On June 9, the front page of this newspaper carried a photograph of a red, white and blue object floating, like some sort of gaily colored raft, in a blue-black ocean. To pilots, it brought a chilling sense of deja vu. In November 2001, a similarly shaped and colored object floated in Jamaica Bay, just off Long Island. It was the vertical stabilizer -- colloquially, the "tail fin" -- of an American Airlines Airbus, Flight 587, that had broken up shortly after taking off from JFK. That fin was practically undamaged; it had parted at the root, each of the massive fittings that attach it to the fuselage torn neatly in half. Here was another such fin: seemingly intact, snapped cleanly from the vanished Air France Flight 447.
The National Transportation Safety Board took almost three years to untangle the mystery of the American Airlines crash. It eventually concluded that the first officer had caused the breakup by stepping too vigorously on the airplane's rudder pedals, and that the rudder pedals of Airbus airplanes were more susceptible to over-control than those of rival Boeing's jets.
The rudder is the movable portion of the vertical fin. Unlike the rudder of a boat, it is not used to turn. In fact, the rudders of jets are seldom used at all, except when landing in a strong crosswind or to hold the airplane straight after an engine failure. In this case, the NTSB thought, the pilot had tried to use the rudder to steady the plane in the wake of a 747 several miles ahead and had managed to break the vertical tail off instead.
Pilots were incredulous. The airplane had just taken off and was climbing; it was flying well under its "maneuvering speed," the speed below which a pilot should be able to use the flight controls in any way without risk of damaging the airplane. How, then, could this pilot possibly have broken the airplane with its own controls?
The New York crash uncovered a gaping misunderstanding among pilots, manufacturers and the FAA, which sets standards for structural safety and certifies compliance. Even though pilots believed in the absolute protection of maneuvering speed, and informational publications from the FAA and from manufacturers supported that belief, it turned out that if you read the certification regulations carefully, you would discover an exception: The vertical fin did not have to be strong enough to allow the rudder to deflect fully when the airplane was in a "yawed" position -- that is, when the back end of the plane had swung to one side, most likely because of a gust of wind.
The common-sense response of any air traveler is, "Why don't they just make it stronger?"
The answer, which will provide little comfort, is that airplanes are not designed to be as strong as possible. They are designed to be as light as possible. The manufacturer who adds extra "beef" to its structure ends up with a heavier airplane that carries fewer passengers or uses more fuel, and loses sales to the lighter and more efficient, though "weaker," competitor. The same is true of cars. We willingly sacrifice greater safety to get cheaper, more fuel-efficient vehicles.
In the aftermath of the American Airlines accident, the FAA and the manufacturers revised their publications to incorporate the bad news about maneuvering speed and rudders. The NTSB recommended changes to the hypersensitive rudder pedals of some Airbus models, but most, including the Air France A330, have control systems in which the pilot's commands are normally channeled through computers programmed to avoid over-stressing any component. (The American Airlines airplane lacked this protection, and on Flight 447, the computers appear to have been off-line at the time of the accident.)
It is far, far too early to analyze the Air France catastrophe. The series of automated messages that emanated from the doomed airplane paint a murky picture of cascading electronic failures. The triggering event -- icing, turbulence, lightning, bomb or cargo fire have been suggested -- is unknown.
But the picture painted by the fin floating almost undamaged in the Atlantic is much clearer. It broke off in one piece, and maps of debris distribution suggest that it could have been the first thing to go. The pilots may have been controlling the airplane manually, having bypassed the computers because of the electronic problems. If, as seems likely, they were in turbulent weather, they would have had their hands full, and they may possibly have over-controlled in just the way that the American Airlines pilot did in 2001.
This is pure -- and premature -- speculation. But it is not speculative to observe that some vertical fins are weaker than they should be, and that this is because of the regulations that govern how airplanes are built. The FAA, together with its international counterparts, should amend certification requirements for future aircraft to restore to the maneuvering speed the safety status that pilots always believed it had. The rules that govern the structural integrity of airliners should not include asterisks.