Here's another excellent resource for timely updates on the Dover trial. The ACLU of Pennsylvania has set up a blog with frequent updates on what is going on in the courtroom. Jonathan Witt of the Discovery Institute is also blogging live from the trial on the DI blog. His post on Ken Miller's testimony yesterday was rather off the mark, as one would expect. He makes the superficially compelling argument that Ken Miller argued both that ID was not falsifiable and was falsified. But this ignores a fairly obvious logical distinction. Witt writes:

In friendly questioning from the plaintiff, Miller asserted that the theory of intelligent design was "not a testable theory in any sense" and so wasn't science. Later, however, Miller argued that science has tested Michael Behe's bacterial flagellum argument and falsified it, by pointing to a micro-syringe called the Type III Secretory System, and arguing that it could have served as a functional step on the gradual, Darwinian pathway to the full flagellar motor.

Did the journalists covering the trial notice the contradiction? Miller tried to provide a fig leaf for it, but the fig leaf was itself a misrepresentation. Miller said Behe's argument was in every respect a negative argument (and, further, that ALL the leading design theorists' arguments he was aware of are purely negative, with nothing positive anywhere). Miller conceded that Behe's irreducible complexity argument was testable, but said Behe's inference to design doesn't follow from irreducible complexity because Behe was committing the either/or fallacy—If not A (Darwinism), then it must be B (design). Miller said there were, in principle, an infinite number of other possible explanations, so jumping from a refutation of Darwinism to design was illegitimate.

He's missing a crucial distinction by conflating Behe's argument for ID with ID itself. The notion that an intelligent designer was involved is not in any way falsifiable. There is no conceivable set of data that could falsify that proposition. But specific arguments that purport to point to such a designer can be falsified, and it's important to distinguish here between facts and theories. Behe's argument offers both factual claims and a theoretical or explanatory claim. It goes like this:

Factual claim: Some biochemical systems are irreducibly complex, meaning that if you took out any single component of the system, the system would fail to function.

Factual claim: Irreducibly complex systems could not have evolved step by step because the intermediate or precursor systems would not have been functional.

Explanatory claim: Therefore, when we find an irreducibly complex system, we know that it must have been designed from scratch and came into existence all in one step.

Only the explanatory claim is an explicit statement in support of ID, but one can still falsify the argument if one shows that either of the two factual claims it is based upon is false. For instance, when we look at Behe's example of the blood clotting cascade, we can falsify it simply by looking at the first factual claim. Is the blood clotting cascade irreducibly complex? The answer is no. There are animals who lack one of the components of the system, yet their blood clots just fine. Dolphins, for example, lack Hageman factor (or Factor 12). By Behe's definition of irreducible complexity, this should be impossible. The fact that it's not shows that this is not, in fact, an irreducibly complex system.

Likewise on the bacterial flagellum, Behe's favorite example of irreducible complexity, the fact that one subset of the system works well for another function shows that the second factual claim in Behe's argument is not necessarily true. We have lots of examples in molecular biology of components for one system being adapted or co-opted for use in a different system. Even Behe would admit as much. Lots of examples, for instance, of a given gene duplication resulting in the production of two proteins, one of which is then coopted for a different function in a system it was not originally involved with inside the organism. So when we see that the flagellum includes a subset that functions well in a different type of system, we can reasonably infer that perhaps it was coopted in exactly the same way. Add this to the fact that we in fact have multiple different types of flagella at work in the bacterial world, suggesting that rather than being irreducibly complex there are multiple different ways to get to the same result, and you have good reason to think that Behe's second example fails because the second factual claim may well not be true.

So Miller is in fact correct when he says that ID is not falsifiable, while specific arguments for ID have been falsified. He's also correct to say that there is not positive theory of ID, only a set of negative arguments or criticisms of evolution. As Miller points out, this is the either/or fallacy at work - "if not evolution, therefore God". Witt tries to debunk this argument by pointing to positive statements from Of Pandas and People, but these are easily debunked. He writes:

"If experience has shown that a certain class of phenomena results from intelligent causes and then we encounter something new but similar, we conclude its origin also to be from an intelligent cause."

It seems a reasonable argument, but the analogy is very poor for one obvious reason: we have no experience with supernatural designers. IDers love to make this analogy between human designers and the "intelligent designer" and to pretend that the "intelligent designer" doesn't necessarily have to be supernatural, but that is completely false. Their own definition of ID shows this to be false, as I've argued many times without refutation (including DI's John West, who ignored it completely). The DI's own definition of ID says:

The theory of intelligent design holds that certain features of the universe and of living things are best explained by an intelligent cause, not an undirected process such as natural selection.

Their own definition combines biological ID with cosmological ID, which means the designer is responsible not only for living things, but for creating the universe itself. The DI's writings on cosmological ID make clear that when they say "certain features of the universe", they mean the nature of the universe itself - the nature of nature. This pretty much closes the door on their mantra-like citation of aliens as possible "intelligent designers". If the argument is that the universe was designed with the ability to sustain life, invoking alien life as an explanation is clearly absurd - alien life would be an effect of that universal design, just as human life is, not a cause of it.

(Incidentally, let me add that I have no problem whatsoever with the claim that the universe is designed to support life. Indeed, this is something I agree with. I am a deist and I do believe that the universe was created with the attributes to allow life to begin, exist and evolve when the conditions are right. But not only does this have little to do with whether evolution is true or not, I would maintain that there is a tension between accepting cosmological ID and rejecting evolution. As my friend and colleague Howard Van Till likes to point out, the ID movement believes in a God who created the universe with the ability to sustain life, but did so poor a job of it that he had to continually intervene to make sure it happened.)

And I am not alone in making this argument. William Dembski himself has admitted that the intelligent designer could not be natural but must be supernatural, in an August 1998 article called The Act of Creation, published on ARN:

The complexity-specification criterion demonstrates that design pervades cosmology and biology. Moreover, it is a transcendent design, not reducible to the physical world. Indeed, no intelligent agent who is strictly physical could have presided over the origin of the universe or the origin of life.

So the only "positive statements" of ID "theory" that Witt can come up with are in fact weak inferences based upon a terrible analogy between objects designed by human beings and objects allegedly designed by a supernatural being unconstrained by the laws of nature. It is the invocation of the supernatural that makes ID outside the realm of science. Science does not deal with supernatural explanations, not because of an a priori rejection of the supernatural, but for the very practical reason that there is no means of testing such explanations, no means of distinguishing between true and false explanations of that type.

And because the history of science clearly shows that phenomena that we once believed could only be explained by the kindness or anger of God were eventually explained, through science, in purely natural terms. Where once we could only explain bad crops or disease as the result of God's anger with us, or good crops and good health as proof of God's pleasure with us, we now know the full range of environmental and microbial causes of those outcomes. Where once we could only explain earthquakes or hurricanes as proof of God's judgement upon us, we now understand the natural causes of such events well enough to predict them with a high degree of accuracy.

And this points up exactly why supernatural claims are not falsifiable - because no matter what the outcome, it can be ascribed to God's whims. Good crops or bad crops, good weather or natural disaster, sickness or health, all can be easily "explained" by reference to the whim of a supernatural being with the power to manipulate nature. But because we didn't stop there, because we didn't accept this unfalsifiable explanation and continued to search for solid and testable explanations for these natural phenomena, we now have modern medicine that has extended human lifespans greatly. We now have modern agriculture that feeds billions, and modern seismology and meteorology that saves countless lives through their ability to predict disasters before they happen.

Science is ruthlessly practical. Give scientists an explanation that works and they'll run with it. But explanations that invoke the supernatural have never achieved anything in science, while rejecting those claims and continuing to search for natural explanations has literally transformed our world. Small wonder, then, that scientists - including that sizable portion of them who do believe that God, and therefore supernatural entities, does exist - insist upon continuing the search for natural explanations rather than giving up and accepting the untestable and unknowable as an explanation and thereby ending the search that has proven so fruitful in the past.