Every few years someone wanders over to the National Transportation Safety Board and the Federal Aviation Administration, collects reports on a decade's worth of horrible aircraft accidents or damn-close incidents, then writes a book declaring that flying, an unnatural act for man, is also dangerous.

"The Unsafe Sky," by William Norris, identified on the dust jacket as a former foreign correspondent for The Times of London, is this year's entry. Norris not only revisits most of the recent U.S. accidents and incidents, but also provides a catalogue of European mishaps. His book falls far short of revelation.

The problem is that Norris never lets us in on the difficult, stomach-wrenching nature of the problems that have to be solved to get a 200-ton aircraft safely into the air, safely through it and safely back on the ground. Instead, he buries us in a blizzard of accident summaries, each followed by a few sentences of outrage but little illumination.

For example, birds sometimes fly into airplane engines. The classic case is that of the Overseas National Airways DC10 (which Norris mistakenly calls Overseas International Airlines) that tried to take off from New York's Kennedy Airport in 1975. The plane struck a flock of seagulls, some were ingested into an engine, and the engine disintegrated. Shrapnel cut hydraulic lines, controls were lost, tires blew, a fire started. The pilot stopped the plane and the passengers--all airline employes trained in evacuation--got out before the flames consumed the plane.

"Is it beyond the power of human ingenuity to think up a device which will keep birds out of jet engines?" Norris asks. "Surely not," he answers. End of subject. Return to anecdote.

His question is a good one; but the answer is not as simple as "surely not." About 840 U.S.-registered commercial airplanes are struck by birds every year, and on about 150 of those occasions an engine ingests a bird. On 75 of those occasions the engine fails, a situation that, while an emergency, almost always has a happy ending. The most effective control methods the Federal Aviation Administration has found are environmental, not technical. At FAA urging, many states now prohibit sanitary landfills near airports, because landfills attract seagulls. Other landfills have been cleaned up. Such actions have reduced--but not eliminated--bird problems.

What about a technical fix? Why not just put a screen over the front of the jet engine? The FAA and the major engine manufacturers tested screens a few years back and determined they were not effective. Why not? Engine intakes for jumbo jets exceed 8 feet in diameter. A screen to cover that opening would be huge and heavy. It would impede the engine's ability to suck air, and thus would reduce engine power, making a bigger engine necessary to get the same payload off the ground. In tests, the screen strained the bird, but did not keep it out of the engine. "There was just too much of a thrust and weight penalty," an FAA engineer said. In English, that means the screen would cost too much to carry.

Outrage.

But is it? Flying must be economically feasible to happen. "Any airplane is a tissue of compromise," a top aviation attorney told me once, and that, of course, is the problem. Weight is the game. A heavier airplane might be "safer," but it would also carry fewer people or less cargo and require more fuel, which means tickets would cost more.

Locating that fine line between responsible risk-taking and reprehensible disregard for human life is not easy, but the way that is done or not done is at the heart of our existence in this technical age. The author doesn't help us understand this.

Norris does repeat the questions that have been asked many times about aviation safety. He tells us about drunken pilots, careless air traffic controllers, inept regulators, venal corporations, lawyers more concerned about limiting liability than finding truth. It is useful to have those problems raised and, as the author points out, they are usually addressed seriously only in the weeks and months immediately following the most recent disaster. But the central point, the intellectual strife at the heart of aviation safety, is missed.

In the meantime, consider this: 508 people died on the nation's highways over the 1979 Memorial Day weekend, which began with a plane crash in Chicago that killed 273 people. Nobody noticed the traffic accidents, because they killed in ones and twos. It was 32 months before there was another major airline crash in the United States, and that was on the 14th Street bridge. In the time between, more than 12 million commercial domestic airline flights took off and landed safely.