It seems elementary, but a science -- like the Party -- must have unity, at least in the fundamentals. Chemistry cannot have two periodic tables. Physics will not permit believers in 19th-century ether. Alchemy is not an elective at MIT.

Indeed, perhaps the most important event for the development of a science is the dying away of its schools. Biology could not mature if split between Darwinian and Lamarckian schools of heredity. It was not until that battle was settled -- the Lamarckians fell on their swords and were carted away -- that biology could take off as a science.

Which brings us to psychotherapy, the science of the talking cure, now turning a ripe 100. Psychotherapy has had quite the opposite development. It opened for business, as it were, a century ago next spring, when Sigmund Freud opened his consulting room at Berggasse 19. What began in one great mind and one great room has proliferated wildly into . . . well, let the Great Phoenix Gathering tell the story.

A few weeks ago in Phoenix an extraordinary conference was held to mark the anniversary of the opening of Freud's office. It was called, optimistically, "The Evolution of Psychotherapy." Seven thousand psychotherapists showed up to see and hear the largest assembly of gurus in history.

It was the greatest concentration of psychotherapeutic talent to gather in one place since Freud dined alone. The leaders of every major school, more than two dozen in all, were there. Rollo May, Bruno Bettelheim, Virginia Satir, R. D. Laing, Carl Rogers. They represented every technique: Freudian therapy, family therapy, behavior therapy, existential therapy. Even Thomas Szasz was there, representing, I suppose, pseudo-therapy, since he believes that mental illness is a myth. (Szasz once outlined his approach to the patient who comes to him and says he is Jesus: "I say he is lying.") For a science, evolution means development toward some deeper unity. This jamboree of jousting sects and one-man shows might more properly have been called the devolution of psychotherapy. It showed what these hundred years have wrought: "a babel of conflicting voices," to quote Joseph Wolpe, a founder of behavior therapy.

Psychotherapy has come upon this state of confusion because, true to its healing, understanding soul, it permits too few deaths among its schools. It is incapable of killing its own. Psychotherapy is dying of dilution.

So what? Business is good and the intellectual ferment brings new techniques ("a new crop every year from California," noted Wolpe wryly) to serve new patients. Who cares whether psychotherapy is a science? Let's see. A few intellectual purists. A few nostalgics, who respect Freud's original vision of psychoanalysis as a scientific technique.

Oh, yes. And one 800-pound gorilla: the insurance companies. As psychotherapy grew more popular, it grew more expensive for insurers. By the mid-'70s, with every psychotherapy school claiming incomparable (in both senses of the word) results, and with bills mounting and premiums rising, insurers began cutting coverage. But finding no way to separate the elite from the quacks, they cut the subsidy to all the schools.

Fifteen years ago in Washington, you could get insurance to cover 80 percent of unlimited psychotherapy. Around then, when I told a psychiatric colleague (at the time, I was a psychiatrist) that I would be coming to Washington to work for the government, he smiled and said, "Now you can get the Big Tune-up." I was puz zled. He explained: With insurance paying 80 percent and no limit on visits, why not go for it Five-days-a-week psychoanalysis. Redo the engine.

I answered that my engine felt okay, and I did not want anyone poking around under the hood. But today it wouldn't matter. The Big Tune-up is gone. You can barely get a lube job. Insurers have generally cut coverage to 50 percent, with severe limits on visits.

As long as psychotherapies resist pressure to produce scientific evidence that they work, the economic squeeze will tighten. After all, if psychotherapy is really an art, it should be supported by the National Endowment, not by Blue Cross.

The first to face economic extinction will be the longer-term therapies, such as, ironically enough, Freudian analysis. Where it ends, though, is not clear. My hope is that society will not totally abandon support for psychotherapy as a form of treatment. In my own experience, some psychotherapies (behavior therapies, in particular) helped my patients, some dramatically. But mine is anecdotal evidence, and there is not a school that cannot produce a bagful of glowing affidavits. What is needed is real science.

Unfortunately, psychotherapy shows little sign that it is inclined to reverse the direction of its disastrous anti-scientific evolution. Phoenix didn't help. In fact, it makes clear that, as an intellectual and perhaps soon as an economic enterprise, psychotherapy in its 100th year is deep into its twilight.