Sean Carroll’s refutation of William Lane Craig- UPDATED (and heavily revised)

Although originally set forth as a refutation of a Kalam argument proposed by WLC, Sean Carroll gives a refutation of any attempt to found a cosmological argument on the findings of modern cosmology:

Why should we expect that there are causes or explanations or a reason why in the universe in which we live? It’s because the physical world inside of which we’re embedded has two important features. There are unbreakable patterns, laws of physics—things don’t just happen, they obey the laws—and there is an arrow of time stretching from the past to the future…. But crucially, both of these features of the universe that allow us to speak the language of causes and effects are completely absent when we talk about the universe as a whole. We don’t think that our universe is part of a bigger ensemble that obeys laws. Even if it’s part of the multiverse, the multiverse is not part of a bigger ensemble that obeys laws. Therefore, nothing gives us the right to demand some kind of external cause.

Carroll wins this debate decisively, and while he may not always respond to Craig formally (which might lead some to think he isn’t responding to him) he clearly refutes the arguments against him. For example, he gives very good reasons why fine tuning can’t be designed, and so must either be a stroke of good luck or follow from some law. Since he sees fine tuning arguments as the best Craig has to offer (and he’s right in this) nothing I say after this, which will be critical of Carroll, need bother him all that much. I concede his refutation of what is, to him, the best argument against him. But I disagree with the universality of the account of cause that Carroll is working from.

Carroll’s argument turns on what can count as a cause, and for him this is an entity that is prior in time (and so in the general procession of entropy) and which is co-ordinated with its effect under some general law. First, my suspicion is that it is impossible to restrict the meaning of “cause” to this, even in cosmology, for two reasons: (I) At some point cosmologists will appeal to a cause and effect happening at the same time, and so the arrow of time will become superfluous, but this is the only connection that Carroll’s account of cause has to concrete objects in the universe (one can sense time and disorder but not “laws”). (II) cosmologists will have to make some use of statistical laws, but a statistical law cannot predict a particular as particular (e.g. the likelihood that I will have a car accident can’t explain my having this accident here and now.) But so far as this is the case, Carroll’s account of a cause can only be exhaustive if  either i.) There is no cause at all of particulars. or ii.) Particular things are not real. Both alternatives commit him to idealism, just as Berkeley proved they would; and “idealist naturalism” is almost certainly an oxymoron.

But the deeper problem with Carroll’s attempt to universalize his account of cause is that if all he means is that there can be no intrinsic argument within cosmology that concludes to God, then it’s not even clear if Craig himself needs to dispute this. If all Carroll means is that theology is not included as a subtopic in cosmology, and so can’t be concluded to by methods that Cosmologists use, then its hard to see what objection Craig would have. But if what Carroll wants to deny that cosmology can provide various data points that provide evidence for God when viewed in light of an account of cause that is broader than the one used by Cosmologists, then he is ipso facto working from the wrong account of a cause. After all, he’s giving an account of cause that is tailored to cosmology, or at least to the science of natural objects.

To his credit, Carroll does offer some dialectical arguments for Naturalism too, which I’ll have to deal with later.


  1. Mark said,

    June 30, 2014 at 4:14 pm

    Hi, thanks for this very interesting review. I thought Carroll won this debate as well. Although when the conversation drifted beyond cosmology, the arguments he presented for naturalism in general seemed to suffer from a number of glaring weaknesses that Craig could’ve easily addressed had he not admirably chosen to stay on topic.

    I was just wondering if you could elaborate on your comments regarding Carroll’s take on fine-tuning. While I agree that Carroll did a great job of explaining how the fine-tuning problem might eventually be resolved within a naturalistic perspective, I don’t see how any of the points he raised decisively answered the problems it presents. At best, he showed why, under the right (often highly speculative) circumstances, it’s not necessarily fatal for naturalism, but that’s a far cry from refuting it.

    Of course, there’s a very real possibility that I’m badly misreading what you wrote and I apologize in advance if that’s the case.

    • June 30, 2014 at 7:55 pm

      I don’t know that I could explain the problems with fine tuning better than Carroll, but most of his arguments arise from the fact that the universe is not designed, or at least not designed once by a single agent who could make things however he pleased. Design is elegant, straightforward, using no more that necessary to get the job done. Fine tuning isn’t like this, nor, more generally, is Nature like this: it is adaptive, at times way too efficient and at times wildly inefficient, making wildly diverse things work together without crashing – it frequently has a mad-scientist look to it, or it looks like one of those movie villains that somehow can think sixteen steps ahead through ten different contingencies to pull off a plan. A friend of mine likes to tell a story about the systems that clot blood: that one system makes the blood clot, another acts around the cut to keep the clots from returning to the bloodstream, but they are both totally independent of each other. Again, would anyone design a universe that was explained by both QM and Relativity? This doesn’t look designed, even though it certainly works. It doesn’t look random either, still less does it look like a machine. All the old Enlightenment metaphors for nature fall flat when we look at them: machine, watch-mechanism, selection.

      This is the real insight that Carroll is working from in his critique of tuning. Things are wildly overtuned to allow for life; there is no good reason to assume that a designer would put together a universe that needed to be tuned (Why not just make things with the values he wants? Why bother to create an infinite possibility out of which one might determine values?) and one doesn’t expect tuning any more on theism or naturalism. The only difference is that I don’t agree with Carroll any more than Craig. The tuning we find in nature, like nature itself, looks neither designed nor random. It points to the simultaneous failure of our competing metaphors for nature as either designed or machine.

      What I’m writing here is really an empirical case: we need to just increase the amount of strangeness and mad-scientist work that nature does.

      All right, so I’m rambling here, but clearly the mad scientist metaphor is popping up a lot. There is an intelligence in nature, but one that doesn’t look at all like a designer. A naturalist can look at this and read it as nature being not designed, and he’s right, as far as it goes. But there’s more to it than that.

      It’s interesting to compare a 18th century atheist-for-science to our contemporary atheists: the former thought they could replace God with neat, elegant equations that would we so much simpler than religion or metaphysics; the modern sort are exactly the reverse – insisting that nature is far too difficult, confusing, and even mysterious to be grasped by “intuitive concepts” or “logic”. In reading them there is a consistent antiphon of just how baffling and counterintuitive nature is, which requires (for some reason) that we use one and only one method and metaphor to understand it (science and “the model” or “machine”) and that any attempt to speak generally about it is impossible. There are things to love about this sort of atheism, and as a theist I think what it is really teaching me is just how frighteningly different and sheerly other the divine intelligence is. Modern scientific atheism is in certain ways very Hebrew: it is striving to be the prophet of the voice of God that spoke to Job out of the whirlwind, or the voice of Koheleth that is simply baffled and overwhelmed by the infinity that he sees in nature.

      • thenyssan said,

        July 1, 2014 at 6:50 am

        Either the universe is a living thing or clearly the work of a divine intelligence utterly transcendent to our own, and the atheists have proven it.

        Where am I? What year is this? 🙂

      • Crude said,

        July 1, 2014 at 9:18 pm

        Design is elegant, straightforward, using no more that necessary to get the job done. Fine tuning isn’t like this, nor, more generally, is Nature like this: it is adaptive, at times way too efficient and at times wildly inefficient, making wildly diverse things work together without crashing – it frequently has a mad-scientist look to it, or it looks like one of those movie villains that somehow can think sixteen steps ahead through ten different contingencies to pull off a plan.

        I don’t think this works. In fact, I think it fails radically, on multiple levels – in a way that undercuts both Carroll and Craig at once, at least as far as I understand them.

        I think it says a lot that your own example for how nature doesn’t look designed actually appeals to two different forms of (pop fiction) design – the movie villain and the mad scientist. Another way of saying that is that if the universe is designed, then it’s designed by a mind far beyond or different from our own… but that’s exactly the sort of mind being put on offer anyway. Still another point would be that the sort of design you speak about is an idealized design that’s specific only to particular desires of outcomes – but once it’s recognized that the possible desires are much broader, the point falls.

        I understand the obvious reply here – that if the sort of mind we’re discussing is distant from our own, then it makes no sense to look for design. But to say that is to undercut both Craig and Carroll at once; it means that Craig’s inference to design doesn’t work (because the mind he has in mind (ha) wouldn’t necessary work in a familiar way), and neither does Carroll’s inference against design (for the same reason.)

        In fact, I think it’s even worse for Carroll for one reason – between the two of them, only one is able to say ‘My conviction on this topic rests outside of science’ without the rest of their stance falling apart, and that would be Craig. Carroll has no scientific test for God’s presence or lack given what we see about nature, and that it seems is precisely what he needs to maintain his stance.

        The best that Carroll can establish here is that science, as science, is ultimately compatible with both an atheistic view that makes room for magic (uncaused causes, inexplicable things that have no explanation or reason, ‘just so’ stories) and a theistic view. But to quote another would-be apologist – Carroll, insofar as he is associated with the New Atheists, has to do better than that. To pull a draw is to lose.

      • Augustine Thomas said,

        September 4, 2014 at 7:40 pm

        I think you’re ignoring that God is infinitely more knowledgeable than you, Carroll or the sum total of human intellect. How could we ever propose to say “well God surely wouldn’t do that”?

        Furthermore, Christianity proposes God and Satan. How do we know the limits of Satan’s power? Why couldn’t Satan be causing the blood clot and God be causing the healing?
        If you asked, “well why doesn’t God just fix everything immediately?” I would answer, similarly as I did above, how can you be so arrogant to believe that you have any idea what exactly God wants or what the best way is for him to arrive at the end he wants?

        I would also put forth that you’re completely ignoring free will. By definition, there needs to be some unpredictability if we truly have free will. If we guess that God wants as many souls as possible to freely choose him (not to be programmed to choose him or forced into it with material proof), then, yet again, how can we know what the best way is to arrive at that end?

        I think one of the main problems with all this is modern arrogance. We’ve done some extraordinary things and we think that must mean we can know God’s mind. Nothing is more absurd or ridiculous than believing we can understand the being who created the Milky way.

        God bless you!

  2. John said,

    June 30, 2014 at 8:27 pm

    I’m not entirely sure that Craig lost the debate convincingly.

    Is it more probably true that the Universe had a beginning, or is the science really that far up in the air where we just cannot call it whatsoever? All Craig needs is that it is more probably true than not, and the Kalam pushes through.

    Why do you think fine tuning is a better argument than the Kalam?


    • July 1, 2014 at 9:02 am

      In order for Craig’s argument to go through, he needs more than just a universe with a beginning, he needs that beginning to be inexplicable with reference to some physical law or model, which is exactly what no cosmologist would ever concede to him, and which certainly isn’t a finding towards which modern cosmology points. In fact, I doubt it could ever point to it, even in principle. This is the sense of Carroll’s continual insistence that there are still a lot of physical models on offer as explanations of the big bang singularity.

      This situation is not unique to contemporary physics. Even Aristotle’s physics, as a physics, can’t conclude to a first mover that is pure act. Act and potency in his physics are aspects of beings, not things themselves. “Pure act” – an act that is not an aspect of some complex reality – makes no more sense within Aristotle’s physics than a waterless ocean wave. In concluding to pure act, Aristotle is demarcating a region of being that is unintelligible according to what he understood as physical law.

      Interestingly, Carroll did something similar at the end of his speech when he marked off a realm called “religion” that was separate from physical law. IOW, Carroll and Aristotle agree that there is some region outside of physical law, but Aristotle sees this region as scientific whereas Carroll doesn’t. Even this difference could admit some nuance – Aristotle did not see science as having a single method, but only a single goal, whereas Carroll sees method as essential to science. Given their respective definitions, it makes sense that Carroll sees no science outside of natural science whereas Aristotle does.

      • John said,

        July 2, 2014 at 8:36 am

        If the Universe had an absolute beginning, as Craig argues, then no “physical” explanation is possible.

      • July 2, 2014 at 9:22 am

        No, he still needs more than this. He wants to argue that modern cosmology gives us a universe with a beginning. But modern cosmology means by “the universe” an object that’s about 13 billion years old, is expanding at an increasing rate, traces back to a singularity, etc. This is not the same thing as “universe” meaning “everything that can be physically modeled”, but Craig needs these two things to be the same in order for his argument to go through. But Carroll gave very good reasons why they are not the same thing.

      • Crude said,

        July 2, 2014 at 6:37 pm

        My understanding is that Craig never says ‘science proves God created the universe’ or ‘science shows God exists’. When it comes to arguments about a past-infinite universe and concluding God from it, Craig’s go-to argument is Kalam – not modern cosmology, much less fine-tuning.

        Science is provisional, and I think Craig admits this. But it seems at least possible that a current cosmological situation, given certain *philosophical* caveats, can at least be evidence, if not proof, of what Craig’s arguing.

  3. PatrickH said,

    July 3, 2014 at 1:09 pm

    Doesn’t Craig cite Borde-Vilenkin-Guth as at least suggesting that a past-infinite universe is impossible from modern cosmology?

  4. Clark Griswold said,

    July 4, 2014 at 8:57 am

    I thought Carroll’s attempt to refute the Kalaam failed on several points, in that it was based on a familiar appeal to gratuitous physics (though better-packaged than, say, Lawrence Krauss has attempted). I’ve not yet seen a credible attempt to avoid an absolute beginning to the universe and the metaphysical implications thereof.

    Carroll is better-mannered and more prepared than Krauss, but in the end his arguments are similarly weak.

    On Carroll’s broader attempt to refute the existence of God, “if God, then I would expect to see…” it seemed as though Craig failed to recognize this for what it was, a version of “if I were God I would have done things quite differently.” All the argument indicates is that Carroll is not God, and that is neither relevant nor is it even new information, which is apparent whether Craig pointed this out or not.

    Carroll did an excellent job in representing the atheist/naturalist position, weak though it is, and succeeded in chipping away the perception of relative certainty regarding the KCA’s central arguments, nevertheless Craig made a convincing appeal that the existence of God remains far more plausible than God’s non-existence on the basis of observation and reason.

    Regarding tuning, even the minimum amount required vastly exceeds the amount of leeway permitted in the gambler’s fallacy inherent in the multiverse.

    No opinion rests totally inside or outside science, which cannot exist apart from philosophy or religion. Science is a methodology, not a set of conclusions or even a type of conclusion. While methodological naturalism is suggested in the conduct of test and observation, you can’t do science without philosophy, and any appeal to truth must be based on religion. This is why Newton didn’t end with God (as some suggest), rather he began with God.

%d bloggers like this: