Over the past weekend, I called into the Atheist Experience, a TV show based in Austin, Texas, hosted by atheist activist Matt Dillahunty. It is a weekly call-in program, airing Sundays from 4:30 to 5:30 pm Central and can be viewed on ustream. The recording of my discussion with Matt (I am the first and longest caller) is accessible online (mp3, video).

Informed and intelligent theists rarely call into the program. It is unfortunate that the vast majority of callers are not knowledgeable about their own beliefs, let alone how to defend those beliefs rationally. I am not quite sure why this is, but I would speculate that most educated people would feel that the Atheist Experience does not offer a fair setting for discussion. It is not easy to have such a debate when your opponent has control of the microphone, when there is no impartial moderator, and when your interlocutor repeatedly interrupts you mid flow, bringing up multiple objections and then giving you about three to five seconds at a time to make your case. I understand that they cannot have someone rambling on and on endlessly, but the limitation ought to be reasonable.

At the end of our conversation, Matt accused me of employing circular reasoning. When I started to respond, he cut me off, saying “We’re done.” But this was a key point in the conversation. Overall, the experience suggests to me that Dillahunty is more interested in making himself look good in the eyes of his fans than in the objective pursuit of truth.

The first portion of the discussion concerned the nature of faith. I objected to the way that atheists typically use the word faith, which is in stark contrast with the way it has been used both Biblically and historically. The word “faith” is really a synonym for “trust,” and indeed the Greek word in the New Testament that is translated “faith,” pistis, has just that connotation. Believers are encouraged to have an intellectually responsible and rational faith.

Beyond this, Matt asserted that science cannot provide support for the supernatural. My response was to say that science can provide support for a theologically neutral premise which forms part of a philosophical syllogism that in turn provides support for the supernatural. Certain observations about the world are more to be expected given theism than given atheism. Such observations include the discovery that the universe has a finite space-time boundary, the finely tuned character of the cosmos, and the role of information in biological systems. We then moved on to discuss specific arguments — the fine-tuning argument and the intelligent design argument. I am not going to address every point here, but I do want to comment on a few things that came up that deserve fuller elaboration.

First there is Matt’s frequentist approach to probability. Whereas (as I pointed out) physicists are in wide agreement that the parameters of physics could have had a wide range of values, Matt argued that this cannot be demonstrated because we only have one universe to observe. Such an argument, of course, presupposes a frequentist view of probability that, although useful in some contexts, in other contexts has its limitations.

For example, if we are considering the cause of the worldwide extinction of the dinosaurs, approximately 65 million years ago, we are dealing with a one-time event. For many events in history, we have a sample size of exactly one. A frequentist approach is not a particularly valuable tool for studying such singularities. The way fine-tuning proponents articulate their argument makes use of epistemic probability, which is what we use when making determinations about whether something counts as evidence. The idea is, given that physicists are right about the number of ways the universe could have been, and the overwhelmingly high ratio of hypothetical universes that would not permit life compared to those that would, the epistemic probability of a life-permitting universe is very, very low.

Matt also pointed out the hostility of most of our universe to life. He suggested that the cosmos is thus really anything but finely tuned for bio-habitability. This is, however, to commit an understated evidence fallacy since the fact that our universe is even here at all is due to very exquisite fine-tuning. By far the most impressive calculation of this fine-tuning was given by Roger Penrose, of Oxford University, who has shown that the probability of the initial low entropy conditions of our universe being obtained by chance alone is of the order of one part in ten raised to the tenth power raised to the 123rd power. The denominator after the one possesses more zeroes than there are elementary particles in the observable universe.

A further point I want to comment on is Matt’s claim that there is no original research being done by the intelligent design community. For a laboratory where such research is being conducted, I would refer him and other readers to the Biologic Institute webpage, in particular their list of selected research publications. For a more comprehensive list, there is one posted on the website of Discovery Institute, which can be located here.

Matt also said that intelligent design is not science because it is unfalsifiable. Leaving aside the limitations of falsifiability as a demarcation criterion for distinguishing science from non-science, I would argue that the theory that there is detectable design in biology can be falsified in two ways. One would be to show that life does not contain a level of specified complexity that is non-amenable to a neo-Darwinian stepwise pathway. For example, in the case of protein structures, you could demonstrate that the prevalence of stable folds in combinatorial space is relatively high. The second would be to show that intelligent agency is not the only cause that can produce specified complexity. If you could show either of those, it would demolish the design inference.

Matt also claimed that specified complexity is a meaningless concept. For a response, see my article for ENV, “ Two Reasons Why Frequently Heard Criticisms of Specified Complexity Are Misguided.”

Finally, I want to remark on Matt’s assertion before cutting me off — that is, that the concept of specified complexity is circular and smuggles in the designer with the word “specified.” His question was, “Is it possible for something to have specified complexity without an intelligent agent doing the specification?” I answered in the negative. Unfortunately, I think I probably misinterpreted this question. If Matt was asking at this point if it is possible in principle, then the answer would be yes because you could have such specified complexity produced by, say, a quantum fluctuation. But would one be reasonable in positing that an observed instance of specified complexity could come about apart from intelligent design? To that, I would indeed answer in the negative.

Image: Matt Dillahunty/Wikipedia.