Shots in the Dark

On a frigid day in Alaska last winter, a rocket, designed to simulate an incoming missile launched at the United States, blasted out of the ground. Fifteen minutes later, an interceptor rocket was to be deployed from a site in the Marshall Islands, knock down the deadly missile, and save America.

That was the early February 2005 flight test of the Missile Defense Agency's (MDA) Ground-based Midcourse Defense, a program at the heart of the Pentagon's vaunted national missile defense that President Bush had promised would be operational by the fall of 2004.

If this had been real life, some U.S. city would have been incinerated: The interceptor rocket never made it off the launch pad.

It was the second test failure of the system in three months. In a December test, which cost more than $80 million, the interceptor rocket had failed to launch because of what the Pentagon called “an unknown anomaly” that it insisted was “a very rare occurrence.”

Only days prior to the February launch, the Pentagon was still struggling to fix at least 20 glitches in the system. On test day, the faulty software created an abort command before the interceptor was supposed to launch. “At the rate they're going … it could take them 50 years to do the kind of developmental testing” necessary for the system to work, said Philip E. Coyle, a missile-defense expert at the Center for Defense Information who performed much of the oversight on the program under President Clinton.

Pentagon officials, however, didn't seem overly worried about the flubbed test. As The New York Times reported, in the wake of the failure MDA spokesman Richard A. Lehner was looking on the bright side. Lehner announced that the agency had learned “quite a bit” from the aborted test and called it “a very good training exercise.”

Though the Pentagon won't admit it, missile defense, one of the bedrocks of the Republican national-security doctrine, is virtually useless today. The MDA has lavished money on a system of interceptors that is nowhere near operational and might never work -- a rushed effort that diverts resources from other vital priorities, like the fighting in Iraq and Afghanistan, while making it harder to develop defense against longer-term missile threats. And it's not just outside analysts, dismissed by the hawks as doomsayers, who realize that missile defense is a train wreck. Throughout the history of the program, the government's own internal reports and analyses, buried out of public view, have shown it could be an abject failure.

The trouble is that the administration is not looking to abandon the program. On the contrary, missile-defense officials have said that testing, halted after the February failure, will resume, and the coming year likely will be crucial to the future of missile defense, which Pentagon officials constantly claim is on the verge of being operational. Secretary of Defense Donald Rumsfeld has been close to declaring the missile-defense system active or deployed. If declared active, missile defense would in theory be responsible for protecting Americans, something the program simply can't do. Declaring it active would ensure there's no turning back -- and that billions of dollars continue being plowed into the program, already the most expensive military research-and-development program under the Bush administration. Ultimately, missile defense could be one of the costliest defense boondoggles in history.

* * *

The current missile-defense program is the offspring of President Reagan's Strategic Defense Initiative, popularly known as “Star Wars,” an anti-missile shield against Soviet ICBMs. It is a program beloved by hawkish Republicans, who for two decades have seen it as an ideal way to protect America without having to rely on allies or treaties. Long-term boosters of the program include Rumsfeld, who headed a 1998 commission overplaying the missile threat, and then–CIA Director George Tenet. In 1999, the GOP–led Congress formally called on President Clinton to deploy a missile-defense system, and Clinton reluctantly complied.

Still, shortly before Clinton left office, a report came out from the Office of the Secretary of Defense, Operational Test and Evaluation Directorate (DOT&E), the part of the Pentagon that performs independent evaluations of weapons systems before they are allowed to keep moving forward. The report, recalls then-DOT&E head Coyle, “said the system was not ready to be deployed.” In the report, Coyle and his lieutenants declared that the system “has not achieved two intercepts nor demonstrated integrated system performance,” used radar whose “performance … was generally poor,” and would not be ready for deployment by 2005, a proposed target date. When the Pentagon had run tests of the program, they were heavily scripted and did not simulate a real threat; defense officials made sure that global positioning devices were placed on the rockets to be intercepted, which helped ensure that they were shot down successfully.

That the new administration ignored these warnings was perhaps not unexpected. What is shocking is just how vigorously and unwaveringly it has stayed the course with missile defense, even in a time of escalating deficits and even as the evidence of the program's abject failure has piled up. Once in office under President Bush, Rumsfeld pounced on the program, while the White House began considering how to withdraw from the 1972 Anti-Ballistic Missile Treaty, a Cold War–era pillar of arms control that outlawed testing ballistic-missile-defense systems. Rumsfeld suppressed the Coyle report; for eight months the Pentagon refused to turn it over to Congress, which wanted to make it public. Reagan's “Star Wars” initiative had crumbled in part because its failings received so much public attention, so, one official remembers, Rumsfeld began developing ways to reduce oversight of the program by preventing the DOT&E from performing independent analysis.

In the Senate, Democrats, still in the majority, planned to block Rumsfeld's gambit. “We were going to fight Rumsfeld taking away the oversight,” says one former congressional staffer. “But [September 11] happened, and [Democrats] did not want a floor fight on missile defense.”

After 9-11, Rumsfeld saw his opening. While Clinton had imagined a small number of interceptors based on land, Rumsfeld's ambitious system would mix interceptors launched from land, sea, and potentially even space, an enormous task that would have drawn criticism inside the Pentagon. But in a memo penned in January 2002, Rumsfeld announced that the Missile Defense Agency would have near-total control over its programs, without having to set exact goals, timetables, or budgets for as long as missile defense was in its “experimental” phase. The MDA would not have to provide Congress and the public with the typical reports required for weapons programs in development, and the DOT&E would not necessarily have access to key missile-defense documents.

Most important, in contrast to the Pentagon's normal “fly before you buy” policy, the MDA now would not have to demonstrate that missile defense actually works before continuing funding for the costly program. For the 2003 fiscal year, the Pentagon requested a total of $7.8 billion for missile defense. Yet beyond offering broad plans for using the funding, the MDA neither had to specify how the money would be utilized nor give an estimate of how much missile defense, once completely in place, might ultimately cost. (One group of esteemed economists, adding up the Pentagon's own estimates, predicted that the system might ultimately cost more than $1 trillion to develop and operate.) For the missile-defense program, the Pentagon could sign deals with defense companies without complying with normal laws designed to ensure financial openness in defense contracting.

Meanwhile, the Missile Defense Agency began classifying more information as secret, making it harder for outside watchdogs to tell what was going on. Before the Bush administration, “we knew a good deal about planned tests,” says Stephen Young, a missile-defense expert at the Union of Concerned Scientists, a scientific nonprofit. “Now we know basically just which quarter of the year a test might occur … this may be the most difficult thing the U.S. government has ever deployed.”

Even within the government, other agencies and groups couldn't get information out of the MDA. “I tried to make a foray [to the agency] to get a briefing … they wouldn't even give us a classified briefing,” says one former intelligence official with security clearance. Instead, the intelligence officer got the same kind of unclassified information available to the public. “The Department of Defense normally has pretty good oversight of key programs,” says one former congressional staffer. “But MDA said, ‘We're going to set up our own regime.'”

Missile-defense experts within the administration itself predicted the lack of accountability would be a mistake, that it would allow the Pentagon to conduct Potemkin tests of the program and then spend on it without limit. One internal Government Accountability Office (GAO) report bluntly warned that “[s]ystem effectiveness will be largely unproved when the initial capability goes on alert,” and that the Pentagon had no idea of the program's cost. Another internal report in 2003, done by the Office of Test and Evaluation, said that “very little system testing was performed,” and that the system might not be able to hit a missile that it didn't know about in advance. In other words, the Pentagon had no idea whether missile defense worked, whether it would ever work, how much it would cost, or what it would even look like down the line.

Still, members of Congress, who had access to some of this internal reporting, said little because the Republicans now controlled both the Senate and the House, and because leading Democrat Evan Bayh, perhaps seeking to boost his national-security bona fides, repeatedly backed missile defense. (In the most recent election cycle, Bayh also was one of the top five Senate Democratic recipients of contributions from the defense industry, according to the Center for Responsive Politics.) According to one congressional source, when missile-defense critics in the House tried to offer amendments imposing some restrictions on the program, the House gop leadership, like Armed Services Committee Chairman Duncan Hunter, the California Republican, prevented the legislation from moving in committee.

Strengthened by Congress' silence, Bush, Rumsfeld, and their allies stepped up the pace. In December 2003, the president promised that the first part of the missile-defense system would be ready for deployment by September 30, 2004. Less than a week earlier, the nascent system had failed an intercept test. Echoing the commander in chief, Lieutenant General Ronald Kadish, then director of the MDA, told a defense writers breakfast, “We believe by September [2004] we should have the hardware and software and communications capability” to be operational. Another top defense official told Congress he was sure that by fall 2004 missile defense would be “90 percent” effective against an incoming threat. Bush was even more boastful, reportedly telling employees at Boeing, a major missile-defense contractor, “We say to those tyrants who believe they can blackmail America and the free world, ‘You fire, we're going to shoot it down.'”

The leadership, however, must have known its system still had no clothes. As Coyle argues, the MDA made realistic testing of the system even less of a priority, both because of Bush's announcement and because it had to focus on quickly building enough ground-based interceptors in Alaska, planned as the first part of the system, for the program to appear operational by September 2004 -- that is, in time for national elections. “The president's decision … has lowered the bar on the acceptable standards,” Coyle notes. Indeed, by fall 2004, the MDA was furiously trying to construct an enormous, and enormously complex, number of components: an air-based laser, fixed ground-based interceptors, a space-based laser, sea-based missiles, and a complicated radar-and-communications infrastructure designed to integrate all parts of the program.

Yet the MDA was able to keep most criticism quiet. The agency postponed several scheduled tests, and the few tests that went forward remained heavily scripted. In them, the agency continued to propel the incoming missiles slower than an actual attacking missile would fly, placed a radar device on the incoming missile to make it easier to shoot down, and gave the interceptor crucial information about the incoming missile, like its trajectory. Even this apparently was not enough: The Pentagon also announced that it would make even more information classified after 2004. When a scientist at the Massachusetts Institute of Technology (MIT) charged that claims of unqualified success were due to fraudulent claims by defense contractors, and the school launched an inquiry into his assertions, the Pentagon classified the information MIT investigators said they needed to complete their task.

“The easiest way to ensure accountability would be to publish the results of your tests,” but there was no desire to do so, says David Mosher, who follows missile defense at the RAND Corporation. Meanwhile, critical GAO reports got little public attention, and one of the office's own top missile-defense experts, Bob Levin, says the Department of Defense initially just replied “nonconcur” to GAO recommendations for making testing more realistic.

To create a facade of accountability, the Defense Department touted a new organization it had set up, the Missile Defense Support Group, as providing advice to the program. Yet members of the group, comprising mid-ranking officials from within the Pentagon and other parts of the armed forces, told The Washington Post that they were not a “critical decision group” and that they learned of some major MDA decisions only after they were made. When any criticism did reach the public, Rumsfeld quickly shot it down. “Did we have perfection with our first airplane, our first rifle, our first ship?” Rumsfeld told the Post. “I mean, they'd still be testing at Kitty Hawk, for God's sake, if you wanted perfection.”

By the fall of 2004, the Bush administration faced a dilemma: Internal reporting showed that missile defense was far from operational, and the program was eating up resources -- the GAO would report that in fiscal year 2004, missile defense ran approximately $370 million over budget -- at a time when funding was needed for the Iraq conflict and homeland defense. “All the other programs had taken cuts to feed missile defense,” says Mosher.

Still, Bush had promised that the system would be up and running, and the president, as everyone knows, doesn't like to back down. So, the White House fudged. The administration announced that it would not officially activate the first part of the system, the ground-based interceptors in Alaska, until fall 2004, but said that, if needed, the Alaska interceptors could be relied on immediately. During a House Armed Services Committee hearing, Lieutenant General Henry A. “Trey” Obering II, head of the Missile Defense Agency, proudly told representatives, “For the first time in its history, the United States today has a limited capability to defend our people against long-range ballistic-missile attack.”

Internally it was a different story. In a Pentagon report, Coyle's successor at the DOT&E, Thomas Christie, reported that the missile-defense system might be only 20-percent effective. A GAO report this spring again warned that missile defense had not demonstrated that it could operate as an integrated system. A report by the Congressional Research Service added, “There is no unambiguous, empirical evidence to support the contention that kinetic kill for [ballistic missile] defense will work.”

* * *

After the February test failure, the pentagon convened an independent review panel to analyze the program and review testing procedures. The panel produced a report for internal use, a copy of which the Prospect obtained from an administration official. Though the panel contained a former leader of the national missile-defense program and a longtime executive in the defense industry, both of whom might be sympathetic to Pentagon programs, the report repeated the same critiques that had been leveled internally at missile defense for a decade. The Missile Defense Agency, it said, had taken dangerous shortcuts in an attempt to get the system running, launching tests without even being sure that all parts were working. The panel predicted future test failures and warned that the current mind-set at the MDA “appears to be ‘Prove why we should not fly,'” not “‘Prove why [we] should fly.'”

Some of this criticism may finally be leaking out of the Pentagon. At the same House hearing where Obering announced confidence in the program, Christie's successor at the DOT&E, David Duma, said, “I don't think you can say the system is operationally ready.” And though congressional Republicans, and some congressional Democrats, have traditionally supported the program, the system's spiraling cost and lack of accountability, especially during a time of war, have begun to alienate players on the Hill. Perhaps because of pressure from Congress, the MDA slashed a billion dollars off its 2006 funding request. The agency also has appointed Kathleen Paige, a respected officer who has overseen the U.S. Navy's more successful -- if limited -- efforts to create interceptors against short-range missiles.

But a program that has been the basis of the Republican national-security strategy for decades, and that is poised for activation, isn't easily halted. Even Obering admitted that he “could not say with confidence” when there would be realistic testing. The Bush administration has requested more than $7 billion for missile defense for fiscal year 2006.

Meanwhile, Rumsfeld still appears to be moving toward more formally activating the system, which would require stepping up staffing at interceptor sites (the Pentagon has said it will need nearly $60 billion for missile defense for the next five years). This spring, the head of U.S. Strategic Command, Marine Corps General James Cartwright, previously known as a skeptic of missile defense, seems to have signed on to putting the system into action, telling a defense trade publication that missile defense is worth the investment, even if it's only modestly effective. Should the Pentagon activate the system, it will be active despite the fact that the MDA has yet to demonstrate even once that the system can shoot down a real incoming missile.

“The only way to fix this is to clean house at MDA, fire everyone,” says one former Pentagon official. “Anyone really aiming for quality in a product has left MDA after so many years of banging their heads on the wall.”

Not likely. Even today, when forthright Pentagon officials like Duma try to get the message out, missile-defense proponents can silence them. Consider the April 2005 hearing of the Senate Armed Service Committee, almost a parody of Congress' supposed oversight function. The chairman, Republican Senator Jeff Sessions, one of the truest believers in missile defense, started the supposedly impartial hearing by announcing, “I feel confident that the last two aborted test launches do not represent setbacks for this important program.” Later, the DOT&E's Duma appeared before the committee. Quizzed by Democratic Senator Bill Nelson, Duma admitted that the Pentagon lacks evidence that missile defense's many parts works together, and Obering again refused to say whether the MDA would conduct realistic testing later this year. But Sessions wouldn't let any gloom into the room. After Duma's comments, the senator quickly added, “Those portions of the system weren't proven to be failures; they just didn't have the opportunity to be successful.”

As one panel member responded, “That's quite a spin.”

Joshua Kurlantzick is The New Republic's special correspondent.