On the evening of July 17, 1996, TWA flight 800, a Boeing 747-100 bound for Paris, took off from JFK international airport in New York. Twelve minutes later, as the plane traversed the southern coast of Long Island and climbed to an altitude of 13,700 feet, it exploded in a tremendous fireball, broke up into two pieces, and fell into the Atlantic seven miles from shore, killing all 230 passengers and crew on board.
Two years and two months later, Swissair flight 111, a McDonnell-Douglas aircraft, left JFK en route to Geneva. As the plane made its way up the eastern seaboard, the pilots noticed smoke in the cockpit and asked permission for an emergency landing. They did not succeed in bringing the plane safely to the ground. At approximately 9:30 PM on September 2, 1998, Swissair 111 plunged into the sea off the coast of Nova Scotia, taking the lives of the 229 people aboard.
One year and one month later, on October 31, 1999, EgyptAir flight 990, a Boeing 767, took off from JFK en route to Cairo. After flying in an easterly direction for 31 minutes, it crashed into the Atlantic Ocean south of the island of Nantucket. The 217 passengers and crew all perished.
The TWA and EgyptAir crashes were the subjects of exhaustive investigations led by the National Transportation Safety Board (NTSB) in conjunction with a range of other governmental agencies and private firms. Because Swissair 111 crashed in Canadian waters, the Canadian Safety Board conducted a no less thorough inquiry into that accident.
The NTSB determined that the TWA 800 crash was the result of an explosion in the central-wing fuel tank. Although the exact cause could not be definitively identified, the board suggested that defective or damaged wiring had short-circuited, causing the ignition of gasoline fumes in the nearly empty tank. Swissair 111 was found to have similarly suffered equipment failure, in this case causing a fire that began somewhere in the cockpit. Owing in large measure to the extreme difficulty of recovering aircraft fragments scattered across the ocean floor, the precise origins of the blaze remain obscure, though a short circuit in the cabin map light has been identified as a likely possibility. As for EgyptAir 990, no evidence of equipment failure was found. All indications pointed in another direction entirely: pilot suicide.
Those, at least, are the official explanations of the three disasters. But another possibility has also been raised: namely, that what brought down all three civilian aircraft was electromagnetic interference (EMI), generated by U.S. military installations and vessels. This sensational and disturbing theory is associated above all with the name of Elaine Scarry, a professor at Harvard who over the last three years has published a series of lengthy and heavily footnoted articles on the aviation disasters in the New York Review of Books. What makes Scarry’s work all the more intriguing is that she is a professor not of physics or engineering but of literature—or rather, to give her full title, of “Aesthetics and Value Theory.” This fact, among others, may account for the wide attention she has received for her ideas, including an adoring profile in the New York Times Magazine last November.
_____________
Scarry turned her attention to aviation disasters after stumbling upon a news article reporting that a number of U.S. army helicopters may have crashed after having encountered EMI emitted by nearby military aircraft. This report came to her with the force of an epiphany. “Immediately,” according to the author of the Times Magazine article, “she thought of TWA Flight 800. . . . Could EMI from military ships and planes in the vicinity of TWA Flight 800 have taken down the plane?” To find out, Scarry began immersing herself in the literature of EMI. After spending months with government reports and engineering and military journals, and assimilating a vast body of highly technical information, she concluded that the government’s inquiries had utterly failed to ask the right questions.
In her first article for the New York Review, “The Fall of TWA 800: The Possibility of Electromagnetic Interference” (April 9, 1998), Scarry began by describing possible EMI-related problems that have plagued military aircraft, and then catalogued the emitters of EMI known to have been located in the vicinity of TWA 800 just as it exploded. “Directly overhead,” she wrote,
was a Navy P3 Orion with its transponder turned off: it was 6,300 feet above the passenger plane and had intersected its longitude and latitude within seconds of the moment the catastrophe began. In the airspace beneath TWA 800 a Black Hawk helicopter and an HC-130 plane were flying at an altitude of 3,000 feet: they were five miles north of the commercial liner. A Coast Guard cutter rescue ship, the Adak, was somewhere on the sea below, its precise location in relation to the falling plane unspecified in the official record but given in news reports as nine to twelve miles south. One hundred and eighty-five miles to the southwest was an Aegis cruiser, the USS Normandy: that mileage places it off Maryland’s Eastern Shore.
These craft, plus an additional five named by Scarry, were all capable of emitting electromagnetic radiation. Together, they “formed a temporary envelope around the passenger plane” that could have interfered with the workings of its electrical systems. Although Scarry refrained from asserting outright that EMI brought down the flight, she held it open as a distinct likelihood and called for a crash program of research:
The NTSB needs to ask: What equipment [on the military aircraft and vessels in the area] was switched on? What instruments were turned off? What was the sequence of those acts? We need to know the answers to these questions if we are to determine whether electromagnetic interference—a sudden pulsing or spiking in the electronic environment—may have caused TWA 800’s electrical systems to go haywire.
_____________
In her second article (“Swissair 111, TWA 800, and Electromagnetic Interference,” September 21, 2000), Scarry pointed to the numerous similarities between the Swissair and TWA disasters. Some of these similarities were truly striking: both planes had taken off from JFK, both took off on a Wednesday, both took off precisely at 8:19 PM, and both were following the same route. These, she thought, might not be mere coincidences: when an “event recurs at an exact time and place, it may become clear that . . . the problem is arising not from some problem inside the affected piece of equipment but from something outside—something on a regular enough schedule that its responsibility can eventually be tracked down.”
What might that something be? Once again, reported Scarry, various military craft, including several P3’s and several Navy submarines, were in the area and may well have been engaged in “electronic warfare practice.” Although Swissair 111 managed to stay aloft a full hour longer than TWA 800, a mysterious thirteen-minute interruption of radio contact between the plane and ground controllers commenced just as the flight approached the location of the earlier TWA crash.
This thirteen-minute break, conjectured Scarry, was the result of an encounter with the same field of EMI that had brought down TWA 800. Although Swissair 111’s computerized control system then carried out preprogrammed acts of self-repair, it was finally overwhelmed by problems that could no longer be contained. As in her earlier article, Scarry concluded with a plea for an NTSB investigation, a task, she wrote, that should be carried out by those in authority rather than being left to researchers like herself, “isolated citizens working through Freedom of Information inquiries.”
Scarry’s third article, “The Fall of EgyptAir 990” (October 5, 2000), took note of some obvious differences between this accident and the previous two. EgyptAir 990 left early in the morning, and on a Sunday, and it followed a more southerly route. Such differences aside, however, one key similarity stood out: the distinct possibility that EgyptAir 990 had passed near U.S. military vessels or ground stations that emit EMI.
It seems that, under the direction of air-traffic controllers, EgyptAir 990 had been asked to fly through a restricted zone that is held open for commercial aviation only when no military exercises are planned in it, as was the case that morning. But “the absence of scheduled military exercises,” wrote Scarry, “does not guarantee that no military craft were in the corridors around the warning zones or inside the warning zones.” Moreover, as became known from news reports, “powerful Air Force radar antennas were tracking EgyptAir 990,” with “capacities not shared by any civilian radars.”
These facts, Scarry once again concluded, were sufficient to warrant another NTSB investigation. Indeed, without such an investigation any suggestion that the crash was attributable to pilot suicide was “incomprehensible.”
_____________
How plausible is any of this? Both Scarry’s text and her extensive footnotes demonstrate that she is capable, if nothing else, of assembling a mass of information from recondite sources, ranging from the Journal of Electronic Defense to the South China Morning Post. They also reveal someone at home in the jargon of fields remote from the usual fare of the Harvard English department. But as for the heart of the matter—the EMI thesis itself—it is not only implausible, it is irresponsible in the extreme.
In the simplest terms, EMI is the effect that the energy or the radiation emitted by one electrical device has on the workings of another electrical device. It is, of course, a real and well-documented phenomenon, and one that can be readily observed when, for example, the operation of an electrical appliance causes static to appear on a nearby television screen.
Electromagnetic radiation runs a gamut, from sluggish, long-wave, low-energy radio waves that are quite innocuous to extremely short-wave, high-energy X-rays and gamma waves that can be lethal to living organisms if absorbed in sufficient doses. What kind of radiation, then, could cause an aircraft to drop from the sky, and by what mechanism would it do so? Though Scarry’s analysis is exhaustive on a good many points, when it comes to these crucial issues, she offers not a single answer. Her omissions are significant.
As it happens, commercial pilots are required to file reports on all significant disruptions that occur in their flights. Yet, in the telling words of Air Safety Week, a leading industry newsletter, “there is not a single documented case of external electromagnetic interference among the 79,500 records dating back to 1988 filed in the Aviation Safety Reporting System. Not one.”
This is hardly surprising. For among the factors that Scarry fails properly to consider in her various published articles are distance, impedance, and coupling. Thus, radiation of a given frequency and intensity may cause considerable damage at a range of 100 feet; at a range of miles, it may have no effect at all. For a plane cruising at, say, 20,000 feet (a little under four miles), any radiation striking its skin would be about ten-thousand to one-hundred-thousand times less intense than what it would be in the immediate vicinity of an emitter on the ground.
If, in addition to the distance factor, one takes into account the impedance of the atmosphere and the shielding effect of the fuselage, the radiation that could actually penetrate to the interior of an aircraft would be attenuated still more. Finally, even the quantity of radiation that did penetrate would have to be of the right frequency in order to “couple,” i.e., resonate harmfully, with the electronic circuitry aboard the plane.
Let me put all this into perspective: the radiation routinely emitted inside an aircraft—from the microwave ovens in the pantry—is more powerful by many orders of magnitude than any radiation affecting it from external sources.
To be sure, the problems I have enumerated could theoretically be compensated for by emitter power. But superpowerful, monster emitters on military aircraft or Navy cruisers and submarines exist only in science fiction—and in the imagination of Elaine Scarry. The actual energy level of radiation from planes and ships, military or civilian, is severely constrained by the amount of power that can be generated on board. In fact, not even ground stations can produce the kind of radiation that Scarry has in mind. If they could, modern warfare as we know it would be altered profoundly. In particular, if all we had to do about incoming enemy missiles was to point at them the powerful tracking radars already in our possession, we would not now be spending billions on missile defense.
If Scarry’s grasp of physics is shaky,1 she does not fare much better when it comes to the treatment of evidence or the logic of ordinary experience. For example: in pressing her case, Scarry relies heavily on suspicious-sounding anomalies and coincidences. Thus, as I have already noted, she points out that TWA 800 and Swissair 111
could have originated from two different airports anywhere in the country, or, for that matter, anywhere in the world. But as it happens, both planes took off from a single airport, New York’s JFK. The two flights could have taken off on any two days of the week and at any two minutes of the day. But as it happens, both took off on a Wednesday at 8:19 PM.
Scarry also makes much of the fact that all three accidents occurred amid a wider interval of relatively incident-free years, which suggests to her that they were not random occurrences: “No other large passenger plane taking off in the United States,” she writes, “crashed during this three-years-and-three-months period.”
It is well known, however, that if one looks closely at any half-way complicated event—the assassination of President Kennedy is a notorious case in point—all sorts of seemingly weird coincidences can be found. True randomness often appears to give rise to just such strange accumulations of “hits,” a phenomenon that never ceases to amaze those unfamiliar with statistics. The myth that randomness implies uniformity is known as the gambler’s fallacy; Scarry has fallen prey to it.
_____________
But the most serious derelictions in Scarry’s reasoning are to be seen in her analysis of the EgyptAir 990 crash. Both the TWA 800 and Swissair 111 accidents were, as she correctly notes, “unmistakably electrical” in origin (though she wrongly identifies the source of the problem). But Scarry also insists that the same might be said for EgyptAir 990. Not only were there those “powerful Air Force radar antennas” monitoring the doomed flight, but, she adds, it is “not hard to think of reasons why EgyptAir 990 might have become an ordinary—or even an extraordinary—target of intense observation.”
For one thing, the plane was “a foreign carrier from a country that is not always regarded as a close United States ally.” For another thing, on board were
thirty-three Egyptian military officers, including one brigadier general. During their visit to the United States, the thirty-three officers were, according to the Defense Department spokesman, Kenneth Bacon, subdivided into distinct groups: the members of one group were here to study communications; a second group was testing recently purchased helicopters; the third group was here to receive “training on high frequency” equipment; the fourth group was learning about telecommunications; the fifth group was studying the repair of Chaparral missiles; the members of the sixth group were carrying out individual projects.
Unfortunately, Scarry never explains the relevance of this detailed enumeration of the Egyptian military personnel and their various assignments. But through her favorite literary devices of the hedged phrase (“persuasive but not unmistakable evidence”) and the subjunctive form (“could have,” “might have,” and their permutations) she makes clear her opinion that U.S. electronic surveillance inadvertently knocked the plane from the sky while eavesdropping on in-flight conversations.
As for the real facts, as opposed to the remotely potential ones, Scarry’s treatment of them is both revealing and scandalous. Thanks to the data recovered from the flight recording systems—the so-called black box—the NTSB was able to piece together a picture of EgyptAir 990’s final minutes. The result so unambiguously conflicts with the EMI hypothesis that it is worth summarizing.2
Shortly before EgyptAir 990 began its dive into the ocean, the pilot, Ahmed El Habashy, excused himself to go to the bathroom, leaving the relief pilot, Gamil El Batouty, alone in the cockpit. At 01:48:40, Batouty is heard saying faintly in Arabic: “I rely on God.” For another minute the plane continues flying normally, but then the autopilot is disconnected. Now manually flying the jumbo jet on his own, Batouty repeats the incantation, “I rely on God.” At 01:49:53, the wing elevators—the adjustable surfaces that control the pitch of the aircraft—are shown dropping, and the plane begins to plunge nearly straight down. Eleven seconds later, Habashy re-enters the cockpit and exclaims “What’s happening?”; Batouty continues to chant. At 01:50:21, both engine levers are moved to the cut-off position, and the elevators split, with the left (controlled by Habashy) moving to raise the nose and the right (controlled by Batouty) remaining in the down position. Batouty continues his chants. At 01:50:24, both throttles are advanced, with no engine response. The captain then exclaims, “What is this? Did you shut down the engines?” Four subsequent pleas by the captain—“Pull with me”—go unanswered as the plane spirals downward into the water.
As NTSB investigators have suggested, this is about as unambiguous a picture of suicide as we are ever likely to see. Scarry, however, relegates the conversation in the cockpit to a footnote, and then proceeds to put a reverse spin on it, once again employing the subjunctive:
“[I]t may have been that all subsequent command actions [after the craft began to dive] were initiated by the pilots, desperately trying to regain control of the plane and making the best decisions they could: those actions may have been a highly competent effort to save the aircraft or they may have been, however admirable, less competent. It is possible that all four of the key events—automatic pilot shutdown, steep dive, split-elevator anomaly, engine shutdown—were caused by electromagnetic interference or that only some of the four were caused by electromagnetic interference, the rest by pilot intervention and attempted rescue.
In Scarry’s interpretation, the man who brought the plane down has thus been transformed into an “admirable” if ultimately unsuccessful savior of terrified passengers. At the same time, it is suggested that unnamed U.S. military personnel were once again recklessly flooding the skies with deadly emissions—emissions that, all told, have taken the lives of some 676 innocent people in three avoidable accidents.
_____________
In the New York Times Magazine profile, Elaine Scarry is described as a “wide-eyed woman with a girlish voice and unruly blond locks” who, despite the lack of “formal training in aviation science,” wrote a “dazzling brief” in behalf of the thesis that electromagnetic radiation may have caused three devastating crashes. The Times of London has similarly praised “Scarry’s meticulously researched conclusions,” and has even speculated on the reason why the EMI thesis has been “downplayed” by investigators: “partly,” according to the Times, “because it intrude[s] on the secrecies of national defense and partly because what science does not fully understand it prefers to disregard.”
This is reprehensible. Far from disregarding EMI, the NTSB dutifully investigated Scarry’s charges after her very first article appeared, consulting with both the military’s Joint Spectrum Center and NASA and performing “worst-case” calculations based upon the strongest transmitters in the area. As James Hall, the acting chairman of the NTSB, has summarized the findings, “the maximum possible signal strength inside the fuselage of the TWA flight 800 from the highest power emitter known to exist in the area of the accident could not have provided sufficient energy to ignite the fuel/air mixture.” Even more decisively, Hall notes, “EMI caused by an emitter capable of transmitting even greater energy than that transmitted by the known emitters could not ignite a fuel/air mixture in a fuel tank or cause the loss of control of an aircraft” (emphasis added).
But this did not deter Scarry or her editors from publishing a second and then a third article repeating her allegations. Today, she still denies that the NTSB has punctured her theory, and she continues to call upon it and other government agencies to perform what she calls “a comprehensive review of the external electromagnetic environment” surrounding each of the doomed flights.
We have been down this road before. Back in 1989, the New Yorker published a series of dramatic articles by Paul Brodeur purporting to demonstrate that people living near high-voltage electric-power lines were at a heightened risk of contracting cancer. After many lengthy investigations and counterinvestigations, it was proved (as Robert L. Parks reports in Voodoo Science) that no correlation existed between power lines and cancer; the electromagnetic fields emanating from the lines were well below the threshold to have any effect on human health. Nevertheless, the scare imposed its costs. Homeowners in some locations saw their property values collapse, utility companies were compelled to reroute their cables at great expense, and the estimated total bill for the fiasco ran in excess of $25 billion.
Are we in for another bout of radiation hysteria, this one focusing on the electronic-warfare capabilities of the U.S. military and complete with a Bermuda Triangle-like scenario of mysterious aircraft catastrophes? If history is indeed repeating itself, the first time was farce, and the second is farce again.
_____________
1 Only after her three articles were published and subjected to critical scrutiny did Scarry begin to muster any sort of acquaintance with the basic physics involved. She did so, however, not in a published article but on the Internet, in a “web-based supplement” to the New York Review of Books dated October 26, 2000. Even this document, however, misunderstands EMI, erroneously holding that it can cause not only commonly observed static but also short circuits in electrical wiring of the kind that probably brought down TWA 800.
2 The complete NTSB report is available at www.ntsb.gov/events/EA990/default.htm. My own brief summary is adapted from Air Safety Week, September 18, 2000.
_____________