It is often said that Intelligent Design (ID) isn't science because it does not lend itself to testing. There is, we are told, no way to falsify the claim that intelligence was involved in the development of living things and therefore the suggestion that it was is scientifically vacuous. This is the conventional wisdom, but it seems to me to be mistaken.
As Karl Popper instructed us, science proceeds by what he called conjectures and refutations. Scientists offer a conjecture that some phenomenon can be explained in a particular fashion and then other scientists set about trying to show that the conjecture is false. If there is no way that the conjecture can be tested then it's considered (by some but by no means all scientists) to be outside the domain of science.
As long as a testable conjecture withstands the scrutiny to which it is subjected, it remains scientifically viable. It remains a legitimate scientific hypothesis until evidence for its falsity, or its unliklihood, accumulates to the point where a consensus of scientists becomes persuaded that the conjecture is improbable. At that point it doesn't cease to be a scientific hypothesis, rather it falls into the category of "bad science." Even this is not a death sentence, however, since future investigators may resurrect it as new evidence becomes available and the old conjecture, perhaps modified, is seen as a compelling explanation for that new evidence.
Intelligent Design, or at least that argument for it based upon irreducible complexity (IC), lends itself nicely to the Popperian model.
Let some structure, system, or process (S) be adduced which exhibits at least prima facie design in the form of what appears to be irreducible complexity. The ID position is twofold: It asserts 1. that there are at least some (S) that are irreducibly complex, and that, 2. if these instances of (S) do in fact exhibit IC then whatever processes might be ultimately responsible for (S) must include as part of the mix intelligent guidance or agency (I). The latter claim is not especially controversial, but the first claim is.
Thus it is this claim that biologists must refute in order to discredit IC (and derivatively, ID), and this claim, that some (S) are irreducibly complex, is testable.
Once the claim that a particular (S) possesses IC is made then biologists set about trying to come up with a plausible pathway or set of mechanisms that could have produced (S) independently of (I). The solution must be plausible and not merely logically possible.
If they are successful in formulating a plausible naturalistic pathway for the construction of (S) then the ID option in this case will be effectively refuted. If they are repeatedly successful over a wide range of examples of (S) then ID, at least in its biological dimension, will, over time, be seen as falsified. Failing such a demonstration, however, the conjecture that (S) could not have arisen apart from (I) should be accepted as a valid scientific hypothesis.
Indeed, it is this very process of seeking to refute the conjectures of irreducible complexity that biologists are undertaking in labs all across the country. The most recent example is the work of Bridgham, et al. at the University of Oregon whose paper is being touted as a refutation of claims of IC in the endocrine system (ID theorists maintain that their work does no such thing).
Whether the Bridgham paper accomplishes what its promoters say it does or not, the important point here is that it argues that ID, at least in this one instance, is false. The critics of ID can't have it both ways, however. They can't say that ID isn't falsifiable while simultaneously crowing that it is false. There is irony in the fact that the very attempt to show that ID is false confirms that ID can be tested and is therefore a legitimate scientific hypothesis.
Judge John Jones might not think ID is science, the witnesses who appeared for the plaintiffs in Kitzmiller v. Dover may have argued that it's not science, but real scientists are certainly acting as if it is. They're trying very hard to falsify it, and this is exactly as it should be. It's what science is all about.