Medical education has always faced a doubly difficult challenge: How to keep up with increasingly specialized medical knowledge without losing sight of the physician's fundamental role of caring for whole patients.

In short, how to teach both the science and the art of medicine -- the mass of facts and the laying on of hands.

Nearly a century after the founding of the prototypical modern medical school at Johns Hopkins, the dilemma persists.

"We've turned medical school into four years of cramming," said Dr. Milton Corn, dean of Georgetown University School of Medicine. "We can't continue to believe we can close our eyes to this and simply throw students into a sea of information probably beyond the power of any human to understand."

The scientific revolution is so rapid that biomedical knowledge has "a half-life of about five years," said Dr. John A. D. Cooper, president of the Association of American Medical Colleges. That means about half of what is taught today will be obsolete in five years.

The solution, experts agree, is not to try to teach it all, but to make doctors life-long learners.

"A brilliant student who read nothing after graduation," Corn said, "would be a hopeless charlatan within five years."

"Perhaps the most important concept emanating from this study," concluded the AAMC's 1984 report on the state of medical education, "is that medical students must be prepared to learn throughout their professional lives."

In many medical schools, "lectures could be reduced by one third to one half," the AAMC report said. It also recommended that schools "limit the amount of factual information students are expected to memorize," require broad study in the social sciences and humanities and emphasize "the acquisition and development of skills, values and attitudes by students at least to the same extent they do their acquisition of knowledge."

That's hardly a new idea. William Osler, the famed Johns Hopkins professor, wrote in 1905 that medical education must be "a life-long process." And a few years later, Harvard's Richard Cabot bemoaned medical schools' failure to give students "the preparedness for treating a human being as if he possessed a mind, affections, talents, vices and habits good and bad, as well as more or less diseased organs."

But the accelerating pace of information gives the idea new urgency.

"Current medical education is rooted in 19th century mythology, where students were expected to learn the sum total of all medical knowledge," said Jeffrey Stolz, a third-year student at the University of North Carolina School of Medicine at Chapel Hill and president of the 28,000-member American Medical Student Association. "There's an amazing overload of information.

"We need more interactive teaching and less time spent in the darkened lecture hall listening to a monotonal voice for hours on end."

Such criticisms echo those of Abraham Flexner, whose landmark report to the Carnegie Foundation in 1910 helped reform medical education and who later warned that "a student staggering under such a burden cannot, except sporadically, stop to read, work or think at random."

The Flexner report was a muckraking critique of inferior medical schools -- the dissecting room of one medical school "did duty incidentally as a chicken yard" -- and hastened the demise of many diploma-mill medical schools.

Flexner's philosophy of medical education was simple: "The student no longer merely watches, listens, memorizes; he does." "Are we overdoing factual information, as opposed to problem-solving?" said Dr. L. Thompson Bowles, dean for academic affairs at George Washington Medical Center. "Most of us think we are."

But the details of the curriculum, he said, are less important than the general environment in which learning takes place: "It isn't how many hours of pharmacology or how many labs that matters, but the quality of the faculty-student interchange."

The number of applicants to Johns Hopkins jumped by 50 percent this year, after Hopkins waived the traditional medical college admissions test (MCAT), which emphasizes scientific knowledge, as an admissions requirement. Dean Richard S. Ross said Hopkins dropped the MCAT requirement because of a belief that not only scientific skill but also "broad general education, humanism and the properties of personality and character" are what make good physicians.

"Medical practice and medical education are intimately related," said Dr. Kenneth M. Ludmerer, professor of medicine and history at Washington University in St. Louis, whose recent book, "Learning to Heal," traces the history of medical education in America. "Doctors practice medicine the way they're trained to do it."

In an era when changing economics and technological and legal pressures are challenging some of the basic tenets of traditional medical practice, that training is all the more crucial.

"Basically there are two ways you can practice medicine," Ludmerer said. "You can order everything there is, because it's there. Or, you can order what is necessary.

"The latter process involves a much more difficult, scientific, critical way of thinking."

But merely tinkering with the curriculum won't produce that kind of critical thinking, Ludmerer said. Medical schools must encourage and reward good teaching as forcefully as they reward pioneering research -- with salary increases, promotions and tenure.

"I don't think the great William Osler, the greatest medical clinician ever, could receive tenure in an American medical school today," said Ludmerer, "because he wasn't an investigator."

A hundred years ago, before Osler and others transformed the modern medical school, American medical education was slipshod at best. Admission requirements were lower than for a good high school. As late as 1900, Johns Hopkins was the only American medical school requiring a college degree. There were no written examinations, which was fortunate for the large percentage of illiterate students. Instruction, all by lecture, took place during two 16-week terms.

"Indeed, except for the enterprising and affluent few who would bribe patients into submitting to an examination," writes Ludmerer, "students would often graduate without ever having touched a patient."

Even at Harvard Medical School, any applicant who could pay the fees got in, and to graduate, a student had only to pass five of nine five-minute oral quizzes at the end of the second term of lectures.

"The ignorance and general incompetency of the average graduate of American medical schools, at the time when he receives the degree which turns him loose upon the community, is something horrible to contemplate," lamented Charles Eliot in his first report as president of Harvard in 1870. Eliot set out to reform the medical school, emphasizing scientific training, critical thinking and "learning by doing."

But when Eliot recommended written examinations as a requirement for graduation, a Harvard professor of surgery vehemently objected, protesting that more than half of Harvard's medical students could barely write.

Today, despite stringent entrance requirements and generally high academic standards, the debate over curriculum reform continues.

"We need an overhaul," said AMSA president Stolz. "Adding courses here and there has been the approach taken by most schools. What you end up with is a very cluttered curriculum."

Some medical schools have tried a bolder approach. Under Harvard's New Pathway curriculum, introduced last fall, a select group of 24 students skip most of the lectures that traditionally dominate the first two years of medical school. Instead, they meet in small groups, led by tutors, to discuss real-life case studies and decide what they need to know and how to go about learning it.

"I think what Harvard is doing is noble," said historian Ludmerer. "But I don't think it's new. They're doing what Charles Eliot wanted to do 100 years ago."

In an effort to broaden medical education and produce more literate, humane doctors, most medical schools have recently added courses in medical ethics, sociology and the humanities. But some critics are skeptical of such piecemeal additions. Dr. Gerald Weissmann, professor of medicine at New York University Medical Center, has dubbed them "remedial humanities" and said they are a "sorry substitute" for a solid liberal arts undergraduate education.

"I have serious doubts that tutored appreciation of the arts, of the humanities, or of philosophy will, like monosodium glutamate, enhance the flavor of a student's humanity," wrote Weissmann in an essay in his recent book, "The Woods Hole Cantata."

Nor would it help to do away with pre-med scientific requirements, said Georgetown's Corn, noting that philosophers are not necessarily any more compassionate or civilized than chemists.

"Well-read compassionate dummies are not what we're looking for," he wrote last year in "The New Physician."

Corn compared curriculum-tinkering to "rearranging deck chairs . . . The response has been to keep adding things. We just added and added. And now, as the [AAMC] report pointed out, the ship is full. There's no more we can add now.

"What we need to do is pare down the total set of facts to something that's manageable and will stand the test of time, and then teach the student how to supplement this and change and be an intelligent, well-informed, flexible physician."