PZ Meyers vs. Alister McGrath

PZ Meyers shows a somewhat tenuous grasp on philosophy of science while dealing with an interview with Alister McGrath. The nice thing about PZ Meyers and his Pharyngula blog is that he tends to be wrong in a remarkably clear and consistent way. No difference here.

McGrath remarks, quite correctly, that:

I think Richard Dawkins approaches the question of whether God exists in much the same way as if he’d approach the question of whether there is water on Mars. In other words, it’s something that’s open to objective scientific experimentation. And of course there’s no way you can bring those criteria to bear on God. I think Dawkins seems reluctant to allow that God may not be in the same category as scientific objects. That’s an extremely important point to make in beginning to critique him.

PZ Meyers proves his point by stating that:

He's actually right on one thing: we are approaching the question of god as a scientific problem, and the question of water on Mars is a pretty good analogy. We can't see it here, we aren't there, we have to build a case on inference from evidence and we have to design tests to evaluate the possibilities. That's been an eminently successful strategy for humanity. So why can't we bring them to bear on the god question? I've highlighted his answer: he says we just can't. He doesn't say why we can't, it's just a dogmatic assertion. Keep this in mind, though, because he's going to contradict himself in a moment.

Also, I'd like to know what he means by this category of "scientific objects". Everything is a scientific object, from distant stars to grains of dirt, from the first picoseconds of the Big Bang to pillow talk between lovers. If we can ask a question about it, it can be science. McGrath may think this is a useful strategy for a critique, but all it amounts to is setting up his premises as unquestionable. We simply do not have to accept that.

Later on in the interview, McGrath states that:

As someone who has studied the history and philosophy of science extensively, I think I’ve noticed a number of things that Dawkins seems to have overlooked. One of them is this: One of the most commonly encountered patterns in scientific development is seeing a pattern of observations and then saying, in order to explain these observations, we propose that there exists something that is as yet unobserved but we believe that one day will be observed because if it’s there, it can explain everything that can be observed.

Of course, if you’re a Christian you’ll see immediately that that same pattern is there in thinking about God. We can’t prove there’s a God but he makes an awful lot of sense of things and therefore there’s a very good reason to suppose that this may, in fact, be right.

At which point PZ Meyers charges him with a contradiction:

Whoa. What happened to "of course there's no way you can bring those criteria to bear on God"? What about "God may not be in the same category as scientific objects"? One moment he's claiming you can't study god like you would the possibility of water on Mars, and next he's claiming the validity of using observation and theory to justify the existence of the remote and directly unseen. How … inconsistent.

Except that he's not. McGrath was talking here about an inference to the best explanation, or about what Peirce called an abductive inference - which by itself is only half the scientific method. The explanation would potentially fall within the domain of science if the explanandum could be observed and studied in an empirical fashion. But in the case of God, the kind of explanandum you'd be dealing with are not scientifically observable things but things which are the preconditions for us making any (scientific) observation at all - the existence of laws of nature (if indeed they are laws of nature - which in itself is not a scientific question); the problem of universals; the existence of mind, etc. There is no scientific proof for the existence of mind in addition to matter. It's highly reasonable to assume there is such a thing (as we have direct experience of mind, and only very indirect of matter, for one) but no empirical proof. Given a world-picture in which both mind and matter are co-existing, one could (reasonably- but without scientific proof) argue mind to be primary, and extrapolate from this to the existence of God. But this is not science.

For McGrath to argue that God is amenable to scientific study, he would have to argue not only the existence of God as the result of an abductive inference, but to argue empirical observables under the operation of regularities of nature - which are thereby falsifiable, and predictable. In other words, even a miracle would not do as a scientific observation, as miracles are pretty much by definition one-off breaches of the laws of nature. In natural science, of courses, empirical observables are automatically assumed to be subject to natural laws as the whole method (of prediction and experimental verification) depends upon it. So supposed empirical facts which, to the believer, call for the activity of God as an explanation, are by definition either outside the realm of science or, to an atheist, no proof of anything except the normal workings of the universe. However, it would not be a justification of atheism without a defense of some kind of empiricism as an underlying metaphysics.

Now, charitably, PZ Meyers could have used science in the sense of "natural science, the humanities, philosophy, and rational discourse in general" which does indeed just about envelop anything we can ask a question about. But he would be very wrong to do so - and I don't think he is.

What has happened here is that PZ Meyers bases his own reasoning upon a philosophical outlook - some kind of very naive, scientistic empiricism - without allowing for the possibility McGrath does not share that outlook. Which is a pattern I've seen a lot of atheists fall into. They start arguing that they are under no obligation to believe in God just as they are under no obligation to argue in the existence of a teapot circling Saturn. At which point one may counter that God is not a spatially and temporally bound object like the teapot, that God is not "in" the universe but rather the other way around, and that God is not subject to the laws of nature but rather the basis of them. And that, therefore, arguments on the existence of God need to take place in philosophical rather than scientific discourse. Usually the atheist will counter that there is as little need for him to acquaint himself with philosophy of religion as there is for goblinology or Saturnian teapotology - and we're back to square one. The atheist might even invoke the Courtier's reply which is another way of stating that the atheist is under no obligation to acquaint himself with what he's arguing against. When an atheist does this, it is of course a sign of the atheist's superior Reason. If a creationist fighting to remove evolution from the classroom were to make the same move, it would obviously be a symptom of the creationist's redneck anti-science idiocy.

Anyway. All this nicely illustrates the dangers of unconscious empiricism (an empiricism that does not know it is one).

And many principles which are quite appropriate for science and empirical inquiry are abused by cajoling them into philosophical service when used as arguments for atheism. Such as Occam's razor - which is just about unusable, as the relative parsimony of atheistic and theistic explanations is hardly statable. The explananda are often not even quite the same. Another is Sagan's claim that "extraordinary claims demand extraordinary evidence". Which tends to make my blood boil, as I've hardly ever seen appropriate use of it. In the normal process of science, scientists are usually quite justified to not abandon the theories they work with unless there is a lot of proof that they are indeed wrong. Suppose a well-argued and superficially plausible claim were to be made that the Americas were colonized out of Polynesia rather than the Bering land bridge. A "wait and see" approach would be very appropriate. This makes the scientific process somewhat conservative, which is probably a good thing.

The principle is, however, wholly inappropriate if used to evaluate "extraordinary" claims which do, in fact, contradict no established scientific theories. Which is the way it is commonly used by professional skeptics. The claim for the existence of ESP needs evidence, but no "extraordinary" evidence. Claims for the existence of God need argument (rather than evidence) but certainly no extraordinary evidence.

Because the problem in the latter cases is that we have no scientific grounds to state that a given claim is extraordinary. Based on the scientific consensus concerning the origins of Native Americans, and the chronology of the colonization of Polynesia, a proposed Polynesian origin would be very extraordinary. However, whether one finds ESP or God extraordinary depends on one's philosophical presuppositions. ESP may well fit easily in the philosophical framework of a dualist or a panpsychist, but would be somewhat more troublesome for a philosophical materialist. Similarly God.

Using heuristic or pragmatic principles such as Occam's razor or Sagan's "extraordinary claims" as principles to evaluate philosophies, world-outlooks, or things quite beyond current science, usually shows some kind of underlying naive empiricism. The problem with philosophical outlooks that are hidden from view is that they cannot be evaluated, and they cannot be criticized. In this sense, the explicit, reflective theist is usually in better intellectual health than the "pro-science" atheist.