Bobby R. Inman

The Walker Case: A Direct Hit?

Q: What's the value of the loss to our security in the Walker spy case?

A: All we know really is where the people served. If you look at their duty stations, there are some judgments that you can make about worst-case kind of losses. First . . . would be service on an SSBN, a U.S. ballistic missile submarine; that would have been at or near the top of Soviet interests. How the SSBN operated, where, when did it come near the surface to communicate, did it transmit communications, if so, exactly what time, when, where, on what frequencies?

A good deal of that fortunately has been ameliorated by time. And because of the advent of Trident, we're operating in entirely different areas -- different missile systems, a great many things are different. While the damage could have been very severe at the time, the prospect of its being useful still to increase the vulnerability of that part of our deterrent force has been substantially reduced.

The second major area of possible loss would be from the service (by one of the suspects) with attack submarines -- the ability to see the communications from the attack submarines. Revealing details on how we went about detecting Soviet units, the effectiveness of it, how we operated, what kinds of tactics we used, what kinds of tactics we might use if we were to move into a time of crisis or hostilities. There could have been insights into our other means of locating foreign submarines, including some impact on both surface and air capability. And there, because systems stay in place a lot longer, the damage could be enduring.

The third area that is still pretty murky is the degree of exposure to surface, amphibious and naval air warfare areas, where there is the potential that message traffic could have been provided on the details of exercises, exercising our war plans, candid assessments of weaknesses of equipment or of tactics and doctrine. All those that have proven particularly valuable.

The fourth area would have to do with the security of communications. Here, because of losses in the '50s, early '60s, the U.S. has moved to vastly more expensive and complex means of protecting its own communications. Variable, changing, keying material every day, designed to withstand the damage from an occasional individual who would sell a code of security.

I inherently worry when I see that one of these indivduals was a crypto repairman, and the prospect that they were in position to provide materials for years. Even the best of systems could come under some attack under those circumstances.

Now, that's clearly a worst-case look. It would require a continuous flow of material for a very long time, and we do not in fact know at this point in time that that did occur.

Q: In the spectrum of cases over a period of time, this is one of the worst?

A: It would not reach the magnitude in my judgment of the Klaus Fuchs-Rosenberg passing-over of nuclear weapons. But it sure stands pretty high on the list of those that come thereafter -- in prospective damage. It may turn out, if we're lucky, that there was substantially less than could have occurred.

Q: How could this have happened? What produced this hemorrhage, and why wasn't it caught in time?

A: We know that during the '70s there was strong bipartisan support to try to improve trade and long- term relationships, and we permitted a substantial increase in the number of Soviet citizens in this country legally and East Europeans and the PRC, other nations as well. In the same time frame, we were drawing down resources across the national security account and elsewhere in government, and the number of counterintelligence agents in the FBI and the military services were drawn down substantially. So there was a mismatch in the numbers of people trying to maintain continuous surveillance of those who could be engaged in recruiting and running spies.

Then, there has been an increase in the total numbers of people with access to classified information, and there have been allegations -- probably reality from time of time -- of information being classified that didn't merit classification.

Finally, beginning in the middle '60s, there was a substantial hemorrhage of classified information being leaked to the press. Newsmen didn't go into the offices and take classified documents off desks. They were given them by government employees authorized to have access to classified information. There was an impact on the judgment of individuals about the value of the information they were charged to protect.

If you've got a young enlisted man who's already got a family who's living close to the poverty line, he sees a classified message, he looks at the front page of a newspaper and sees the same information, he knows that someone provided that. And if he's been approached, the temptation is to go ahead -- since there's already been a damage -- to sell the information for cash. Certainly all this increased the hazard that people would elect to sell.

In looking at the personnel security system, you've got different categories of information to protect. You've got, we're told in recently published figures, perhaps 100,000 people with the most sensitive level of clearance. Those people are subjected to very detailed background investigations every five years. A substantial percentage -- civilians at the National Security Agency, all employees of the Central Intelligence Agency -- are subject to polygraph examinations as a part of that process.

The next largest category, 600,000 or so, the figures indicate, have access to top secret including cryptographic materials. There again, background investigations are conducted -- polygraphs are not used -- and they are less rigorous than those for the most sensitive level of clearance.

The vast majority are at secret level and below, and there much less stringent procedures have been in place both for conducting investigations and for determination of who might have access.

In the current case, John Walker and Jerry Whitworth were in the category of those with access to top secret and crypto clearances. So simply reducing sharply the number of people cleared would not have impacted on their access.

Q: What about remedies?

A: One problem is the basic attitude of the society at large about selling secrets. It's a basic question of ethics. It impacts too on the sale of industrial secrets in corporations, but of course much more significantly on the damage done to the country at large from the sale of the government's secrets.

We're going to have to increase the penalties. I'm inclined at this point to mandatory life sentences without parole for those who have sold the country's secrets.

We need to find some way to create incentives for those who suspect that an individual is selling secrets to alert the authorities -- and not wait years. I don't have any good answer what those incentives might be. They could range from being an accessory to the crime if they don't report it all the way on to pandering to the same greed that creates most of these cases in the first place -- offering rewards for reporting if it leads to an arrest and a conviction.

What we don't want to do is to create a climate where there are all kinds of mindless allegations, and a person has a cloud over his head until he proves himself innocent.

On the question of what is classified, obviously the system could stand a good rigorous wirebrushing and an examining both of what is classified and of how many people need to have access to it. But my sense is that's not going to have a major effect. There clearly needs to be more resources applied to the job of screening individuals who have access.

In the area of personal security investigations, I believe we should consider substantially wider use of polygraph in the process of trying to get through clearance of a much larger group of people. Not polygraph to inquire into everybody's personal private life: after all, blackmail hasn't shown up as an element in any of these cases. But there are two fundamental areas that wouldn't take very long to look for some kind of difficulty.

One is financial status: does the person have unaccounted-for outside income? Is he living substantially beyond the means of the salary he's being paid or other financial arrangements that he's prepared to account for?

The second one is: does he have any kind of contact with individuals from other countries, or whom he believes may be acting on behalf of other countries, who have attempted to solicit classified information? Certainly as you look at this case, one of the early areas for these approaches would be those who handle communications, cryptographic equipment.

When one looks further out, there's clearly also an advantage to random rechecks. One of the great success stories of the last couple of years has been the program to drastically reduce drug use within the military. And its answer has been the urinalysis tests, officer and enlisted, and then the random re-analysis along the way. I believe the reasonable prospect that there would be unscheduled recurring checks of polygraph -- all we'll need is two simple questions -- could have a dramatic impact.