Of course, it's a good question, a question that Darwinians should continue to struggle with (as they have), notwithstanding the inherent difficulty of discerning an evolutionary path that has been lost in the mists of prehistory. Still, there is about this basic question. It is straight out of Creationism 101.
Doesn't Behe deserve some credit for applying the question to new things—such as blood clotting? Sure. Such new applications can be productive. In this case Behe's probing led the Brown University biologist Kenneth Miller to show that, actually, there is a fair amount of evidence—in our species and others—suggesting how the blood clotting mechanism could have evolved incrementally. Unfortunately for Behe, Miller showed this in the course of a powerful critique of Behe's overall argument, in a chapter of Miller's book Finding Darwin's God. (For a book-length critique of the intelligent design movement see the philosopher Robert Pennock's Tower of Babel. For Behe's reply to Miller, see Science and Evidence for Design in the Universe, co-edited by Behe.)
Behe's (and Darwin's) basic question—What good is half an eye, or half an anything?—is now getting more tractable thanks to genome studies. For example, scientists have found a gene involved in eyesight that is shared by humans and one-celled creatures. Apparently, in one-celled creatures the gene confers nothing like vision, but does confer a vague sensitivity to light—thus showing that, actually, much less than half an eye can be good for something.
Incidentally, the first time I opened Behe's book, I came upon a about Darwinian theory—a confusion on par with, and in fact related to, the confusion of Phillip Johnson's described above. It's no wonder Behe can't imagine how natural selection could create complex things if he hasn't bothered to find out how natural selection is supposed to work in the first place.
Intellectual father No. 3: William Dembski, a mathematician at Baylor University. Dembski, according to the Times, has developed a "mathematical 'explanatory filter' that he asserted can distinguish randomness from complexity designed by an intelligent agent." And Dembski, applying this litmus test to organisms, finds them to fall in the latter category. Now this, unlike Behe's argument, does sound new. Is it significant?
First of all, devising a test that shows that organisms aren't randomly arranged molecules is a curious way to spend time. After all, no one ever said that natural selection produces random conglomerations of matter. Rather, it is said to produce complex, functional arrangements of matter. In fact, according to evolutionary biologists, it produces arrangements that look for all the world as if they were composed by an intelligent designer. So, even if Dembski does have some test that can determine whether a being's complexity is of the precise sort that an intelligent designer would produce, that won't help his cause. For that is evolutionary biologists expect to find in the first place. (Of course, you can argue that they're wrong—that natural selection can't produce this kind of complexity. But then you're back to the Behe-esque arguments—Creationism 101—and Dembski's mathematical rendering of the issue hasn't changed the state of play.)
Surely, you say, Dembski must be saying something more sophisticated than this. After all, he is at an accredited university! Well, I've now spent about an hour and a half interrogating him, trying to find the sophisticated thing that he's saying, and I have failed. (Click for details on my exercise in futility.) So far as I can tell, Dembski's argument is just an example of something demonstrated time and again in various disciplines at various accredited universities: If you phrase your argument in mathematical symbolism and technical terms, some people, including other academics, can be counted on to lose track of what the exact connection is between the symbolism and the reality it's supposed to represent. Then they may conclude that your mathematical model proves something—e.g., that natural selection couldn't have produced life as we know it—when in fact that's what your model assumes.
In sum: So far as I can tell, all the major components of "intelligent design theory" are either not new, not significant, or just wrong.
The Times piece was a legitimate news story. The "intelligent design movement"is having impact—getting the attention of school boards, legislators, and, obviously, journalists. And the Times is right to say that intelligent design theorists are "more sophisticated" than past creationists in the sense that most of them don't believe the Earth was created a few thousand years ago as described in Genesis. Some of them even believe evolution happened—albeit with. Still, in the movement's critique of Darwinian theory, there is no sign of any new sophistication—at least, not in any positive sense of the word. "Intelligent design theory" is just a fresh label, a marketing device—and, evidently, an effective one.