StarDestroyer.Net BBS

Physicist blows whistle on US missile defence
From Roland Watson in Washington

THE credibility of President Bush’s multibillion-dollar missile defence plans are being questioned by leading scientists after claims that the results of key tests were falsified.
The Massachusetts Institute of Technology (MIT) is considering an investigation into accusations that fundamental flaws in the proposed “Son of Star Wars” system have been covered up.

The criticism is led by Theodore Postol, a physicist and missile defence critic at MIT, who has said that the institute is sitting on what is potentially “the most serious fraud that we’ve seen at a great American university”.

After months of demanding an inquiry into the affair, Ed Crawley, the chairman of MIT’s aeronautics and astronautics department, has reversed previous refusals and recommended an investigation.

The issue in question goes to the heart of missile defence technology, an article of faith among right-wing Republicans and a key plank in Mr Bush’s 2000 presidential manifesto. The United States unilaterally withdrew last year from the 1972 Anti-Ballistic Missile Treaty with Russia in order to pursue the controversial proposed system, which is designed to intercept enemy warheads in flight, a feat likened to hitting a bullet with a bullet.

Dr Postol and fellow critics say the ability of an interceptor missile to distinguish between an incoming warhead and the decoys likely to accompany it is deeply suspect. Any such doubts would cripple the credibility of the system.

Such questions date back to mid-1997 when the military contractor TWR Inc was accused by one of its employees, Nira Schwartz, of faking test results on a prototype anti-missile sensor meant to tell hostile warheads from decoys.

The company and its system was given the all-clear by the Lincoln Laboratory, a federally funded research centre at MIT. But subsequently the General Accounting Office, an investigative arm of Congress, accused TWR of exaggerating the sensors’ performance, saying its conclusions had been “highly misleading”.

Dr Postol has written to 20 members of Congress saying that MIT’s reluctance to investigate the role of its own research centre “may indicate an attempt to conceal evidence of criminal violations”.

Critics say that MIT’s independence is compromised by its interest in maintaining hundreds of millions of dollars in annual government contracts.

The missile defence system, the first steps of which Mr Bush announced in December with the aim of having ten missile interceptors in Alaska by 2004, is being built by Raytheon, which beat TWR to the contract. But Dr Postol said the TWR test, which offers a rare glimpse into the highly secretive world of missile testing and is based on the same infra-red technology used by Raytheon, suggests some flaws that challenge the overall feasibility of the entire project.

Dr Postol, a persistent missile defence critic who is accusing MIT of a “serious case of scientific fraud”, cannot be lightly dismissed. After the Gulf War he challenged the Pentagon’s claims for the success of its defensive Patriot missiles, saying they had intercepted few if any Iraqi Scuds. Despite initial ridicule, his assertion is now accepted.

Since 1999 three of the eight tests of “hit to kill” interceptors have failed. Critics say that wrapping a nuclear warhead in radar-absorbing rubber foam or releasing thousands of small pieces of metal would be enough to fool an interceptor.

Separately the State Department yesterday charged two US aerospace companies with illegally supplying China with satellite and rocket technology that could be used for intercontinental missiles.

Hughes Electronics Corp and its parent company, Boeing Satellite Systems, stand accused of 123 arms control violations by helping China with technical data after failed rocket launches in 1995 and 1996. Hughes said that it had done nothing wrong.

Vympel wrote:http://www.timesonline.co.uk/article/0, ... 47,00.html

Let's talk about the more recent tests between 1997 and 2000. What did they reveal about the physical possibility of hitting a bullet with a bullet? Just what have we learned from those tests?

Well, I think from my point of view, there's never been a question in my mind that you can hit a bullet with a bullet. The problem is - can you hit a bullet with a bullet when the adversary is making a determined effort to hide the bullet as it comes at you with various means available to them and that's a very different problem. And what the Ballistic Missile Defence Organization learned in 1997 when they did their first, what's called fly by, was something that any competent military scientist would have known and should have known. It was simply that if you look through a telescope at distant objects in space, all of which simply appear like points of light, that many of these objects are going to look similar to each other and they're going to look in many cases, similar to the warhead and you're not going to be able to reliably tell one from the other and they really should have known this. This accident occurred because of some combination of ignorance and incompetence on the part of the highest level of management and in part because of the competence of lower level people involved in the programme, because what happened I believe, although I don't know for a fact, is that the people in charge have no idea what they're doing, by all the evidence I've seen. But at lower levels there are people who are competent doing different parts of the scientist programme and what happened is, there was this group at Sandia National Laboratory, one of our weapons labs, who was asked to build a sweep of decoys and these characters went out and built this sweep of decoys that were good, were credible, pretty much not knowing that they were expected to build decoys that are supposed to be non-credible and, and the people at the top were so ignorant and incompetent that they didn't understand that it was possible to build these kinds of credible decoys and all of a sudden this stuff was flying in front of their their kill vehicle and they couldn't tell one object from the other. And then the question how do we cover this up began to be a big issue.

So how were the findings disguised?

Well, let me back up and give you a little primer on the question of what's called discrimination, telling the warheads from the decoys. These objects are all in the near vacuum of space because the warhead is launched at hundreds of kilometres altitude, there are other objects surrounding it that are supposed to fool the defence into thinking they're also warheads. Now in the near vacuum of space if I have a rock and a feather, the rock and the feather will travel along together because there's no air to cause air drag to slow up the feather relative to the rock, so it just never slows up. And if the rock is tumbling slowly and the feather is tumbling slowly, they'll both again have their tumbling motion - will not be modified by air drag and an object can tumble in any way, in any - of a wide range of ways ah which are not determinable from just looking at, it doesn't tell you anything about the physics of the object. So, if I have a warhead that might look like an ice cream cone with a nose front, I can make a balloon that's shaped like a warhead and in the near vacuum of space that balloon that's shaped like a warhead is going to travel along with the warhead and if the warhead is slowly tumbling, and the balloon could be slowly tumbling, more slowly than the warhead or more rapidly than the warhead - because there's no way to know because it's an accident of the motion that's imparted when things are pushed off the upper stage of the rocket. So we have all these objects and if you were floating along in a space suit and could look at them with your eyes, you would have no way of knowing which was which by looking at their shape and their motion - and in fact, for example, just to show you how complex you could make it - if you thought the warhead were more distinguishable you could put a balloon around it and you know, so you could essentially make it fundamentally impossible for the human eye to select the object based on what you can see. Now, if you can make it fundamentally impossible for the human eye to not be able to determine which object is which, then a sensor that's operating thousands of kilometres away that has much lower resolution than a human eye, a radar or an infrared telescope has even less of a chance of being able to tell one object from another. So that would come to discrimination, how would they discriminate, by that I mean pick one object from the other. What they do, is they construct a template, by that I mean a set of estimates of what they think each object is going to look like when they're looking at it. So this by itself is a complicated thing to do because if you think of each object as acting like a light bulb, it's lit up under its own temperature. But I can paint the stripe on the light bulb and make it look different and um, so what I do is I have this template and I look at how bright each object is relative to each other, of course there's no physics in that I have to know what each object is prior to looking at them, and then it turns out what they also use is how much the object scintillates as it's seen from a great distance. And a way to understand why an object would scintillate is to imagine a pen as a light bulb and it is so distant that it just looks like a point of light to you, but when it's nose- on you see less bright area than when its side on - so it looks brighter side- on than nose- on, and if it's tumbling end over end, you could see from a great distance a point of light that got brighter and dimmer, brighter and dimmer, and that might indicate that this was a tumbling pen. Of course, if the pen were tumbling straight, it wouldn't scintillate at all - so the orientation of the object is important, too. So, what they did is they had these ten objects, a mock warhead, the upper stage of the rocket that deploys the warhead plus eight decoys and they looked at these objects, basically what happened is there was a bunch of these objects couldn't distinguish one from the other, in spite of the fact they expected them to look somewhat different. And the reasons for that have to do with, well actually we don't even know now all the reasons, we don't know how much of the failure was due to bad physical modelling which is one possibility, another part of the problem is that objects didn't deploy as they expected them to. Bear in mind that, if you're looking at an object that's scintillating because you think it's tumbling, if you deployed it you expected it to tumble end over end but instead because of some accident of what you didn't expect, it was tumbling like this, then it doesn't look as you expect it to. So they had these ten objects, all of which they expected to look a certain way, some of which did not look as they expected and led to a situation where they couldn't tell the warhead from these other objects with reliability. So what they did is they changed the template. They took the data and this template where they amass these things, they took the data, changed all the positions in the template and then they claimed that they could match. Well that's fraud. This is like um, this is like me having a computer programme that I claim will predict the price of stocks at some future time and I tell you that um you give me the price of different stocks on the stock market at this time and I'll put in some other parameters and it will predict the price of the stocks as time goes on and you'll know what stock to buy. And it turns out that the programme does one thing and the stock prices do another but at some small interval they happen to overlap, so what I do is cut out the prediction in in these other areas. I don't tell you that they didn't match in these other areas and I say, here, see I got a match. So it's fraud.

So they dumbed-down the test to begin with, they manipulated the data and they covered up.

Yeah, in fact they removed data from the experiment in addition to what I was describing to you. They only did this game of changing their template during the period in which they had data for the experiment, about one fourth of the data was used, three fourths of the data was censored, not talked about and basically the fact that it existed as far as I could tell, was hidden, so we still don't know everything about what was going on with the data - the general accounting office and agency of the US Congress is in the process of trying to find out.

The Duchess of Zeon wrote:He's incorrect; they didn't change the parameters of the test. There are ways to discriminate between the objects in question, and so the tests which were initiated weren't cheating, but rather a feasable estimate of our capabilities.

The Duchess of Zeon wrote: The missile defence is not supposed to be perfect, especially during testing WHEN YOU CORRECT FLAWS IN THINGS. Any engineer can tell you that about a system. The Physicist ought stick to theory where he belongs.

SWPIGWANG wrote:

The Duchess of Zeon wrote: The missile defence is not supposed to be perfect, especially during testing WHEN YOU CORRECT FLAWS IN THINGS. Any engineer can tell you that about a system. The Physicist ought stick to theory where he belongs.

Therefore you can lie about having flaws?

Vympel wrote:
How do you get that from 1/4 of the data being used and the other 3/4 being tossed out?

SWPIGWANG wrote: Therefore you can lie about having flaws?

The Duchess of Zeon wrote:
I'm saying that didn't happen.

The Duchess of Zeon wrote:
Of course it has flaws; they're just within the acceptable margins for deployment. Nothing will work perfectly. All it has to do, though, is work well enough to force the enemy to deploy decoys, and then still shoot down a few after that, and we've made a tremendous gain. And it will do that.

Vympel wrote:

The Duchess of Zeon wrote:
I'm saying that didn't happen.

And you know this ... how?

The Duchess of Zeon wrote:

Vympel wrote:

The Duchess of Zeon wrote:I'm saying that didn't happen.

And you know this ... how?

I've been told it. You probably wouldn't consider the explaination reliable so I'm not going to try and bother convincing you, nor to appeal for you to believe simply on my word alone: Continuing believe this scientist and the other sources as you desire, and under strict debating rules you would correct to do so. They are, however, incorrect.

Ted wrote:
Hearsay isn't acceptable in a debate, especially if you are trying to disprove stated facts of someone INVOLVED in the testing.

Darth Wong wrote:Not to jump in here, since I haven't researched this particular incident, but on the subject of engineering tests, it is completely possible for an engineering test to be conducted in a scrupulous manner, with full reporting of all mitigating variables and special circumstances, but then to have management and/or publicists report its results in a HIGHLY misleading manner.

Contrary to whatever DOZ said, this is actually very common practice, and happens in all manner of industries. Part of the problem is ignorant recipients; they do not read between the lines of what is given to them. If you perform a test with 6 important caveats and report them as such to your superiors, those superiors can then go and report a selected portion of your results without bothering to explicitly mention the caveats. Or they might downplay them. The result is that decisions are made based on distorted information.

So why did scientists at the BMDO go along with such a disguise?

Well, I'm not aware of any scientists at the BMDO myself. I mean I never met anybody at the BMDO that I would call a scientist. I know scientists who are involved in Missile Defence activities - some of them are even my friends, although I don't admit to it in public, but nobody at the main office appears to know anything about either science or engineering. It seems to be a big public relations activity.

MKSheppard wrote:In real life, Mr North Korean ICBM would be targetted by not one but FOUR or more missiles.....
One is bound to get the kill....
NMD missiles ain't cheap, but compared to the costs of cleaning up after
a WMD attack.....

Ted wrote: Yeah, but MRV's, you don't know which is real, which is decoy.

MKSheppard wrote:Like I said, shoot them all down with multiple missiles

My solution to missile defense is simple:

The US has ~100 interceptors on hot standbuy at any time, and
assuming 4-1 ratio per missile...you can shoot down 25 missiles with
that, and that's pretty much it for China's ICBM force, North Korea's
force, etc

It's a lot more rational than expecting 1 missile = 1 kill

Ted wrote:

MKSheppard wrote:In real life, Mr North Korean ICBM would be targetted by not one but FOUR or more missiles.....
One is bound to get the kill....
NMD missiles ain't cheap, but compared to the costs of cleaning up after
a WMD attack.....

Yeah, but MRV's, you don't know which is real, which is decoy.

Admiral Piett wrote:

MKSheppard wrote:Like I said, shoot them all down with multiple missiles

My solution to missile defense is simple:

The US has ~100 interceptors on hot standbuy at any time, and
assuming 4-1 ratio per missile...you can shoot down 25 missiles with
that, and that's pretty much it for China's ICBM force, North Korea's
force, etc

It's a lot more rational than expecting 1 missile = 1 kill

With North Korea,Pakistan & Co,yes a brute force approach may just be enough to do the job.China is an other story.Their current ICBM force suck,but do not worry,they are upgrading it as the development of the CSS-X-9 has clearly demonstrated.Beyond certain numbers it is unavoidable that with the help of decoys at least few warheads will get throught.
[chinese communist party boss mode on]Surely the USA are not going to risk a couple of their cities in order to mess with the chinese political goals in Asia.[chinese communist party boss mode off]