Atheist "Refutation" of the Cosmological Argument? Not Here, Not Now

A reader asks: “Professor Stackhouse, Do you have any comments on The Apostate’s post about William Lane Craig?

I reply:

I stopped reading The Apostate’s first post after he made such a hash of “refuting” the cosmological argument. I’m not saying he doesn’t raise good points later on: I just don’t have time to read everything, and the opening is so bad it doesn’t give me good grounds to keep reading. Here’s what I mean.

His first argument is to cite mumbo-jumbo about cosmological theories that are, to put it kindly, in very early days and (to his credit) the Apostate doesn’t claim they have a lot of empirical verification because they don’t. (I figure if even I know that, and I’m no scientist but have a lively amateur interest in such things, then everybody talking about such matters should know it, too!).

That’s the best of his three arguments. The second one shows that he doesn’t understand the cosmological argument. (I just heard a University of Toronto professor emeritus of philosophy make the same embarrassing mistake at a debate in the University of Ottawa.) Just look up “cosmological argument” on that noted philosophical reference tool, Wikipedia, to see what The Apostate fails to see, namely, this basic, basic qualification: everything that is finite and contingent ultimately must stem from something that is neither finite nor contingent. “Super-intelligent aliens,” the Apostate’s alternative hypothesis, are finite and contingent. Ergo, they don’t help his case.

This point perhaps deserves underlining, because it’s common for Christians to encounter critics–even educated critics–saying something they think is quite clever: “You say everything has to have a cause. Well, who caused God?” They then congratulate themselves on scoring some important point here, but really they just show that they don’t know what they’re talking about.

The argument is indeed fairly simple, but not stupid! “Everything we know about is both finite and contingent. An infinite regress of finite and contingent beings is impossible. So there must be something that is infinite and necessary”–which is a good beginning toward a definition of God. (Thomas Aquinas said it better, of course.)

The third “refutation” says that the act of creation requires both space and time, but space and time are properties of the universe–the thing to be created–so creation before there is space and time is impossible.

This kind of thing also shows up a lot, and just stumps me as to why people think it’s a good argument. First, some of these people want to avoid the idea of divine creation by positing multiple universes–but suddenly space and time is limited to this and only this universe. Second, why is space and time dogmatically confined to this universe? Why can’t there be Someone who exists on his own, in his own space and time, who creates this universe (and perhaps others before, or alongside) when and where he pleases? Not only is this not an incoherent idea: It seems pretty obvious.

(Perhaps the ghost of Plato haunts us still, since both Augustine and Leibniz–each Platonic in his own way–actually argue something similar. They posit a timeless God, since they, too, have trouble imagining a time-space situation in which God exists before this universe is created. But if God simply is temporal and spatial in and of himself, as many of us Christian theologians and philosophers think makes the most sense of the Bible and of philosophy, then the problem vanishes.)

I like Bill Craig personally and I think he does a lot of good. I don’t agree with, or even personally like, all of his approach, which sometimes strikes me as flinging a bunch of arguments at one’s opponent and one’s audience too quickly for anyone to really be persuaded of any or all of them. But these arguments deserve intelligent treatment, and they don’t get them from the Apostate or his ilk.

0 Responses to “Atheist "Refutation" of the Cosmological Argument? Not Here, Not Now”

  1. Patrick Oden

    Seems pretty obvious? How could anything that is infinite and necessary be bound by any sort of space or time?

    How could something exist in time and space, much less be temporal and spatial in and of itself, while still having an infinite character?

    How is that obvious at all?

  2. Ed Gentry

    John, I agree with Patrick. It does not seem obvious that God is spacial and live in his own space-time. The idea does seem intriguing and it does make some sense of some of the Biblical data. (I have always thought that the common idea that God stands outside space-time is theological speculation with little support from the Biblical text.) But have you, as Patrick asks, now made God bot contingent and finite?

    You said “if God simply is temporal and spatial in and of himself”. Perhaps you did not mean that God is subject to space-time as we are, but rather that somehow space-time is intrinsic to God?

  3. robahas

    Thank you for posting this very interesting response!

    “Super-intelligent aliens from another dimension” – An attempt to win the argument by changing labels? This debate is all about what possibilities our world view allows us to imagine. Religious parlance says “God”, Science fiction says “Super-intelligent aliens from another dimension”. Apostate is willing to speculate on quantum states and redefinitions of “time” which then create “shaky ground” for the theistic argument. Speculation becomes becomes data. But the theistic argument does not get any flexibility. You need time to create the universe and so creation is impossible. Period. (His quantum argument could actually cancel out his time argument, right?) It seems to me that the possibilities here are actually manifold. I wish we could just talk about them instead of using them as pawns in a game that is already won in own minds. Still, I agree that apologetics is necessary…

  4. John Stackhouse

    The issue is the definition of “infinite” and its relevance to God. The Bible itself doesn’t call God “infinite,” and that word is fraught with difficulties in both mathematics and theology. But I suggest that “infinite” does not entail timelessness or spacelessness. “Infinite” means “without an end” along whatever analytical axis one is referring to at that moment.

    God is not “bound” by space and time, as if God is subservient to something greater than Godself. But God might be, as I suggested, both spatial and temporal. There is thus no contradiction, and if you prefer using “infinite,” then say that God is infinite in space and time. I prefer to say that God is everywhere and everlasting.

  5. Mike D

    Well! I had no idea I was being referenced, if not for an alert reader sending me an email. It’s late, I’m tired, it’s been a rough night, and I’m a little drunk. But I will respond as best I can:

    The “super-intelligent extra-dimensional aliens” is simply to postulate a similarly unfalsifiable hypothesis, to demonstrate the fallacy of that line of reasoning. If I were to attribute the exact same characteristics attributed to any theistic god to those aliens, it would be equally impossible to “disprove”.

    Secondly, it is simply a fact that the laws of physics allow a universe to be created from “nothing”; we know within a 1% margin of error that the universe is flat and has zero total energy. In quantum physics, if you have nothing, you will always get something. And finally, it’s worth pointing out that the point is not necessarily to “prove” beyond a reasonable doubt that there is a clear scientific explanation for the universe, but that the origins of the universe are in fact a topic within the realm of scientific inquiry.

  6. John Stackhouse

    Mike D, good to have you aboard!

    First point: Provability isn’t the point of the cosmological argument. And you can hardly just “attribute” such qualities as “infinite” or “necessary” to your aliens, can you? Then they would just be “God” or “gods” by another name and you would have shown the cosmological argument works. The cosmological argument as it stands, stands. (But please see below for an important qualification.)

    Second point: If in fact (rather than today’s or this year’s version of quantum theory) you can indeed get something from nothing, then that’ll be cool and certain theistic arguments will fail. On that we agree. But let’s let the cosmologists investigate and theorize and dispute to something approaching consensus, shall we, before we announce that indeed the universe emerged from nothing.

    Even if it can be shown convincingly (I don’t say “proved absolutely”) that the universe might have emerged from nothing, the question of whether God so arranged things such that the universe did emerge, and this (kind of) universe in particular, would not be answered. To be sure, one wouldn’t NEED a God to explain the origins of the universe and, again, certain theistic arguments would thereby fail. But the question of whether God did supervise this process or whether it was something all by itself would have to be resolved on other grounds.

    That is, to show that one can produce an effect (say, “a feeling of oneness with the supernatural”) by electrically stimulating a brain says nothing about what the cause of such a feeling is in any other given case. One can cause people to “smell” various objects not actually in the lab by similar means, but that says nothing about whether there are real-world causes for such sensations when one is walking down a street or purchasing food in a market. You can perhaps make me smell onions artificially, but that doesn’t mean there aren’t any onions–only that we don’t need onions to account for that smell.

    So theists do need to argue carefully here, as do our atheistic counterparts. Perhaps natural science can indeed account for the origins of the universe without the “God-hypothesis.” I don’t think it has done so yet, but maybe it will. Again, if it does, that will render certain theistic arguments otiose. But it won’t render THEISM otiose, since theism doesn’t depend on those arguments anyhow–or shouldn’t.

    Thus the cosmological argument continues to work so long as the universe is not indeed infinite and necessary. But if it were shown to be, then the cosmological argument would fail. To date, however, I don’t think it has been shown to fail (so long as it is properly construed). That was my original point.

    (For responses to the Guth/Hawking tradition of a universe popping into existence out of nothing via a quantum fluctuation, see Dr. Hugh Ross & his team at Reasons to Believe, and particularly here.)

  7. Ranger

    I’ve got a few comments/questions:

    1. I’m not sure that the “cosmological arguments” would be shot even if whatever latest theory showed that the universe self-created out of nothing or similar. It would still be a contingent entity. Aquinas’ original argument assumed an eternally past universe anyways. An updated form would be something like Mortimer Adler’s cosmological argument.

    2. I’ve only read a few books dealing with quantum fluctuations and theoretical physics. One was Smolin’s “The Trouble with Physics” that sort of jaded the way I view the field now. My question has to do with a quantum vacuum. This isn’t anything like the void we normally think about when we argue for something out of nothing, correct? I mean, this isn’t technically “nothing,” is it? Zero-point energy seems to be a measurable “something” with descriptive qualities, and thus contingent.

    3. Just for those interested, the American Scientific Affiliation had some great discussions this summer on multiverses and Christian theology which are available on their website ( under the “Annual Meeting” tab.

  8. Mike D

    Dr. Stackhouse, thanks for the detailed reply, and I respect you for challenging me. Thanks for the link too.

    The problem with the cosmological argument is something that I actually discuss in my most recent post, when I talk about the Kalam argument. “Causality” is a process that we observe from the material universe; matter and energy are not created or destroyed, but rather reconstituted in a complex chain of cause and effect. This is completely different than “creation ex nihilo”, or “something out of nothing”, which is the kind of supernatural causality that the cosmological argument is meant to infer. It’s an inductive fallacy to use observable, material causality to make inferences about speculative, unfalsifiable, immaterial processes.

    It will always be possible to contort a theology of God to fit with the inexorable forward march of science. But is that the kind of God anyone wants to believe in? One that science renders unnecessary, or is constantly having to be re-imagined to fit into a scientific description of reality? I’m struck for example at Christian biologists like Kenneth Miller and Francis Collins. Both recognize that natural selection is a blind, indifferent process which shows that the traits of neither species nor individuals has to be chosen with great care. Yet they suggest that God “directed” or perhaps “initiated” the process. I’m at a loss to see what, exactly, is gained from such a bald and unfalsifiable assertion. It seems to me that the only thing worse than a God that might not exist is a God that doesn’t need to exist.

    Ranger: zero-point energy is indeed measurable, but it is not contingent, nor would be the kind of timeless, pre-existing quantum vacuum that may have given rise to the universe. The universe would not be infinite as Aquinas had presumed, but be finite with no boundary or edge. Thus the universe would not be created or destroyed. It would simply BE.

  9. Gavin Polhemus

    I took Prof. Stackhouse’s advice and looked up Cosmological Argument on Wikipedia. I found this brief argument.

    1. Every finite and contingent being has a cause.
    2. Nothing finite and contingent can cause itself.
    3. A causal chain cannot be of infinite length.
    4. Therefore, a First Cause (or something that is not an effect) must exist.

    I have a Ph.D. in Physics and study quantum mechanical of black holes. I do not see why I should believe any of those first three statements.

    There are cosmological models where the universe has no cause (violating 1), causes itself (violating 2), or is infinite in time (violating 3).

    A model known as eternal inflation (which violates 3) makes predictions which have been confirmed by observations of the cosmic microwave background.

    According to the theory, prior to the big bang the universe had a very large cosmological constant, causing space to expand exponentially. The universe has a cosmological constant and is expanding today, so this idea isn’t very controversial.

    A large cosmological constant is unstable, allowing bubbles of lower cosmological constant to form spontaneously in the expanding space. These bubbles are expanding too, but with without the big cosmological constant to push them along, they will decelerate. The interiors of these bubbles would look very much like our universe, which has been expanding, cooling, and decelerating for most of its history. Density fluctuations in these bubbles would come from quantum fluctuations in the inflationary stage. Quantum fluctuations have a Gaussian distribution, which is what we see in the cosmic microwave background.

    Inflation models also solve a number of other problems, like the horizon problem and the lack of monopoles. They may even solve the fine-tuning problem, but there is still argument about that.

    Connecting this to the cosmological argument, there is no reason that the inflationary stage cannot be infinity long. The large cosmological constant is unstable, so bubbles are constantly forming, but these bubbles never gobble up all of the high cosmological constant region. The high cosmological constant causes the space between the bubble to grow faster that bubbles can form, so it never runs out.

    Eternal inflation is speculative. We need more data, and data is very hard to get. We have just put another microwave telescope in space to take more detailed measurements, which will take a couple years. Initial data shows that it is working well.

    My point is that while the cosmological argument states that “this can’t happen, that can’t happen, and the other is impossible, therefore: God,” scientist are busy making real mathematical models in which this, that, or the other do happen. Scientists are doing the math to generate real predictions and are then doing experiments to test those predictions. Amazingly, we are getting some good results.

    To the community of scientist doing this work, the cosmological argument looks very silly.

  10. John Stackhouse

    Thanks for these responses, guys, but I think what I say in #6 and Ranger says in #7 still respond to the basic points you’re making. The cosmological argument, properly construed, still stands.

    The open question–“open” in terms of the cosmological theories being formulated and tested, and none of them yet proved–is whether the universe somehow qualifies as being non-contingent, thus rendering the cosmological argument otiose. The fact, Brother Gavin, that there are models currently available that seem to flout this or that element of the cosmological model is not impressive, is it? What will be impressive is when one of them is proved, and we’re a long way from that.

    As for why theists keep modifying their views, well, that’s because everybody is modifing their views in the light of new knowledge. Theists believe that God made the world and keeps it going, so as we learn more about the world, we will modify our understanding of God’s creating and sustaining of the world.

    And why would we keep doing that instead of just abandoning theism or, better, Christianity in particular? We would do so because we believe we have very strong grounds on which to believe that Christianity is essentially true.

    I don’t think the strongest of those grounds is anything like the cosmological argument, however. I think stronger grounds are to be found in the discipline of history, per my post on the resurrection of Jesus. Others will be impressed by powerful spiritual experience. Still others will be convinced by theoretical philosophical arguments. And most will have read the Bible and become convinced (later they will understand that this is the work of the Holy Spirit) that it is the Word of God and truly does proclaim Good News.

    Still, I think the cosmological argument is not, in fact, easily dispensed with in the light of current cosmological theories. I’ll stay tuned, though, in case it finally is.

  11. Ranger

    Thanks for these very insightful comments, and whereas I don’t have time to fully keep up with the future discussion, I hope I have something to share, and know that I have plenty to ask!

    I’m thankful for Gavin’s explanation of cosmological studies. It took the difficult and put it in plain language, which the rest of us in this discussion (who are not experts in this field) can understand. Hopefully, I can do something similar for the philosophical side of the discussion and explain why I think we are talking past each other in regards to contingency (have in mind that in a field with four thousand years of history and lively discussions concerning the very earliest philosophical claims…nobody is an expert and I wouldn’t dare claim to be one!).

    A basic definition (from the Stanford Encyclopedia of Philosophy) is that contingent means that “something could have been different than it is.” Thus, whenever there is something that is measurable (and lends to the question, “Why is this measurement ‘x’ and not ‘y’?”) it would be philosophically contingent. Things stop being contingent when we are forced to say “This measure ‘x’ must necessarily be ‘x’ in virtue of its nature.”

    Furthermore, existence is a category of contingency in philosophical terms. Therefore, if it is possible that something not exist, then it is contingent. Therefore, if anything came into existence, whether by self-causation or external causation, then it is contingent, because there was a time that it did not exist. There was a point where “something [was] different than it is [now].”

    Someone earlier asked why God could not be contingent. Assuming the standard definition of God as a maximally great being (i.e. the God of classical theism from the ancient philosophers to the monotheistic religions to the one that most people worship today) then it would be impossible for God, as an infinitely maximal being to not exist. Therefore, according to philosophical terms, if God exists, then his existence is necessary. Inversely, if God does not exist, it is not possible that God exists, because such infinite maximal qualities entails existence. That is why the contested premise of most recent formulations of the ontological argument (ala Plantinga, Maydole) is the first premise “It is possible that a maximally great being exists.” According to modal logic, if it is possible that a maximally great being exists, then a maximally great being exists.

    Finally, I think there is a common mistake made in regards to these arguments from those on both sides. As Dr. Stackhouse said in his previous comment, most people believe for other reasons, or for a combination of reasons that largely fall under what theologians call special revelation. The point of theistic “proofs” is not to prove that God exists as though they were arguments that nobody could deny. Their actual intent (and of natural theology in general) is to show that it is rational for someone to believe that God exists apart from special revelation.

    If you can find a copy of Alistair McGrath’s latest book “A Fine-Tuned Universe?” that was compiled from his Gifford Lectures last year, it makes a similar point that is insightful for both theologians and atheologians. The actual presentations are available at the Gifford Lectures website here.

    Thanks for such an insightful reply. I do have questions though, but unfortunately may not have time to keep up in a conversation. Still, I’d be happy to hear your answers because I’m wanting to learn more.

    1. I’m interested in knowing the difference between models that have no cause, and those that are self-caused. Honestly, I don’t see how the former is philosophically possible, and why it wouldn’t simply be self-caused. Could you explain this further if you get a chance?

    2. Also, an eternal inflation model (is this the same as the chaotic inflation model that Andrei Linde has proposed?) would still be contingent in the philosophical sense (I hope I showed why this is the case above, but know that the discussion in blog comments is usually oversimplified and rarely helpful!). Please correct me if I’m wrong, because my reading in this area has been limited. Don’t Borde/Guth/Vilenken propose a point in the (unknown) distant past of an ultimate beginning, whereas Linde suggests otherwise? Either model would fall prey to a cosmological argument. The Guth/Vilenken model would fall prey to the Kalam (and other variations as well). The Linde model would fall prey to the cosmological arguments that are more Leibnizian (Mortimer Adler, Richard Taylor and Alexander Pruss are three major philosophers who have worked on this argument in the past twenty or so years). I don’t have time to formulate the entire argument, but I’m pretty sure you can find simplified formulations online, and I know the American Scientific Affiliation mentioned this latter argument multiple times during its discussions of multiverse models this past summer.

    3. You said, “we need more data.” I’m glad you admit that many of these models are still in early stages, as some people on both sides of these discussions overspeculate as to outcomes. Still, you also said that there are scientists making models, doing the math, testing and getting good results. In what direction are the results tending toward? Obviously, the data is similar in each of these models, whereas the models are radically different. So I’m just wanting to know which side has the upper hand in your thinking thus far. Thanks!

    4. I hope you don’t see the cosmological argument as you said at the end. I’ve never heard anyone make that argument or anything like it as an actual argument, and it comes across as patronizing to the thousands of living philosophers who study the latest cosmology and spend their lives working through these arguments (whether for – like Bill Craig, or against – like Quentin Smith). Wikipedia and YouTube three minute videos don’t quite ever allow people to say what they are intending to say! A better article (although now a little dated) would be Bill Craig’s essay from Astrophysics and Space Science. I’m sure there are more recent discussions and formulations, but it should take you a little past Wikipedia’s formulation.

    These topics will not be settled over blog comments, or even in forums because there isn’t enough space and the interaction never reaches that of two people sitting over a beer. Thank you though for such an insightful comment. I hope my comment helps clarify the philosophical side of things, as yours has helped clarify the scientific side.


Comments are closed.