Psychiatry and psychiatrists have been besieged by critics on all fronts in recent years, but particularly since the not-guilty-by-reason-of-insanity verdict was decided against John Hinckley last year.

While criticism of the insanity plea both from within and without the psychiatric profession has been heard frequently, it is only recently that the attacks pose a real threat: lawsuits for medical malpractice.

In what must be the classic case of the chickens coming home to roost, the medical pretentions of the psychiatric profession may now result in psychiatrists' having to take some of the legal responsibility as well as some of the glory for their particular brand of health care.

While psychiatrists have long maintained that psychiatry is a medical specialty that properly defines its clients as "patients" who need "treatment," they have been less than anxious to accept the legal consequences of having their mistakes qualified as malpractice. Critics have countered that if patient misbehavior is a result of a pathology that can be diagnosed, and that if psychiatry is a medical practice, then bad psychiatry amounts to medical malpractice.

For many years only certain types of psychiatric errors--those most analogous to medical errors-- have resulted in verdicts against psychiatrists. For example, failure to obtain informed consent and failure to give proper tests prior to prescribing powerful drugs have been accepted as malpractice by the courts whether committed by psychiatrists or other medical professionals.

But, interestingly, suits based on alleged misdiagnoses have been restricted primarily to other medical specialties because in psychiatry a variety of patient behavior could indicate, say, schizophrenia. Therapies--talking and drug--are similarly nonspecific. This makes it difficult for other "psychiatric experts" to dispute diagnoses or even demonstrate that damaging consequences could have been avoided by a more accurate diagnosis.

Thus, it is not surprising that the new legal threat to psychiatrists is not based on their vague "diagnoses," but their not-so-vague "prognoses," which means in this area of pseudo-medicine, patients' behavior.

Psychiatrists, then, are being sued for the wrongdoing of their patients:

In Virginia, a teen-ager who was paralyzed by a gunshot fired by her former boyfriend filed a multimillion-dollar damages suit against the psychiatrists who allegedly misdiagnosed and negligently treated him.

In Pittsburgh, a kidnap/rape victim filed suit against two psychiatrists at the Mayview State Mental Facility for negligence in their handling of the accused.

And in the most famous suit to date, the victims of the Hinckley shooting (except the president) have joined together in an action against Hinckley's psychiatrist for failure to recognize that he "knew or should have known" that Hinckley would make an assassination attempt and for failure to "warn law enforcement officials of Hinckley's dangerousness."

The notion that a non-psychiatric physician might be held liable for the criminal acts of his patient would be regarded as nonsense, for these doctors make no claim to special knowledge of human behavior--only human physiology. But psychiatrists, claiming medical and scientific legitimacy, make such a claim.

Thus we have the dilemma: if psychiatrists claim to be able to predict by virtue of medical expertise who will be dangerous, they should be held liable if they fail to act on what they purport to know scientifically. On the other hand, if they deny liability on the grounds that they have no such special knowledge, they risk losing their medical identity.

How have they handled this choice? Typically, they have asked the public to accord them the status of doctors, but withhold from them the legal liabilities of doctors.

The American Psychiatric Association, in its Statement on the Insanity Defense, has indicated that perhaps psychiatrists do not have the ability to predict dangerousness. Therefore, by implication, no one (including attorneys for victims of psychiatrically treated criminals) has the right to expect psychiatrists to know, much less to warn, when patients will physically harm others.

So what is the answer? Shall we simply exempt psychiatrists from the standards to which other doctors are held? This would increase their authority and would probably permit them to exercise more freely their too-often erroneous judgments in criminal cases. Such judgments are often horrifyingly unconstrained.

Recently, for example, Ronald Hoffman, the "Sunshine Sniper" of Chicago who randomly shot and killed his neighbors five years ago while living with his parents, has been granted permission to spend weekends at his home by a judge on the recommendation of psychiatrists who had "treated" him.

On the other hand, psychiatrists disinclined to release their clients have been upheld by the Supreme Court, which ruled that it does not violate the Constitution to hold a person found not-guilty- by-reason-of-insanity longer (perhaps indefinitely) than he might have been held if tried and found guilty. Such wide-ranging powers of social control are granted to psychiatrists for no other reason than that their judgments are perceived to be the diagnoses and prognoses of scientific and medical practitioners.

Can we not finally admit that such responsibility for predicting people's behavior cannot be entrusted to psychiatrists--not because the so-called medical science of behavior is underdeveloped but because it is not and never has been a medical science at all? Perhaps the law as it develops over the next few years will prove to be the final chink in the medical armor of psychiatry, and society will finally have to face the fact that we really do not and cannot know medically or scientifically why Hinckley emptied his gun in Washington or why a boyfriend maimed his girlfriend in Virginia or why the mental patient committed a brutal rape in Pittsburgh--or even whether any or all of them might act dangerously again.

So let us support the psychiatrists who want to be free of liability for the crimes of their patients; but let us insist in return that they give up the pretense of medicine and simply acknowledge their limited (and non-medical) ability to help some people who wish to understand and change their behavior. After all, such prediction is no more than soothsaying.