Winston Churchill called it his "black dog." British medical biologist Lewis Wolpert has described it as "the cancer of the emotions." Once known politely as melancholia, it is more often referred to these days as clinical depression, and it has been estimated that as many as two-thirds of sufferers, even in developed countries, go undiagnosed and untreated. Here in Japan, concern about the steadily rising suicide rate has prompted the Health, Labor and Welfare Ministry to sponsor studies of better ways to identify and combat depression.

But there's the rub. Precisely because the causes of depression are unclear, there is still no real consensus about the most effective treatment for it. The man in the street is likely to advise a depressed friend or colleague to "pull himself together," take a vacation, or get more sleep and exercise. Some old-fashioned doctors may tell patients the same thing. Others advise alternative remedies: acupuncture or herbal regimens. Most, though, prescribe antidepressant drugs such as Prozac and Paxil, either alone or in combination with therapy. Numerous studies to date have indicated that the drugs do help people feel less depressed, although why they do is not clearly understood. At any rate, they are the closest thing we have to a generally agreed-upon treatment for depression.

That is, until this month, when a news story shattered -- or seemed to shatter -- even that emerging consensus. Last month, a Seattle psychiatrist published the results of his analysis of the placebo effect in scores of clinical trials undertaken between 1979 and 1996 by U.S. companies seeking government approval of antidepressant drugs. According to Dr. Arif Khan, sugar pills worked as well as the antidepressants more than half the time, and, what's more, they triggered changes in the same parts of the brain as the medicines did.