COUNTDOWN: 3, 2, 1, mark and visual confirmation.
MARGARET WARNER: It was test number four for the controversial missile defense system championed by President Bush. And this one was declared successful when a 120-pound interceptor found and destroyed its target over the Pacific Ocean late Saturday night. Here's how the test unfolded. A modified minuteman intercontinental missile carrying a warhead and a large balloon decoy was shot off from Vandenberg Air Force Base in California at 7:40 p.m., Pacific time.
Twenty minutes later, the interceptor, or kill vehicle, was launched into space from the Marshall Islands, some 4,800 miles away. Just ten minutes after that, at a point 144 miles in space, the kill vehicle hit the target warhead and demolished it. The news was welcomed at the Pentagon, after two failed tests last year. The head of the missile defense program spoke to reporters shortly after midnight.
LIEUTENANT GENERAL RONALD KADISH: The early indications that we have... there's a lot more work to be done, but the early indications that we have is that everything worked in what we call a nominal mode, which is what is expected. However, these tests take many weeks to deduce the data. But we believe we have a successful test in all aspects at this time.
REPORTER: Was this by more the most rigorous and difficult test, or was it just slightly more realistic than the last one? How would you rate it?
LIEUTENANT GENERAL RONALD KADISH: This test is the same test that we did the first time, that we tried the second time, and we tried the third time. So this... this is an idea that we want to replicate the difficulty so that we can get confidence in the reliability of the hit-to-kill technology.
MARGARET WARNER: The Pentagon is asking Congress for $8 billion in missile defense funding for next year -- a 57 percent increase. About half the increase would go for the ground-based program that was tested Saturday night. But that's only part of what the administration has in mind in creating a shield that would protect the entire United States, and perhaps its allies, against ballistic missile attack. Last week, Pentagon officials laid out plans for a multi-layered system that would also launch interceptors from the air, sea and space.
The Pentagon plans to pursue all the different technologies simultaneously. The hope, according to a recent Pentagon memo, is that the deployment of an interim ground-based system could be completed in Alaska as early as 2004. Ground-clearing for that ground-based component is set to begin next month at Fort Greeley Army base in Alaska. Another, more complex anti-missile test is scheduled for October.
MARGARET WARNER: For more on what the weekend test tells us, and what it doesn't, we hear from three scientists with extensive backgrounds in weapons development.
Theodore Postol is professor of science, technology, and national security policy at MIT. In the early 1980s, he was the principal weapons technology adviser to the chief of naval operations. William Graham is chairman of National Security Research, Inc., a consulting firm. He was a member of the Ballistic Missile Threat Commission headed by Donald Rumsfeld in 1999. He's also a longtime member of the Pentagon's ballistic missile defense advisory committee. And Philip Coyle is a senior adviser at the Center for Defense Information, a think tank. He was the Pentagon's director of operational testing and evaluation during the Clinton administration. Welcome, gentlemen.
Mr. Coyle, beginning with you, tell us a little bit more about this test this weekend. What exactly did it demonstrate in the way of capabilities?
PHILIP COYLE: It demonstrated the first capabilities that you would need to hit a bullet with a bullet, as they say. It's like doing a hole in one when the hole is going 15,000 miles an hour. But in the future, we'll have to do that same sort of thing where the hole is in the green with lots of other flags and holes that look just like the real thing. So these are decoys and countermeasures that an enemy might produce.
MARGARET WARNER: All right. But going back to this weekend's test, so how much help did the interceptor, the kill vehicle, have in homing in on the target?
PHILIP COYLE: The kill vehicle is guided early on by a satellite system and by radars on the ground, and then by a radar beacon, which is actually on the target. This is a necessary thing at this point because we don't have a forward-based radar such as you would have in a real operational situation. But that radar beacon obviously having an active beacon on the target is not something you would have in a realistic war-time situation.
MARGARET WARNER: Then how about the balloon dummy, the balloon decoy that was sent up? Did it also have a radar beacon so at least the kill vehicle had to differentiate or did it not?
PHILIP COYLE: No, it did not but it's a large balloon, larger than the reentry vehicle target, the supposed enemy target. So it's not a very challenging decoy and doesn't need to be in such early tests.
MARGARET WARNER: So, Professor Postol, what do you think this weekend's test told us about the feasibility of a missile defense system?
THEODORE POSTOL: Well, I agree with Mr. Coyle completely. I think the test does show the very high level of guidance and control capability to hit what I would call a cooperatively... a cooperating target. And that is an impressive technical achievement. But it doesn't in any way address the very fundamental problem that Mr. Coyle already pointed to, which is the problem of decoys. For example, the balloon that was deployed along with the warhead is nearly ten times brighter than the warhead. And there is no discrimination involved here.
Basically what's going on is the kill vehicle sees the balloon and it sees the warhead and it says, oh, this object was very bright relative to this other object. I have been told to home on the less bright object. So it homes on the less bright object. If you took the balloon and you put the warhead inside it, for example, and then took another balloon that was empty -- or for that matter dozens of other balloons that were empty -- and each of these balloons you intentionally put stripes on or had slightly different coatings, what the kill vehicle would see is a bunch of objects in front of it all of which looked slightly different and the kill vehicle could measure these slight differences but it would have no way of knowing why these slight differences occur.
So essentially it would essentially have a probability of one over n, where n is the number of objects you're trying to hit, in terms of hitting the target. It would only hit one out of 20. It would have a very low probability of picking the right target.
MARGARET WARNER: Okay, so are you saying, and I probably won't use very accurate technical language here but essentially that what the test showed is the ability to guide this kill vehicle but it really showed no capability yet for the kill vehicle to discriminate among targets?
THEODORE POSTOL: That's right. And, in fact, the first test that was done showed that they had a very, very difficult time discriminating between objects. There were a series of objects that looked quite alike.
MARGARET WARNER: You're talking now about the very first test, what, two years ago?
THEODORE POSTOL: This was in 1998.
MARGARET WARNER: Three years ago.
THEODORE POSTOL: This was a test that was prior to the four tests we have just had. There were two tests that were just intended to understand what objects look like in space. Basically one of the tests, the one test that we have public data for, shows that it's quite easy to make objects that look credibly like, for example, a tumbling warhead.
MARGARET WARNER: Let me get Mr. Graham in here. William Graham, tell me what you thought this weekend's test showed us.
WILLIAM GRAHAM: First it showed that the systems that we had developed to execute this intercept worked as the models predicted and as we had had anticipated. That's very encouraging news. We're going to go forward with a very intense series of tests, each of which will be more complicated than the last over the next two or three years. You don't start testing the most difficult problem in your early tests.
You move forward in a step-by-step fashion. But at least we now have a situation where the administration is supporting an active program to make a best effort in developing a ballistic missile defense system. That's something we haven't had before. I think we'll find, going forward, that it's not easy to make the countermeasures that others have described perhaps except as pieces of imaginary engineering. Certainly when the U.S. has made countermeasures in the past, we found them very challenging engineering problems.
But it's worth noting that if people test countermeasures, we have the best observation and data collection resources in the world to watch those countermeasure tests. We're going to know more about the performance of their countermeasures than they will by the time they deploy them. If they don't test them, they have a very low chance of working if used in a real military scenario.
MARGARET WARNER: Do you agree with Professor Postol that the testing so far doesn't show one way or another, doesn't demonstrate an ability to discriminate among any kind of realistic sets of targets?
WILLIAM GRAHAM: Well, let me go back one step and say what we're doing. I wouldn't characterize these tests as hitting a bullet with a bullet. I'm an engineer and I would characterize them a little more accurately as saying we're hitting an offensive warhead about the size of a garbage can with a super computer. That interceptor is not a bullet. It's a very, very intelligent robot that's maneuvering and navigating itself on the signals it sees with its optical sensors right up to the intercept.
We have an initial decoy, a balloon. It's not the most complicated decoy we can think of but it's certainly a decoy that the world knows about and we're going to continue the tests with more and more complicated countermeasures. But the Defense Department's also pursuing a ballistic missile defense program that looks at other phases, boost phase, for example.
If we can intercept ballistic missiles before they finish their thrusted flight, before the engines shut down, they won't have a chance to deploy countermeasures. We can certainly overcome countermeasures in that way even more effectively.
MARGARET WARNER: Just to explain to our viewers because we didn't illustrate this in the set-up. Right now you've been testing just hitting targets at their mid-range point. What you're talking about is attacking them just shortly after they've taken off.
WILLIAM GRAHAM: Yes, or at least before the propulsion engines shut down.
MARGARET WARNER: Philip Coyle, address that aspect of the Bush plan which is that it's much more complex than previous administrations had pursued or at least actively tested; that is, it's going to be on sea, air, space, as well as ground and in all these different phases. How feasible an approach is that and how does it... I mean, do you think it's the right approach?
PHILIP COYLE: Well, the idea behind a layered system is that if you miss within the boost phase you get another chance in mid-course and perhaps a third chance in the terminal phase. But each of these phases and the technologies which are intended for them are very challenging, very difficult. For example, in the boost phase, to go back to my golf analogy, this is like hitting your partner's drive out of the air shortly after he has driven the golf ball with your drive. Obviously to do that....
MARGARET WARNER: Right off the tee, in other words?
PHILIP COYLE: Practically. And so obviously to do that you have to be very close and your drive has to be very fast.
MARGARET WARNER: And Professor Postol, what do you... what's your assessment of this multi-layered approach that the Bush administration's pursuing?
THEODORE POSTOL: Well, I think if you take zero capability and you multiply it by a number of layers you get still close to zero capability. I agree with Mr. Coyle that the boost phase is a very difficult and challenging area to work, but I think there are some limited capabilities that might be put in place. It's not clear they're at a useful level of capabilities so I agree that it's ambiguous. But certainly the midcourse and the reentry phases of defense that the administration is talking about have really essentially no chance of working.
I think it's worth pointing out that the last test series that were planned in May of last year revealed that the entire test program had been rigged to avoid simple decoys that were observed in the first test. We know that's the case. That's a matter of fact now. That also indicates that there's a problem with the program. People look like they're trying to be too optimistic and make claims that I don't think science supports.
MARGARET WARNER: Mr. Graham, your response to that?
WILLIAM GRAHAM: Well, I don't know what claims Mr. Postol is talking about. This is one step, one early step, in a systematic and rigorous engineering program that will go forward dealing with more and more complex and difficult situations. At the same time, it's one phase of a three-phase defense capability, and personally I'm not overwhelmed by the technology of North Korea or even China or Iraq or Iran. If I had to put my money on the technological capability of a country I think I'd put it on the United States.
MARGARET WARNER: Very briefly because I want to get to all three of you on this, starting with you, Mr. Graham, how soon do you think there could be a system operational that would protect the United States from ballistic missiles?
WILLIAM GRAHAM: I don't know but I do know that the law that the president and the secretary of defense are carrying out says it's the policy of the United States to deploy as soon as technologically possible a national missile defense capability against limited threats. They're trying to determine when that is. That's the purpose of these tests.
MARGARET WARNER: Professor Postol, your thought on the timetable here?
THEODORE POSTOL: Well, I think research is certainly a good thing to be doing. So I am an advocate for research. But my expectation is I cannot foresee a future time at this point where the technology will support what the people claim they can do.
MARGARET WARNER: And, Philip Coyle, your thought on the timetable and address this aim at least that was laid out by Pentagon briefers and so on. I'm turning now to Mr. Coyle, that they hope to at least have part of the ground-based system operational by 2004. Do you think that's doable?
PHILIP COYLE: It will depend on how rapidly the tests can be done. If there are going to be another 20 developmental tests before we even get to realistic operational tests. And if each of those 20 tests are a year apart as the last two were, that would take 20 years before you'd even get to realistically operational tests. If -- as the administration is proposing -- they can do those 20 tests in four or five years then you might be able to begin realistic operational testing at that time. But it still would be a number of years away.
MARGARET WARNER: All right. Well, gentlemen, thank you all three very much.