Well, not only is that not a prediction
It sounds like a prediction to me: if we look hard enough, we will find an unmistakable message encoded in DNA that intends to tell us that only God can make a tree.
(because it is not derived from a logical argument)
I don't think Virtually anything of any great general interest, that I am aware of in science, is derived from logical argument.
but it isn't testable because the notion is too ill defined. Is he being more specific than I infer from your comments?
You could read for yourself the publishers (I presume) comments just above. I'm undecided, but this is probably just a sort of inverted variation on Behe's defense: just because you can't find this message from God in the DNA, will never prove it ain't there.
When Isaac Asimov used to get cornered at booksignings by creationists, he would begin to feverishly pitch the theory that animals spoke the King's English 10,000 years ago. Being Isaac Asimov, he could, of course, present reams of evidence in support of this theory, which he insistently presented, until his protagonists retreated. Who knows, perhaps animals did speak the king's english in 8000BC. After all, evidence is evidence, right?
I might not have been clear. I was saying that it isn't a *scientific* prediction because it doesn't follow from any theory (or at least I doubt he has so derived it). Obviously there is more to science than predictions, things like facts, theories, laws, experiments.