Debunking is to science as criticism is to art—not useless, but not the thing itself, and often requiring a cast of mind opposite to what it tries to protect. A debunking is a rhetorical technique; a disproof is a logical achievement, or a scientific consensus. In the history of science there is hardly an idea worth general attention that has not met with a wave of debunkings. They rise in jealousy from the center of science, and recede in time to the bewilderment and resentment of the fringes.
A debunking, like any persecution, only strengthens what it attacks; and debunkers, in the long run, are enemies of science, because they belie its attractions. They make science seem progress through mutual abuse, where only the most smug, mordant, and venomous are worthy to possess truth. Debunkers, by their defense of science, do all to make science attractive that pundits do for politics. Most of their criticisms are reducible to copyediting. Take a scientific paper; confuse the punctuation and capitalization; add exclamation points; put an uncredentialed named on it—and they will crush it. Take a crackpot theory; copyedit it to a scientific style; abbreviate the name (J. Smith); add the name of a university or an institute, and they will not even to allow themselves to wonder about it.
Two prominent exceptions, where criticisms are substantial and urgent, are intelligent design and climate change denial. But these are not really pseudosciences; they are themselves debunkings. These two movement have rudely proven that the effect of a debunking has nothing to do with the truth of the proposal, only with the skill, prestige, and power (or powerful alliances) of the debunker. I do not believe that a debunker can defeat truth in the long run—corroborating evidence in time must strain the skill, undermine the prestige, and sour the alliances of the denier—but it can be held off for a very long time. A history of science can be constructed in which truth always promptly wins, only because, in the past, truth opposed in one place has always had somewhere else to win. Until the last few decades there has not been a single, international scientific community and consensus; only several separate national scientific establishments, which have served to correct one another. Only since the end of WWII has there been a single Western scientific establishment; and only since the end of the Cold War (or since Khrushchev, in some degree) has there been a worldwide scientific community. The great disadvantage of this single system is that it is difficult to shame it. For example: the wave theory of light originated in England (with Young), but was not accepted in England until after—through the work of Fresnel—it had been elaborated and accepted in France. At the extreme, in France, Voltaire had to satirize the Cartesian mechanical ether to advance the theory of gravity. Competition between universities can in some degree replace this international competition; but only in disciplines which do not depend on centralized sources of money—where independent budgets allow for independent thinking. Consider the slow acceptance of the role of angiogenesis in the development of tumors, in the costly world of medical research.
In the history of science, it is rarely well-known anomalies that necessitate new theories. Remember that Copernicus kept epicycles in his heliocentrism; that Kepler set out not to discover the shape of orbits, but the spacing between them; that Newton "made no theories" for the material basis of gravitation, at a time when the great question was that of the ether—a question which Einstein also fruitfully ignored. Cranks are drawn to easy solutions for most or all problems; scientists are drawn towards mines of new problems. It is easy to multiply after the fact theories of what is, or is not, science. But the behavior of scientists shows only one rule: scientists go where the work is. And while it is unusual for scientists to have to step back, and declare a body of work nonsense (caloric fluid, for example), it is almost the rule that as a new science advances, it goes from vague pretensions of revolutionary importance, to mere usefulness, or even footnote-filling triviality; and that, as theories mature, they surrender their ambitions, and ceasing to be projects of their own, end serving as instruments of old projects.
For the practical recognition of crankery and quackery, it is not necessary for the borders of science to be patrolled and enforced by debunkers—it is enough to avoid easy answers. The question to ask is not some berating, trolling "Why" ("If God designed us, why the appendix, coccyx, recurved spine?" "If global warming is natural, why do so many climatologists think it is not, why do the models overwhelmingly refute it, why changing patterns of vegetation, glaciation?")—because these are the same kind of questions that the other side is asking ("If evolution is random and undirected, why such useless complexity, why so many broken chains, why dogs can still interbreed?" "If global warming is anthropogenic, why no rush of disasters, why still harsh winters, why no one can agree on what would happen even if it were true?")—and because the answers would not be scientific: "To keep us humble," "Institutional veniality, overconfidence, bad records," "Because so little survives, so little is seen in so short a time," "Because we've been lucky so far." The right question is simply—"And?" Science needs problems, science needs questions; so science cannot abide easy answers, science cannot settle for dead ends.