MIT team tied to questionable missile studies
by David Abel, Boston Globe, 4 Mar 2002
Pentagon agency, two major military contractors, and an independent research team led by MIT scientists produced flawed studies that exaggerated the success of a key test used to justify spending billions of dollars on the fledgling national missile defense program, according to two reports obtained by the Globe.
The long-awaited reports, to be released today by the General Accounting Office, detail the flawed analysis of critical missile-defense technologies provided by the contractors, Boeing Co. and TRW, verified by senior researchers at the Massachusetts Institute of Technology's Lincoln Laboratory, and hailed by the Pentagon's recently renamed Missile Defense Agency.
In reports about a highly sophisticated sensor used in the first test of the missile-defense program - a technology similar to one now designed for the vital task of distinguishing decoys from warheads - contractors described its performance as "excellent" and the overall test as a "success." The team directed by two MIT scientists, which evaluated the contractors' reports of the test, pronounced them "basically sound." And officials in the Missile Defense Agency called the first test of the technology in space "highly successful."
Yet the review by the GAO, the investigative arm of Congress, found that crucial elements of the 1997 test failed - prompting investigators to raise questions about the oversight of a program that has already cost billions of dollars and could, if the Bush administration has its way, ultimately cost taxpayers as much as $238 billion, according to a recent estimate by the Congressional Budget Office.
"The data are garbage - they had to use all these software shenanigans and throw out two-thirds of the data to make it look like a success," said a congressional source close to the GAO investigation. "Up to now, there has been no independent verification of the contractors' claims. This pulls out the rug from those calling the test a success. By any definition, there's no way to call it a success."
The main defect in the test, according to the GAO, was that the infrared sensor built by Boeing failed to cool to a sufficient temperature to function properly. Also, the power supply of the sensor turned out to be much louder than expected. The excess heat and noise, missile specialists said, caused a significant distortion, by a factor of up to 200 times, in the ability of the sensor to detect targets. As a result, the sensor often detected targets where none existed.
The performance of the sensor is crucial because the planned land-based national missile defense system might have only one chance to hit its target. And once the military launches an antimissile against an incoming ballistic missile, military analysts say they believe it would almost certainly face a barrage of decoys. Moving at great speeds, it would have to distinguish the fake from the real in a matter of minutes.
Regarding what became known in defense circles as the "MIT study," a review of the contractors' findings that the Pentagon used to champion missile defense spending, the GAO faulted the team led by scientists at Lincoln Lab for relying on data processed by TRW - instead of seeking the contractor's raw data.
Although the team reported that TRW's sensor contained a few software glitches, GAO investigators said the scientists' use of the processed data allowed them to review only 14 of 54 seconds worth of data. The limited look at the sensor's performance, according to the GAO, skewed the scientists' review and led them to pronounce the sensor's software well designed and say it worked properly.
The failure to review the raw data, investigators wrote in the report, means "the team cannot be said to have definitively proved or disproved TRW's claim that its software successfully discriminated the mock warhead from the decoys."
For MIT physicist Theodore Postol, a frequent critic of the Pentagon's missile defense plans, the omissions of his colleagues and their stamp of approval for the Missile Defense Agency amounts to scientific fraud. Postol recently lodged complaints with the MIT Corporation about the study - charging that the university's president, Charles M. Vest, knew of the alleged misconduct and did nothing about it.
"This certainly has the appearance of a well-orchestrated fraud," Postol said. "The managers at Lincoln Lab either knew or should have known that this experiment was a total failure - and they falsely represented it as a success. The implications of that deceit could cost taxpayers hundreds of billions of dollars."
MIT officials did not return calls for comment. But Roger Sudbury, a spokesman for Lincoln Lab, told the Globe last month that the Lexington-based research arm of MIT received no complaints from contractors or the Pentagon about their review, and he said, "There is no evidence of fraud."
Lieutenant Colonel Rick Lehner, a spokesman for the Missile Defense Agency, which oversees the Pentagon's effort to develop an overlapping air, land, sea, and space-based missile shield, insisted that, as far as he knows, the sensor guiding Boeing's "kill vehicle" worked as planned.
Still, in the scheme of the overall missile defense plan, he said, the 1997 test is irrelevant. Not long after the test, the Pentagon decided to use a sensor built by Raytheon Corp., one with "totally different" technology than the one designed by Boeing.
"I would guess our people will take issue with this report," Lehner said. "At face value, the only thing I was told was that the Boeing kill vehicle did discriminate against the decoys and warhead. Until the agency tells me otherwise, I have to go with that."
The GAO reports, requested by Representative Edward J. Markey, a Massachusetts Democrat, and two other members of Congress, were sought nearly two years ago after Postol sent the White House a detailed analysis of the 1997 test, alleging both Boeing and TRW misrepresented the results.
The MIT professor analyzed the raw data of the test, which he obtained through Nira Schwartz, a senior staff engineer at TRW who was fired after she reported that the software her company developed would not distinguish decoys from warheads. Schwartz, who is suing TRW, and Postol insisted it's a fallacy to say the 1997 test is irrelevant.
Because both the Boeing and Raytheon sensor use "infrared eyes," "It's the equivalent of looking at a bunch of suitcases with only your eyes and trying to find a bomb inside," Postol said. "If I give you a telescope, a microscope, or dark glasses when you look at the suitcase, none will tell you which has the bomb."
Despite the allegations, the GAO studies stop short of calling the reports and exaggerated results fraud. Unlike most GAO reports, and despite congressional requests for them, they don't include recommendations.
The reason, another congressional source close to the investigation said, is political. The reports, delayed by sluggish responses from the Pentagon and contractors for documents, were vetted very closely to avoid casting too much blame on any one party, the source said.
"Much of the findings were buried inside the text and purposely written in technical language so as not to highlight many things," the source said. "There are many political pressures, and the report was certainly edited for political reasons."
With billions of dollars at stake and $100 million a pop for each antimissile test, a lot is riding on whether it is technically possible to build a national missile defense that works. Over the past five years, three out of the five antimissile tests hit their targets. But during that time, the tests have been downgraded in complexity, now using only one decoy that is much larger and brighter than the mock warhead.
For the Bush administration, which vowed to build a robust national missile defense during its campaign two years ago, fielding a viable system is one of its highest priorities. In December, President Bush announced the United States would withdraw in June from the 30-year-old ABM treaty, which bars a nationwide missile shield.
In a statement about the GAO reports, Markey, who has proposed a bill calling for independent oversight of the missile shield, cautioned that relying on questionable technology could amount to a massive waste of taxpayer dollars.
"The national missile defense program needs independent oversight and testing milestones to ensure that it works before we spend countless billions of dollars deploying it," he said. "If it can't tell the warhead apart from a decoy, what good is it?"
© 2002 Boston Globe
Reprinted for Fair Use Only.