"The National Transportation Safety Board determines that the probable cause of the accident was the failure of the flight crew to (fill in blank)."

That is bureaucratese for pilot error. Such sentences often begin the key paragraph of the safety board's formal report on an airline accident, and no sentence is more likely to infuriate a professional pilot.

Author John J. Nance, who flew for Braniff International before it went belly up, has made a strong case for his fraternity's belief that blaming accidents on pilot error does little to prevent similar crashes in the future and is too simplistic to explain what really happened.

To be effective, he argues, investigators must find out why the error was made, through a careful analysis of how humans might be induced through airplane design, management pressure, training failures or personal problems to perform the way they did. Then, airplanes must be designed and pilots trained to keep the same accident from recurring, even if somebody does make a mistake.

That debate is long running in aviation and provocatively presented. Unfortunately the book is being marketed to make another point, a more politically topical one that is really its second theme. Deregulation, the dust jacket screams at us, "has jeopardized airline safety."

Nance does not make as persuasive a case for that as he does for investigations of human factors. Nonetheless, his book comes after the worst year, in terms of loss of life, in civil aviation history, and he reminds us of real aviation safety problems.

They include a Federal Aviation Administration that is never out front but always reactive and defensive, and a government (both executive and legislature) that has not provided the necessary personnel and equipment to guarantee that safety is adequately monitored.

Those problems existed before deregulation and can hardly be blamed on it. The real question since the Airline Deregulation Act of 1978 is whether the air transport system is as safe as it was before.

Nance's negative conclusion rests on careful research into three postderegulation accidents, the best known of which occurred on Jan. 13, 1982, when Air Florida Flight 90 took off in a snowstorm from Washington National Airport and crashed into the 14th Street Bridge, killing 78 people, including four who were driving across the bridge.

The key paragraph of the official safety board report lists the probable cause as pilot error, including the failure to use engine anti-ice heaters, a decision to take off with snow and ice on the wings and the captain's early failure to abort the takeoff when weird engine gauge readings should have warned him that something was wrong.

The board listed as contributing factors the prolonged delay between the time the plane was deiced and the takeoff, "the known inherent pitch-up characteristics of the Boeing 737 aircraft" when the front of the wing is contaminated with even slight amounts of snow or ice and "the limited experience of the flightcrew in jet transport winter operations."

In other words, a many-factored accident, a typical accident. Nance writes, "The element the safety board failed to spotlight was the primary force that permitted all other factors to come together to cause the accident: Congress's 1978 passage of deregulation."

The point is difficult to debate. Without deregulation, Air Florida would never have been flying in a northern snowstorm; it would still have been a small operation carrying people around Florida in propeller planes. Without deregulation, the young Air Florida captain would never have been in command of a high-performance jetliner; at best he might have been a second officer waiting to move up the endless seniority lists at United or American.

The benefit of deregulation -- and the one that economists and Congress promised -- is one Air Florida undeniably provided: cheap fares.

Nance wanders back and forth between his two points -- the need for better understanding of human factors in accident investigations and the perils of deregulation -- and unsuccessfully tries to marry them. Nonetheless, this is one of the best and most troubling aviation safety books of recent years because it is careful, not shrill. Nance has his facts in order.

He understands that the FAA's people work as hard as they can to make a deregulated industry safe, but he asserts that because of the proliferation of airlines since deregulation, the FAA has little control over safety. "In terms of the vital influence of management, the airlines are on the honor system," he writes.