I've been reading The Testosterone Advantage Plan, a diet and exercise book by Lou Schuler (yeah, I'm trying to get in better shape), and I came across a short passage that, surprisingly enough, is relevant to the subject of parapsychology. It involves the value of anecdotes in the scientific process.
Realize, too, that almost all important scientific discoveries have begun with small pockets of anecdotal data. For example, in the early 1980s, doctors in urban centers on both coasts noticed that male patients were trickling in with a strange disorder that seemed to be zapping their immune systems and, often, causing strange red blotches on their faces and other parts of their bodies. The doctors took personal histories and learned that almost all of these patients were gay. As the early trend swelled into an alarming epidemic, doctors theorized – based on the anecdotal information available to them at that point – that some terrible new bug was afoot in the gay community. As months passed, having little else to go on, researchers further theorized that this new syndrome was spread by homosexual contact. This, of course, was medical science's first hazy awareness of the modern plague that more intensive research later diagnosed as the AIDS virus. [pp. 46-47; emphasis in original]
As you know, one of the most popular criticisms of paranormal phenomena is that the evidence is largely anecdotal. This isn't quite true, since there is an impressive library of carefully documented case histories investigated by the Society for Psychical Research and its counterparts, and by other researchers. Still, it's undeniable that a good deal of the evidence for psi and life after death is anecdotal. And as the saying goes, "The plural of anecdote is not data."
Fair enough. But as the example from Schuler's book suggests, anecdotes are not necessarily worthless, either; in fact, they can be very valuable. In some cases, they provide important hints about what's going on – hints that, when followed up, can lead to harder data. Although Schuler gives only one example, he does note that "almost all important scientific discoveries" begin with anecdotal cases, and I suspect this is true. For instance, the discovery of meteorites began with farmers' anecdotal accounts of stones falling from the sky; the discovery of fossils began with anecdotal accounts of workmen digging up old bones; the discovery of citrus as a preventive measure for scurvy began with anecdotal accounts of sailors who avoided the illness by eating fruit; etc.
Anecdotes may not be data in the strict sense, but we ignore them at our peril.