Fantasy of the Gaps

KEVIN HARRIS: The god of the gaps fallacy came up again when our podcast discussed a dialogue between Neil deGrasse Tyson and Sean Carroll. Let’s look at an article by Canadian professor Kirk Durston who says he’s noticed something similar that secularists do. He calls it the fantasy of the gaps.^[1] Before we look at the article, review for us what god of the gaps reasoning is.

DR. CRAIG: This is a form of reasoning whereby the theist uses God to plug up the gaps in our scientific knowledge. So if there is something that we cannot currently explain scientifically, the theist says, “Oh, well God did it. God is the explanation for that.” And the unfortunate thing about that type of argument is that as science then progresses and closes the gap, God gets squeezed out and looks more and more irrelevant. This is not an argument against the existence of God, I think you can see, but it would be, I guess, a form of advice to the theist not to simply use God as a stopgap for the gaps in our scientific knowledge.

KEVIN HARRIS: The article begins,

Several weeks ago I was invited to participate in an atheist’s livestream discussion on the topic of the ‘god-of-the-gaps’ fallacy. Afterward, as I reflected on our discussion, I noticed something rather remarkable.

Just as some theists may be tempted to invoke a god-of-the-gaps argument as an explanation for something that we may not yet understand, so a non-theist is often prone to making a fantasy-of-the-gaps move when the science and logic begin to actually point to God. For the fantasy-of-the-gaps move, a person denies what science and logic say and, instead, appeals to various imaginary scenarios that sound scientific but do not map to reality in one or more essential aspects.

We'll look at some of his examples of this in a moment, but what do you think about what he's saying? Is the multiverse hypothesis a fantasy of the gaps response to get around the evidence for God?

DR. CRAIG: I think Durston is quite right about this appeal to fantasy of the gaps by certain secularists. When the evidence is pointing in a certain direction, it will deny what the evidence is pointing to in favor of some exotic, ad hoc, and unsubstantiated metaphysical hypothesis. Whether or not the multiverse hypothesis qualifies for that is disputed. Some scientists think that it is a sort of fantasy of the gaps, but others would say, no, this is a tenable hypothesis. If it's not a physical hypothesis, it's at least a metaphysical hypothesis on a par with the hypothesis of God and therefore needs to be taken seriously and adjudicated. So I think you can have metaphysical hypotheses that are not scientific or physical, and that doesn't make them fantasies. According to Durston’s definition, a fantasy of the gaps is an appeal to something that goes contrary to what science and logic indicate, and appeals instead to these imaginary scenarios that really are unsubstantiated and unreal.

KEVIN HARRIS: He continues,

The casual listener, not versed in distinguishing between science and science fiction, is often enthralled by what sounds like the latest in scientific theories. The result is that positive scientific and logical evidence for theism is, in their mind, assumed to be neutralized by what, in reality, is imaginary.

It is important, therefore, to tune one’s mind to recognize the ‘fantasy-of-the-gaps’ move so that we do not set aside evidence that might point to something far more important than we had realized.

Scientific explanation vs science fiction: A valid scientific explanation must be testable, observable, and repeatable. If it is not, then it is, at best, an inductive leap and it reduces to “fantasy” or science fiction.

So he suggests that we become good at spotting the fantasy of the gaps. Any thoughts on how we can do that and not be snowed by popular media reports of some exotic hypothesis?

DR. CRAIG: Well, I think that as Kirk says we should always ask, “What is the evidence supporting the hypothesis in question?” Is the hypothesis even testable by the evidence? Or is this just an ad hoc device adopted arbitrarily? And I also think it requires a certain sense of credibility. There are some things that are proposed that are just incredible, and it's not enough just to say, well, this is possible. Anything is possible if it's not a contradiction, but that doesn't make it a credible suggestion.

KEVIN HARRIS: Continuing, Kirk writes,

Given the track record of science in explaining how nature works, there is reasonable justification for assuming that a natural explanation will emerge for some phenomenon for which we currently have no good scientific explanation. There is a significant problem with maintaining that explanation, however, if, as science advances, the gap becomes larger or the imaginary scenarios and mathematical models are forced to become increasingly more creative and less credible. This is especially problematic when the evidence starts substantially tipping in the direction of an intelligent creator.

When the gap gets bigger and more imaginary, I think you call this bloated ontology, Bill. What does he mean by this becoming problematic when the evidence starts tipping toward an intelligent creator? Problematic for secular scientists who resist it?

DR. CRAIG: I think so. To give an illustration of this sort of fantasy of the gaps, do you remember the debate that I had with Lewis Wolpert in Central Hall, London, where I presented the kalam cosmological argument. And his response was to say, “Oh, well, I think it was a computer that created the universe.” And I said, “Well, it couldn't have been a computer because that's a physical object and the origin of the universe represented the beginning of all matter and energy, space and time.” “Oh, but this is a very special computer. It's immaterial and exists beyond space and time.” And I said, “But in that case it wouldn't be able to cause the physical universe to come into being.” And again he resorted to another ad hoc hypothesis. It was purely fantasy of the gap to avoid the creator of the universe. And as I said to him, “It seems to me that what you're calling an immaterial transcendent computer is just another name for God,” and the audience really got the point. I think that these fantasies are especially evident when you are using the scientific evidence to establish a religiously neutral proposition in an argument. For example, the premise that the universe began to exist – that is religiously neutral. And the refusal to follow the scientific evidence where it points in support of that would, I think, represent a kind of fantasy of the gaps where all sorts of fantasies are postulated including the arrow of time running in the opposite direction or time-like loops at the beginning of the universe. Just fantastic proposals for which there's no evidence which are incredible just to avoid a religiously neutral statement that is supported by the scientific evidence and logic. Another example would be the premise in the fine-tuning argument that the fine-tuning is not due to physical necessity or to chance. That again is a religiously neutral statement that is strongly supported by the scientific evidence. And someone like Richard Dawkins, who is no friend of theism, argues against the explanation of physical necessity as an account of the fine-tuning. On the other hand, Roger Penrose (who is also not a theist) argues against the possibility of chance as the best explanation for the fine-tuning of the universe. So secular scholars who are themselves scientists and who are objective can recognize the truth of this premise, and therefore it would be a fantasy of the gaps to try to appeal to something else to try to avoid the fine-tuning of the universe toward which the evidence is firmly pointing. So this fantasy of the gaps, I think, becomes especially evident when we have good scientific evidence for a religiously neutral proposition, and yet that proposition is denied.

KEVIN HARRIS: Before we continue the article, I wanted to ask you about that exchange with Wolpert. Every day it seems somebody new clips that one little segment and puts it online. I mean, people circulate and repost that little clip all the time. And it's got me wondering. Did you see where he was going with this, or were you having to think on your feet based on what he was saying?

DR. CRAIG: I was thinking on my feet. I mean, this is so fantastic. Inside me, when I'm listening to him, I'm thinking to myself, “You can't be serious,” and yet you cannot be disrespectful toward an eminent scientist. So you need to try to treat his hypothesis with respect. But it is a fantasy of the gaps just like Durston describes.

KEVIN HARRIS: Next up, Kirk gives two examples of fantasy of the gaps: one from the origin of nature, and the other from the origin of life. He recounts a conversation he had with theoretical physicist William G. Unrau at the University of British Columbia. A new mathematical model was announced for the origin of the universe and Unrau told him no mathematical model for the origin of the universe works in reality. Durston writes,

To clarify, every mathematical model will have one or more essential components. To map to reality, each one of those components must have a counterpart in the real world. For every mathematical model designed by cosmologists, each one has at least one essential component that has no counterpart in material reality or is incompatible with/contradicts material reality.

The example he gives is the attempt to traverse an actual infinite by subsequent addition. He says,

Perhaps the single greatest mistake in thinking a mathematical model of an eternal universe shows that the actual universe could be eternal, is that in mathematical models, a completed countable infinity (e.g., an infinite past) can easily be mathematically dealt with as a mathematical object. In the natural world, however, the past does not elapse as a single mathematical object in ‘one fell swoop’, but must be traversed one interval at a time (e.g., one year at a time or, in an expanding and contracting universe, one expansion and contraction at a time). Thus, although it is relatively easy to create a mathematical model of a universe with an eternal past, none of them work in reality.

He continues, referring to mathematical models that contain an infinite past, we can do that in mathematics because,

. . . we can treat completed countable infinities as a single mathematical object whereas in reality, the real-world past is an arrow of time that moves forward, one interval at a time, in sequence. The result in reality is that an infinite past would never, ever elapse through a countable infinite number of intervals such that we could arrive at this point, or any other point in actual history.

Now we're in William Lane Craig territory.

DR. CRAIG: Yes. This sounds like the kalam cosmological argument, doesn't it? I do want to issue one correction. I think that Durston claimed too much when he said that every aspect of a scientific theory must have a counterpart in reality. That is clearly not true. In scientific theories we often find what are called useful fictions, which are sometimes idealizations or are simply fictitious entities that don't really exist. For example, in modeling water waves, scientists assume the existence of an ocean that is infinitely deep which obviously is a fiction. It's unreal. And yet this enables them to model genuine waves. Or another example would be treating gasses as what are called ideal gasses – gasses that are made up of mathematical point particles. And that's clearly an idealization that hasn't got a counterpart in reality. But nevertheless these ideal gas laws enable us to approximate the behavior of real gasses. Indeed, I would say spacetime itself is an example of a fiction. It is a geometrical four-dimensional representation of space and time that I do not think actually exists. It is a useful fiction for modeling the behavior of things in space and time. And so these mathematical models that feature such entities still work, and they do make testable predictions. So it's not the case that every single feature of a mathematical model needs to have a counterpart in reality. But it does need to have some sort of testable predictions. Now, having said that, let me agree with Durston with regard to Roger Penrose's conformal cyclical cosmology. In Penrose's model of the origin of the universe you do have featured an infinite regress of expansions of the universe one after another, each one of which is infinite. The universe expands to infinity and then somehow starts over again and expands again into infinity, and there's an infinite regress of such infinite expansions. This is, I think, clearly a fantasy. In reality you would never be able to even get through one of those infinite expansions much less be able to arrive at its successor. So Penrose's conformal cyclic cosmology is not taken very seriously by professional cosmologists today. It doesn't really qualify as a serious scientific hypothesis.

KEVIN HARRIS: The second example he gives is an origin of life study. He writes,

A rather shocking example of a wholesale dive into fantasy is found in a paper by internationally respected evolutionary biologist Eugene Koonin. One of the more popular scenarios for the origin of life is the RNA world hypothesis held by many scientists. Koonin, however, has correctly pointed out that the probability of a natural occurrence of RNA replication is so small, that we should not expect it to occur anywhere in the universe over its entire history. . . . His solution, therefore, was to propose an infinite multiverse. With an infinite number of universes, in his mind, the probability for the emergence of RNA replication by chance is inevitable. The final sentence in his paper starkly exposes his motivation for appealing to an infinite number of unseen, untestable entities when he states that it “leaves no room whatsoever for any form of intelligent design.” So when forced to choose between just one unseen mind that might have encoded the genomes of life (and intelligent minds are the only testable, repeatable option science actually has for generating and writing significant levels of functional information), or an infinite number of unseen entities, Koonin sets aside Ockham’s razor and goes for the infinite number of unseen universes option, which is not testable and which has been labelled “fantasy” by other scientists simply for that fact.

A lot of this was brought up by James Tour recently.

DR. CRAIG: Now, I have, in fact, read this very paper by Eugene Koonin that Durston cites as part of my research on the origin and evolution of life. And he doesn't even mention the most outrageous aspect of this hypothesis. What Koonin realizes is that if you do resort to the infinite universe hypothesis then this actually completely undermines the theory of biological evolution. For why couldn't you just say that fully formed organisms just popped into being in the past? That there never was a history of evolution – these organisms just popped fully formed into existence. If one says, “Well, that's fantastically improbable. That can never happen,” then the proponent of the infinite multiverse will say in an infinite universe everything that is possible will happen somewhere. Indeed, it will happen an infinite number of times. So if you postulate the infinite multiverse in order to explain the origin of RNA replication, you undermine the entire theory of evolution because the organisms could just pop into being full-formed out of nothing. There doesn't need to be any evolution at all, and in an infinite universe this will happen infinitely many times over. So this is a version in biology of the famous Boltzmann Brain problem. If you appeal to the infinite multiverse then we have no way of knowing that our observations of the universe are, in fact, accurate rather than just hallucinatory. And Koonin doesn't understand this, and has no answer to this Boltzmann Brain problem that reappears here in evolutionary biology by this misguided attempt to appeal to the infinite multiverse in order to explain the RNA world. It's a great irony.

KEVIN HARRIS: A couple more excerpts. Next, Kirk asks whether God could be considered a fantasy. He gives two responses. One,

There is a strong correlation between those who insist God is fantasy and those who are most guilty of using the fantasy-of-the-gaps dodge to escape the evidence for God. If the evidence points to an intelligent creator, then it is no longer fantasy.

And two, he says,

The key reason God cannot simply be relegated to the category of fantasy is the extent and scope of the evidence and arguments for God’s existence and involvement in human history. These arguments and data may not be ‘proof’ in the checkmate sense, but they are substantive enough to be seriously discussed at the highest level in academic institutions around the world.

So he's saying there's a good cumulative case for God. Bill, your thoughts?

DR. CRAIG: I would simply want to reiterate the point that this evidence is especially powerful when it is evidence in support of religiously neutral premises in a philosophical argument leading to a conclusion that has theological significance. Then one is not guilty of postulating a god of the gaps. One is not appealing to God to plug up gaps in scientific knowledge. Rather, one is appealing to the best evidence of science – what we do know on the basis of contemporary science – to establish a religiously neutral premise in an argument that leads to a conclusion that has theological significance. And in that sense, one cannot be accused of postulating some sort of fantastic god of the gaps.

KEVIN HARRIS: Here's how he wraps up the article.

The evidence for the existence of God has sharply increased over this past century with the advance of science to the extent that the two largest gaps in scientific knowledge (the origin of the universe and the origin and diversity of life) have become much more difficult, if not impossible, to explain without resorting to the fantasy-of-the-gaps move. A critical part of rational thinking is the ability to separate fact from fiction. Fiction is rife in modern science which often confuses creative story-telling and imaginary mathematical or computer models with the real world. The takeaway is the importance of recognizing when a fantasy-of-the-gaps move is being made and improving one’s skill in identifying instances of fantasy and imagination when it is presented as science, or used as a dodge when the evidence for God is becoming stronger by the decade.

Do you agree with all or most of that, Bill?

DR. CRAIG: Yes, I would. And I would add two more examples to the two illustrations that Durston gives. I would add the fine tuning of the initial conditions of the universe for the existence of embodied agents like ourselves, and then, secondly, the problem of consciousness which remains inexplicable on physicalist or materialist grounds. Those two problems as well, I think, are a real challenge for any kind of materialistic metaphysic and would result in resorting to fantasies to try to explain them away.^[2]