from china with love
homeparlor maidchina spyingcounterintelligence
interview: edward appel
photo of appel

He was an FBI special agent from 1973 to 1997, where he specialized in counterintelligence. He also served as FBI liason to the National Security Council from 1996 to 1997. In this interview, Appel explains how an FBI counterintelligence agent develops an asset and talks in detail about China's methods of espionage. He describes Bill Cleveland, who worked for him, as "one of the nation's most expert counterintelligence officers with regard to China." Appel also discusses why prosecuting espionage may not always be the most attractive option for the U.S. government, but argues that the FBI prosecutes "an awful lot of messy cases," and that procedures are in place to protect classified information. This interview was conducted on Aug. 21, 2003.

Just to begin with sort of an overall question, the China program, the FCI [foreign counterintelligence] program in the FBI, what had been the goals, long-term, and how had we achieved those goals?

The long-term counterintelligence goals against China are about the same as any other country -- to discover the degree of intelligence operations that they focus on the United States, to determine what they're after and obviously, if somebody is violating the law, to bring them to justice. …

How is [the threat of Chinese espionage] any different, any more special than what we've seen in the past?

Again, with the caveat that I'm not an expert in the overall threat, China is trying to build its military power. So it benefits when it can basically steal the plans to weapon systems to improve their own weapon systems, based on things they find out. They actually want to acquire weapons that other countries have. …

The United States has never expended very many resources on the People's Republic of China. We've never thought it represented all that big of a threat.

China still thinks of itself as a Third World country in many respects. But they basically manufacture a great deal of what we use here, and that's their goal. They want to be the Japan of this era. They want to be able to dominate world manufacturing. So a lot of their espionage, or a lot of their intelligence work is focused on that economic advantage that they might gain.

There is very little distinction in China, even today, between the government and its interests, and Chinese businesses and their interests. There's a great deal of information sharing. There is a great deal of collection that's based on making Chinese business competitive and, in fact, dominant in some of the areas where they want to be dominant.

Tactics they use, what we're used to, did we need to change our thinking?

The word about the Chinese is that they do things differently. They've been doing this for 4,000 years. They operate on the basis of relationships, interpersonal relationships. They call it "guanxi." They call it developing a superior subordinate relationship with someone, to trade information back and forth with someone.

At the end of the day, though, intelligence collection methodology relies on someone who knows telling someone who is trying to collect information. So whether the person is paid or not, whether the person owes a debt of gratitude or owes some kind of guanxi to the other person, at the end of the day, the intelligence service succeeds by establishing sources. That's the way the Chinese work, just like everybody else. …

They do run agents. They do collect intelligence through classic operations. But by and large, they rely on a large number of what they call "overseas Chinese" and a few non-Chinese to provide them with information, basically, because of the relationships they build over time.

Basically, it's a game played on the use of assets?

Right. An asset is somebody you know that can give you what you want. Obviously, you develop a relationship where that person wants to give you things, where that person actively goes out and collects things for you.

In our thinking here in the West, we tend to think about an employer/employee relationship or a contractor relationship, so that, in a secret way, the contractor goes out and collects information, and then brings it, for money, to the intelligence officer who is running the operation.

But oftentimes, the Chinese don't feel it's necessary to pay that person, or at least not to pay the person for intelligence. They might get paid because they have a strong business relationship with the Chinese government. But they might not get paid specifically for the intelligence that they bring. The idea is that the person is so obliged to the person who has the guanxi with them, that, out of a sense of duty, a sense of obligation, a sense of togetherness, on behalf of the motherland, this person will take risks and bring information as a part of the overall relationship. …

So we found that frequently the Chinese intelligence services were practically invisible to their sources, because the relationship between the intelligence officer and the source was a personal relationship, and was one of mutual obligation. The source didn't think of himself as an employee or contractor. The source thought of himself a person who should go out and do for a friend, for a colleague. …

Speaking as someone on the inside, who's seen this for a lot of years … have we been winning this game of chess, whatever it is, with the Chinese? Or are they playing a different game, or do they understand things we don't understand? What's your view on that?

I think it's very hard to say whether you win or lose. We're always talking about the current espionage case and trying to see these things through the light that that case sheds on a counterintelligence issue. Frankly, that's not a very clear way to see the whole picture. However, I do think that, to some extent, we have lost the counterintelligence war against the Chinese intelligence services, partly because we haven't had major counterespionage successes.

There are really two ways you want to play this game. The first way is to know what the other people's intelligence service is doing -- who are their sources -- and to either neutralize those sources, in some cases, arrest them, or control the information that they're reporting back. You want to know more than the other side knows. Of course, it's very hard to know exactly what the other side knows.

The other aspect of it is to simply uphold the law, and to arrest people, particularly Americans who are stealing American secrets, classified American information, to put them in jail and to bring them to justice for committing espionage. I don't think we've done a good job in that particular respect with regard to the Chinese intelligence services. I think there are quite a few cases in which the Chinese have succeeded in collecting classified information from U.S. government sources and, for whatever reason, we haven't brought those people to justice and put them in jail.

So on an absolute scale, if you would say, "Show me the numbers. How many times did you put a person in jail for this?" I think it's an incontrovertible fact that we don't have many numbers. We didn't really do a very good job of putting those people in jail. …

Take us sort of to Quantico 101 here. Somebody comes in the door who's going to work FCI. What are they taught about assets? What's an asset?

… In an ideal sense, I guess, a good counterintelligence asset would be a person who is very familiar with the foreign intelligence officer and tells you all about who they are and what they do. But it really runs the gamut from someone who simply observes the person to someone who's really a part of that intelligence service and has elected to be recruited in place, if you will, or provide information, while still acting in a capacity of a foreign intelligence officer. So an asset can be any of the above.

How do you develop one?

Obviously, it relates to personal relationships. You might identify the people that a foreign intelligence officer is in touch with and, from that group of people, select one or more that you would approach, and you would find out what they're willing to tell you. Over a series of interviews, a series of contacts, you might simply ask them questions and see how far they're willing to go in answering them, and compare what you know from your investigation against what they tell you, and you balance the two.

Obviously, if they continuously lie to you, if they're not willing to talk to you, if they're motivated only by personal greed or money, you might want to take what they say with a grain of salt. Hopefully, over time, you could develop a trusting relationship with that person, where you're convinced that they are telling you more than they are that foreign intelligence officer. So that's a balance you also have to maintain.

How do you know? How do you know if someone is acting more as a double agent, or how do you know for sure?

It's very, very important, and I think deception analysis is at the root of this. Basically, I think most of us have a gut feeling one way or the other, and a good detective has a better gut feeling. FBI agents are excellent detectives. But you also have to be systematic about how you find out what the other side is doing. What's their order of battle? Who are they? Where are they? What are they doing? Compare that with what you're told. You continually want to assess what your source tells you against your other sources. It means that you have to have a range of knowledge.

This frequently does not take place in the hands of the handler -- in the case of the FBI, the FBI agent who handles the source -- but rather it takes place in the analytical unit at the FBI headquarters in Washington. [T]here they compare everything they know from all of the sources with what this particular source is telling us, and then try to see, "Is this person telling us the truth? Is this person giving us anything of value? How can we weigh that against the overall pictures of the Chinese counterintelligence operations?"

… How does that relationship between Washington and the handler work?

It works very well. Actually, in order to make the source have some value, the handler has to document what the source says. There are some cases in which the source has to provide documents from an intelligence service, if they have access to them. If they can't provide those documents, or they won't provide them, that's a clue that they might be under the control of foreign intelligence service, more so than the control of the counterintelligence service.

In some instances, we want to direct the activity of a source, to tell them to find out something, to ask a question of their handler on the other side. In those cases, obviously, you already know what you want to find out. Hopefully you ask the question directly, and at the end of the day, you weigh the answer against things you already know. The more trusted the source, the more sensitive the questions.

In many cases, you can't ask a question in a counterintelligence investigation without revealing classified information. You hope that that classified information isn't turned around and then provided to the other side, to the other intelligence officers that might be in touch with your source. That's something that you always have to be careful about, because if you ask that question, you may be telling the other side a great deal -- more than you want to.

The way you define it, a lot of times it does turn into a game of spy versus spy, in a way. A lot of times it isn't the hot dog dealer, it is people that have relationships with friends in China with MSS or other organizations. Is really a large proportion of this game being played between the two intelligence services?

Sure. It's the way the game is played. Obviously, it is spy versus spy. That's what counterintelligence is. At the end of the day, the better your operation as a counterintelligence service, the more you know about that foreign intelligence service. Hopefully, they don't know that you know it. Hopefully, you can to some extent contain their operations, and understand more about their operations than they want you to, so that you know what they're looking for and you know who they're using. You know their order of battle. You know what their targets are. You know what information they need.

I can remember a time many years ago, when the Chinese intelligence service sent a high-ranking intelligence officer to discover what this "Supreme Court thing" is that the U.S. has. Well, that's the kind of intelligence operation we want to see. Frankly, we'll help that guy find out all about the Supreme Court. But when the intelligence officer comes after weapons systems, that's a whole different thing, and we want to know what weapons systems they're coming after. We want to know where they're seeking market intelligence about particular companies or about particular opportunities they might have to compete in the economic sphere. …

So a J.J. Smith would see a Katrina Leung. What would he see in Katrina Leung [as a potential asset]? Why would she be of interest? What would he have seen in her?

Obviously, it's all about access. In the Chinese system, high-ranking Communist Party officials and intelligence officers are the people that the counterintelligence service is interested in. Without addressing this particular case directly, because a great deal of it is classified, I would say that what you really want from a person like this is that they have an in with intelligence officers, they have an in with high-ranking Communist Party people.

They can tell you what those people are talking about, what their motivations are, what kinds of things they want to find out, what questions are they asking. Where are they going? What are they doing? It can help the counterintelligence service understand what is that foreign intelligence service, what is that foreign Communist Party doing that would be of interest to counterintelligence.

These are all the same talents, or connections, that would also probably make that asset look like a very good prospect for the Chinese also.

Right. Obviously, somebody who can play in an influential circle, shall we say, becomes of interest to both sides. Frequently, in the counterintelligence business, the degree to which an asset or source has access to the intelligence officers on the other side will help determine how valuable they are as a source. So the other side's interested, because of this person's access in the United States, and we're interested, because of this person's access, in this case, in China.

[It] gets to be a very complicated game.

It's very complicated. Frankly, a person who's worked it for many years as a street agent and then has become a supervisor and basically is looked to as a mentor by others, as someone who has developed a great deal of expertise in this game, that's a person who certainly has gone beyond counterespionage 101. That's a person who really knows the rules of the road. In a complicated affair like Chinese counterintelligence, a person who's a supervisor, a long-time worker, a China hand, as we call them, that's a person who would be looked up to as an example [of] how to do it.

One of those people was Bill Cleveland, who adds another level of complexity when it comes to this case. He's involved with Leung. Why? It seems like a very simple thing. You've got a handler. You've got a source, an asset. Where does somebody like Cleveland come in, and why the complication of that relationship, both specifically professional, but also personal?

Let me just say I've known Bill Cleveland for many years. He's a great person as far as I know. This whole case comes as a bit of a shock to me. But frankly, in any kind of long-term relationship that exists between two individuals, you have a certain amount of trust, you have a certain amount of friendship, you have a certain relationship.

I always say it takes two to tango. If one person in the relationship wants to take it to another level, wants to make it physical, for example, it takes that other person's willing agreement to do that.

In a counterintelligence operation, you have a situation where deception is going on all the time, where clandestine meetings are happening, where secret information is exchanged, where confidential relationships are built up. In those kinds of situations, it's very possible for a personal relationship of trust to develop into a personal friendship as well. These are things that you try to avoid in a professional counterintelligence operation. …

When two people are intimate, they're intimate under either one of two circumstances -- either it's legitimate or it's not. It's either licit or illicit. If it's illicit, in the case of a person who's sleeping with a married person, then obviously there is a blackmailable offense going on. There is an illicit aspect to it that creates, for counterintelligence purposes, an impossible relationship.

So if you have something like that going on, it's obviously very undesirable, from a counterintelligence point of view. It would call into question the objectivity and the veracity of the information produced. You'd never really know whether the person was being objective and truthful, or whether the person was limiting in some way the value of the operation to your side, because of that relationship -- either because of excessive love or because of fear of exposure. …

Let's talk about Bill Cleveland for a second. Bill Cleveland worked for you. Describe him. Describe the kind of agent that he was.

Bill Cleveland is, to my knowledge, a very professional, very neat, very together, very personable individual. He always conducted himself in a proper way. He was always well spoken. He was well thought of by his colleagues. He was well thought of by those who worked for him. He was well thought of by those who worked with him. I worked as a colleague of his and a superior of his for quite some time. It was very hard to find anybody who didn't like Bill Cleveland.

After Bill left the FBI and went with Lawrence Livermore Laboratories, he was still very well respected, very widely regarded as one of the best China hands in the United States government. Certainly fluently spoke Chinese and had many friends in the Chinese community. Was always very discreet and proper. Was a good family man. Very religious guy, and a very well respected person.

Sometimes a lot of people will say that the program attracted a geekish group of people in some ways. But Cleveland seemed to be different than that. How was he different, and how did that come into play with how he used assets?

Bill Cleveland is the son of a very high-ranking FBI agent, a person who is renowned in the FBI, a person who, even in retirement, was renowned among his colleagues and the Society of Former Special Agents of the FBI. So Bill had a legacy, a personal legacy.

Bill became a China hand and became one of the nation's most expert counterintelligence officers with regard to China. … If you were going to have a conference in the United States government about Chinese intelligence, he was the person that you should have in the room. The operations that he controlled were numerous, were voluminous. They produced a great deal of very valuable information for the U.S. government, and they were highly successful. In fact, if you would balance Bill's productivity as a group supervisor against anybody else's, Bill would come out on top. So he was very highly thought of and a very successful professional.

His understanding. What made him better with gaining assets, with getting information from assets? What did he get that other people didn't get?

I think it's always good to be the right person in the right place at the right time. Bill was well trained to do this. He was highly motivated, and he had good people working for him. He was very selective if he could be, in trying to pick the best people. Frequently those who worked in Chinese counterintelligence had some language or other skill that made them valuable to it, and Bill knew how to exploit that.

Chinese counterintelligence work in the FBI was not the number one most attractive thing that people thought of to do. But those who worked it, even though they were perhaps geekish, they were very smart people, and they were very good at what they did. …

One of the cases he got a lot of kudos for was Tiger Trap. Explain this to us. In some ways, this long-term investigation, which didn't ever really go anywhere -- that's the common belief. But yet a Cleveland gets a lot of respect and a lot of kudos for that investigation. Why?

I don't want to talk about investigations that are still classified, even if they've been written about. But let me just explain the basic approach that you might take as a counterintelligence officer to a case where you think espionage is being committed. Perhaps someone is trying to steal U.S. nuclear weapons secrets. So you have some goals. The first goal is to find out who's responsible, who's got access to those secrets that might be giving them to the other side. Having developed a primary suspect, you have to find out if there's concrete evidence that this person did it.

Are you going to go about collecting the evidence in a way in which it would be most conveniently produced in a court of law? If you conduct intelligence operations through close-in surveillance of this person, you might actually produce evidence that's not usable in a court of law. If you use sources and methods that are not usable in a court of law, you might very well make it impossible ever to prosecute that person, but you would find out exactly what they're doing, exactly what secrets they're finding out, exactly what secrets they're passing to the other side. That sort of thing.

At the end of the day, you have to make a decision. Are you going to prosecute this individual? Are you going to see to it that this person's fired? Are you going to remove their access to classified information? Are you going to interfere with their relationship with the foreign intelligence service, make it impossible for that person to ever really help that foreign intelligence service again?

In the overall scheme of things, you have a couple of big goals. One is to cut off that foreign intelligence service from its access to classified U.S. government information.

Another is to neutralize that source of that foreign intelligence service, so that person can't ever be a threat to the U.S. again. I would say that in most of the cases that Bill Cleveland or anybody touched in my time in California, we succeeded in removing the access that the foreign intelligence service had through U.S. sources to the classified U.S. government information. It wasn't always pretty, and it did not result in as many prosecutions as I would have liked to see. It didn't mean we weren't successful. …

[How do you think Cleveland felt about not getting a confession out of Gwo-Bao Min?]

I think he was extremely disappointed. I think anybody who interrogated an espionage subject and didn't get them would be extremely disappointed. Those interrogations and those confessions have been absolutely essential to the prosecution of espionage cases. Sometimes the reason a case doesn't get into court is because that kind of confession is not forthcoming, and obviously, that's a measure of success. It's not the only way that an espionage case is successfully concluded, because you don't have to prosecute somebody to remove the source of information from the foreign intelligence service. …

So in your view, Tiger Trap was a success?

Let's take a big step back and ask ourselves, how many people from several U.S. government agencies have we ever heard of being charged with espionage? State Department? Department of Energy? Pentagon? How many times? I mean, we're talking about FBI agents that are being charged with espionage. How many agencies really want to see their employees brought to justice for espionage?

There's a huge belief, and it's based on many proper factors, that it doesn't do an agency any good to bring a person to justice for espionage. It drags you through the mud. One of the worst things about success in counterintelligence in the FBI, as in the Cleveland/J.J. Smith/Leung case, is -- guess what? Our success is our failure. It points out just how human and how frail we really all are. Does that mean it's not happening in other agencies? Oh, yes, it's happening in other agencies. You don't see how the case is disposed of.

So I would say that there's a real argument to be had here about whether it suits the government in every instance to have an espionage prosecution, and whether or not success would be judged by what gets into a public courtroom.

… Does that mean to some extent we ignore it, that we allow it to happen?

No, and I think you need to look at the demographics. We've gotten very focused in recent years on this racial profiling thing. But let's look at it from the point of view of the Chinese intelligence officer. One of the things you find out if you're a Chinese intelligence officer is that the Asia-Pacific population in America has gone from less than 1 percent to more than 3 percent of the population. [It's] considerably more than that in West Coast cities like San Francisco and Los Angeles, where there's a huge amount of U.S. government classified contracting.

Then you find out that, in the physical sciences and engineering, at the head of the most complex and most valuable research and development that the U.S. is doing, you will find a very high percentage of Chinese. Five to 10 percent in some cases. Why is this? Well, it's their chosen profession. Coincidence? Who knows? Who cares?

Now the point of the matter is, if I'm a Chinese intelligence officer and I'm looking at this group of people who know about, let's just say nuclear engineering, who am I going to look at? Am I going to look at the Italian-American? Am I going to look at the Irish-American? Or am I going to look at the Chinese-American, who still speaks my language, who still has an aunt over in Shanghai?

So if I'm a Chinese intelligence officer, I'm not thinking about infiltrating somebody into that nuclear engineering facility. I'm thinking about how do I go about getting access to that Chinese person who's already there. How do I recruit that person? How do I develop guanxi with that person? How do I assume a relationship in which I have the upper hand and I can convince that person to share information with me?

At the end of the day, it's pretty elementary from the standpoint of the Chinese intelligence service. They do practice racial profiling, and it's very successful for them.

Your answer to the claims out there that what that does is it leads to a stereotype, a double standard, where it's not only a few people, but anyone who's of Chinese ancestry is viewed with a different lens -- how does one get around that, that there might be a double standard?

It really isn't a double standard when you stop and think about it. It's basic methodology. It's basic counterintelligence methodology and security methodology. The first question you ask yourself is, does that foreign intelligence service have power over that person? If they have control over the relatives of that person, the place where those relatives live, the welfare of those relatives -- that's a tremendous leverage, or potential leverage over that person, obviously a source of potential influence.

The other aspect of it is, how can I best find out what I need to find out? Maybe English is my second language. If I have somebody I can talk to in my own native tongue, that's a huge advantage to me. From the standpoint of the security guy, if you were looking at three candidates for a position at HP, and one of the candidates was coming from Dell Computer, would you automatically put him at the head of the line, or would you put him last in line, because that's a competitor of yours? You don't want to hire a competitor. So it's the same kind of thing in the intelligence game. You want to hire somebody to do nuclear engineering. Do you really want to hire somebody with a whole bunch of relatives in a hostile foreign nation, or not? It's a call that you have to make.

If you're a counterintelligence officer, it gets even worse, because you look at basics like, where are these people? Where do they go? For example, in a small group of nuclear engineers, in a facility that has the technology that you think has been stolen by that foreign power, how many of them ever went to that foreign power's home country? Well, guess who it is? It's the people who have a reason to go. It's people who have relatives there. It's people who have colleagues there that welcome them and want to discuss with them their latest science and technology research and development.

So it gets pretty elementary. It has really nothing to do with racial profiling, it has to do with the facts. You start with a small group of people that the foreign intelligence service has access to, and then you develop a profile based on potential influence, not their ethnicity.

That brings up, of course, the Wen Ho Lee case. Did that sort of analysis lead us wrong, at least in the judgment about the W88? Did we go down a wrong route, and why?

It's interesting, you know. If you're going to work in counterintelligence, you have to take your clues where you get them. So it comes down to the same kind of thing that occurs in almost every other kind of criminal investigation. You look at motive and opportunity.

If you look at the very small list of people who have motive and opportunity and then you look at their pattern of behavior, do you think it's suspicious for somebody to download voluminous quantities of classified information and to move that data onto another type of magnetic media? Do you think that's a violation of the rules?

I think, in the case of Wen Ho Lee, he pled guilty. Why did he plead guilty? Well, his actions made it impossible for him to skate on that particular charge. Did he commit espionage? It doesn't say so in the records, does it? From a counterintelligence point of view, he's no longer a potential threat to the United States.

But I think it comes down to a simple question of, who had access, who had motive, who had opportunity, and what did they do with that opportunity? Did they do things that are either highly suspicious or certainly prove that they didn't handle classified information correctly? It's a pretty open-and-shut case to me. You can talk about racial profiling, or you can raise a smokescreen about this cruel FBI out there. But at the end of the day, I think it's clear that the right thing was done -- perhaps not in the right way -- but I think it was the right thing to do. …

What brings the few cases that do come forward [for prosecution]?

There's a federal process for bringing a case. That is, that the federal investigative agency, in this case the FBI, brings the case to the designated federal prosecutor, who gives a prosecutive opinion. Usually it results in saying, "Yes, I'll prosecute," or "No, I won't." Frequently it results in a long list, a laundry list, of things that the investigator has to do in order to prove the case -- even to get it indicted, much less into a court.

In the case of espionage, you will find that the United States government has lost one espionage case since World War II. One. … Every other case has been a victory for the United States government. Why? Because all of the dots have been connected. All of the evidence is there. In almost every instance, the individual charged has confessed to espionage and there has been incontrovertible proof. We have a very high set of standards, historically, in the United States for prosecuting espionage. We don't take the 80 percent cases. We take the 99.9 percent cases, and those are the ones that are brought.

In espionage, you have a big complication that frequently occurs. That is that the agency from which the classified information was stolen has no particular interest in seeing the case get into court. The intelligence and counterintelligence agencies have good reason to prevent the case from getting into court, because all of their methodologies are going to be laid open for everyone to see.

In the old days, before the Classified Information Procedures Act, you had to lay out the secrets in front of the court as well, and they became part of the public record. So that was a good reason not to bring an espionage case. But today there are ways the judge can seal those records, so that the actual classified information stolen never gets into the public media.

Nevertheless, I would point out to you that all of the major espionage cases in recent years have resulted in books and articles and shows on TV. Those shows and books, the media, have made it very easy for foreign intelligence services to see exactly how the U.S. operates and, in fact, who was doing the operating. So it opens up counterintelligence and counterespionage to public scrutiny and to the scrutiny of foreign intelligence services. It makes it very easy for them to see. …

So it's a weighing procedure that has to go on by the prosecutor who decides whether or not to prosecute the case. In the case of espionage, all of those cases are brought to the Internal Security section of Justice, where they decide whether the thing will be prosecuted or not, and, if so, who will actually prosecute it. Frequently the attorney who prosecutes it is somebody with a lot of notches on their belt, who has put a lot of people in jail for espionage. That person is in Washington, D.C., and they actually go to the location and prosecute the person wherever that is.

They're very, very careful. If you think that a bank robber is 93 percent bound to be convicted when the FBI arrests him for bank robbery -- and the stats were like that at one time, I don't know what they are right now -- it's 99.9 percent that they're going to be convinced of espionage when they're indicted. …

Are we having success in the China program?

I think the bottom line is that we've got limited success in the China program. It's a shame that more of the successes can't come out in public, and they shouldn't. But they are limited. They are limited in numbers.

I guess the thing I'm afraid of is, when you look at a nation that's that large, that has that large of a presence in the United States, and that has, basically, blood in its eye as far as modernization and military and market success, competitiveness, if you will, that the United States has never expended very many resources on the People's Republic of China. We've never thought it represented all that big of a threat. As a consequence, we've never had quantitatively as much success as we have against other foreign intelligence services that were operating against the U.S.

So the recent statements by Mueller and others to say that China is foremost in the future, [and] that counterintelligence, counterespionage is now number two priority, what does that say?

It says that there's a recognition that this is important now. Frankly, I think this has come over the last few years from success.

Not failure?

Not failure. Right. If you failed, then their secret services would be succeeding without you knowing. Once you realize what their secret services are doing, that's when they become a higher priority. That's when you beef up your operations. That's when you improve your operations.

I think it's clear from my view of public statements, that in the last few years the U.S. has come to an increasing recognition of the potential danger posed by the intelligence services of China here in the U.S., and the necessity to do good counterintelligence operations, whether or not the other side is committing espionage. I think you have to know what they're doing. You can't judge if what they're doing [is] threatening or not unless you know what they're doing.

That comes back to assets and sources a lot.

Right. It comes back to all of the methods that you use to conduct surveillance on the foreign intelligence service, and people are one of the key areas, especially for the FBI.

A couple of areas to clean up on the Leung case. In 1991, there's a trip over to China by Cleveland. He comes back and he hears the recording, the famous recording, and he recognizes Leung's voice on it, and he calls J.J. Eventually they go to Washington. How was that handled? Was that handled correctly, and why was it handled in that way? That was surprising.

I don't recall all the details of the way those things went down, the way that was handled. But what I do recall is there was a great deal of scrutiny given to that case, not only in 1991 or 1990, but at other times. The scrutiny was given to the case because of the quality and the quantity of the information being obtained, the costs of the operation, and the perception that we needed to be winning; that we had to make sure that we were in control of the operation and that the operation was sufficiently productive to justify itself.

There's always the question of, are you telling the other side more than they're telling you, or is this operation producing more information and better information for us than for them? Every time you have a source who has admitted that they are in touch with foreign counterintelligence, you have to wonder whether they're not getting more out of it than you are. …

How much attention would then be paid to a situation that's a problem, even if the handler is then given responsibility to resolve it? How many red flags go up in Washington? ...

There are two things that are really pertinent to this case, and these are things I can talk about without getting into classified information. Now the first thing is, the case did receive additional scrutiny, but actually it's the same kind of scrutiny that's given to any major source case. So you're always trying to measure the case and weigh it. The more sensitive the case, the more likely it will get very intense scrutiny, and in this case obviously it did. So there was a continual weighing of it.

The second thing is what I would call secret side activity, which, if known, would certainly have made a difference in the way the case was administered, supervised. That is, that there was a personal relationship that was not known to the FBI, at least not to the people in the chain of command. So at that point in time, you have to judge the facts and circumstances quite differently, don't you? To my knowledge, that was not done, but that's because people involved were deceptive. They continually deceived their supervisor. They continually deceived the agency. They made it impossible to objectively judge the case, because they basically doctored the facts.

How easy was that to do in this case? A lot of people are saying that you've got a good source, you rise up the ladder, and people just look at you differently, and so J.J. was bulletproof.

Nobody's bulletproof. The fact that somebody would get called back to Washington shows that they're not bulletproof. In fact, it's always possible to judge a case one way or the other. It's like, should we arrest somebody, or not? You can believe that a value judgment was made about whether or not to arrest these people in this case.

Certainly it's not good from a PR standpoint. So why would you do it? Well, you do it to do the right thing. You do it to clean up messy operations. You do it to set a standard for the future. You do it because somebody did a wrong thing, and should be put in jail for that wrong thing. …

In this case, did Cleveland run it in the correct fashion except for the fact of hiding his relationship also? Did he do it according to the books?

I think history is going to show one way or the other, and the court case is going to bring that out. To my knowledge, Cleveland did the right thing. But I think this case shows that a supervisor can only know so much. Somebody in charge of operations can only know so much about each individual operation. Even very important operations can be conducted in a manner that's not only unorthodox, but basically illicit. And that can happen. …

[We're] talking about a situation where it's very difficult, if not impossible, to ascertain the truth. In this situation, some people have said that Leung, as the important asset that she was, involved for so long with two agents, knew an awful lot, so that, in some way, if this asset was tainted, to some extent the whole program was tainted?

It could very well be true. Frankly, when a critical asset is tainted, it makes you go back and take a look at the damage. I'm sure that that's been done in this case. I think it brings up the Chin case. Here's a situation where, for 30 or more years, [you have] a person working for the CIA, ostensibly as a person who was merely translating the media from Chinese into English for evaluation, open source intelligence, essentially. He was such a great linguist that many, many very sensitive cases were brought to him to translate from the Chinese to the English. There were so few people like him available at the time, in the CIA, that that made him aware of way more cases, and way more sensitive information than he should ever have been made aware of.

So it meant that, as an asset of the Chinese intelligence services, he was far more valuable than anyone had a right to assume he ever could be. It also means that more damage was done to U.S. intelligence, to U.S. counterintelligence, than would have been done had the source been a different source.

In the case of Leung, what's the risk? What's the damage that could have been done?

Well, very grave damage could have been done -- partly because of her ability to manipulate, partly because of her ability to elicit information and the possibility that, as charged, she's actually copied very highly sensitive documents and actually turned them over to the other side.

That could result in compromise of sources and methods. It could even result in death or execution. It certainly could result in compromise of U.S. government interests and intelligence interests with regard to China. So this could be extremely damaging. …

Are we playing a game of checkers, while they're playing a game of chess?

There are those that would say that, because the Chinese are renowned for their thoughtful application of principles. I'll give you a current example that's kind of scary to me. The Chinese command and the military have actually written books saying that the great equalizer is that the U.S. is so dependent on its computers that we can even the score by basically using the Internet against the United States -- that it is the great equalizer in a conflict.

Obviously, we worry about nuclear weapons. But I think we're in a new era now, where we have a lot of other things to worry about. How many manufacturing jobs move to China? How is it that they can produce something identical to our products, whether they be CDs and movies, or whether they be automobiles or aircraft?

I think we have to decide, what is it we're going to do from a counterintelligence perspective to protect our economic interests, to protect our national security interests? I think that we simply can't let them win this game this easily. They've been winning, at least in the public media. I think we have to make sure that they don't win it. …

One of the other cases that were going on was the Peter Lee case. What were the allegations there? What did we fear had happened?

Again, I don't want to get into stuff that was classified when I left the FBI. But, again, let's just take a step back and look at the kind of allegations that we're talking about. What we're talking about is people who have to highly classified and sensitive U.S. government weapons information, and that might have been transferred to the People's Republic of China. Obviously, there are three major goals: Find out the extent of the loss and the evidence of it; to present the case as a potential espionage prosecution; and to neutralize the leak, if you will, the loss of classified information. So it may be that you can only do two of three, but in so doing, you would actually succeed. ...

In that case, a lot of people say that he was rapped with a ruler on the knuckles, and that's about all here. Your view sort of as a historian also of these events -- did we go wrong? Where did we go wrong? Should something else have happened? Would it have been more effective to have thrown the guy in jail? I mean, how important a case was this?

One of the great frustrations of my career was to have seen a number of cases that were presented for potential prosecution, where it was decided not to prosecute them. Many times an agency that was in charge of the classified information didn't want the case prosecuted. Many times the prosecutor demanded additional evidence or additional facts that were not there to be used in a courtroom. Many times, out of a desire to find out where the leak was occurring, we conducted operations that, later on, the prosecutors didn't want to bring into a court, for many reasons.

So it's clear that, in some of these cases, it would have been better historically for us to prosecute them and maybe take our lumps. Certainly, somebody made the decision, we're going to take our lumps with J.J. Smith and Katrina Leung. Even if it hurts, we're going to prosecute these people. That kind of decision, that hard decision, has not been made in every case in the past; too bad that it hasn't been. Hopefully, we've learned from that.

And in the Peter Lee case, specifically?

I can't tell.

That's on the public record, though. That's what happened. Are you happy with the eventual decision to not prosecute?

No. I'm not happy that he wasn't prosecuted.

And the reason for that is?

Because I think we prosecute an awful lot of messy cases. The FBI investigates theft of government property. It investigates highly sensitive frauds from within government agencies, and we put politicians in jail. Those are horribly painful cases. Why can't we take a case like this and do it? And that decision was made by a prosecutor that knew exactly what he was doing. So he's the judge of that situation. He's the lawyer. I'm not a lawyer. I have to live with the results, but I'm not happy that we haven't had major Chinese espionage prosecutions. …

We've already talked about the fact that the FBI has sort of noted that this is now number two priority. We've had some problems, it seems now with the assets; we were worried about that. How do we expect to win in this situation, looking at the track record and what it's been?

Let me again just take a big step back. When the FBI says counterintelligence is its number two priority, I think it makes it very clear that, next to terrorism, this is the most important thing the FBI does. The real reason is because it's only the FBI that does this. Customs doesn't do this very much. Immigration doesn't do this very much. Alcohol, Tobacco, Firearms Division doesn't do [this]. So besides the CIA, the FBI is the one that does this. …

I think a foreign intelligence service working against the United States has a decided advantage. We have a free country. We think information is free. We don't restrict our citizens very much, even those with highly classified access. So I think we have to be very careful how we do this.

But I think we have to be very good at penetrating that foreign intelligence service in order to find out who their agents are here. We have to do a better job of security within our agencies, and within really business in America, to protect the things that are essential to us, our intellectual property, our research and design and things like that. We're not good at that as a society, and that's what we have to improve. …


home + introduction + the "parlor maid" case + how china spies + inside fbi counterintelligence
interviews + join the discussion + producer's chat
tapes & transcripts + press reaction + credits + privacy policy
FRONTLINE + wgbh + pbsi

posted january 15, 2004

FRONTLINE is a registered trademark of wgbh educational foundation.
web site copyright 1995-2014 WGBH educational foundation