115 years of debate about evidence for God

October 21, 2010 • 5:34 am

In correspondence with a Maarten Boudry, a Belgian philosopher of science, I’ve learned that my contretemps with P.Z. about the nature of empirical evidence for God is actually more than a century old. (As you may recall, I listed some evidence that would convince me of the existence of a god, while P.Z. argued that no evidence, however bizarre, could support the existence of a divine being.)

In my strident and militant New Republic article “Seeing and believing”, I wrote the following:

Many religious beliefs can be scientifically tested, at least in principle. Faith-based healing is particularly suited to these tests. Yet time after time it has failed them. After seeing the objects cast off by visitors to Lourdes, Anatole France is said to have remarked, “All those canes, braces and crutches, and not a single glass eye, wooden leg, or toupee!” If God can cure cancer, why is He impotent before missing eyes and limbs?

I’m now informed that this sentiment was actually expressed not by Anatole France, but by one of his friends.  Moreover, France—playing the role of either P.Z. or Massimo Pigliucci—impugned the probative value of those wooden legs at Lourdes.  Boudry sent me the quotation given below (my translation follows), taken from France’s Le Jardin d’Epicure (1895).  I’m putting the whole thing down for the sake those who want the correct reference.

Étant à Lourdes, au mois d’août, je visitai la grotte où d’innombrables béquilles étaient suspendues, en signe de guérison. Mon compagnon me montra du doigt ces trophées d’infirmerie et murmura à mon oreille :

— Une seule jambe de bois en dirait bien davantage.

C’est une parole de bon sens ; mais philosophiquement la jambe de bois n’aurait pas plus de valeur qu’une béquille. Si un observateur d’un esprit vraiment scientifique était appel constater que la jambe coupée d’un homme s’est reconstituée subitement dans une piscine ou ailleurs, il ne dirait point : « Voilà un miracle ! » Il dirait : « Une observation jusqu’à présent unique tend à faire croire qu’en des circonstances encore indéterminées les tissus d’une jambe humaine ont la propriété de se reconstituer comme les pinces des homards, les pattes des écrevisses et la queue des lézards, mais beaucoup plus rapidement.

My translation (excuse the poor French):

When I was at Lourdes in August, I visited the grotto where innumerable crutches had been put on display as a sign of miraculous healing. My companion pointed out these trophies of illness and whispered in my ear:

“One single wooden leg would have been much more convincing.”

That seems sensible, but, philosophically speaking, the wooden leg has no more value than a crutch. If an observer with true scientific spirit witnessed the regrowing of a man’s severed leg after immersion in a sacred pool or the like, he would not say “Oy vey, it’s a miracle! ” Rather, he would say, “A single observation like this would lead us to believe only that circumstances we don’t fully understand could regrow the leg tissues of a human—just like they regrow the claws of lobsters or the tails of lizards, but much faster.”

Here we see two types of methodological naturalism: a brand that dismisses the possibility of supernatural explanation a priori, and another brand that’s open to it.  This is the basis of my debate with P.Z.  Boudry is publishing on this distinction, and we’ll have more on that soon.

143 thoughts on “115 years of debate about evidence for God

  1. Here we see two types of methodological naturalism: a brand that dismisses the possibility of supernatural explanation a priori, and another brand that’s open to it.

    The problem is that you haven’t defined “supernatural” yet. It can’t mean “that which is beyond our current understanding”, because with this definition there would be no contradiction. It also can’t mean “that which is known to be impossible”, because then you’d be right to reject supernatural explanations. It can’t mean “that which is beyond all possible understanding”, because then it can’t ever explain anything.

    You can’t be open to supernatural explanations if you can’t tell us what they are.

    By the way, a wooden leg would only be evidence that someone left a wooden leg there, not that someone got their amputeed leg regenerated.

    1. Deen is absolutely right, Jerry.

      I’ve yet to come across a definition of the term, “supernatural,” that doesn’t reduce to either “impossible” or “paranormal.” That is, it either violates the laws of nature in an absolute manner (such as by drawing an Euclidean triangle with angles that sum to 360°) or it “merely” violates the laws of nature as we understand them (such as by allowing humans to communicate without the use of known technology, aka telepathy).

      Similarly, I’ve yet to come across a definition of the term, “god,” that doesn’t reduce to either “an entity that can do the impossible” (draw a square triangle) or “natural entity that is the object of reverence” (Hirohito, for example).



      1. You would be well-advised to drop the “square triangle” as an example of the impossible.
        For in non-Euclidean geometry, it is a trivial bit of work.

        1. That would be why I’m (usually) careful to specify Euclidean geometry.

          It’s equally trivial to do the trick by drawing a “triangle” with four lines, or with curvy “lines,” or by redefining the “degree” to mean a different fraction of a circle, or any of another variety of gimmicks.

          All such examples, including resorting to non-Euclidean geometries, are equally banal.



    2. Yes, I’m with Deen.

      I thought my opinion was more in line with yours, but now I’m not so sure. There are two assumptions in play here (I’ll phrase them as questions).

      1) Is the supernatural by definition impossible?

      2) Does it have to be supernatural to be God?

      PZ answers “Yes” to both questions. I answer “Yes” to the first and “No” to the second. It seems Jerry answers “No to the first! Which I find rather surprising… but I agree with Deen, perhaps it is because you have not clarified your definition of “supernatural”. By my definition of “supernatural”, a being endowed with the power to make up the laws of physics as She went along would still be “natural” if such a being existed.

      1. How would your creator goddess differ from the programmers of the Matrix or Alice’s Red King? Do they deserve deification? If so, why? If not, why does your hypothetical goddess?

        And it still doesn’t address the fundamental question that seems to bug people so: “Where’d it all come from?” It would tell us where the observable universe came from, but that’s much like discovering the origins of the solar system or the Milky Way. In your scenario, the observable universe would be but a small part of a larger reality, so where did that reality come from? A Hyper-Matrix? Alice’s Maroon Emperor?



        1. Are the motivations of the programmers of the Matrix really all that different from Yahweh of the Old Testament? Depending on humanity’s history of interaction with them, it might be fair to call them gods.

          Now, an interesting point you mention and that I’ve heard elsewhere (though not from either Jerry or PZ!) is that you can’t call ’em “god(s)” unless at least one of them is responsible for Creation. There is some validity to this restriction. If we accept that, then I would modify my stance slightly, but still not necessarily reject the possibility of any evidence convincing me.

          The existence of an external Creator of the universe seems to me to be a contradiction, a philosophical impossibility. It doesn’t make any sense to me — wouldn’t that Creator’s “house” have to be considered part of a larger super-universe? It would be like if you told me you could prove that the color green was the most overweight. The statement doesn’t even make sense, so I can’t even imagine what evidence would “prove” that to me.

          OTOH, I would have reacted exactly the same way if you told me a year ago that you could prove there were no hidden variables that went into the state of a particle that, if you could measure them, would predict the result of a quantum interaction. I would have mounted a philosophical argument that such a thing could never be proven, because it could only ever entail an Argument from Ignorance. That you could never know if some brilliant researcher was less than a week away from uncovering that hidden variable.

          And yet, physics and math conspired to give a big fat middle finger to that philosophical argument. Which is why, even if I think a thing is logically contradictory and therefore inherently impossible (not just empirically impossible), I still don’t rule out the possibility that there might be evidence that would convince me otherwise. I just can’t say what it would be, because I’m not as clever as John Bell.

          1. A valid argument certainly can be made for Matrix programmers being gods, but I’ve found all such unconvincing. If they are gods, then so are we, compared to our ancestors or even non-human primates. By my book, though, they would be idols. Very impressive idols, to be sure, but idols nonetheless. No different except in scale from, say, The Amazing Randi were he to use his prestidigitation to con a back-bush tribe into worshipping him.

            The logic that demonstrates that an “external creator of everything that exists” is meaningless is exactly as sound as — indeed, almost indistinguishable from — the logic used to prove that the Halting Problem is insolvable. And that logic itself is a minor variation on the theme that Gödel used for his Incompleteness theorem or Cantor used to prove the uncountability of the reals. I’d believe claims of a perpetual motion machine long before I’d believe claims of a creator of Life, the Universe, and Everything.



          2. Your analogy between “external Creator” and the Halting Problem is intriguing. Perhaps I am thinking about it backwards then…

            Now I need to decide if I think there is, even in principle, the possibility that some evidence could convince me that the proof of the Halting Problem was incorrect. Hrm….

          3. And BTW, if we except that in order to be a “god”, you must either have created everything that exists (including yourself I guess?) or be part of a pantheon which was either collectively responsible for all-creation or had a member that did it… and if we accept your assertion that the logic refuting this possibility is as sound as the Halting Problem… then I might be convinced to swing over to the “there could be no evidence even in principle” side.

            (Though not PZ’s side per se… I do not think he is arguing the case well, to be frank. At this point, I don’t think Jerry is either! hahaha…)

          4. Okay, given this exchange with Ben Goren, I will modify my stance slightly…

            If our definition of “god” requires that She or one of Her counterparts was externally responsible for the creation of everything that exists (an apparent oxymoron), then I would say that “I cannot imagine any evidence that would convince me of the existence of such a god.” I will stop just short of saying that there can be no evidence that would convince me, because it’s possible that there is some evidence and I just haven’t imagined it yet — just as Bell’s Theorem convinces me that the Hidden Variables Interpretation of QM cannot be correct, even though prior to hearing of Bell’s Theorem I could not myself imagine any evidence that would convince me of that.

            In terms of evidential requirements, I would put “A creator God exists” on par with “A Halting Oracle can be implemented in a Turing Machine”. I don’t quite rule out the possibility — but any attempt at mounting a case would have to begin by establishing why it is not a self-contradictory idea after all, i.e. by refuting the proof against it, before you could even begin to mount an evidentiary case.

            It is the same token by which any consistent theory which purported to be able to predict the outcome of a quantum interaction would have to include as its central feature a damn good explanation for why Bell’s Inequalities are violated. The theoretical flaws in the “proof against” must be addressed before we can even start making an evidentiary case.

          5. Fair enough. I don’t personally see any point in even granting the possibility, any more than I see the point in granting the possibility that maybe 1 + 1 doesn’t equal 2 after all…but I will concede that there are some people out there much smarter than I who might invalidate diagonalization. And then immediately go on to invalidate arithmetic as a necessary logical consequence. And, I must add, convince me that I’ve gone nuts.



          6. I think we are saying the same thing, really. We are both saying “The ‘evidence’ that would convince me of a Creator god would have to be on par with the ‘evidence’ that would convince me of 1 + 1 = 2.00001” It’s probably meaningless to even ask.

            BTW, you inspired me to finally clarify my thoughts in a blog post.

    3. One possible definition:

      A supernatural being is one that has mental attributes, notably volition, which are not grounded in or dependent on the physical world. A supernatural event is one brought about by such a being.

      N.B. Independence does not imply that such a being cannot act on the physical world: supernaturalists believe the mental to be prior to the physical, so the mental can cause physical effects.

      There is, of course, no evidence for the existence for the supernatural. But there could have been; and it is conceivable that some could be found in future. Jerry is right, P.Z. is wrong.

      1. Okay, you need to define “physical world” then. I would argue that a being that actually existed whose volition could influence the physical world independent of any other aspects of the physical world would itself be part of the physical world — otherwise it wouldn’t exist.

        But that’s because my definition of “physical world” is “everything that exists.” Maybe you have a different one?

        1. Of course if you define “physical world” as “everything that exists”, then everything is part of the physical world. I mean what physicists mean by it: the spacetime we inhabit, together with the particles and fields it contains. This does not allow for beings with mental attributes such as volition that are not made of the same fundamental constituents as stars, rocks, wombats, etc.

      2. So the paranormal would be supernatural? If telepathy were possible, it would involve the supernatural?

        supernaturalists believe the mental to be prior to the physical, so the mental can cause physical effects

        You may need to refine your definition — in a former life I did neuroimaging studies where people were told to imagine a sad event, and this volitional mental act produced physical effects in brain function.

        1. Telepathy would be supernatural if it was not using some form of “radiation” that could be studied by physicists in a non-mental context. Of course we wouldn’t necessarily know whether it was but if, for example, it turned out not to be attenuated by distance, not to be blocked by any physical barrier (or only by barriers made of living tissue), then it would be reasonable to hypothesise that it was “supernatural” in the sense I defined.

          Your neuroimaging studies are as far as I can see quite compatible with my definition, as long as volitional mental acts are themselves grounded in and dependent on the physical world. The point of the definition is that those who believe in the supernatural generally believe that minds are not in any sense reducible to, dependent on, or grounded in, the physical world.

      3. Thanks to Claude Shannon, it is as firmly established in physics as anything else, including Einstein and Newton, that communication requires energy. And, thanks to Turing, we know that computation requires communication.

        For your scenario to be true, either you have to outright reject basically all of physics or posit that we’re in a Matrix-style simulation. The former is not a tenable proposition worthy of serious consideration. The latter is unprovable for the same reasons the Halting Problem is unsolvable…but, even if it is the case, it simply broadens our horizons.

        I’ll let others comment on the well-trod ground of the immaterial interacting with the material, if they so choose.



        1. either you have to outright reject basically all of physics

          So there’s absolutely no conceivable evidence that would convince you to do so?

          Forget the God angle. What if tomorrow, you woke up and all of the laws of physics were different… coincidentally, they were just similar enough to allow us all to go on surviving and studying it. But it was a completely different set of equations, not even just different constants, but different equations altogether. Over the years, we start to understand these new equations through testing and observation.

          Then, some smarty-pants comes along and shows how both of the two apparently independent systems of equations — the ones that worked before Oct 22nd 2010 and the ones that worked after — were both special cases of a more general system. And that furthermore, the system will change again in another couple millenia or something, due to some predictable naturally occurring phenomenon.

          That wouldn’t convince you to reject all of physics?

          It’s a flight of fancy, obviously. And bringing it back to the God discussion, clearly pretty much any evidence we can imagine for it is such a ridiculous flight of fancy that we would never seriously consider any chance of it actually occurring (being atheists and all, y’know…) But we can still discuss it in principle.

          1. What if tomorrow, you woke up and all of the laws of physics were different

            I’d first conclude that I’d gone nuts. I’d be willing to tentatively concede that the programmers of the Matrix were playing a cruel joke, but I’d place more weight on the theory that somebody had extracted my own brain and put it in a vat. And I’d place even more weight on the theory that I was experiencing trauma- or chemical-induced hallucinations. To be honest, I’d probably be too scared shitless to think much of anything coherent.

            And, in any of those scenarios, I’d be much more interested in attempting to enjoy the ride than worrying about he ultimate nature of reality. If the very foundation of existence, or at least experience, is capricious…well, any attempt at understanding would be doomed to failure.

            In stark contrast to your hypothetical scenario, the universe I observe is nothing but rational. So long as it continues to be thus, I will continue to assume that it is. Should it ever cease to be thus, I will (hopefully) conclude that I am experiencing the well-known phenomenon of psychosis.



          2. It would require a *lot* of evidence to convince people that the laws of physics had suddenly shifted. No single experiment could do it. Even more evidence is needed to figure out what the new laws of nature would be. More evidence still would be required to find a possible higher-order system explaining both sets of laws.

            But if all that evidence is there, we wouldn’t be rejecting all of physics. We’d still be doing physics, and we’d have just adopted a richer understanding of it. The “old” physics would still be valid, but only on a more limited range of circumstances – just like Newton’s laws are still valid for speeds much slower than the speed of light.

        2. You seem to be under the misapprehension that I consider the existence of the supernatural plausible and worth investigating. I don’t; I simply consider it to be logically possible, and that its implausibility is established precisely by considering the available evidence. This evidence could, logically, have been different; and it is conceivable that future evidence could lend the supernatural, in the sense I defined, plausibility.

          The fact that it is unprovable that we’re living in the Matrix (or, of course, that we’re not) does not mean there could not be evidence that would raise the plausibility of the proposition that we are: for example, an amputated leg instantly regenerating. If we were living in the Matrix, then the actions of whatever agents are running the simulation would be supernatural by my definition, because they would indeed be independent of and not grounded in what we currently understand to be the physical world. We would be able to study the effects of their actions (which might be of the nature of run-time interrupts to the simulation), but would have no access whatever to the processes leading to those actions. Indeed, the Abrahamic ideas of God and creation can be interpreted as saying that we are living in the Matrix: that the physical world of the spacetime we inhabit is not the fundamental level of reality, but a construct of intelligences external to it, and able to change its rules at will.

      4. But if such a supernatural being observes our natural world (which it must, otherwise it can’t know how to bring about events in it), then how can its volition be independent of the natural world?

          1. But a programmer must also be aware of what’s going on in the simulation if she wants to interact with it. The events she’s observing in the simulation will change her mental state, which may cause her to make changes in the simulation. Her decision-making process may run on different hardware from the simulation, but since she’s actively interfacing with the simulation, she’s not independent of it.

            Now you could try to get around this by defining “volition” as a supernatural phenomenon, rather than as some sort of causal relation between internal mental states, observations and actions. However, in that case your definition of “supernatural” would just become circular.

          2. I think you’re just quibbling about the meaning of “independent of”, giving it a meaning I clearly did not intend. If I substitute “not entirely dependent on”, will that satisfy you?

          3. Then what do you mean by “independent”? And no, substituting “not entirely dependent on” would only make it less clear what you mean.

            And I don’t think it is a mere quibble. If it is not clear what “independent” means, your definition falls apart. And since you clearly want to leave open the possibility of supernatural events, it’s going to be hard to come up with a meaning of “independent” which does not imply a separation that prevents interaction between the natural and supernatural.

  2. The desperate faith of the mind to push the body and use the excuse of God for what the person doesn’t understand.

    Trading one crutch for another crutch, religion.

  3. I see this point, but:
    Possible evidence for the supernatural is not possible evidence for godhood.

    Some claims currently defined as “supernatural” can be reduced to a testable form. Thor, Zeus and Yahwe in his earlier days could be reduced to a testable form. Modern religousity can’t. If we are going to argue wheter gods can be the subject for evidence we it makes no sense to use out-dated views on what makes a god.

    The un-detectable, all-powerful god of modern abrehamites makes no sense in a discorse about evidence.

    Elements like faith-healing and even creation makes sense to test, but to sum from faith-healing and creation to a stealth-interventionist god is a non-sequiteur.

    And personally I find “supernatural” to be a carnival mirror and some smoke. Any phenomenon sufficiently understood will be very hard not to call natural. To claim something is supernatural is to claim there are things we cannot understand. While teists do this all the time (and it’s pretty much a part of the diagnosis), we should know better.

    Every time a claimed supernatural effect has been show and later understood, we have found a natural explaination for it.

  4. You know, Mythbusters should do a Lourdes episode – or Jerry should envision one for them. I actually think a lot of Religious claims are amenable to the Mythbusters approach – though not all.

    1. This would probably be more likely to show up in an episode of Penn and Teller’s Bullshit. I’ve never seen Mythbusters take on something with even a hint of religious overtones (please correct me if I’m wrong). I’d love to see it, but I think that’s a topic they’re afraid to touch.

      1. Adam is a pretty outspoken atheist on teh interwubz, but yeah, their show has never touched it. I don’t think it’s fair to assert without further information that the Mythbusters themselves are afraid to touch it — it seems just as likely to me that the network is ready to nix anything of that nature. (Remember, their show is on the same channel that makes mad bucks off of Ghost Hunters and other such nonsense)

        1. I was thinking about this after Bryan Dunning was writing some articles against the Mythbusters at skeptiblog. They could spend time trying to debunk paranormal claims, but its hard to do that as definitively as, say showing how to escape from a sinking car. A true believer will believe his stuff still works, because his stuff is different from what was tested (even if it is not).

          I’d like a paranormal debunking show, especially since Penn and Teller pissed me off really early in the run of BS by trying to claim that a 10 year EPA study was invalidated by a paper that was made in Winston Salem, NC, home to all cigarette companies ever. Seeing James Randi do his prize testing might be a fun show.

          1. I like the idea of a Randi paranormal debunking show, but he’d have to lower his standards a great deal just to let people to show up. A big part of the challenge is the development of the testing protocol, and even that’s a hurdle too high for anybody to meet.

            And…he’d only get rank amateurs as contestants. The pros would avoid him like the plague. And then he’d get all the lunatics…and a never-ending supply of dowsers…and, really, I’m afraid it’s one of those ideas that sound wonderful but never would actually work in the real world.



          2. To be fair to Penn and Teller, they later accepted that they were wrong and that they no longer support what they said in that episode.

  5. There are plenty of supernatural claims that could be verified by evidence. Look at the James Randi challenge. Any psychic that passes the test will have given reasonable (although not absolute) evidence for an ability that is – as currently defined – supernatural.
    The same thing goes for things like ghosts – a empirical evidence for poltergeist activity would go some way towards giving us a degree of confidence that such ‘supernatural’ things do exist.
    The same thing should go for Gods.
    To use the excuse that there is no consistent definition of “God” is merely skirting the issue. It is the same rhetorical trick that Massimo Pigliucci uses to argue that science cannot have any say on whether God acts on the world. Fingerprints left by a designer God (say a group of galaxies that, seen from earth, spelled out the first sentence of the book of Genesis, in English) would be a decent start. OK, we might not be able to rule out super intelligent aliens who were communicating a religious message but at least we have a piece of evidence that has a narrow range of possible explanations – one of them being that it is a message from a designer God. I am certainly willing to say that a piece of evidence that could only be explained by a ‘God’ or by god-like aliens would go pretty far in my estimation towards meeting the burden of proof for a God.
    Could it be the Christian God? – Well if it is, then it will not have the generic described characteristics of that God – since they are inconsistent and contradictory, but the contradictory nature of the various descriptions of that God doesn’t make a ‘God’ inherently impossible – just not one with the contradictions (meaning that even if the Christians are right(!) they are going to be still wrong about a lot.
    And to round off I can give a very simple thing that might convince me of God and its something that happens all the time, to even the brightest of people.
    A mental breakdown.

    1. an ability that is – as currently defined – supernatural.

      But discovery of it’s actual existence would make it no longer supernatural.

      Either that, or you have to call X-rays supernatural, because the ability to see through a person’s skin to their bones without making an incision used to be defined as supernatural.

      1. “discovery of it’s actual existence would make it no longer supernatural.” Indeed. But that’s not the point. The discovery of some plausible evidential explanation for God would make God natural (or at least part of nature) but the question here is not prove something that ‘cannot exist’, rather its whether it might be possible, in theory, to prove (or give good evidence for) something that currently lacks ANY evidence.
        The X-ray thing is interesting but a rather weak analogy – there are no historical stories of something supernatural that corresponds to the major features of X-rays.

  6. I am reminded of the Q from “Star Trek: TNG”; they were for all intents and purposes gods, but they had natural origins. We don’t know if anything like the Q exist. “God” is the one we all know about, and, if you will, the hypothesis that would to us best seem to fit an omnipotent John de Lancie descending from the heavens.

    Were we to see a “miracle” such as the instant regrowth of a leg, we could conclude it’s a divine miracle only because we’ve read of such things in the bible. Without that reference, we’d say simply that something incredibly unlikely has happened that we’ve never seen before; mass was apparently created in seconds, an adult leg was contructed, et cetera. But only those who know of the fairytales will attribute these events to those fairytales.

    Feel free to replace the above with “what could God do that the Q couldn’t?”

    1. Exactly. If we look at the traditional question in a different way, we see the problem: “What evidence would convince you that a being that claims to be an unimaginably powerful alien is in fact a god?”

    2. The god envisioned by the “sophisticated” theologists is omnipresent. The Q were not.

      So, a “true” god would be able to function in all places at all times. The Q were omnipotent, but not omnipresent, nor omniscient (nor, it must be added, omnibenevolent).

      I think my “test” is still the best…a god that could infer consequences of future acts that happen randomly everywhere around the world and act to mitigate those consequences would be acting in a “godlike” and not a “Qlike” way.

      Announcing that no weapon would work if the outcome were to be harm to a human would do it. Guns fire fine, but not if the outcome, intentional or not, would be harm to a human. Bombs not work, mines not explode, swords turn to nerf and then back to steel. Drugs used for lethal injection would have benign consequences. The hangman’s rope would turn to dust. Everywhere and always around the world.

      There are no physics that could account for that outcome. No technology that could instantly predict the outcome of a drive-by shooting.

      No “Q” would be able to accomplish that because they’re neither omniscient nor omnipresent.

      The discussion keeps getting mired in the omnipotent plane, and that’s only one of the attributes of a god. At least of the type of god we’re told demands our worship and 10% of our income to its messenger.

      We really do need to move past the simplistic notion of god-as-omnipotent to be able to understand just how goofy the modern conception of such a thing truly is.

      1. A Q-like being who demanded ten percent of our income and made us sing shitty music… I’d call that a God, for sure. Especially if he played mean tricks like, “Stab your son… psyche!”

        How would that not be the OT God?

        1. Oh sure. The OT god is definitely a Q-like being. Because despite the assertions of the “sophisticated” theologian-philosophers, it doesn’t demonstrate anything close to the attributes they assign to it.

          In Genesis, its powers are seen to be limited…it takes 6 DAYS to create the universe, and then it has to REST for a day. That’s implicit limitation on its potency. Then it can’t find Adam and Eve. There goes both omniscience and omnipresence. And then it kicks them out of the garden because they demonstrated its own incompetence and lack of omniscience. Out goes omnibenevolence with them.

          You can’t get out of the first book and all of the attributes assigned to “god” have been demonstrated to be false.

          So, we either have to redefine what a god “is” in terms of its attributes (and let’s not even start on ontology) or we have to acknowledge that the modern definition of “god” is most definitely not the creature described in the holy texts.

      2. Omnipotence, of necessity, subsumes all other omni- properties except omnibenevolence.

        An omnipotent being can trivially use its power to learn all there is to know and thus become omniscient. It can make itself present anywhere and everywhere at once, and so on.

        It’s also unable to make itself powerless, and therefore not all-powerful, so can we please stop pretending that the word has any meaning?



  7. Our ordinary concept of the supernatural includes the idea that it isn’t explicable in terms of natural laws and other empirically-derived explanatory relationships. Moreover, the supernatural is ordinarily thought of as being beyond our ken: we can’t understand or predict it; it has powers and capacities that aren’t reducible to, or explicable in terms of, known laws and mechanisms. So if we encountered something that defied natural laws in ways beyond our ken it would be fair to categorize it as supernatural. It would stand in stark contrast to the natural as we ordinarily conceive it.

    Since we can’t rule out the possibility of such an encounter – of the 4th kind, let us say – the claim that the natural world is all there is can in principle be falsified. So although metaphysical naturalism is an extremely good bet given all the available evidence thus far, although it’s an empirically well-substantiated conjecture about what is ultimately the case, we can’t take it as a foregone, a priori conclusion about reality. To admit this keeps naturalism consistent with its own epistemic foundations in science, and makes naturalists non-dogmatic fallibilists about their worldview. We must remain cognitively humble, not arrogant, given the possibility of someday being proven wrong by our own standards of evidence.

    From “Close Encounters of the 4th Kind:
    Metaphysical Naturalism as an Empirically Plausible Conjecture”


    1. the supernatural is ordinarily thought of as being beyond our ken: we can’t understand or predict it; it has powers and capacities that aren’t reducible to, or explicable in terms of, known laws and mechanisms

      Doesn’t that describe quantum mechanics? Or at least didn’t that describe quantum mechanics when it was first postulated? Didn’t that describe radioactivity? Viruses? Electromagnetism? How do we know something is permanently “beyond our ken”?

      1. How do we know something is permanently “beyond our ken”?

        I would have made similar arguments about both the Hidden Variables interpretation of quantum mechanics, as well as the idea that we could ever “know” that any two electrons were in principle identical. How would we ever know in principle that there wasn’t some discovery just around the corner that would allow us to predict the result of quantum interactions by measuring some other aspect of the particle, or that we wouldn’t suddenly find out that each electron had a very slightly different charge from each other electron?

        And yet, philosophical arguments aside, quantum mechanics has turned out to rather handily prove both of those things to be impossible. I would not have imagined what type of evidence would be sufficient to convince me that those two propositions were proven, and yet there it is.

        So, although I’m not a fan of Tom Clark’s definition in general, I think that we could know that something was “permanently beyond our ken.” Hell, we are virtually certain that there is plenty more stars ‘n’ galaxies ‘n’ shit out there beyond the known universe, but we are equally certain that we can never observe it due to the increasing rate of expansion of the galaxy. Again, I say I’m not a fan of Tom Clark’s comment because I don’t think that makes the superclusters that presumably exist beyond the edge of the known universe “supernatural”… but they are “permanently beyond our ken,” so to speak…

        1. And yet, philosophical arguments aside, quantum mechanics has turned out to rather handily prove both of those things to be impossible

          Impossible given our current understanding. There were plenty of things in physics that were previously thought to be impossible (e.g., the sun burning for billions of years), but which our later understanding has shown to be incorrect.

          The issue I’m trying to get at is deeper than just “what physics tells us know” — it is the meta-epistemological question of how we can know what things we can’t know. The claims of the supernatural essentially boil down to “we could never in the future in principle come up with a way that X could happen naturally” — I just don’t see how we can have that kind of certainty.

          1. Given our current understanding, sure, but the point is that our current understanding would have to be not only incomplete but wrong in order for the Hidden Variables interpretation to be accurate, or for two electrons to be distinguishable. It’s not like the difference between Newtonian physics and relativity, where the former ends up being a special case of the latter — two distinguishable electrons would disprove quantum mechanics.

      2. Tulse: “How do we know something is permanently ‘beyond our ken’?”

        We don’t, which is why I go on to say in the article that the supernatural claim is defeasible since our understanding and theories might end up incorporating what now is beyond our ken.

        1. The better term for what you’re describing is “paranormal.”

          ESP as commonly portrayed in fiction would fit your definition of supernatural…but, then again, a century ago people using smartphones to the same effect would also have been supernatural. We now know that there’s nothing at all supernatural about observing events at a distance and can even theoretically posit ways that it could be done in the same style as in fiction with only relatively minor advancements in technology.

          Using “paranormal” to describe such phenomenon is very accurate: they are now beyond our ken, but there’s nothing that says they always will be. They’re also not well supported by evidence.

          Using “supernatural” to describe such phenomenon is misleading, as it implies not only that we don’t currently understand it, but that it is something that is impossible to understand. And, again, of course, evidence is lacking.

          If we have evidence, calling it a natural phenomenon is perfectly fair game whether or not we understand it. Gravity, for example, still eludes our full understanding — and what could be spookier than two objects inexorably and endlessly pulling at each other, silently, over distances that defy imagination?

          Even if it requires an intelligence beyond the physical capabilities of a recognizably-human brain to understand a phenomenon, I still wouldn’t want to label it with the term “supernatural.” We can, after all, easily envision synthetic brains far superior to our own that will one day be built as the result of a process set in motion by humans. Indeed, a fair chunk of mathematics is being done today in precisely that manner…and yet the results are clearly not supernatural. Or even paranormal, in that particular case.



    2. I would say that supernatural is not beyond our ability to understand it…actually, I think implicit in any definition of the supernatural is that we do understand it as a violation of our knowledge of the natural world.

      It’s not “we can’t explain it with our current level of scientific sophistication”.

      It’s “we understand the scientific concepts underpinning the phenomenon, and there’s no way this can happen, therefore some other-worldly power that can alter the known forces of nature was at work.”

      If you look at the so-called modern miracles, they’re all very ordinary in terms of the physics. A pilot lands his disabled plane on the Hudson River, saving the lives of all aboard — well, that was following the ordinary laws of aerodynamics. Nothing “miraculous” about it.

      If the plane had hovered above the water without being under power, or teleported instantly to the hangar, or if the birds had been sucked into the jet and out the other side without harm to themselves or the airplane, or if only the faithful who prayed were deposited instantly safely ashore while all else perished…well, that would take a little bit of explaining.

      It’s all about breaking the laws of physics in ways that we know isn’t possible.

      Frankly, whenever I’m asked about the miracles of the bible and the NT in particular, it’s not the “impossibility” that I object to. It’s the fact that the laws of physics being violated are so substantial that the miracle is banal by consequence.

      Violate the laws of gravity and surface tension of water without technology? Well, OK then. But why just for a few seconds in front of only a handful of witnesses?

      Violate the laws of nucleogenesis? OK, then. But why do that at a wedding by turning water into wine? Where the evidence disappears in a few hours (and can’t be verified or tested in any way)?

      The power implicit in the miracles is truly awesome. The miracles themselves? They’re the output of First Century primitives.

      1. It’s all about breaking the laws of physics in ways that we know isn’t possible.

        But how can we ever “know” this? Radioactivity broke the laws of physics in ways that, at the time, weren’t possible. The orbit of Mercury broke the laws of physics in ways that, at the time of Newtonian mechanics, wasn’t possible. Most modern gadgets break the laws of nature as understood by our 18th century forebears.

        How do we know what we can’t know? How do we know what we will always think is physically impossible?

        1. I think we do know a lot about the laws of physics. Enough to know when they’re being violated. That’s the point.

          We know that it takes the power of a fusion reactor to create heavier elements from lighter elements. Being able to create complex carbon molecules (ie, wine) in a vessel containing ordinary water requires a certain amount of power no matter what way you look at it. If there’s no fusion reactor in evidence, then the power to change water into wine violates what we know about physics.

          That’s why the miracle itself doesn’t work. The power implicit in having a fusion reactor being used in such a trivial, ephemeral way? Really…it’s quite silly to even consider.

          It’s not the power that’s the issue — it’s the use of the power in a manner that only someone without an understanding of the physics involved would dream up.

          1. Being able to create complex carbon molecules (ie, wine) in a vessel containing ordinary water requires a certain amount of power no matter what way you look at it. If there’s no fusion reactor in evidence, then the power to change water into wine violates what we know about physics.

            “An object the size of the sun being able to burn at its temperature requires a certain amount of coal or other combustible substance. Given that, the ability of such an amount to burn for millions of years violates what we know about physics.” That was pretty much our state of knowledge prior to the discovery of fusion. Indeed, it was an important piece of the argument about the possible age of the earth and evolution.

          2. Okay, but some of the proofs we are talking about are more strenuous than that one. “No combustible fuel could generate that kind of energy for that length of time with that mass” was something that was known as fact, and that was correct. What was wrong was whether there was some other hitherto unknown type of energy.

            Some of the things we have been discussing in this thread, it would be more like if 18th century scientists thought they had a proof that “Energy can never be generated more efficiently and rapidly than with combustion, with any process, even if it were a hitherto unknown form of energy.”

            The discovery of nuclear reactions did not invalidate the chemistry of combustion reactions. But, if we found two electrons that were subtly different, that would invalidate quantum mechanics. If we found a hidden variable that would allow us to predict a priori the result of a particular quantum interaction, that would invalidate quantum mechanics. It wouldn’t just add to it.

            You’d have to read the explanations for why this is so in order to be convinced, I imagine… I dunno if you are aware of them?

          3. The “any two electrons must be identical” thing in a nutshell:

            You can do an experimental setup where you have two electron emitters, two electron detectors, and a bunch of half-silvered mirrors in between them. If the two electrons are indistinguishable, then it turns out the amplitude of the quantum wavefunction for “Emitter A hits detector 1 and emitter B hits detector 2” exactly cancels out the amplitude for “Emitter A hits detector 2 and emitter B hits detector 1”, so the prediction would be that we’d only ever see two electrons are detector 1 or two electrons at detector 2 — not one at each. But if the electrons were distinguishable, then the quantum amplitudes represent two different outcomes, and they don’t cancel, they balance. So the experimental outcome we’d expect would be half the time seeing one electron at each detector, 1/4 of the time seeing two at detector 1, and 1/4 of the time seeing two at detector 2.

            We can do this experiment. And the experimental results reflect the first outcome. This proves that — unless quantum mechanics is not just incomplete but wrong — any two electrons are indistinguishable.

            You can’t add something on like, “Oh, but quantum mechanics didn’t take into account phenomenon X, and that explains how two electrons might be distinguished.” You’d have to say, “Oops, QM is completely wrong. Toss it out.” That’s totally different from adding nuclear reactions to chemical reactions… when nuclear reactions were understood, we didn’t say, “Oops, that whole combustion thing was an error, toss it out.”

          4. By the way, I am not quite as familiar with the physics, but I think the certainty with which we believe practical FTL travel to be impossible is on par with this. Relativity would have to be not just incomplete, but wrong. Which is why I would say that the presumed superclusters beyond the known universe are “permanently beyond our ken”. That might be wrong… but I’d be as likely to believe that the inverse square law of gravity was just plain wrong (not just an approximation, but wrong wrong wrongety wrong)

          5. James, if I may simplify your explanation a bit:

            Observing distinguishable electrons would be comparable to observing stones falling up, discovering a rabbit in the precambrian, or watching the Sun rise in the West. (And, please, everybody, spare us bad the jokes about hidden air vents or supersonic westbound aircraft or the like.)

            Again, if we’re in a Matrix-style simulation, it is conceivable that the programmers could press some buttons and suddenly change everything we thought we knew about the universe. All bets are off. But I’d still be thinking, “So this is what an acid trip is like. I wonder who slipped me the Mickey Finn, and what’s happening to my body back in the real world?”



          6. Wait, so if a supersonic jet is flying west, which will it encounter first: A Precambrian rabbit, or a pair of distinguishable electrons?

            And what kind of sadist would stuff a Precambrian rabbit into an air vent?

    3. Again, if “supernatural” simply means “beyond our ken”, then a supernatural explanation isn’t an explanation at all, and must therefore be rejected. And by the time we can actually explain the phenomenon, it is no longer considered supernatural.

  8. If cosmologists were to make a discovery tomorrow that demonstrated that the universe had to be caused by some entity’s action it would do little to persuade me that repentance of my sins (sins being defined by the Bible) and trust in Jesus as my savior. So there is a Creator. Absolutely central to Christian belief is the idea that there is an afterlife that will be experienced in either heaven or hell for eternity. Coyne’s example of a mass regrowth of limbs that can be traced to the Creator or its agent would lead me to believe that the deity is capricious and mean – why not save Haitian earthquake victims, why allow severely damaged children to be born, suffer mightily, and then die?
    The demonstrated existence of hell is the only evidence that would cause me to want to change my thinking and align it to some theology. Other than that, bupkus!

  9. I think the existence (or not) of evidence is the trivial part of the question – despite the many comments arguing the matter.

    There are recorded instances of people being ‘struck’ with belief through seeing frozen waterfalls, or hearing God etc.

    Since there seems to be little rational control over the switch into belief I wonder if anyone, even the most uber-rational, can assert that they will never experience being god-struck. Most unlikely to be god-struck because of atheist pre-disposition, sure. But never?

    So I think the quality of evidence need only be trivial – when the switch happens. Before then it needs to be really strong.

    Some people are raised god-struck. Some people will never be god-struck. But some will switch on the flimsiest of evidence.

    I think that is more of a concern to those who value rationality.

  10. I’m wondering: if God is omnipotent, and he can regenerate a leg in seconds, could he teach us how to do it with technology instead of magic? If so, is regenerating limbs really a supernatural process? If not, is God really omnipotent?

  11. My simplistic way of thinking about the “supernatural”: everything in and of the universe if “natural”, but anything from “outside” it would be “supernatural”.

    Is there evidence of any such thing? No. Could there be? Maybe.

    Suppose something — let’s call it Ceiling Cat — existed in some dimensions or whatever outside this universe. Through His whiskery magic, He inserts His paw into this universe to muck about with something.

    Now, the effects created/left by His mucking about would be natural and could in principle be studied by science, even if CC could not be studied directly.

    One could of course say, “There’s no ‘place’ outside of the universe — the concept makes no sense.” This blasphemes Ceiling Cat, but perhaps He cannot hear despite being supernatural (or imaginary).

    On the other hand, the concept at least seems coherent to me. That’s why I’m the second sort of naturalist: “open” to the possibilty of something outside “nature”.

    1. Through His whiskery magic, He inserts His paw into this universe to muck about with something.

      But doesn’t this admit that the supernatural is not separate from the natural, but is in fact in contact with it?

      1. Presumably it would be at least temporarily in contact with “nature” to muck about with it, but because His Whiskerliness was generally “outside” nature it would not be subject to natural principles.

        Now, just how this might be possible is entirely conjecture, and it could very well be impossible. But it seems at least conceivable and could point toward a testable hypothesis. The origin and nature of Ceiling Cat might remain inscrutable, but His mucking about would be detectable and could be studied.

        If no evidence were ever found, it would be a failed hypothesis.

    2. I’d say that, given your example, languages is inadequate.

      We still need a word for the set of all that exists and is real. I’m still quite fond of “universe” for that term, but perhaps Dr. Sagan’s Cosmos is better suited.

      Put in that perspective, Ceiling Cat is no more a god than a ten-year-old-boy with an ant farm and a magnifying glass.



    1. However, the supernatural quickly retreats to hypothetical situations and the gaps in our knowledge of nature. There is only the natural. There is no supernatural. There isn’t so much as a single sentient deity. The universe was not created by a sentient being. I disagree with Jerry’s argument against the apriori dismissal of the possibility of a deity.

    2. Except (probably) some deranged trekkies, no body is worshipping Q.

      The question is not “could there be evidence of a entity that people would view as a god”, it is “could there be evidence for a god”.

      And by the emic definitions of gods used today, there can be no evidence. You can’t have evidence of the deist god, the stealth-interventionist god or the vague god (ie. “god is love” and suchlike).

      If you know of a defintion of god used by educated people today that is testable: ‘Fess up! I don’t.

  12. If we want predictive power and control over nature, then we must seek naturalistic explanations.
    However, religious explanations do not have that goal.
    Moreover, in philosophy of science, there is much discussion about the reality of scientific entities. See the entries in the Stanford Encyclopedia of Philosophy about “scientific realism”, “structural realism”, “constructive empiricism” and antirrealism, for example.

  13. IMHO Lourdes is a great example.

    How many people come away from Lourdes healed? How many people not in Lourdes get healed?

    The factor (according to Sagan) is 10X. That’s some pretty strong evidence right there. Except not in the direction the godbotherers want.

    1. I don’t understand your logic. Now you require a certain “healing rate” in order to concede that there was a miracle? How much? 10%? 50%? If miracles were not exceptional, they would be “natural,” don’t you think?

          1. Let me rephrase that: I’d go with one modern, well-documented, well-researched restored amputee.

      1. The chance of a “miracle” is much higher (by an order of magnitude) any place EXCEPT Lourdes.

        So why is Lourdes viewed as a place for miraculous healings?

      2. I think you misunderstood the point, Carlos. It seems that even tap water dripped on an injured leg in New Jersey (a sham control) has a better chance of spontaneously healing than Lourdes water.

  14. I’ll still go with PZ. Assuming a ‘miracle’ is demonstrated (miracles have never been credibly demonstrated), how do we even attribute this to a deity? We would be setting up control groups and trying to discover if these ‘miracles’ are repeatable and under what conditions. I would be interested in hearing what conditions would satisfy a scientist that something is indeed a miracle brought about by divine intervention.

    The “I’d like to see a leg grow back” is simply facetious; it never happened and unless humans somehow evolve to regrow limbs, a leg will never grow back on its own.

    Perhaps we should abandon the infantile arguments such as “I’ll be convinced if I see an eye grow back” and stick with reality: upon conducting a scientific investigation of a reported miracle, what will it take to convince you that it is indeed the intervention of a deity? I think that is the relevant question. Just look around – Mary McKillop has been declared a ‘saint’ for being involved in at least two miracles. The Vatican’s investigators assert that there were indeed miracles while sensible people laugh with derision or shout angrily at the pious frauds.

    1. “I would be interested in hearing what conditions would satisfy a scientist that something is indeed a miracle brought about by divine intervention.”

      Certainly the case if by “scientist” you mean a ‘scientismist’, because I suspect Francis Collins or Ken Miller could tell you under what conditions they’d accept something as a miracle.

      1. You know what is most important for this kind of experiment:


        I could lend you a hand here: I’ll chop off your leg and you pray really hard (it has to be this way around since i don’t BELIVE) – and we’ll see if it grows out or if you’re just listening to undocumented tales from 300 years ago…

        1. If miracles were repeatable they would be natural phenomena, not miracles.

          Now, I am curious: do you just accepts truths that can be verified by repeatable experiment? Do you accept that Caesar was assassinated on 3/15/44 BC? That is much less thoroughly documented that the Calanda case…

          1. A roman politician dying of natural causes (a dagger in the back being pretty natural) and the statement having no implications beside one being slightlty altering his state.

            I got no problem with that. It could bloody well have happended the day before or the day after of course, but guess what: It doesn’t really matter. Juilius Gauis death has no direct influence today except as a reference. And for that use the myth is far more used than the (historically accepted) facts.

            Most importantly: The whole story rings true and there is no reason to suspect people are playing silly buggers.

            Extraordinary claims require extraordinary evidence. Learn to live with it.

          2. Right, so no amount of historical documentation will ever convince you of any extraordinary fact, because by definition the witnesses must be “playing silly buggers.”

            This discussion started because somebody asked for miracles about amputated legs. I provided and historical example and predicted that people like you would never believe it REGARDLESS of the good or bad quality of the historical records. You just proved my points (and so did Rob in his reply)

          3. “Right, so no amount of historical documentation will ever convince you of any extraordinary fact, because by definition the witnesses must be “playing silly buggers.” ”

            Aye – that pretty much nails it.

            You know – history isn’t science. Historical facts can only be trusted as far as they play by the rules.

          4. ‘cept of course: It needn’t be the audience who’s playing silly buggers, most likely it’s the show.

          5. Except, you are the one setting the rules. What if reality does not like your rules?

            Just hope that never in your life anything important depends on ascertaining the truth of any extraordinary historical claim, because then you will be introuble.

          6. Any amputee restorations in the 20th century, one that can be examined by scientists?

            No? Quit trotting this one out.

          7. Oh, excuse me. And don’t you think that 20th centuries scientists could be bribed to write false reports (like all the Lourdes stuff)? I honestly don’t think you would be convinced. But perhaps we can arrange for a little miracle to take place in your living room?

          8. Oh, and how do you “demonstrate” a miracle? I assume you don’t give any credence to the reports of the Lourdes scientific commission, so what else would convince you?

          1. You said “Historical facts can only be trusted as far as they play by the rules.”

            What rules?I thought you meant that reports of miracles in general could NEVER be trusted because they do not play by what have you decided a priori to be plausible. If that’s not what you meant I stand corrected but your previous post was unclear.

          2. If we had miracles properly described by scientific method, it could somewhat be trusted (certainly controlled rigorously of course).

            The main thing is, history isn’t science. So when a single piece of history contradicts known science, it has to be ignored until science can confirm it.

            Science works, in this context, history don’t.

          3. I thought we agreed that science cannot confirm historical facts, because historical fact are not repeatable.

            So my point stands: you decide a priori what historical facts may or may not have happened based on your conviction that natural laws as we currently understand them are absolute. I only accept natural laws inasmuch they reflect experience, and as such I am perfectly willing to see them violated if my experience (or the experience of a trustworth witness) says so. Even if the experience is not repeatable.

          4. You’re right, science can’t confirmn history.

            Science can disprove history by demonstrating that the accepted fact isn’t physically possible thought. Or make a historical truth less plausible by demonstrating that another explaination is more probable.

            This has happened time and time again. There was a time the flood was accepted as a historical fact. Science and archeology has disproved it.

            As I said before – it would serve you well to actually learn what science is and what it does. It doesn’t seem like you understand it.

          1. There’s documentation, and it was done in front of a lot of people.

            How is it any less credible than a restored limb?

          2. Perhaps, but I don’t go around calling people names.

            And by the way, the David Copperfield example was also not very smart, you know?

    2. “The Vatican’s investigators assert that there were indeed miracles while sensible people laugh with derision or shout angrily at the pious frauds.”

      Puff yourself up how you like, ‘Mad Scientist’, but you’re not much of an example of the sensible. Either the ‘pious frauds’ believe in the existence of miracles and that subject events are miracles in which case they’re not frauds or they don’t believe in miracles but just know which side their bread is buttered on and thus aren’t pious. But thanks for playing ‘So you think you can reason’.

      It’s been asked elsewhere whether the Gnu Atheism is anything more than mere emotionalism. I don’t think that is the case but rather that, as demonstrated by Mad Scientist above, the distinguishing feature of Gnu-A is its emotionalism.

      1. Wait, you’re getting hung-up on the fact that, if you parse everything as literally as possible, “pious frauds” is a contradiction, and then getting all up in somebody’s face about that… and calling them emotional?


  15. You forgot to mention the part where *nobody* has ever regrown a leg from having it immersed in Lourdes’ healing waters; nor are they ever likely to. Ever.

  16. @ Ben Goren, I’ve yet to come across a definition of the term, “supernatural,” that doesn’t reduce to either “impossible” or “paranormal.”

    You’re conflating logically impossible with physically impossible, and without an axiomatic explanation for “nature” (define “nature” too, please), there is in principal no reason we know of that a “supernatural” entity could not break the physical laws known to us.

    The simplest of known physical laws that affect most of what happens to us simply says that the probability of an event equals exp(iS[x]/) summed over all paths x(t), where S is the classical Lagrangian in units of Planck’s constant . Where the hell does that come from? No one knows. It simply works, to 10 decimal places, but no one knows why. Richard Feynman on this unsolved problem: “There is no theory that adequately explains these numbers. We use the numbers in all our theories, but we don’t understand them—what they are, or where they come from. I believe that from a fundamental point of view, this is a very interesting and serious problem.”

    Feynman expresses the distinction between logically derived principles (sum of angles in Euclidean triangle equals 180°) from physically derived principles (QED) as the difference between Greek mathematics (axiomatic) and Babylonian mathematics (whatever works). We may may be able to achieve an axiomatic foundation for known physical laws which would make physics simply another branch of arithmetic, but this is not guaranteed, either because it is not true or because we are unable to make the physical measurements necessary to test any such explanation.

    The principles you’re using to exclude absolutely the possibility of a “supernatural” god—invoking Shannon, Einstein, and Turing—all depend upon the absolute truth of QM and GR. Fine, but remember, QM and the no-communication theorem were only recently(!) validated in the Bell test experiments, QM GR is a no go, and we have absolutely no idea why QM works in the first place or even where it comes from.

    So from a dorm-room bull session philosophical point-of-view, there is no scientific reason I’m aware of that would preclude an all particle-controlling deity that could either move all the air into the corner of the room or restore eyeballs and limbs to blind amputees.  Absolutely precluding such a deity on these grounds necessarily needs to say why QM must be an absolute logical truth, not merely a well-validated physical truth.

    1. Your hypothetically-omnipotent deity would be incapable of physically drawing, on a flat sheet of paper, a geometric figure that is the representation of a triangle with angles that sum to anything other than 180°. Or, at least, it would be unable to do so without bending spacetime in fascinating ways that would nevertheless constitute cheating exactly as much as drawing a four-sided triangle or a triangle with curved edges or stretching the paper over a globe or whatever.

      It is a physical impossibility as well as a logical one.

      If the god is allegedly omnipotent, then it is physically impossible for it to commit suicide. Committing suicide would render it incapable of doing even the most trivial of things after the moment of death, which hardly meets the criteria of omnipotence in any form. Either the god is capable of suicide, in which case the god is not omnipotent, or the god is incapable, in which the god can’t do something almost any human can do and therefore not omnipotent.

      The god is also physically incapable of either convincingly causing another entity to falsely believe that the other entity or is incapable of determining whether or not s/h/it is the one being duped. The ability to do either precludes the ability to do the other; in either case, the god is limited and thus not omnipotent.

      Further, any claim that a meaningful distinction can be made between different types of impossibilities with regards to omnipotence renders the terms meaningless.

      No god can do the square triangle trick, but that doesn’t count against the god’s alleged omnipotence. I can’t do the square triangle trick, either, but that can’t count against my own alleged omnipotence without resorting to special pleading. Assuming GR is absolute, the god can’t exceed the speed of light, but neither can I. If the god can, after all, exceed the speed of light, it’s entirely plausible I (or a distant relative of mine in the future) will do so, too, so how is the god’s ability to do so significant in light of all it can’t do?

      Finally, there is an empirical experiment you can perform right this very instant to determine whether or not there are any omnipotent beings in the universe.

      We know that, if one searches the entire range of possibilities and finds something missing, that something doesn’t exist and is, by definition, impossible. For example, it is impossible for any digit other than “3” to appear in the decimal expansion of “1/3.” We have sound methods for exploring that infinite space, and the fact that nothing other than 3s can be found there is, in and of itself, sufficient proof that only 3s can exist in that number.

      So, here’s the experiment: are you reading these words?

      If so, then we know, without doubt, that no omnipotent beings exist.


      Because, if you’re reading these words, then you know, without doubt, that there is no point in all of past, present, and future history (or outside of time or whatever) in which an omnipotent (or even merely very powerful) deity retroactively erases you from history. If unimaginable aeons from now a god decided to rub you out and prevent you from even being born, you would not be able to read these words. (Maybe the god does so to some alternate you in an alternate universe, but that has nothing whatsoever to do with the original you reading these words right here and now.)

      In exactly the same way as we know that anything other than a 3 in the decimal expansion of 1/3 is impossible, we know that it is impossible for gods to retroactively cause our own nonexistence. We have thoroughly searched the entire possible space and found no evidence of such a thing occurring.

      The particular reasons why it’s impossible are irrelevant. It might be that the power is beyond even the most powerful god; it might be that the gods are too kind-hearted to carry out such an action. Regardless, it is an impossible action, even for the gods.

      And, if it’s impossible, then the gods can’t, again by definition, actually be omnipotent. And you’ve got the experimental proof staring you right in the face.

      I can continue, if you like….



      1. You lost me when you started talking about God as an “omnipotent being.” I subscribe to the classical Thomistic notion that God is not a being, He is Being. So, I agree with you there is no omnipotent being, but that does not make me an atheist.

          1. You’re too generous. Personally, I’d be more inclined to ask Carlo what kind of dressing he’d like to go with his word salad.

            Maybe some Sangre de Christo vinaigrette, and topped with some Jesus Cracker Croutons?



          2. Apparently, it means he is a fan of Aquninas.

            You know – unmoved mover and all that BS we’ve falsified long time ago.

      2. You’re still missing an important point about _current_ knowledge. We KNOW that it’s impossible for any entity to violate logic: no square Euckidean triangles and no combined omnipotence and omniscience.

        We do _not_ KNOW if it’s possible to violate QM. We don’t even know why QM describes reality. Therefore, we do not now have a scientific basis to NECESSARILY dismiss the possibility of a particle manipulating deity, even though we CAN dismiss a deity Who is both omnipotent and omniscient, or possesses any other logic-violating capabilities.

        So the question stands: if, in accordance with Jesus’ words in John 14:14 “You may ask me for anything in my name, and I will do it,” we observed that everyone was granted whatever they asked of Jesus (constrained by logic, but not necessarily QM), would that be sufficient evidence for the Christian god?

        1. First, there’s the matter of quantum mechanics the theory versus the observations of quantum phenomenon.

          Even if (as is almost certainly the case) our understanding of quantum phenomenon is incomplete, any new theory will have to be at least as good at explaining the overwhelming mounds of evidence as quantum mechanics is.

          Your question in the last paragraph borders on nonsensical.

          If the two millennia we’ve been running that experiment with negative results isn’t enough to disprove the hypothesis, then what in Jesus’s name is?

          Even if Christians started running around performing miracles all of a sudden, it would in no way invalidate the twenty centuries of observations that Christianity is bullshit. It would, of course, be evidence of some incredible new phenomenon, but we can be quite certain it’s not the Jesus of the Bible. Aliens taking advantage of a preexisting popular mythology would be overwhelmingly more likely, and a personal hallucination even more likely still.

          But why give the Christian mythology so much credit? Why not ask if a young man with a peculiar scar on his forehead who rode a flying broom into the middle of Trafalgar square while waving a stick would be sufficient evidence for Harry Potter? Why not ask if an X-Wing and a TIE fighter landing in the middle of Times Square and the pilots emerging to have a Force-powered lightsaber duel is sufficient evidence for Star Wars? Why not ask if a cyclops, a woman with snakes for hair, and a winged horse wandering around downtown Athens is sufficient evidence for ancient Greek paganism?

          I fail to see how the exercise is anything but a variation on a child’s “What if?” game coupled with the even younger child’s infinitely repetitive “But why?” game. There comes a point when pondering what color an invisible unicorn would be if monstrous spaghetti were capable of flight is pointless nonsense.



        2. This would be evidence that prayer (to Jesus specifically) works, yes. It doesn’t prove that Jesus is also God, and therefore the Creator of the universe.

          Besides, nobody can ever live up to “You may ask me for anything in my name, and I will do it,” for the simple reason that two people can ask for mutually contradictory things.

          1. Besides, nobody can ever live up to “You may ask me for anything in my name, and I will do it,” for the simple reason that two people can ask for mutually contradictory things.

            “Oh Jesus, I ask in your name that you be unable to do anything asked for in your name.”

      3. @Ben,

        What are your thoughts about other definitions of omnipotence? That is, those which do not grant the power to do the illogical.

        1. I’ve written on them elsewhere, including in this thread. Short version: they render the term meaningless.

          If Jesus can still be omnipotent without being able to draw a square Euclidean triangle, then I can still be omnipotent without drawing a square Euclidean triangle. And if Jesus can be omnipotent without being able to time travel, then I can be omnipotent without being able to time travel. If Jesus can be omnipotent without violating the conservation of mass/energy, then so can I. And but what is my inability to leap tall buildings in a single bound but my failure to violate the conservation of mass/energy?

          Jesus can’t commit suicide and remain omnipotent. Either there’s an infinite variety of things he can’t do after his death that I can, thus demonstrating his impotence; or he is unable to do something I can (though certainly don’t want) to do.

          Omnipotence is as much a self-contained contradiction as “the set of all sets.”



          1. “Pure agency” or “pure omnipotence” (i.e., the power to do all things including that which is illogical) is certainly illogical. Thus, I would be in complete agreement with your sentiments on that. I am not sure I follow on the alternate definition(s) though.

            Before I continue, if I have overlooked a post which addresses the ideas below feel free to simply point me to that post. (I have no desire to waste your time.)

            If we follow in Aquinas’ footsteps, then we have a definition which states omnipotence is “the power to do all which is logically possible.” This does not seem to work well with the “meaningless” analogy. That is, I can imagine many things which are logically possible, but which I am incapable of doing. Thus, I would be unable to consider myself as omnipotent following this definition. What are your thoughts?

            The second definition holds that omnipotence means “maximal power.” That is, a being who is omnipotent is more powerful than all other beings. This would also seem to effectively circumvent the “meaningless” criticism. What do you think?

            I can certainly see your point about rendering the term meaningless. If an omnipotent being is indistinguishable from any other being, then it makes no sense to use the concept to begin with. Though, to my unsophisticated mind, it would seem the two definitions above work around this issue. Let me know your thoughts (or kindly point me to a post or even another website). Thanks.

          2. Neither re-workings of the definition hold, I’m afraid.

            Can Jesus commit suicide?

            If so, he’s powerless to do anything after the moment of his death, and obviously not even remotely omnipotent by any stretch of the term.

            If not, he’s unable to do something that all too many humans have tragically demonstrated we are capable of doing and, again, obviously not omnipotent.

            (I should note: the Gospels describe Jesus as having a long nap, not as actually dying. For the purposes of this exercise, death is eternal and irrevocable.)

            I’ve written lots more in recent threads here on WEIT. For example, is it physically impossible or logically impossible to draw, on a flat sheet of paper, a three-sided figure with angles that add to 360°? How does such a distinction even make sense?



          3. @Ben,

            From your perspective, would a deity exist within nature, beyond nature, both, or something altogether different? I ask because as I have attempted to unravel your thoughts it seems you view a deity as existing within nature and therefore necessarily bound by physical limitations. (This conclusion was drawn from the references to the “suicidal omnipotent god” [i.e., a god which can live and die] and the connections made between a deity and non-omnipotent humans.) This seems to be the point of disagreement between your conception of an omnipotent deity and stvs’ “all particle-controlling deity.”

            With regard to Jesus, the idea of Jesus as an omnipotent deity which is bound by physical limitations (i.e., can live and die) is inconsistent with Christian theology, specifically the idea of kenosis which states that Jesus “emptied” or “divested” himself of omnipotence, omniscience, and other god-attributes.

            Finally, your response does not seem to address the “maximal power” conception of omnipotence. Perhaps I am missing some connection?

      4. Ben, actually the sum of the angles of a triangle depend on the curvature parameter of space, and so that’s not necessarily the best example of an impossible task.

        However, as closely as we can tell from cosmological data, that curvature parameter does appear to be very close to zero (although it could have been otherwise).

        1. “sigh”

          You know, it’s getting tiresome.

          Drawing a square triangle in non-Euclidean space is no more noteworthy than drawing a square “triangle” with four sides, or with three curved “lines,” or with “degrees” that aren’t 1/360 of a circle, or any of the rest.

          Do I have to preface this example with a complete recap of Euclid’s definitions every time I use it?

          If I were to change my example to the fact that it’s impossible to draw parallel lines in a spherical geometry, I’m sure everybody would want to jump in with how easy it is to do in an Euclidean geometry, or how you could change the definition of a line, or how you can’t draw parallel lines but you can draw an infinite number of perpendicular lines, or….

          People, I get it. You’ve all passed high school geometry. Can you please assume that I have, too?



    2. I fail to see how the exercise is anything but a variation on a child’s “What if?”

      This is the entire point. What if there is a set of observations that would disprove the atheist’s position. What would they be? Unless those “what ifs” are precluded by logic, any hypothesis is valid against the position that we would a priori reject all evidence for a deity.

      And this is not child’s play, as there is a world of difference between “We almost certainly know that there is no god” and “We can reject a priori all evidence for god.”

      Without an axiomatic understanding of nature, attempting to take the strongest position is, right now, a bridge too far, and makes you vulnerable to many factual and rhetorical attacks.

      It is rather a stronger position to observe, based on current knowledge, that however unlikely it may be, we do not know with absolute certainty that Jesus couldn’t fulfill his promise of John 14:14, and ask why doesn’t He?

      1. Once more, into the breach.

        The whole point of gods is that they can do the impossible. What else is a miracle but a manifestation of the impossible?

        Except, of course, by doing the impossible, a god demonstrates that the phenomenon is, actually, possible — and, therefore, not miraculous after all. And, further, it should be obvious that the very premise of deities entails a necessary contradiction — exactly like the square (Euclidean) triangle.

        If all it takes to be a god is to do the improbable, then, by any rational definition, I am a god. Do you have any idea how improbable, how miraculous it is for me to type these words on this here keyboard in front of me, a machine completely disconnected from the screen in front of me, and have those words appear instantly on the screen? And for me to move a magical hockey puck, also unconnected to anything, wave it and press it in just the right way, and have these words nearly-instantly displayed on thousands of other computers all across the globe?

        If I wanted, I could put on my shoes, step out the door, and by this time tomorrow be on the other side of the planet. How is that not a miracle? I’ve seen pictures of my insides. Again, how is that not a miracle? If my ability to press a button and have pure ice come out of a magic box isn’t a a miracle, then what is?

        As to why Jesus doesn’t do diddly-squat…isn’t it obvious? Jesus doesn’t do diddly-squat for the exact same reason Santa Claus doesn’t deliver presents. They’re both fictional characters. Jesus doesn’t fulfill any promise he ever made because he never made them in the first place.



        1. Except, of course, by doing the impossible, a god demonstrates that the phenomenon is, actually, possible — and, therefore, not miraculous after all.

          Exactly. I think some are being mislead by the alleged infrequency of miracles, and confusing that with their ontological status. If we, for example, consider a pantheon (e.g., Greek or Roman religion, or ancient Egyptian, or Norse), there are huge numbers of entities that can, on a regular basis, perform what some would define as “miracles”. But doesn’t the very regularity and frequency undercut the whole notion of the miraculous? If all thunder and lightning really was caused by Thor, would such commonplace occurrences still count as “miracles”? If literally hundreds of beings can violate what we understand to be natural law, does that mean that they are violating natural law, or that our understanding of it is flawed. If ghosts existed by the millions, would we consider them a “supernatural” phenomenon, or merely some furniture of the universe that needed explanation?

          If it is really the case that miracles are simply the violation of natural laws, then it shouldn’t matter if they are as common as raindrops. But surely, if they were that common, we would see them as merely a currently unexplained part of the natural world.

  17. I find the Coyne-Myers debate illuminating on both sides and very valuable for that.

    I do accept that observations can raise my posterior probability for some kind of super-powerful-being explanation of some phenomena, but on the other hand it’s difficult to see where we would clearly be able to distinguish natural-but-very-powerful (*sufficiently advanced* aliens, say) from some undefined ‘other-natural’, and Occam would lead me to prefer the natural explanation unless I could identify a clear way to eliminate that explanation, or at least make it much less likely.

    I see invoking a “non-natural” explanation as the ultimate in ‘extraordinariness’ and so I need clearly extraordinary evidence to prefer it.

    Even the *existence* of a specific set of predictions that could rule out some unnatural explanations would help me to treat it as a serious alternative.

    I guess I’m inclined to try to test, whenever it’s possible, but as things stand I feel there are substantial barriers that must be overcome.

    Certainly an extraordinary event alone – or event several – would not lead me to rule out natural alternative explanations. I’d want something more – something that would really say “this cannot be natural, so we must invoke something outside of nature” — whatever that means.

    I see it as distinctly nontrivial, and so find myself agreeing and disagreeing with you both.

    1. “this cannot be natural, so we must invoke something outside of nature”

      To be able to say this with a clear conscience – we must be sure we understand nature perfectly. We don’t. Its possible we will, but I seriously suspect we will not. So until we got that understanding, the most probable explaination for any seemingly unnatural phenomenon is that it is our understanding that’s faulty.

      Either that, or the phenomenon is not what it seems (ie – somebody is playing silly buggers).

      1. “this cannot be natural, so we must invoke something outside of nature”

        To be able to say this with a clear conscience – we must be sure we understand nature perfectly.

        Exactly — the claim that something is “outside of nature” in principle is a claim that our scientific understanding is finished, that there is nothing more we will learn. Surely that is false.

Leave a Comment

Your email address will not be published. Required fields are marked *