The Giunta-Dillahunty debate

In the End it's just Magic

In #articles

I was listening to the Giunta-Dillahunty debate, "Does God Exist?", and the subsequent appearance of these two on the Dogma Debate show and felt there was enough there that required comment. I've commented a bit on the structure here and on the terminology here, but took down a lot more notes listening to the debate again. In this post, I refer to the debaters as Blake (Blake Giunta) and Matt (Matt Dillahunty) because it is a lot easier to type. My notes are just my responses while listening to the debate.

Causation

In Blake's debate he says that the Big Bang is plausibly our "first contingent event" and that "this is part of the standard cosmology" - however, he is wrong. There are many cosmological models. Some have the Big Bang as a recurrent event in an otherwise eternal universe. Some have the Big Bang as a recurrent event in a non-eternal universe, but the original event is lost to direct observation. Then there are multiverse models where the Big Bang is just one of many events. Since there are many competing models, each of which is consistent with the current observations but differ in the Big Bang being the first contingient event, one cannot state that it is plausible that the Big Bang is the first contingient event.

Another thing I noticed is that Blake talks a lot about "this causing that" and about "first causes" when referring to cosmology. Sean Carroll in his excellent debate with William Lane Craig says that this sort of Aristotelean vocabulary was "cutting edge stuff 2500 years ago", but "today we know better". The entire structure of the argument doesn't resemble anything a cosmologist would say, and I tend to trust the cosmologists on issues in cosmology.

Why focus so much on causation? Something like the universe may be uncaused. There are models of uncaused universes, and there are models of eternal universes - these are not a problem. Blake says that his argument "does not require that the universe has a cause. The argument only requires that a caused universe (contingent) is the kind of thing that could exist. It’s not a square-circle." I don't think this adds anything useful, but I'd have to go into that more with Blake.

Even quantum field theory has the production of pairs of particles, apparently uncaused. If you're going to say that these particles are caused in someway, then you need to put forward a model where are you demonstrate what the causation is. The current understanding is either these effects are uncaused, or the notion of the causation vocabulary doesn't enter in. So far I've never seen a causal a model put forward. Blake states later, and is correct in saying, that model equations in physics are not causally related to anything and that's exactly Sean Carroll's point - modern cosmology doesn't use the vocabulary of causation.

Philosophy as Word Salad

Blake takes a lot of time talking about "contingent" and "necessary" things. An argument he uses is the following, with several alternate forms found on his website:

  1. It is possible that the first contingent thing is caused to exist. (Premise)
  2. In the possible case where the first contingent thing is caused to exist, a causally powerful necessary being must cause it to exist. (Premise)
  3. A causally powerful necessary being possibly exists. (From 1 and 2)
  4. A possibly causally powerful necessary being necessarily exists. (From 3)

He refers to S5, another philosophical term I had to look up. It is a modal logic system that implies the claim that "if something is possibly necessary, then it is necessary".

Blake talks about the necessary entity who "can possibly cause things" and because this entity is necessary, it can exist without space and time and that sounds "suspiciously like God". Further, he talks about the "non-necessitating nature of the causation" and thus claims that the action of this entity "must be nondeterministic", concluding that "this is exactly the sort of thing we get with a free will choice."

Blake says that the universe with all its 17 particles and all of the constants would be unlikely to be necessary itself (another version of the fine-tuning argument). However, Blake fails to realize that the same argument could've been made for the spectra of stars in the 1800's. The spectra of stars are very specific, and complex. It was known that the patterns in the complexity could be used to classify stars and even infer some of their contents. It would have been very hard to believe that all of that complexity, and the specific patterns observed, would follow from a single equation - but that is in fact what is true. This is one of the reasons why these philosophical arguments for the existence of things is unconvincing to me - I simply do not trust human intuition when I hear the universe "must be" a certain way.

Aristotle made philosophical arguments for the existence of the four elements, the motion of objects relative to their composition, and an Earth-centered universe. We find these notions of Aristotle quaint now, not just because they were wrong (which they were) but because he never tested his ideas. That is a critical point here - you can make all the philosophical arguments you want for the existence of whatever you want but you need to close the loop and actually test something. It's not enough to show something is plausible (which I don't think Blake has even done) and that there are philosophical reasons for why you think it is there.

So here is a challenge to Blake, and similar theists making these philosophical arguments. Can you give me one prediction that the philosophical models actually make? I'm not talking about something that they explain that we already know, but something we haven't seen yet that differentiates their model from other models. We ask the same for the Big-Bang cosmologists, and the multiverse cosmologists, and the string theorists. I have yet to see that even attempted by the philosophers and that is why scientists just don't take them seriously.

Probabilistic thinking

Blake says that conceivability is a good guide to possibility - if you can conceive of something, then it is likely that something is at least possible. His notion of "possible" he states is "metaphysical possibility" which, as a scientist, I read as "useless possibility". Just show me whether there is evidence for something - I don't care to know if it is possible beforehand. If you show evidence, then the notion of possibility is answered automatically. If you don't, then who cares whether something is possible? A unicorn is possible, but does anyone really care?

Through purely philosophical arguments, Blake believes he has demonstrated that the prior probability of theism is high (which he of course hasn't). Then he devotes his attention to the likelihood, and focuses on the moral arena. The structure of the argument can be written mathematically like:

\begin{eqnarray} P(\mbox{moral arena }|\,{\rm theism}) &\sim& [10\%,90\%] \\ P(\mbox{moral arena }|\,{\rm atheism}) &\sim& P({\rm universe}\,|\,{\rm atheism})\times P({\rm life}\,|\,{\rm atheism})\times \nonumber \\ &&P({\rm consciousness}\,|\,{\rm atheism}) \times P({\rm morality}\,|\,{\rm atheism}) \end{eqnarray}

Then what happens is that, unless some of the terms in the atheist calculation are very large, the long product will give you something quite small. There are many problems with this line of reasoning, as one can imagine. First, why isn't the calculation symmetric? For example, shouldn't we really do:

\begin{eqnarray} P(\mbox{moral arena}\,|\,{\rm theism}) &\sim& P({\rm universe}\,|\,{\rm theism})\times P({\rm life}\,|\,{\rm theism})\times \nonumber\\ &&P({\rm consciousness}\,|\,{\rm theism}) \times P({\rm morality}\,|\,{\rm theism}) \\ P(\mbox{moral arena}\,|\,{\rm atheism}) &\sim& P({\rm universe}\,|\,{\rm atheism})\times P({\rm life}\,|\,{\rm atheism})\times \nonumber\\ &&P({\rm consciousness}\,|\,{\rm atheism}) \times P({\rm morality}\,|\,{\rm atheism}) \end{eqnarray}

Let's take one term, but similar things can be done with all of them. To demonstrate \(P({\rm universe}\,|\,{\rm atheism})\) is low, Blake brings up the fine-tuning argument and the Boltzmann Brain problem. Let's flip this around, and apply it to the theist part, \(P({\rm universe}\,|\,{\rm theism})\). The universe we observe should be very unlikely on theism for at least three reasons:

  1. A perfect being is complete, and thus does not need to create anything. In fact the creation of imperfection is less than perfect, and thus cannot contribute to the greatest possible good.
  2. A perfect being need not create a finely-tuned universe, so the existence of a finely-tuned one out of all possible not-finely tuned universes is unlikely.
  3. A non-physical mind would actually prefer Boltzmann brains over brains with the particular physical limitations ours have.

One can construct such arguments quite easily, which leads to Sean Carroll's observation that theism is not well defined. It also leads to my challenge above - give me an actual prediction or I don't need to take the arguments seriously.

Much of the atheist side of the calculation is a hidden God-of-the-gaps argument. Blake only includes terms for which we have the least understanding, and selectively omits all the things that used to be attributed to gods but no longer (e.g. weather and disease). He is almost explicit about it, when he says that he references the list of 25 top questions in science, and focuses only on those terms which would be different between the theist and atheist models. This list is:

  1. What Is the Universe Made Of?
  2. What Is the Biological Basis of Consciousness?
  3. Why Do Humans Have So Few Genes?
  4. To What Extent Are Genetic Variation and Personal Health Linked?
  5. Can the Laws of Physics Be Unified?
  6. How Much Can Human Life Span Be Extended?
  7. What Controls Organ Regeneration?
  8. How Can a Skin Cell Become a Nerve Cell?
  9. How Does a Single Somatic Cell Become a Whole Plant?
  10. How Does Earth’s Interior Work?
  11. Are We Alone in the Universe?
  12. How and Where Did Life on Earth Arise?
  13. What Determines Species Diversity?
  14. What Genetic Changes Made Us Uniquely Human?
  15. How Are Memories Stored and Retrieved?
  16. How Did Cooperative Behavior Evolve?
  17. How Will Big Pictures Emerge From a Sea of Biological Data?
  18. How Far Can We Push Chemical Self-Assembly?
  19. What Are the Limits of Conventional Computing?
  20. Can We Selectively Shut Off Immune Responses?
  21. Do Deeper Principles Underlie Quantum Uncertainty and Nonlocality?
  22. Is an Effective HIV Vaccine Feasible?
  23. How Hot Will the Greenhouse World Be?
  24. What Can Replace Cheap Oil— and When?
  25. Will Malthus Continue to Be Wrong?

I could easily come up with terms which might be more likely "on atheism", if I wanted to approach the problem like that. Numbers 13, 14, and 15 would be pretty easy, I think. Much of this calculation, beyond being God-of-the-Gaps, simply lacks imagination.

One thing that I notice about Blake is his confidence in things that are well outside of expertise. For example, he confidently says things like "spontaneous combustion is just as common under multi-verse as anything else" or "there is a Boltzmann Brain problem in multiverse models". It's clear that he does not know what he's talking about in these cases yet his level of confidence does not scale with that. Again, reference the Craig-Carroll debate.

Finally, on the moral issue, I would ask Blake what properties of the moral world would not be consistent with the God as Blake describes? I'm pretty sure everything that he is arguing for is post-hoc and that there is a rationalization for nearly everything - making the definition of God so plastic as to lack simplicity (see a previous post when I talk about the notion of simplicity).

The limits of science

Here's another example of the kind of arrogance of the Blake's position when he says science cannot explain the fundamental laws of nature because it requires initial conditions, and those initial conditions cannot be given or justified by science itself. I don't think this is a good argument. For example, maybe the fundamental laws will be self-explanatory. We just don't know. However, I'm certainly not about to go claiming that I know what the form of those laws is going to take. I certainly won't trust that my intuition will accurately track all the way back to the fundamental laws - they haven't been a great guide for the past 100 years in physics, why should I expect them to work in the next 100 years?

In order to demonstrate the limits of science, he asks about the difference between "methodological naturalism" and "methodological almost-naturalism", where science would be correct 99.999% of the time, except for the occasional supernatural intervention. Is science limited to excluding the latter as a matter of principle? I don't think so, or at least not entirely. I've written about science testing the supernatural elsewhere, and in summary I state

science cannot test supernatural explanations, because they are not defined. They can test specific predictions that incorporate unknown, and possibly unknowable, entities that have direct physical effects. In most cases of religious claims including supernatural effects there are no predictions made, and thus the term is empty and useless - an excuse for the intellectually lazy. Where specific predictions have been made (e.g. efficacy of prayer, existence of specific miracles, etc...) the predictions have not been supported. In this way, science can test God...and God has failed the test.

So, again, I say to Blake - show me a prediction that almost-naturalism gives you, and then we can test it.

Blake says that fundamental particles cannot have deeper mechanisms, that there is no mechanism for, say, the electron and the charge pulling or pushing on another charge. That's just flat-out wrong. There is a mechanism, part of QED, either with fields or the exchange of virtual photons. There may be a somewhat separate explanation in string theory. The problem isn't that he is wrong, but that Blake is showing his trust in philosophical arguments to try to tell us what forms physical theories must have and it just comes off wrong because he is not proposing a model, he's not making any predictions, and in my book that means he's not even in the discussion.

Blake says that you can't have an infinite regress in mechanisms but I don't see how he could say that. Some infinite series actually converge. In the best theory that we know of, QED, we have Feynman diagrams including infinite number of finite interactions, but the collection gives a meaningful result. Just like the origin of the universe, I do not presume to think that my intuition will hold well for infinite things. For example, you would think the series \(1+2+3+4+5+\cdots\) would not have a finite value...but it does...and it's negative!

Some theology

On the topic of divine hiddenness, Blake says something like "you have to demonstrate that God doesn't have morally sufficient reasons to stay hidden" and I think that's really a dodge, especially given that there is basically no moral act one can think of that God couldn't do - just read the Bible for some whoppers. Another idea he proposes is that maybe you're in a position right now where you would just reject him if he showed up, or maybe more people would be saved if he doesn't show himself. Of course, as Matt pointed out, God didn't seem to have a problem with showing himself in the Bible stories.

When Blake talked about these different "theories" (using "theories" in the colloquial sense, of a hunch or idea), and when he did the same thing with different notions of Hell, the first thing I thought of was "how could you possibly confirm these?" He makes no attempt to do so, as far as I can tell. One thing that surprised me when I was first working with biologists was the length to which the biologists would try to eliminate alternative hypotheses and to verify directly any extra elements that they were positing. If they said that some effect could be this gene expressed, then they would have to go and verify that that the particular gene not only did that function but did it in the specific case that they were looking at etc... I don't see the same thing ever happening with these theological arguments. No attempt is ever made to directly confirm the hunches.

Matt states, correctly, that we always explain things in terms of other things that we understand and that "explaining" things in terms of things that we don't understand isn't an explanation. Robert Price puts it like "explaining X by double-X". Therefore, the God explanation isn't really an explanation. I would add that, because there are no predictions made, it is an unuseful idea anyway, whether you choose to call it an explanation or not.

Blake says that he believes in interventionist God, so I would ask him for his evidence for the interventions, and further, for his evidence that those interventions are caused by God.

On predictions...again

On the origin of life, I would love to hear what predictions the Blake would make given his God model for starting life and what the researchers should look for? What should be the next test, with various proposals like RNA world or early biotic chemistry or replicating chemistry? There are predictions that these ideas make, even when they incomplete, so tell me what should we do with theistic models?

To make the comparison more explicit, let's look at two examples from physics: the Higgs boson and string theory. In the case of the Higgs boson, it was proposed 50 years ago to complete a theory that already explained every known experiment to the accuracy of every known measurement. The Higgs took 50 years of very hard work to confirm, and although most scientists were confident that we'd observe it (that confidence came from the other predictions of QED), the direct confirmation was necessary to rationally accept the model.

In the case of string theory, scientists are not confident, even though there are no experiments currently which contradict it. The problem is that there are also no unique predictions from string theory which can distinguish it from others. Lacking those predictions completely undermines the confidence one has in the theory. Think about this: string theory is consistent at the equation level with our best understanding of gravity and quantum mechanics (which is way better than any theist model), and it still doesn't convince scientists due to the fact that it makes no predictions.

Summary

So in summary, I am not convinced at all that philosophical arguments - by themselves - can give us any knowledge of the external world. The only methods that work involve prediction, and are the methods of science. When I say science, I generally mean the rational enterprise, combined with the empirical foundation of observational science (this includes history and possibly economics, for example). As far as I can tell Blake doesn't offer any real evidence for the existence of the interventionist God, nor does he make any predictions from his theistic model. Essentially, as far as I can tell, his arguments have the same content as "magic did it".