Evaluating a Debate on the Proposition: Science Refutes God

This post is the second in a series of posts evaluating Intelligence Squared U.S. debates. I will consider a debate on the proposition “Science Refutes God.”

A couple of stereotypical stupid, dogmatic, flying spaghetti monster atheists called Lawrence Krauss and Michael Shermer argued for the proposition. Against the proposition were Dinesh D’Souza and MIT nuclear science and engineering professor Ian Hutchinson. The team arguing for the proposition, as you would expect, had no understanding of the philosophy of science, and no interest in learning about it, and therefore no idea that they could be wrong, while the team arguing against was well-versed in such matters and so very nondogmatic. The following is a pretty typical exchange:

D’Souza: But is there a logical case you can make against [miracles]?

Krauss: Our logic is determined by nature, not by what we’d like. And nature has told us that miracles don’t happen. That’s it. It’s not what I want or what I think should be rational. It’s do they happen? And there’s no evidence that they’ve ever happened.

D’Souza: I think we have here a deep fallacy. And the fact that it remains a fallacy shows that what is being called science is actually hiding behind a philosophical principle. It was the philosopher Hume who pointed out 200 years ago that from no amount of empirical generalizations, however large, can you draw a general law that is true as a matter of logic. What I mean is, it doesn’t matter how often you measure something. Let’s say you measure the speed at which this pen falls down. You can measure it a million times. But you don’t know that on a star ten light years away or in some other condition where you haven’t measured it, that if you drop this pen, it will fall at the same speed. Science presumes it. It doesn’t prove –

Krauss: We measure. I’m sorry. We measure it.

We measure! Hume stands refuted! This is honestly how the entire debate went. And the proposition team won. They went from 37% support to 50%, the opposition only from 34% to 38%. It just goes to show, don’t doubt yourself and you will go far in life.

Krauss opens the debate by claiming that the proposition side has “evidence, reason, logic, rationality and empirical methods,” and the opposition has nothing of the kind. How does he deliver on this claim that all logic and evidence is on his side? What evidence can there be for God’s non-existence, or what logical proof can one offer?

Krauss does not attempt a logical proof, but turns only to evidence. This is appropriate, since the resolution is “science refutes God,” and science is about the empirical testing of hypotheses, not about deducing what is logically necessary. But then, he shouldn’t have promised logic.

But let us turn to his evidence. In science of course evidence strictly speaking can only refute a theory, not confirm it. If a theory explains observations, these observations are evidence for the theory, while if there are observations it fails to explain, this is evidence against the theory. So, what observations are inconsistent with God’s existence? Krauss doesn’t explicitly mention any.

What he does instead is note a set of observations for which God is a possible explanation, and argues that there is another explanation available. Specifically, he argues that God is not necessary to explain the existence of our universe, or the fact that it is fine-tuned for life to a degree unlikely to occur by random chance. Rather, the multiverse explains both these things.

Krauss then turns to an analogy. People used to believe life was fine-tuned for its environment, and this was evidence for God, but Darwin removed the need for this explanation through his theory of natural selection, combined with genetic variation. Darwin didn’t know the details of this process; specifically, he didn’t know about DNA and genetic replication. Similarly, the multiverse hypothesis may eventually explain the existence of our fine-tuned universe without recourse to God. It’s true that don’t have a full explanation yet, but our position is analogous to where Darwin’s was when he didn’t know about genetic processes.

Now, where does this analogy get us? Darwin is to multiverse as (Darwin + modern genetics theory) is to (the multiverse with some actual proof for its existence.) Well, yes, but that doesn’t give us a proof of the multiverse’s existence, now does it? It doesn’t get us to an explanation for our universe’s existence. It puts us in a mood to expect that such a proof or explanation will eventually be discovered. A naturalistic explanation proved adequate for one thing, and a religious explanation proved unnecessary,* so it sort of feels like it might happen again. But this is reasoning from the very particular to the very general.

Supposing, though, that the multiverse does explain our universe. What then? “That removes the need for God as an explanation.” Yes, but God removes the need for the multiverse as an explanation. Which is better?

Krauss says the following: “my opponents might argue that the multiverse, which our universe might have spontaneously been created in, was created by physicists because they don’t like God, because it’s eternal and exists outside our universe, those same characteristics that God is supposed to have. But it wasn’t created because we don’t like God, although I don’t like God. It was — we’d been driven to it by measurements. In fact, I don’t even like the multiverse, but I’ve learned to force my beliefs to conform to the evidence of reality.”

But if measurements take us to something outside the universe and eternal, do they really tell us what that entity should be like- whether it should have other characteristics attributed to God, or another set of characteristics attributed to the multiverse? Wouldn’t someone trying to argue “Science Refutes God” want to show us in some way how measurements favor the multiverse over God?

Krauss has satisfied himself that his motives behind his metaphysical belief are pure, so concludes that he must have sound reason for believing in it. Scientists are apparently embracing the multiverse because they can’t explain this universe as self-sufficient; in that sense, yes, evidence is forcing them in a direction they don’t like. But the question is whether this embrace is based on an assumption that God doesn’t exist. We don’t have to speculate about the psychological motive behind the assumption (that it comes from a dislike of God, for instance), only to determine whether the assumption is part of the argument.

Now, science strictly speaking excludes metaphysical explanations. Thus it would exclude God, and the multiverse. (Intelligent scientists recognize that this is an assumption, and that their very commitment to metaphysical neutrality requires that they don’t take the metaphysical position that God doesn’t exist, either. Thus they are carefully to adopt agnosticism, even if they are atheists as a matter of belief or emotional inclination.) On the other hand, if “we’ve been driven by measurements” to the conclusion that the universe can’t exist on its own, that it requires something “eternal and outside our universe” to explain it, where does that leave us? Again, God is one possibility, the multiverse is another.

Some scientists are willing to overlook their commitment to avoid metaphysics in order to embrace the multiverse. But this isn’t just a matter of embracing something they “dislike” or find aesthetically displeasing or unwieldy, thereby showing heroic scientific integrity. Rather, science itself is based partly on rejecting metaphysical explanations, so accepting the multiverse and calling it science is a problem for scientific integrity. But why are some scientists willing to bend the rules for the multiverse and not for God? Aren’t they after all putting their theological preferences ahead of methodological integrity? (I mean the sort of people like Krauss who use the multiverse to support theological beliefs, not those who believe in it for other reasons or find it an interesting hypothesis.)

I said Hutchinson and D’Souza showed a deep understanding of the philosophy of science, an ability to grapple with what evidence, proof, science and rationality mean, that the proposition team lacked. How is it that they lost? My theory is that the New York audience wanted to take what they perceived as the most sophisticated position, agnosticism, going in, and so most voted either undecided or against the proposition (the undecided total is on the high end for these debates.) However, once the games began, the tribal, cultural warrior part of their brains kicked in. Much as they tried to do without metaphysical assumptions, once the debate started they appeared to be presented with a choice between such assumptions, and voted accordingly.

Making matters worse, Hutchinson and D’Souza, for reasons perhaps best known to themselves, defined God as the Christian God.** The proposition team, for their part, made simplistic arguments of a sort that would have encouraged the audience in its worst tendencies; bringing up seven day creationism, for instance. And of course the Darwin analogy served a similar purpose. When the theist team argued that atheism couldn’t account for the existence of morality, the atheists countered that the Bible is immoral, so there!

The Theists’ Arguments

Hutchinson pointed out that what the proposition team really means by “Science Refutes God” is that there is no scientific evidence for God. He rejects this (“[T]here are some things we’ve learned about the universe through science that are highly suggestive of a creator.”) But more importantly, science is the effort to “describe the universe insofar as it is repeatable and follows universal laws.” There are other kinds of evidence that we rely on to, for instance, determine what is just or unjust, or whether Caesar crossed the Rubicon. God’s existence just mostly isn’t a scientific question, and the evidence for God’s existence is mostly not scientific (Hutchinson did not, however, elaborate.)

I’d add that, strictly speaking, saying that an observation is evidence for a theory means only that the theory explains the observation. So the evidence for God is just the existence of the universe, unless creation by God doesn’t explain the existence of the universe. The question is whether some other theory is preferable on some epistemological grounds.

D’Souza’s began his speech by noting that the proposition required proof that belief in God is unreasonable, not that it might do so at some point in the future. This of course should have badly damaged Krauss’s whole argument, which was based on the idea that the multiverse will someday be verified and adequately account for reality, but D’Souza didn’t make this clear to the audience, relying on it to do too much work for him.

He argues that science can’t give answers to questions of morality and purpose, but doesn’t show how this relates to the existence of God (although it’s a potentially intriguing argument, and one I think St. Thomas has made.)

D’Souza then quickly dispensed with the second atheist speaker, Michael Shermer, who gave psychological and sociological explanations for belief in God. These explanations undoubtedly served the purpose, though, of taking the audience away from rational evaluations of truth claims and toward tribal attitudes (ironically, something that would be predicted by the very theories Shermer mentioned.)

D’Souza then argued that the Big Bang favors Judeo-Christian cosmology by doing away with the previous assumption of an eternal (and, I might add, self-sufficient and non-contingent) universe.
In the cross-fire style section of the debate, in addition to the amusing discussion of Hume I recounted in the opening, we had some other interesting exchanges. There was discussion of D’Souza’s point that science can’t account for morality. The proposition was like “yeah, well Biblical morality is pretty bad, so you can’t get morality from the Bible, either.” D’Souza responded as follows:

“First of all, we don’t get, none of us, morality from the Bible. It’s not like I read the read the Ten Commandments and went, ‘Oh, stealing is wrong, wow. Didn’t know that before. Killing is wrong, incredible.’ I already knew that. How did I know it? I knew it because of what Adam Smith calls the impartial spectator, the voice of conscience. So it is the contention of religious believers that the voice of conscience has been implanted in us by God. Now, you say it comes from evolution, and we say it could have been implanted in us by God through evolution.”

Another fun exchange came on the issue of Biblical literalism:
D’Souza: If you look at the bible, it is not a science manual; it doesn’t claim to make scientific proofs. In fact, what’s interesting about the bible is it doesn’t make proofs at all…Why? Because for the most part, what the bible is doing is it’s in the province of revelation. Now, you’ll notice that neither of us have ever in this debate appealed to the bible to support any argument that we made. We have argued on the basis of reason alone. Now, the bible does make certain claims about the world and about man. So, for example, the bible doesn’t say that God made man out of nothing, but it does say that God made the universe out of nothing; it says that it made man out of some other stuff. Now, so the Bible says God created the world and God created man, but the bible doesn’t say how.

So that’s where science comes in. Science can, and the scientific answer often changes, attempt to give explanations that actually don’t refute the bible. You could refute the bible by showing that, actually, the universe has always existed and doesn’t require any explanation whatsoever….
Krauss: Well, I mean, I — you know, you don’t really want to — you don’t really want to go here, Dinesh. So I would say, for example, we refute the bible by arguing that creating day and night before you create the sun is pretty silly.
D’Souza: Well, Lawrence, in that case — here’s the problem: you have a fundamentalist reading of the bible that has…that is maybe subscribed to by 3 percent of Christians and 100 percent of atheists.”

And of course, it isn’t science but logic that refutes the claim that there were days and nights before the sun existed.

The key exchange of the debate for me, though, came in response to Krauss’s claim that science requires the presumption of atheism, since it relies on material explanations rather than recourse to God making things happen. Here it goes:

D’Souza: I think there’s a presumption that science explains the material world, and science does it with material explanations. [The moderator asks if the proposition agrees, and they say yes.] So it is a presumption of modern science, not a proof, but a presumption that beyond the material world, there is nothing. So take, for example, the fact that as humans we experience consciousness. We know that there’s something in us, consciousness, that can’t be reduced simply to materialism.
Krauss: How do you know that?
D’Souza: Hold on a second. When you say “how do you know that?” you’re presuming that it is. Do you agree that the cause of consciousness is not known?
Krauss: The fact that something is not known does not imply it’s God.
D’Souza: Right. But neither – so consciousness is an immaterial thing that may have a material explanation. And the key word is “may.” And yet you as a scientist believe it does.
Krauss: I don’t believe anything. I just want to learn how the world works.
D’Souza: You just said all causes are material.
Krauss: I’ll wait for the experiments and the theory.
D’Souza: I’m simply saying, as a scientist you are closed off to the possibility of nonmaterial explanations. True or false?
Krauss: No. I told you if the starts moved around today, I’d be really thinking there’s some intelligence in the universe. There’s just never been such an observation. So until there is, I’ll assume the reasonable logical thing, since there’s never been such an observation, there’s unlikely to be one. That’s all. As a scientist, I can say what’s likely and what’s unlikely. I don’t believe anything.

A few notes on the exchange to this point:
1) Krauss does believe. He assumes- he says so himself. He assumes “the reasonable logical thing.” But he doesn’t examine why he considers his assumption reasonable and logical; in other words, it is a dogmatic belief.

2) If science starts with the presumption that God doesn’t exist, then it can’t very well prove that God doesn’t exist. Krauss’s whole narrative revolves around scientific discoveries driving God out, but if these discoveries are predicated on the assumption of God’s non-existence, then so much for that. If God’s existence or non-existence is a question on which we have to rely on reason, before getting to empirical science, then so much for the proposition “Science Refutes God!”

3) Krauss realizes this eventually and so backtracks: if the stars moved around, that is evidence for God. Of course, the stars do move around, but he’s referring to something he said earlier: “if tonight when I looked up at the sky, the stars rearranged themselves [to say] ‘I am here,’ in Aramaic or ancient Greek or whatever you want, then I might say, you know, there’s something to it.” But of course there have been observations of miracles. Krauss just doesn’t believe in them, because they’re not replicable and therefore not scientific. So his reasoning is still just as circular. He appears to be demanding as proof of God’s existence that he perform replicable miracles, which is conceptually impossible, as Hutchinson points out.

4) But even if Krauss were genuinely open to the miraculous as proof of God’s existence, that’s just moving from one arbitrary a priori position (God doesn’t exist, everything has a material explanation) to another (God doesn’t exist, everything has a material explanation, unless miracles are real.) But why assume that if God existed, he would arrange the stars a certain way to make words in a human language, tonight, more or less at his demand? I can see where such an event would be evidence of God’s existence, but is its absence evidence of God’s non-existence? Just picking something that could happen to make you accept God, to avoid the accusation of circular reasoning, is cute. But on this argument, science is no longer disproving God because of the multiverse or any scientific explanations of anything. It is disproving God by “disproving” miracles. Yet Hutchinson’s replicability argument goes unanswered throughout the debate (really, you can look it up- it gets stone cold dropped.)

5) Now to return to Dinesh D’Souza’s point. He actually does raise the question of what a logical set of first principles would be, and naturally Krauss wants nothing to do with the debate. The proposition team assumes that everything has a material explanation. But why? On the face of it, you have matter and you have something non-material, thoughts, which can’t straightforwardly be reduced to matter. You might believe that thoughts and ideas are ultimately reducible to matter. Someone smarter than Krauss and Shermer might even be able to give a reason for believing this.

But on the other hnad, one might believe that matter is reducible to ideas; or, with Berkeley, say that there is no matter behind our observations. After all, our observations are never of matter, but are entirely reducible to ideas (color, number, shape, size and so forth) involved in our perception. Or one might believe any number of other things.

6) “The fact that something is not known does not imply it’s God.” No, if science doesn’t presently have an explanation, that doesn’t mean God has to be the explanation. But then again, it doesn’t mean there’s a physical explanation, either. Krauss simply believes science will eventually come up with an explanation for everything.

The resolution, let’s remember, is “Science Refutes God.” The side refuting God has the burden of proof. The opposition has to refute their claim to disprove God, and that may well involve showing the incompleteness or inadequacy of scientific explanations, even where one would not use this same incompleteness as an argument in favor of God.
And if (which is true) “the fact that something is not known does not imply it’s God,” what, for the thousandth time, is the point of the “progress of science” narrative, featuring Darwin and the multiverse? Why should the increase in scientific explanations tend to oppose God, if the scientifically unexplained was never an argument for God in the first place?

The limits of science in principle are another matter, which brings us to the remainder of the exchange:

D’Souza: I think this is very important because throughout your book– this is called “A Universe from Nothing.”
Krauss: Can you hold it up?
D’Souza: I recommend it…It’s a very useful primer on the limits of science. Here’s why.
Krauss: Absolutely, the universe.
D’Souza: So to quote a sample line, and there are many like this. You say “getting something from nothing.” And then you say, “This is how our universe could have arisen.” And then you say, “I stress the word could here because we may never have enough empirical information.” Next page you say, “These are useful operational efforts to describe how our universe might actually have originated.” Here’s may point. You’re invoking “coulds” and “mights” and “maybes” to provide a refutation. Now, even if you were successful in saying that “I have an alternative possibility to God,” you haven’t refuted God. You’ve simply given an alternative possibiltiy.
So you have all these qualifications in the book, but then when you step up here, you act as a science that’s demonstrated [sic] that the universe did in fact come from nothing, whereas you say, it never did. You have not made that demonstration, and you admit it.
Krauss: I was up there, I said it was plausible. And plausible is remarkable because everything that you’ve talked about in terms of religion is implausible. So the point is that just like with evolutoin, a simple plausible assumption appears to work is remarkable and worth celebrating. And so the reason I say “could” and “might” is because I’m honest. And also because I haven’t presumed the answer before I asked the question[!]

So the plausible explanation, the multiverse, refutes God only because God was already implausible; in other words, God already stood refuted; in other words, the multiverse has nothing to do with it; in other words, science does not refute God.

In his speech, Krauss was indeed careful to say that the multiverse was only plausible, so one couldn’t hold him to anything. On the other hand, he also said “science has taught us that we don’t need God to create a universe.” And he said it is an “illusion” to think our universe’s hospitability for life is evidence for God, because “what we’re seeing is a version of natural selection,” which only makes sense if there are multiple cosmoses and eventually one comes along that supports life. Not there is another plausible explanation that would make God unnecessary if it’s right, but “it is an illusion” to think God is necessary.

Moreover, Krauss’s Darwin analogy implies that the multiverse is as established as natural selection was with Darwin, that all that is missing is the mechanism that makes it work, just as Darwin didn’t know about the mechanism of genetics. But the difficulty of the multiverse is nothing to do with the lack of a mechanism, but the lack of empirical proof. Isaac Newton didn’t have a mechanism by which gravity worked, either, but his theory of gravity was pretty solid.*** So the analogy would have implied that the multiverse is a pretty well-established scientific theory. It certainly implied, and was intended to imply, that it is possible for it to eventually be vindicated, just as, in the rather dubious story Krauss tells, genetics vindicated Darwin. Indeed, the audience was supposed to extrapolate, although Krauss did not say this, that because Darwin was eventually vindicated, we can expect that the multiverse will eventually be vindicated. There’s no logical reason for this, but it fits a powerful narrative.

But in the book, we find the gem D’Souza quoted: “I stress the word could here because we may never have enough empirical information.”

* For fine-tuning of life to its environment. The existence of life itself, and of thought, is another matter.
** “The god I’m going to discuss is the God of Christianity, because I’m a follower of Jesus, but my side’s job is not to convince you of the truth of Christianity, although I think it is true. Our job is just to show that Christianity’s God is not refuted by science.” – Hutchinson Intellectually this is a defensible approach, to bring clarity to the debate; but it obviously would tend to contribute to tribal thinking on the part of the audience.
*** And strictly speaking, science doesn’t have causal explanations for anything anyway, only correlations, observations that match models.