Friday, June 17, 2005

Pinkerton and the Argument from Ignorance

In his Tech Central Station column "The Real Intelligent Designers", James Pinkerton characterizes and rebuts Michael Behe's argument in Darwin's Black Box like so:

Behe argues that it just isn't possible that random evolution could have produced the flagellum -- the propeller/tail -- on a bacteria. Such an organ, he concludes, is "irreducibly complex," which is to say, only a Master of Complexity could have created it.
But it's a fallacy to argue that just because one person -- or even all the people of an era -- can't figure out how something works, therefore such mysterious workings are beyond any human comprehension, ever. To take one humble example, years ago I saw Siegfried and Roy perform their tiger-based magic in Las Vegas, and was frankly astonished at some of the illusions they generated at the aptly named Mirage casino. I had no idea how they did their tricks, but since I knew that they employed mechanics, not metaphysics, to do their show, I was content just to enjoy it, marveling all the while at human ingenuity. And of course, if one waits long enough, he will get a peek behind the conjuring curtain, learning how tricks are done and also that like the rest of us, Siegfried and Roy suffer from Murphy's Law, too. And so it is with science: eventually, some scientist will figure out how the "trick" of the bacteria's flagellum is done.

According to Pinkerton, Behe argues something like this:

(1) Behe "can't figure out how something works" (in this case, how gradualistic natural selection could produce the flagellum);
(2) therefore, no-one can ever figure it out.

Though Pinkerton doesn't name this fallacy, it is clear that he is accusing Behe of offering an argument from ignorance--one of the fallacies most commonly attributed to critics of Darwinian evolution. In fact, the charge of arguing from ignorance is a mainstay of those critiquing ID (see, e.g., 'anonymous's comment at 6:40 AM on this posting of Macht's). But is that really Behe's argument? I don't think so.

It is one thing to assert that a certain explanation of a phenomenon should be rejected because one cannot figure out how it can be true; it is quite another to assert that it is impossible for that explanation to be true, and then to give positive reasons why it is impossible. Behe does the latter. And one does not refute the latter sort of argument by claiming that the arguer is arguing from ignorance--one refutes it by (a) showing that the reasons given are false, or by (b) showing that the reasons given don't entail the impossibility of the explanation in question, or by (c) showing by demonstration that the explanation can indeed account for the phenomenon, that is by reproducing the phenomenon according to the explanation.

Suppose person A asserts that it is impossible for person B to jump unaided over a 6-story building, and A argues for her assertion by citing what she takes to be the relevant laws of physics and what she take to be the relevant properties of the objects concerned (B, the building, the earth, etc.). B doesn't refute A's argument by declaring that she is arguing from ignorance; B refutes it in one of the ways (a), (b), or (c) I noted above. Either he defeats one or more of the reasons A gives in her argument, or he shows the form of A's argument to be logically invalid, or he demonstrates that she is wrong. Suppose B is from planet Krypton and can easily leap over buildings much taller than 6 stories. B could provide A with evidence that he is from Krypton, such that A is persuaded that her views of the relevant properties of B were false, and that B does indeed possess the strength to overcome gravity and friction to the point of being able to leap over the building. Or, alternately, B could simply leap over the building in A's presence, thus obviating any need for argumentation (and compelling A go review her premises and the logical form of her argument). But as a matter of discourse, B doesn't refute A's argument merely by asserting that she is arguing from ignorance.

But the form of Behe's argument is exactly like A's. Behe asserts that something is impossible (namely that step-wise gradual evolution accounts for certain phenomena). And he gives reasons in support of his conclusion. Perhaps some of his reasons are untrue. Perhaps the form of his argument is invalid, such that even if his premises are true, the conclusion doesn't follow from them. Perhaps someone is prepared to show by demonstration that he is wrong. One would refute his argument by addressing any of these flaws. But if one simply declares that Behe is committing a fallacy by arguing from ignorance, then one has failed to address his actual argument, and it seems to me you can't refute an argument without addressing the points it makes and the way it makes them.

In fact, Pinkerton in his TCS column mischaracterizes Behe's argument so badly that he fails to identify it altogether. What for Behe is an argument about the way things are and work (an argument in the domain of ontology, if you will), becomes in Pinkerton's column some kind of epistemological train wreck: a bad argument about what we can and cannot know. It's as if he went out to catch the 10 lb. catfish lurking at the bottom of the pond, and returns with a brace of quail: he may be having dinner tonight, but he didn't do what he set out to do. I do not attribute this to malice on Pinkerton's part--this kind of objection to intelligent design, creationism, and to other movements critical of Darwinism, is so prevalent as to be a almost an instinct, a reflex of mind and speech among critics of ID. One observes that they cannot restrain themselves from transmuting their opponents arguments (be they good or bad) into something else altogether. Perhaps anti-creationists do believe in alchemy after all.

One can imagine ways to formulate an valid kind of argument against Behe's position. Here is an example:

(3) Behe claims that there are reaons that structures like the flagellum cannot be produced by gradualistic evolution;
(4) However, there are reasons (of which Behe is ignorant) that make it possible for structures like the flagellum to be produced by gradualistic evolution;
(5) Therefore, Behe is wrong in his claim about structures like the flagellum;
(6) Therefore, either some of the reasons Behe gives for the impossibility of gradualistic evolution's accounting for the flagellum are wrong, or his conclusion does not follow from his reasons.

Perhaps Pinkerton really has something like this in mind. Note that it (unlike Pinkerton's TCS argument) addresses Behe's actual argument, if only in general terms. It does have one shortcoming: (4) does not follow from (3), and therefore it needs a sound sub-argument of its own to establish it, and thus to make the whole argument sound. If one could produce a sound argument for (4), then Behe's claims would be refuted. I suspect that people have provided reasons that are candidates for supporting (4), and I suspect that Behe and others have responded trying to show why the reasons given don't amount to a sound argument for (4). I know little of the scientific reasons themselves on either side, but I do know that the real action in this question lies in their neighborhood, and not in the neighborhood of epistemology, as Pinkerton seems to think.

The upshot is that Pinkerton in his TCS column so badly mischaracterizes Behe's argument that his critique of it misses the point entirely and is thus without force. He does give brief mention to Behe's concept of irreducible complexity, but Pinkerton seems to say that "irreducible complexity" is just another way of saying "only a Master of Complexity could have created it". Perhaps he is just being careless with his words. Perhaps we are meant to take "which is to say" to mean "from which Behe infers that", which would be a more accurate, if still very incomplete characterization of Behe's view.

The un-aptness of the analogy of the Siegfried and Roy show that Pinkerton uses to refute the argument from ignorance would also be worth a discussion, but I'm not sure when I will get around to that.

I regard this entry as a first draft of my treatment of the trope of the argument from ignorance. I welcome comments by email that will help me improve my analysis.

For some reason, Timothy Sandefur at The Panda's Thumb thinks that Pinkerton's article is excellent.