High-Tech Horror

By GEORGE J. CHURCH

At first, it seemed like a triumph of high technology. Supersophisticated radar aboard the U.S.S. Vincennes picked up the airplane almost as soon as it took off from the Iranian airport of Bandar Abbas, on the shores of the Persian Gulf. Within moments the radar received enough information about altitude, speed and flight path for Captain Will Rogers III to reach a conclusion: the plane was a hostile fighter flying an attack pattern. An IFF (Identification, Friend or Foe) signal bounced back by the approaching aircraft seemed to confirm that conclusion. Two missiles launched by the Vincennes were electronically guided precisely to the target. A mere seven minutes after the plane had been detected, it was blown to bits before coming close enough to do any damage -- or even be seen by the unaided eye. The Aegis system, the most sophisticated battle-managing array of radars, sensors, computers and automatically guided weapons ever put together, had worked under combat conditions exactly as it was supposed to. Except . . .

Except, of course, that the plane identified by the Vincennes as a 62-ft.- long F-14 Tomcat fighter turned out to be a 177-ft.-long Iran Air Airbus carrying 290 civilians on a regularly scheduled flight to Dubai on the other side of the gulf. As a horrified world last week watched the pictures of torn bodies displayed by Iran on TV screens, questions mounted. Outside Iran, hardly anyone seemed to doubt that the shootdown had been a genuine mistake. But how could so sophisticated and costly ($600 million a copy ) an intelligence-and-weapons system, and the highly trained men who operate it, have gone so terribly wrong?

A reasonably complete answer will have to wait several weeks until experts finish analyzing tapes from the Vincennes and other U.S. Navy vessels in the gulf. And some questions about the affair may never be resolved. Why, for example, did the Airbus pilot not answer the warnings issued in the last minutes before the shootdown? But enough has become known in the week since the tragedy to suggest a terrible conclusion, one with dismaying implications for a nuclear-armed world: the U.S., and by extension other countries using high-tech weapons, may have become prisoners of a technology so speedy and complex that it forces the fallible humans who run it into snap decisions that can turn into disaster.

To be sure, that is not the only consideration: simple inattention on both sides also figures into the tragedy. On the American side, the military claims that it does not systematically monitor civilian air traffic over the gulf. In fact, a Pentagon official told TIME that the Navy had not even provided the Vincennes with a schedule of Iran Air flights. Captain Rogers did ask a crew member to look into whatever material on civilian flights he had aboard. But none of it mentioned Iran Air Flight 655. Had Rogers known that a commercial flight was scheduled overhead at that time (Flight 655 was only 27 minutes late), he might not have concluded so quickly that the aircraft was hostile.

But how is it possible that the Navy does not track civilian air traffic in the gulf region -- particularly regular flights like the 655, which must have appeared on U.S. radar screens hundreds of times before? The answer seems to be simply that nobody thought it necessary to do so. The Navy is just not used to operating in the half-war, half-peace atmosphere of the gulf, where harmless passengers and deadly enemies all whizz through the same cramped airspace. The Aegis system is designed for the open seas, where Pentagon planners mistakenly thought that wars would be fought.

On the Iranian side, both civilian and military jets take off from Bandar Abbas airport. Military traffic controllers keep close watch on ship movements in the gulf; they must have known that the Vincennes was engaged in a gun battle with Iranian speedboats (two were eventually sunk) only twelve miles offshore at the southern end of the gulf at the very moment that Flight 655 took off. Yet apparently nobody warned the civilian traffic controllers that Flight 655's path would take it directly over a developing firefight; had the controllers known that, they say, they would have delayed the takeoff. Why the foul-up? The military and civilian controllers at Bandar Abbas, it seems, did not talk to each other.

Enter technology, in the form of the Aegis system. It is designed to enable a single vessel to protect an entire Navy battle group from all sorts of attack. The Vincennes is one of eleven U.S. cruisers equipped with the system, and the first to be deployed in the Persian Gulf. Phased-array radars constantly sweep the skies over a vast swatch of ocean. They can track more than 100 aircraft, surface ships, submarines, missiles and torpedoes simultaneously. All show up as white symbols on one of four blue screens; each symbol is in a particular shape, identifying the object as airplane, missile or whatever. Computers can direct the simultaneous firing of missiles and other weapons over enormous distances at every form of threat. Aegis radar can supposedly spot a basketball at 150 miles and a high-altitude aircraft at more than 1,000 miles. One thing Aegis radar cannot do, however, is reliably distinguish the size and shape of an aircraft. Sideways, a longer plane might give off two blips to a shorter plane's one. But head-on, Aegis radar cannot tell an Airbus from an F-14. No radar can, the Pentagon insists.

How, then, can the Aegis system or its operators tell what kind of aircraft they are tracking? One method is flight pattern. Although the Pentagon at first asserted that the Airbus was outside the normal pathway for airline flights over the gulf, it has since conceded that the plane stayed within the 20-mi.-wide corridor all the time. The Pentagon claimed, however, that the pilot had wandered toward the western edge of the corridor and corrected that by veering back east toward the center line. As fate would have it, that turn headed the plane in the direction of the Vincennes.

According to the Pentagon, radar also showed the plane traveling about 500 m.p.h., faster than passenger jets generally fly on so short a hop as Bandar Abbas to Dubai (125 miles). Further, the plane was supposedly between 9,000 ft. and 7,000 ft. and descending at a time after takeoff when a commercial jet would ordinarily still be climbing. A nearby ship, however, the U.S. frigate John H. Sides, reported the plane at an altitude of 12,500 ft. and either flying level or ascending. Iranian air-traffic controllers, who have offered to turn their taped records over to U.S. investigators, claim that the jetliner was flying at about half the speed the Pentagon claimed and climbing from 12,000 ft. to 14,000 ft. The Pentagon cannot now explain these discrepancies. In any case, what Captain Rogers saw on the radar screen was a combination of direction, altitude and speed that suggested to him a plane preparing to attack.

The Aegis system is programmed, and its operators instructed, to consider hostile any craft that cannot be positively identified. The main method of identification is to send out an electronic beam that automatically triggers a transponder (a combination of a transmitter and a responder) in the belly of the aircraft; the transponder sends back a coded signal that is read by an IFF device. From the symbols on the radar screen, the Vincennes got back confusing signals: one was in a mode used by both civilian and military planes, but another was in a mode used exclusively by military craft. This second signal supposedly was identical with signals sent out by Iranian F-14s.

According to a Pentagon official, in fact, the transponder signal was identical to that of one specific F-14 known to be based at Bandar Abbas. One theory espoused by Ohio Senator John Glenn (among others), who attended several briefings on the disaster last week, is that the Aegis picked up and confused transponder signals from the Airbus and an F-14 at the Bandar Abbas base or flying out of it at some distance directly behind the Airbus.

. This theory, however, seems unlikely. The Vincennes IFF would send a beam about three degrees wide, interrogating every transponder in its path; by measuring the time it takes to receive the answering "squawk," the system can calculate the range of the answering transponder. The Vincennes' computers can then select and display responses from transponders at a given range. According to Pentagon Spokesman Dan Howard, the radar would have distinguished among 200 planes unless they had remained within 8 ft. of one another at all times and had simultaneously been destroyed -- a virtually impossible scenario.

But that raises the question of how the radar could have received two different signals from the Airbus. One possible explanation is that besides its civilian transponder, the Airbus carried a military transponder that it used to identify itself to military flight controllers at bases to which it sometimes ferried Iranian troops. Even then, though, there is no obvious explanation of why, as the Pentagon maintains, the military transponder gave out a signal that had been used only days earlier to identify an F-14. Iranian aviation officials, for their part, insist that the Airbus carried only a standard civilian transponder.

Even given the identification of the mystery plane as an F-14, there is some dispute as to whether an unmodified version of the craft would be capable of doing much damage to the Vincennes. The planes, built in the U.S. and sold to Iran in the 1970s during the reign of the Shah, are designed to fight other aircraft and are ordinarily equipped only with air-to-air, not ship-killing, missiles. The Pentagon retorts that Iran is known to have Harpoon antiship missiles and could have fired them; other experts doubt it. In any case, say some pilots, an F-14 trying to sink the Vincennes would probably have been flying much faster and much lower than the plane the Aegis system spotted. "No pilot in his right mind would attack a ship that way," one American F-14 pilot told the Washington Post.

The trouble is that Rogers and his crew had no time to reflect on such considerations. A ship nowadays can easily be sunk by a missile delivered from a plane that no one on board ever sees. In the open ocean, a possibly hostile plane can be tracked over hundreds of miles. But Admiral William Crowe Jr., Chairman of the Joint Chiefs of Staff, has likened combat in the Persian Gulf -- only about 25 miles wide at the narrowest point of the Strait of Hormuz -- to "fighting in a lake." A plane can reach a ship's missile range in minutes or even seconds after it first appears on a radar screen; a captain who hesitates too long while trying to identify conclusively that radar-screen blip could lose his ship and the lives of all those aboard. That almost happened in May 1987 to the frigate Stark. It was hit by two missiles launched in error by an Iraqi plane. The ship was severely damaged, 37 crewmen were killed, and Captain Glenn Brindel was pressured into retirement for failing to take defensive action.

Since then, the U.S. has promulgated new, hair-trigger "rules of engagement" for the gulf. They specify that commanders need not wait until their forces are fired upon before unloosing their own weapons. All they need is some convincing indication that a ship or plane is approaching with hostile intent. Doubtless influencing Rogers' decision was the fact that his ship had just been engaged in hostilities. Following reports of Iranian speedboat attacks on two neutral ships, the Vincennes sent a helicopter to investigate. The Iranians fired on the helicopter, triggering a firefight that Flight 655 had the foul luck to wander into.

Under intense pressure, Rogers had the Vincennes beam seven warning messages -- three on a civilian radio frequency, four on a military one -- at the approaching aircraft; the nearby frigate Sides chipped in with five more. The pilot of the Airbus never answered -- although he had been chattering away to the control tower at Bandar Abbas throughout his brief flight. His last words: "I am at level one-two-zero ((12,000 ft.)), climbing to one-four-zero ((14,000 ft.))." The last words from the controller at Bandar Abbas, who was about to turn over control to a center in Tehran: "Goodbye, have a nice flight."

One Iranian official suggests that U.S. ships so often query civilian airliners, and even try to order them to change course, that some pilots have got in the habit of ignoring such communications; until July 3, they got away with it. But Rogers requested and received permission from Rear Admiral Anthony Less, aboard the command ship U.S.S. Coronado, to fire. The Vincennes then launched two Standard SM-2 missiles that found their target with unerring precision.

In retrospect, the most frightening part of the tragedy is its seeming inevitability. Rogers may have made the only decision he could, given what he knew and when he knew it. The U.S. rules of engagement are not unreasonable, | considering the situation in the Persian Gulf, and the Aegis system apparently worked as it was supposed to. The tragedy seems to have resulted from a collision of random events (an airliner taking off at the moment a naval battle was beginning, for example) with inflexible technology in a pattern that could conceivably happen again. The Navy immediately began searching for ways to guard against that possibility, including the obvious step of feeding information about scheduled civilian air flights into military computers.

But these are at best partial and local answers to a problem that goes much deeper. The central question is whether technology may be pushing the fallible humans who operate it beyond their ability to make wise judgments instantly on the basis of what, with even the most sophisticated systems, will often be ambiguous information. This question applies not only in the Persian Gulf, but wherever there are fingers on buttons that can launch deadly weapons.

With reporting by David S. Jackson/Tehran and Jay Peterzell/Washington