• Colonel Jim Burton, another Air Force colonel, forced into retirement in 1986 after he took a tough stance on tests of the Army’s Bradley Fighting Vehicle; and
• Tom Amilie, former technical director of the Navy’s Special Weapons Laboratory at China Lake, California, who joined Fitzgerald as a cost fighter.6
Most Pentagon dissidents, however, prefer anonymity. These people find a Denny Smith, or they pass their material to intermediaries such as Dina Rasor, a thirty-one-year-old former news assistant for ABC News who now runs the Project on Military Procurement. The project became a major channel for the dissident triangle, gathering and disseminating inside information—usually documents—on Pentagon weapons. It is passed along by as many as one hundred Pentagon sources, Rasor said to me, “from the airman on the flight line who sees a spare part he thinks is too much, clear up to people who are working very closely with the secretary of Defense.”7
Denny Smith’s feuds with the army and navy over multibillion-dollar weapons systems are case studies of how the dissident triangle works. They offer insight into the inner politics of the Pentagon, the bunker mentality at the top, the cover-ups of weapons failures, the stubborn inertia, and the ingenious game of the middle echelons to expose, even undercut, their own top brass. These examples are all the more striking because Denny Smith was no Democrat out to score partisan points against a Republican-run Pentagon; nor was he a liberal ideologically opposed to big military budgets. He was largely reacting as a citizen-politician, innocent about the ways of Washington, at first upset and later angered.
His first tangle came with the Navy: In June 1983, he became suspicious of Pentagon claims that the new super-high-tech, guided-missile Aegis cruiser, the Ticonderoga, had hit thirteen out of thirteen target planes in a simulated test attack.
From combat experience, Denny Smith was convinced that no defender could have such perfect results in realistic tests. He asked the Navy to see the test report. He also asked his own mole network about the tests. Denny Smith was embarking on a familiar path, for Congress usually tackles the Pentagon’s weapons policy with two major questions: 1. Does the weapon work? and 2. Does it cost too much? The deeper questions of whether it is really needed and how it fits into an overall strategy are rarely addressed in earnest. Few members of Congress have a sure enough grasp to handle those questions. What’s more, questions like that could open a Pandora’s box, a free-for-all debate on national strategy that most military commanders, congressmen, and policymakers want to avoid.
The Pentagon moles provided shocking confirmation of the Oregon congressman’s suspicions about the cruiser’s tests. Later, a press article reported that the Ticonderoga had hit only five of twenty-one targets, and Smith indicated that was pretty accurate. The Navy stonewalled for five months on Smith’s request to see the test report. It was a typical bureaucratic reflex: Keep an iron grip on all information so that policy cannot be effectively challenged.
“The reason the Pentagon doesn’t like testing is that testing may interrupt the money flow to its programs,” an Air Force colonel explained to me. “That’s the strategy in the Pentagon: Don’t interrupt the money flow.”
Denny Smith’s bout with the Navy was a vintage example of the clashing political cultures of Congress and the bureaucracy, typical of their power games. The congressman was trying to open up the policy debate; the Navy was keeping it shut tight. If Smith had been on the Armed Services or Appropriations committees, his vote on military programs would have given him leverage with the Pentagon. But as a freshman who was not on those committees, Smith had no political clout with the Navy. He had to appeal for help from more senior congressmen.
Finally, in December 1983, the Navy sent a six-man delegation to appease and silence Smith by offering him a quick, temporary peek at the voluminous technical report on the Ticonderoga’s test results. But they had underestimated their man. Glancing through the report, Smith immediately spotted that page A-29 was missing. From his own secretly obtained copy, Smith knew that that page contained the test report.
“Where’s page A-29?” the congressman demanded.
Naval faces blanched white as naval uniforms. “Oh, isn’t it there?” a Navy captain said, simulating innocence.
“Well, I don’t find it here,” Smith insisted.
A civilian engineer with the Navy cadre offered Smith his copy. “Here it is, right here in mine,” he said, and Smith took permanent possession of the damning test report.
“Gosh, it must have been the Xerox,” one of the Navy men said. Later, Denny Smith told me he felt the vital page had been purposefully omitted “because the rest of the report is about as dull as toilet paper.”
Armed with the damning data, Smith called on the Navy to hold more tests of the Aegis cruiser. Fellow Republicans suggested he was out to “get” the military. Trent Lott, the House Republican whip, asked Smith if he knew that killing the Navy’s Aegis cruiser program could affect sixteen thousand jobs at Ingalls Shipyard in Lott’s home state of Mississippi.
“Hey, listen Trent, we’re not trying to cancel the program,” Smith replied. “What we’re trying to do is get the Navy to be honest, number one, and, number two, if there are flaws in that ship, let’s fix them.”8
The stakes were enormous because the Navy planned twenty-six Aegis guided-missile cruisers at $1.25 billion apiece, and sixty destroyers, with similar technology, costing $1 billion each.
“I decided to go after them to prove that the ship could survive,” Smith explained. “If we were going to spend $90 billion on this huge armada of radar ships to go out there and try to protect the fleet, let’s be sure they work. The MX missile program is known by everybody in the country; it’s about a $20 billion program. It’s peanuts alongside of this thing.”
Grudgingly, the Navy called the Ticonderoga home from the Mediterranean for further testing, in April 1984. This time, the Navy reported ten out of eleven hits, but the mole network passed word that the tests were too easy because there had been no low-level attackers and no saturation attacks by several planes at one time. Once again, Smith asked the Navy for the test report but never got it. The moles shifted him to another target.
Divad: The Gun with Nine Lives
In the Aegis cruiser episode, Denny Smith had been a green congressman who did not know how to gain political leverage through the press and allies in Congress. But by the time he went after the Army’s Divad antiaircraft gun, Smith had political allies. He had become one of four cochairmen of the Military Reform Caucus, a bipartisan group of more than fifty senators and House members, ranging from Senator Gary Hart on the Democratic left to Representative Newt Gingrich on the Republican right. This group was pressing questions about military strategy, not to oppose defense but to make it more efficient. Linked to the caucus, Denny Smith’s voice had more weight.
Divad, moreover, was a more vulnerable target. By mid-1984, it was deep in trouble, plagued by technical snafus, facing some high-level opposition within the Pentagon, and wounded by news leaks of rigged tests and embarrassing failures. Still, the Army top brass clung to it, and Weinberger sided with the Army.
Divad (short for division air defense) had been conceived in the mid-1970s to provide antiaircraft protection for Army tank divisions against Soviet fighters and helicopters. By most estimates, more modern air protection was needed. But Denny Smith, as an old fighter pilot, thought an expensive high-tech gun such as Divad was unnecessary and ill conceived. Flyers, he told me, have greater fear of traditional antiaircraft batteries, which are harder to evade.
With all its gear and ammunition, Divad cost upward of $6.3 million per gun, more than three times the cost of the M-1 tank it was supposed to protect. The Army ultimately intended spending $4.5 billion for 618 Divads. To speed up Divad’s development, the Army combined several proven components: the chassis of an M-48 tank, two Swedish forty-millimeter cannons, radar adapted from the F-16 jet fighter, plus a one-million-dollar computer and other
fancy electronics. But the real speed-up, and one major cause of Divad’s problems, was the Army’s policy of building and producing Divad while it was being tested, rather than testing it first.
Some strange decisions were made along the way. In a shoot-off competition between Ford Aerospace and General Dynamics in November 1980, Ford Aerospace scored worse but got the contract. General Dynamics hit nineteen targets and Ford only nine. The Army later said that Ford had a lot of near-misses which were counted. High Pentagon civilians on Weinberger’s staff such as David Chu, director of Program Analysis and Evaluation, and Lawrence Korb, assistant secretary of Defense for Manpower, Installations and Logistics, opposed Divad. They warned that future Soviet helicopters would be able to stand outside Divad’s best theoretical range of four thousand meters and fire at American tanks. Chu’s staff also pointed out that Divad’s reaction time was too slow, and its odds of killing Soviet planes only one half to one third of what the Army claimed.9 Nonetheless, Frank Carlucci, who was Weinberger’s deputy at the time, signed a $1.5 billion contract in May 1982 to buy 276 Divads.
Any new weapon has kinks, but Divad’s were comic omens: In one check-out test in February 1982, top American and British Army brass went to Fort Bliss, Texas, to see Divad perform. Suddenly, Divad’s turret swerved away from a target drone back toward the reviewing stand. The brass all ducked for cover. The gun did not fire at them, but it spent the rest of the day missing targets and lobbing shells into the weeds. Then in January 1984, the first full-fledged production model that Ford was proudly preparing to turn over to the Army made an embarrassing test debut: The radar-guided, computer-operated fire-control system focused on a false target—a rotating latrine fan in a nearby building—which the computer singled out as the closest threatening target.
This produced guffaws within the Army. One hand-drawn Army cartoon showed two GIs, one pointing to the sky and saying, “The Soviets have come up with a new way to foil the Divad.” It pictured a Soviet helicopter towing an airborne outhouse to distract Divad.
The incident showed that Divad’s radar was still having great difficulty distinguishing the right targets from “ground clutter” (other objects on the terrain). Senator Warren Rudman of New Hampshire, another Divad critic, said this problem highlighted a defect in the gun’s basic design. The designers had used a radar system built for jet fighters and for operating against the clutter-free background of the sky, not on the ground. From the outset, Rudman told me, the Army set unrealistic requirements for Divad, dooming it to failure. Nonetheless, the Army stubbornly pressed on, partly out of need, partly out of pride, mostly out of bureaucratic momentum.
“The Divad is a classic example of how the military system keeps alive a weapons program that doesn’t make any sense,” Denny Smith remarked. “Once the system buys onto the program, there’s almost no way you can stop the program. If you try to, you’re either unpatriotic, you don’t understand the situation, or you’re out for publicity. They try to go after you. You can almost tell when they have a bad system because they get so defensive and come after you.”10
Even so, Congress was growing wary of Divad. More awkward disclosures got into the news. Divad flunked cold-weather tests in early 1984. It had to be heated for six hours with the field equivalent of a hair dryer before it was ready to fire. In another test, the Army had to attach four large metal reflectors to an old target helicopter to help Divad’s radar find the target. By late 1984, Ford Aerospace was months behind its production schedule, and Congress had barred further purchase until Divad passed realistic operational field tests. Congressional pressures forced Weinberger to take a personal interest.
In the spring of 1985, the Army ran a massive monthlong mock battle in the California desert with tanks, Bradley fighting vehicles, and Divads opposing A-10 and F-4 fighter planes and AH-64 Apache helicopters. The finale was the “live fire” tests at White Sands, New Mexico, in May 1985.
Afterward, the Army brass jubilantly proclaimed that Divad had hit and destroyed its targets. Jack Krings, civilian head of the Pentagon’s new Office of Operational Tests and Evaluation, telephoned Denny Smith. “Boy, really impressive,” Krings said. “Blew those mothers right out of the air.” Army Secretary John Marsh, Undersecretary James Ambrose, and General John A Wickam, Jr., the Army chief of staff, all recommended that Weinberger move ahead with Divad. Ambrose, a former Ford Aerospace vice president who had helped launch the Divad program while still at Ford, told me he felt Divad was a big leap forward, a ten- to twenty-percent improvement on existing antiaircraft weapons.11
But the Dissident Triangle had a very different story: It informed Congressman Smith that the mock battle showed Divad’s range was inadequate, and the live fire tests were unrealistic; the Army’s claims of success were misleading. What Denny Smith learned, he told me later, was that the target fighter planes were patsies. They were flown right past the Divad guns “at a suicide elevation of four hundred to five hundred feet, flying straight and level at 420 knots with no jinking [pilot talk for no evasive maneuvers]. The helicopters were flown up to a higher elevation than any sane person would ever do in a combat zone. What they set up was a shooting gallery and, even then, there were no direct hits—none!”12
If so, I asked, how could the Army be claiming success?
Moles at the test site had tipped off Smith to shenanigans on the firing range. For proof, he went after videotapes. The Army happily supplied tapes showing Divad firing and target drones exploding. “You could see that they had been destroyed almost immediately, and you thought maybe the guns had done that,” Smith told me later. “The picture would be on the airplane. It would show maybe a couple of sparks. And then almost immediately, they’d blow up, looking like they’d been hit. But we knew better. We’d been told. The range-safety officer destroyed every one of the drones from the ground. None of them were destroyed by hits from the guns.”
Others were less categorical than Smith. Two Pentagon skeptics told me that gunbursts showed a few Divad kills but asserted that on the large majority, the range-safety officer had been unusually quick to detonate safety charges on the target planes. Safety measures are routine, but Denny Smith and Lieutenant Colonel Tom Carter, a top Pentagon test analyst and a Vietnam veteran with 408 air missions, told me the safety officer used a fast trigger to make it look as though Divad had scored hits.
“We felt they were certainly flawed tests if they destroyed the drones that quickly,” Smith told me. “Why didn’t they let them go on for twenty more seconds?” Smith fired off protest letters to top Pentagon officials. The Army brass fought back, defending its weapon.
Smith’s blast that Divad had not made “a single direct hit” touched off a firestorm in the media. The Early Bird gave hot running coverage to the charges of the maverick network for Weinberger’s ride-to-work reading. The Schattschneider dynamic was at work: Television networks and news weeklies became seized with Divad. The videotapes of the live fire tests, and Smith’s charges about how the targets were destroyed, gave the whiff of scandal and rigged tests to Divad.
Inside the Pentagon, the final test evaluations were being drafted for Weinberger in mid-August. One of them, done in the Office of Developmental Testing and Evaluation by Colonel Tom Carter, was a blistering and fatal indictment of Divad. “My worst suspicions were confirmed,” Carter later told me. “The Divad gun couldn’t detect and track and engage and shoot down enemy aircraft, unless the enemy’s aircraft were using unrealistic tactics which no pilot—Russian or American—in his right mind will fly. The weapon failed miserably to perform.”13
What happened to Carter’s official report was an amusing wrinkle of the Dissident Triangle operations—not leaking, but flooding. My sources told me that the original draft of the second report, prepared by Jack Krings, director of the Office of Operational (as opposed to “Developmental”) Testing and Analysis, was nowhere near as harsh as Carter’s. On August 22, nine copies of Carter’s no-nonsense report were circulated
to top Pentagon officials. The next day, the top Pentagon echelon tried to squelch it.14 Orders were given to retrieve every copy, but it was too late. Instead of nine copies, thirteen copies came back.
“That’s what we call the flood strategy,” one Pentagon gadfly told me with a grin. “Never leak anything yourself, but make plenty of copies. Flood the building. God will take care of the rest. As soon as Krings’s people saw those thirteen copies, they knew they had a P.R. disaster on their hands, because the test results had gotten out of the building.”
It was Friday afternoon, and Weinberger had already headed for a weekend in Maine. Krings’s office spent the weekend redrafting its report to toughen it, more in line with Carter’s.
The “flooder” was right. On August 22, Denny Smith wrote Weinberger a letter to say that he had “obtained and reviewed” the test reports on the Divad which “verify the same criticisms of the weapons flaws leveled over and over again since the inception of the program.” He urged Weinberger to cancel the program, and on Monday, from his home in Oregon, Smith telephoned Weinberger to underscore the fact that he had the damaging report in his possession. “I hope you’ve seen that report, Mr. Secretary, and I just urge you to read that before you make your decision,” he said with an implicit threat to go public if Weinberger did not act on the negative report.
The next day, Weinberger announced that he was canceling Divad because “operational tests have demonstrated that the system’s performance does not effectively meet the growing military threat.” What Divad would offer over existing weapons, he said, was “not worth the additional cost.” He identified its main problems as “the lack of range and the lack of reliability.… The system didn’t work well enough.”
At that point, the Pentagon had sunk $1.8 billion into the program. Weinberger’s decision to kill Divad marked a rare victory for the dissident triangle—one case in hundreds. I have heard many tales of other weapons systems having serious flaws, but they roll on. A few get stopped in the research-and-testing phase. Senator Warren Rudman, a combat infantry captain in the Korean War, fought three years to block production funds for the Viper, a defective antitank weapon with skyrocketing costs. He finally won before production was started.
Power Game Page 24