INT: When you were standing, I should say, in there talking to the President, did Oleg Penkovsky flash through your brain? Did you think...?

SG: Oh, not in a specific sense, because there was no mention of Penkovsky at the meeting. Penkovsky was extremely sensitive and only the President and a few people knew who he was, so I was not going to use Penkovsky's name. What was going through my mind is what information do I have collectively to answer this question and, of course, knowing Penkovsky's information was an essential part of that, but I did not identify Penkovsky, I did not tell where the source of this information was for the President. But there's a very important lesson here, which people should remember, is that what you don't know is as important as what you do. So what I was trying to communicate to the President and McNamara and Rusk was what we knew about these missiles and how they operated, but what we did not know that these missiles had moved over. So it would have been easy to get up and say, well, yes sir, Mr. President, it takes six hours and thirty minutes to set this missile up, we know that because we watched 'em operate. That was not a fair and would not have done justice, because it was important to know that we are uncertain how long it's going to take these missiles to be set up in the conditions in Cuba, which will be different to the conditions in the Soviet Union.

INT: On the documents that circulated to Ex-Comm the word Ironbark appears and I wanted to ask you a question which would allow you to tell me about Ironbark being on the documents and about whether the President actually knew the reliability of Ironbark information and knew of Penkovsky's existence?

SG: The President definitely knew Penkovsky's existence. The President knew who Penkovsky was. The President knew his knowledge base. The President had been fully informed on Penkovsky, but he was informed on a different classification level, known as Chickadee, and that information was only made available to the President, Secretary of Defense, Secretary of State and few senior people in CIA that needed to know Penkovsky's background information. Ironbark was the classification that was used for the data that Penkovsky was providing and so the Ironbark did not identify Penkovsky as to who he was, but the Ironbark was used because his information was so good and so sensitive we wanted to protect it. So the Ironbark was given a slightly wider distribution, but still was not sent to everyone in the intelligence community, it was operated on a strict, what we call, need to know basis. Those people who had a real need to know of this information were provided it, but there were very strict caveats as to what the distribution should be on the Ironbark. So Ironbark was the product, but the Ironbark did not identify Penkovsky, who he was and where he fit into the hierarchy.

INT: Now,... you said something very interesting earlier on. I'd like you to develop it, because it's probably quite relevant in terms of the time frame, which is... one of the things you were trying to do is to use your own knowledge of missiles and missile procedures, as I understand it, to make sure that what Penkovsky was telling you was not coming from a man who had been compromised and was being fed bad information or shall we say sort of deceptive information by the KGB. At a certain point you became suspicious that that might be the case. I want you to tell me about that, you know, that process.

SG: Well, you read the data that Penkovsky is providing. It was clear the data that he was providing by the photographs of the manuals, providing complete manuals, was authentic information that was not fabricated. Penkovsky was providing a lot of other information, some of which was first hand, some of which was second hand. For example, his knowledge of the Soviet ICBM program was extremely limited, but he was telling us what he had heard about the program, which was that it was not near as far along as we thought it was, but we did not know that secondary source, so you have to be careful how much you put into it. So you read his raw reports, what he is informing you - and I'm dealing strictly now with the missile part, because he's providing information in other areas - so those of us reading his raw reports before they were published for the whole community would look at 'em in terms of the quality of the information, the type of information, the uniqueness of the information. Late in Penkovsky's time frame some of us became suspicious that some of the information was not similar to what he had been providing previously and some of it was not as useful, so then you become suspect. As an analyst, you merely tell this to the people that are operating Penkovsky. I told to Jack Maury that we were concerned about this type of information. The details of that information that I don't think I should go into as to which piece was questionable and which piece wasn't, it's a more the nature of the type of information he's been providing and you sense that something is different here. We'd been getting high quality, specific, unique, now you're beginning to get something that's not that useful, not that unique.

INT: OK. I certainly don't want you to stray into any areas to do with that, so just forget about that, but I still don't think you've made it quite plain and I want you to go back on it, because what you're saying is, at a certain point you grew suspicious that he had been compromised. I mean, you don't say that he was compromised, but the information seemed to add up to a situation where he was not the old Penkovsky you knew and could trust and therefore you grew worried about him and about the status of that information, because it needs spelling out to an audience that doesn't know all this stuff, you know, what that might mean. It means he's been got at, maybe he's no longer his own man. Maybe he's been caught in some way or he thinks... do you see what I mean?

SG: Well, being one of the intelligence officers who was reading Penkovsky's material as it came in raw before it was disseminated, sanitized and disseminated , you look at this information in terms of the quality of the information, the uniqueness of the information, something, the sensitivity of it too from a Soviet perspective, so you're following this, you're getting a variety of very good, useful information. Data on manuals which are unique and exercised. But if you follow through this, eventually sometimes some of this information doesn't appear to be quite as unique, quite as sensitive. It's something that is more routine. So when you see this, compared to what you've been getting, all you could do is suspect that maybe there could be a problem here. He may have been compromised. Not having any first hand information, my responsibility was only to tell the people who were operating Penkovsky, Jack Maury, that some of this information doesn't look the same quality that we've been receiving before. This was not an individual judgement I worked with Doctor Scovill on a couple of us. I remember one night we sat and talked for an hour about this information, is this the same, is it not quite as good? Well shouldn't we alert Jack Maury that something may appear amiss here and in hindsight it turned out that things were amiss.

INT: That was good I thought, is there anything happening on that ......


INT: Explain to me how you used to look at Penkovsky's information and how you became familiar as it were with the way in which he spoke about things which struck you as ringing true, but then how historically, after a certain period, you started to worry and about the meeting that you had, worry about it for an hour and passing it on and that you had some concerns that maybe he'd been compromised, maybe he'd been got at.

SG: At the time, I was one of the few people who had the opportunity to actually read Penkovsky's raw information when it came in, working with Doctor Scovill, my immediate boss, and with Jack Maury, the man in the Soviet Division running the Penkovsky. We would read these raw reports. At that time, those reports appeared to be very accurate, very good. The Penkovsky reports, particularly the manuals, were unique. They were hard to come by. He had acquired very good, useful, sensitive information. In other aspects of the missile program, he had second hand information but it appeared to be reported accurately. So his information was unique, it was sensitive, it was something extremely valuable to us. So you're reading these reports, at the time we were reading them each night when they come in, then the report would be sanitized, published out under the proper classification, Ironbark or one of the others, but over the period of time, when reading these you start getting a feeling, well that report that just came in does not appear to be of the same caliber and uniqueness that the reports that we'd been receiving had been from Penkovsky. So your mind gets suspicious a little bit. So a couple of these reports came in and Doctor Scovill and I spent roughly an hour one evening discussing this and questioning ourselves, is this really something to worry about and the two of us felt that, yes, this appeared to be sufficiently worrisome that we should speak to Jack Maury, that our concern that maybe Penkovsky is being compromised, maybe they know him, and when you have a source like Penkovsky and if you're a Soviet, this becomes a very good mechanism for feeding false information to the U.S. So our concern was, has he been compromised and as this information is not as unique, not of the same caliber, not of the same nature that we'd been receiving before. So you get a suspicion, which we got at that time, which in hindsight proved to be correct. Now Penkovsky was providing a lot of other information besides just ballistic missile information and all I'm focusing on is the ballistic missile part and whether or not his other information on political or military activities or things of that nature was changing, I cannot answer that question.

INT: I think that's fine.