they would be very skeptical about AI's ability to deliver anything. The other reason why is because there is a process to intelligence analysis, and that process requires a great deal of cooperation and knowing people, knowing people who you trust and asking for them their opinion. And AI has a way of undermining that because it's taking away, first of all, your understanding of how you got to the conclusion. The computer gives you the answer that says this is the conclusion. And also it takes away that sounding board thing that's so important where you say to somebody, I have a feeling this isn't right, or I have a feeling we're going in this direction and working that out together. As I say, it's not a perfect process, but I think you would find that Langer and Kent and Milliken would have been much more comfortable with that process than one that just you push a button, spits out an answer.
- [33:06]
A
Well, Peter, we talked about stomach problems, men who act as benevolent dictators, ignorance and its inventory. And I really appreciate you sitting down with me today.
- [33:17]
B
Thank you for the invitation.
- [33:20]
A
Thanks for listening to this episode of Spycast. If you like the episode, give us a follow on Apple, Spotify or wherever you get your podcasts and leave us a rating or review. It really helps. If you have any feedback or you want to hear about a particular topic, you can reach us by email@spycastpymuseum.com I'm your host, Sasha Ingber, and the show is brought to you by N2K Network's goat rodeo and the International Spy museum in Washington, D.C.