To the surprise of no one, John W. Hinckley Jr. will plead insanity in the shooting of President Reagan and three others last March. From the defense point of view, it's the only plea that makes sense.

With the evidence of eyewitnesses, TV footage and Hinckley's own statements, it would be tough to convince a jury that somebody else did the shooting. If that were the only question at issue, conviction would seem assured, and Hinckley would face up to life imprisonment.

But if the defense can demonstrate that Hinckley was insane at the time of the shooting and that the assault was the product of his mental disorder, then he would be innocent of the charges against him. He could be a free man in 50 days.

If an insanity plea is a reasonable strategy for the defense, it strikes many of the rest of us as insane. Yet it is excruciatingly difficult to come up with an approach that makes sense in these cases. Clearly there are people who are so deranged that they are unable to control themselves, or who are so deluded that they don't know what they are doing. It is unfair on its face to conclude that such people are "responsible," in the legal sense, for their actions. And if they are not responsible, they are not criminals deserving of incarceration but sick people in need of treatment.

The difficulty is in determining legal responsibility. Sen. Orrin Hatch (R-Utah) outlined the basic problem on March 26, four days before Reagan and the others were shot, when he introduced legislation to limit the insanity defense.

"The traditional insanity defense," he said, "is both a legal anachronism and a concept ill- suited to modern psychological theory. It presents issues--important issues--that are not susceptible of intelligent resolution in the courtroom environment. Trials in which the insanity defense has been raised have often degenerated into swearing contests between opposing teams of expert witnesses, all of whom are forced to translate the language of the psychiatric profession into the quite alien language of the legal profession. This takes place within the context of convoluted, hypothetical questioning that yields responses that can only be confusing and misunderstood. They certainly contribute extremely little toward the pursuit of truth."

He is right. It seems disconcertingly easy to find reputable psychiatrists to take either side of the question of responsibility, and non-expert jurors are forced to choose between conflicting offerings of expert testimony.

The present state of the matter, based on court decisions (the matter has never been legislated) is that:

A person is not responsible for criminal conduct if at the time of such conduct as a result of mental disease or defect he lacks substantial capacity to appreciate the criminality of his conduct or to conform to the requirements of the law;

The terms "mental disease or defect" do not include an abnormality manifested only by repeated criminal or otherwise antisocial conduct.

Hatch's proposal would add this language:

"It shall be a defense to a prosecution under any federal statute that the defendant as a result of mental disease or defect lacked the state of mind required as an element of the offense charged. Mental disease or defect does not otherwise constitute a defense."

As Hatch puts it, "an individual who believed that he was throwing darts at a board instead of stabbing a victim to death would not be guilty of murder. But--and it is the key distinction -- he would not be guilty because he did not 'knowingly' kill another, not because he was covered by a separate insanity defense."

His amendment would come into play only in borderline cases. The obviously deluded defendant would, in all likelihood, be found incompetent to stand trial in the first place. This, in my view, does not mean that he ought to go free. Non-responsible does not mean non-dangerous.

Hatch believes that his proposal would help to rationalize the present formulation that depends on the willingness of a psychiatrist to label a defendant "mentally diseased" and the willingness of a jury to agree.

It helps in another way: by clearing up the confusing distinctions between one defendant who is "mentally ill" and another--the ghetto youth given to random, senseless violence, for instance--who could be said to be the product of "crimogenic social or economic conditions."

Hatch is trying to nudge us toward the truly relevant questions: Did he do it? Did he know what he was doing?