In unusually harsh language, the Court of Appeals for the District of Columbia recently criticized the National Highway Traffic Safety Administration's decision to rescind a rule requiring air bags or automatic seat belts in full-size 1983 cars.

The court reversed NHTSA's decision to abandon an effort to require passive restraints because, the court believed, the agency had failed to consider alternatives, such as requiring only air bags instead of detachable automatic belts, or requiring only certain kinds of automatic belts.

The court has reopened a controversial rule-making that, as The Post said in its June 4 editorial ("Unbuckling the Seat Belt Rule"), is so old it's almost ready for its own driver's license: rule-making began in 1969, leading to a 1972 requirement that was modified in 1975, then suspended in 1976. The latest court decision gives NHTSA 30 days to produce a schedule showing what happens now.

The best next step would be a field test like one proposed in 1976 by then secretary of transportation William Coleman. He proposed that auto manufacturers put 500,000 cars with passive restraints on the road, beginning with 1980 models, and pushed the auto makers into "voluntary" agreements.

But a few months later, the agreement was nullified when Carter administration regulators, wanting quicker action, proposed a mandatory rule for passive restraints.

The impatience of the Carter regulators in 1977 is all too likely to be repeated now, so it's important to consider just what we don't know about passive restraints, which would have have been implemented a decade ago if there had not been serious questions:

* How will the public respond to mandated passive restraints? Public acceptance is a key element in the "practicability" of auto safety rules. The case for passive restraints is not as clear and simple as it was, say, for wearing motorcycle helmets, another controversial rule. Repeal of a passive restraint rule would mean not only a large investment wasted, but also a major setback to the effort to reduce our annual slaughter on the highways.

* How will passive restraints, particularly air bags, actually operate in an incredible variety of situations and after a few years' use? Nothing short of actual experience will show. For example, how many times will air bags accidentally inflate? If we mandate air bags, and our early experience with them is blotted by a few accidental deployments causing injuries or deaths, who would bet against congressional repeal of the requirement and maybe even the gutting of NHTSA?

* Will our never-to-be-underestimated imaginative citizenry reject passive restraints, leaving us no better off than we are now with the far cheaper, highly effective, but little-used, manual seat belts?

* Are all the problems with the technology solved? Is there danger to an out- of-position child from deployment of an air bag? Can you get out of an automatic seat belt easily enough in an emergency?

The questions are well known, but more public hearings and more debate will not produce answers.

Sen. Danforth has held hearings on his bill to provide a tax incentive for producing air-bag-equipped cars. With or without such facilitation, which we applaud, we need to know how passive restraints actually work. Without public acceptance, no restraint system will succeed.