Sean Carroll on the “supernatural”

At yesterday’s Cosmic Variance, physicist Sean Carroll entered the fray about whether science can find evidence for the supernatural (aka “God”).  Instead of fumbling about with the term “supernatural,” something I’ve preferred to avoid in favor of specifying characteristics of a god that one could look for, Carroll gets more specific than I by defining, helpfully, three types of phenomena.

  1. The silent: things that have absolutely no effect on anything that happens in the world.
  2. The hidden: things that affect the world only indirectly, without being immediately observable themselves.
  3. The lawless: things that affect the world in ways that are observable (directly or otherwise), but not subject to the regularities of natural law.

The first category is not, of course, subject to scientific investigation, and if that’s your kind of deistic or apophatic god, then of course science has nothing to say about it.

In the second category Carroll includes dark matter and quarks, which aren’t observable but have predictable consequences that can be tested scientifically.

Dark matter is unambiguously amenable to scientific investigation, and if some purportedly supernatural concept has similar implications for observations we do make, it would be subject to science just as well.

I’m a bit worried about the notion of a predictable and regular “supernatural” force that interacts with the world, because many of the commenters on this site and others will never accept regularity as evidence for anything we could conceive of as a god.  It could, for example, be space aliens or some previously unknown physical law about which we should suspend judgment, although of course there’s no reason why a god couldn’t act with some regularity.  It could, for example, answer the prayers of Jews but not Christians.  And if it rains far more frequently when Native Americans do rain dances than when they don’t, well, it’s just some natural phenomenon we don’t understand.

Number three, the “lawless,” take us into the real of the real popular idea of “supernatural”: miracles and the like.  I like the term “lawless” instead of “supernatural.” Carroll doesn’t see science as impotent before lawless phenomena, and I agree:

Let’s imagine that there really were some sort of miraculous component to existence, some influence that directly affected the world we observe without being subject to rigid laws of behavior. How would science deal with that?

The right way to answer this question is to ask how actual scientists would deal with that, rather than decide ahead of time what is and is not “science” and then apply this definition to some new phenomenon. If life on Earth included regular visits from angels, or miraculous cures as the result of prayer, scientists would certainly try to understand it using the best ideas they could come up with. To be sure, their initial ideas would involve perfectly “natural” explanations of the traditional scientific type. And if the examples of purported supernatural activity were sufficiently rare and poorly documented (as they are in the real world), the scientists would provisionally conclude that there was insufficient reason to abandon the laws of nature. What we think of as lawful, “natural” explanations are certainly simpler — they involve fewer metaphysical categories, and better-behaved ones at that — and correspondingly preferred, all things being equal, to supernatural ones.

But that doesn’t mean that the evidence could never, in principle, be sufficient to overcome this preference. Theory choice in science is typically a matter of competing comprehensive pictures, not dealing with phenomena on a case-by-case basis. There is a presumption in favor of simple explanation; but there is also a presumption in favor of fitting the data. In the real world, there is data favoring the claim that Jesus rose from the dead: it takes the form of the written descriptions in the New Testament. Most scientists judge that this data is simply unreliable or mistaken, because it’s easier to imagine that non-eyewitness-testimony in two-thousand-year-old documents is inaccurate that to imagine that there was a dramatic violation of the laws of physics and biology. But if this kind of thing happened all the time, the situation would be dramatically different; the burden on the “unreliable data” explanation would become harder and harder to bear, until the preference would be in favor of a theory where people really did rise from the dead.

There is a perfectly good question of whether science could ever conclude that the best explanation was one that involved fundamentally lawless behavior. The data in favor of such a conclusion would have to be extremely compelling, for the reasons previously stated, but I don’t see why it couldn’t happen.

. . . if the best explanation scientists could come up with for some set of observations necessarily involved a lawless supernatural component, that’s what they would do. There would inevitably be some latter-day curmudgeonly Einstein figure who refused to believe that God ignored the rules of his own game of dice, but the debate would hinge on what provided the best explanation, not a priori claims about what is and is not science.

This is where I agree with Sean, the philosopher Maarten Boudry, and, I think, Brother Blackford, and where we part company from P.Z Myers, The Great Decider, Eugenie Scott and the NCSE—and nearly everyone else.  At least I (and probably Sean) could envision theoretical cases where we’d see behavior as sporadic and lawless—and provisionally indicative of a god. Others would not.

And, lest we despair at all having settled into positions that are refractory to further change—which is probably the case—this debate has not been futile. It’s enabled us to clarify our own notion of what we mean by “supernatural,” what we mean by “the domain of science,” and what, if anything, could it mean to test for a god—or if even whether we find the idea of a god coherent.  I’ve much appreciated the back and forth we’ve all had, and am grateful to Sean for weighing in.


  1. Sigmund
    Posted November 2, 2010 at 9:35 am | Permalink

    I find it difficult to separate out such hypothetical evidence as being indicative of either a ‘God’ or an ‘alien so advanced it is Godlike to primitives like us’.
    That is not a problem to me. I’d be willing to accept evidence that left me with those two possibilities. It wouldn’t answer the overall question of how everything began – that is a different question that I think is getting unnecessarily confused with the ‘supernatural’ question. It would, however, answer the question of whether something with the ability to cause apparent ‘miracles’ existed. At the moment I see no evidence for such a scenario but I think its unwise to rule it out completely.

  2. Jimbo
    Posted November 2, 2010 at 9:35 am | Permalink

    I’m with Coyne and Carroll on this one. For a scientist to be a 6.99 on the Dawkins scale of atheism is not to admit agnosticism, rather it admits that natural laws could, IN PRINCIPAL, be violated which is de facto supernaturalism. As a scientist, I can easily imagine what would convince me of a supernatural deity. I am 100% sure that it won’t happen but not 100% sure that it couldn’t happen in principal. My atheism is strong but is not certitude beyond reproach.

    • Tyro
      Posted November 2, 2010 at 9:45 am | Permalink

      What evidence would demonstrate omnipotence rather than merely being merely very powerful? Omniscience rather than merely very smart?

      What do we do about all of out history of failed prayer, suffering and general absence of the benevolent, powerful intervention which gods are supposed to exhibit?

      • Posted November 2, 2010 at 10:07 am | Permalink

        As a science-minded person, one might provisionally accept the existence of some entity that seemed to be “supernatural” based on some overwhelming evidence of its existence. We might not be able to distinguish between something omipotent from something that was merely very powerful or capable of deceiving our senses and measurements, for example, but many would go for a parsimonious explanation at least tentatively.

        That’s why for the science guys, a belief would always be tentative — some new evidence might overturn what seemed to be a parsimonious explanation.

        So far, we don’t have any evidence of anything supernatural, and so it’s just a speculative question anyway. But I think if we argue that we reject theism (or other woo) because of a lack of evidence, then we have to be willing to say what sort of evidence we would find convincing.

        • Tyro
          Posted November 2, 2010 at 10:19 am | Permalink

          I don’t have any a priori reason to reject a god hypothesis and I could imagine worlds and universes which could support it. But every scientific discovery seems to attack rather than support this hypothesis.

          So given what we know of the universe and what we know of our past, I find it hard to imagine what could support the god hypothesis. We know any god has acted with indifference to us which rules out virtually all conceptions of “god” so I think we’re left with redefining “god” or, more honestly, saying that God has been falsified and new evidence won’t change things.

          • Posted November 2, 2010 at 10:31 am | Permalink

            Tulse: “God has been falsified and new evidence won’t change things”

            Yeah, I think that’s the reality.

            Victor Stenger’s God: The Failed Hypotheis constructed some hypotheses of what a creator god would have done in creating the universe and examined these against cosmological data — and found that such evidence, which in principle could have been there, was not.

            Hypothesis tested, and failed.

            Tests of the “prayer answering God” hypothesis, the “intervening God” hypothesis, the “all loving God” hypothesis, etc. have all failed, too.

            So, we pretty much have to test different gods than these to have any hope of success, but why bother since these failed gods are the ones that believers advocate?

          • Myron
            Posted November 2, 2010 at 11:57 pm | Permalink

            “I don’t have any a priori reason to reject a god hypothesis…”

            I do! The concept of an incorporeal spirit or an immaterial person is nonsensical.

        • Patrick
          Posted November 2, 2010 at 10:57 am | Permalink

          I agree with your conclusion in general, but I have two problems with the “what would it take to convince you” challenge.

          1. It usually smuggles in the idea that you are considering what would convince you of the supernatural as if you were a blank slate with no prior knowledge of the subject, and as if the supernatural claim in question were similarly without track record.

          2. This challenge almost always involves an elision between “the supernatural” and “the theistic beliefs of the person you’re talking to.”

          The reality is that convincing me of the truth of any given claim about the supernatural would require not only providing me with decent evidence for the claim (taking into account what we know about human psychology and our tendency to get these things horribly, embarrassingly wrong), but it would ALSO need to overcome all existing information I have that militates against that particular supernatural claim.

    • Tulse
      Posted November 2, 2010 at 9:58 am | Permalink

      it admits that natural laws could, IN PRINCIPAL, be violated

      But we can never know, IN PRINCIPLE, whether natural laws are being violated or whether instead we simply don’t fully understand natural laws. In other words, there is a fundamental epistemic problem here, one that isn’t answerable. We can never be certain that we have complete knowledge of the world, therefore we can never be certain that some phenomenon or event isn’t in fact a product of natural laws.

      Ordinary people today routine use technology that would have seemed restricted to witchcraft not more than 200 years ago. Clarke’s Law is completely true, any sufficiently advanced technology is indistinguishable from magic. We will never be in a position to state with certainty that something is magic and not technology.

      • Dan L.
        Posted November 2, 2010 at 11:02 am | Permalink

        People seem to use “natural law” as if it meant something very specific and obviously fundamental, but that’s not really how it’s been used historically. I was trying to get Pigliucci to answer that question a while ago — if you observe an apparent lapse of a law of nature, there are two possibilities: a) the laws of nature were broken or b) your understanding of the laws of nature are flawed. How can anyone ever conclude (a) is true? Isn’t there always the possibility of (b)?

        It’s not enough to say that the laws of nature could be violated in principle. How would such an event affect human beings? Unless we can recognize the correct laws of nature as being correct, we can’t possibly ever know that they’ve been broken.

        There’s a deeper problem, I think. Let’s suppose that there are fundamental laws of nature (whether or not we can recognize them as correct) and that they can be broken. Presumably since we’re talking about fundamental laws, we’re not talking about contingent, derived laws like Boyle’s law, which can be demonstrated to hold only in certain contexts. The fundamental laws, then, can’t break with respect to the context. If it did, we’d have a form of regularity in the validity of the law, and we would have to say that this new regularity we’ve discovered is the more fundamental law governing the behavior of the higher-level law that we had thought to be fundamental. In fact, if natural laws fail consistently in the presence of any particular experimental variable, then we’ve simply shown that our understanding of natural law was mistaken.

        To truly constitute a break of natural law, there must essentially be no reason for it — if there was a reason for it, that reason would be the law. Thus, the “lawless” category collapses neatly into the “hidden” category. The assumption of a “lawless” universe is nothing more or less than positing a natural order that we simply cannot (in principle) understand. Or rather, from a human being’s perspective, there is no way to tell “lawless”ness and “hidden”ness apart.

        • Dan L.
          Posted November 2, 2010 at 11:09 am | Permalink

          To qualify that a little, “lawless” means “hidden in principle.” We can study those events were natural laws fail, but we won’t be able to draw any conclusions about why they failed or under what circumstances they fail — again, by assumption.

          The upshot is that I still don’t see how “the supernatural” can mean anything other than “the unknowable” or “the unobservable” or “the incomprehensible” or some combination.

      • Tim Harris
        Posted November 2, 2010 at 4:41 pm | Permalink

        I think you will find that however supposedly ‘primitive’ people are, they are well able to distinguish between magic and technology, and do so, and, once introduced to it, would also rapidly incorporate advanced technology into the ‘technology slot’ in their brain. The days of Levy-Bruhl (whom I admire) are long over.

      • qbsmd
        Posted November 3, 2010 at 2:24 pm | Permalink

        I think a more useful distinction would be laws that act on the physical properties of particles: position, velocity, energy, charge, etc., vs. laws that operate on abstract concepts, such as karma differentiating good and bad, or results of prayers being different based on whether they’re Christian or Jewish. The laws that operate on abstract concepts require the existence of something intelligent enough to comprehend them, so the only epistemic problem is finding what\where that intelligence is.

    • Myron
      Posted November 2, 2010 at 10:44 pm | Permalink

      A strong atheist, i.e. an antitheist, is defined as someone who believes that there are no gods, not as someone who is certain, i.e. convinced beyond doubt, that there are no gods.

    • Posted November 6, 2010 at 7:39 am | Permalink

      And just how would a human be able to perceive behavior that was “sporadic and lawless”, when our very perceptual apparatus is formed by the natural laws? This is mere speculation and does ZERO to answer atheistic arguments.God still doesn’t exist, except in your mind, as a thought or more likely a wish.

  3. Tyro
    Posted November 2, 2010 at 9:42 am | Permalink

    I’m not convinced that Sean is actually agreeing with you here, Jerry. When he says:

    There would inevitably be some latter-day curmudgeonly Einstein figure who refused to believe that God ignored the rules of his own game of dice, but the debate would hinge on what provided the best explanation, not a priori claims about what is and is not science.

    I think he’s talking more about whether science can investigate questions about the “supernatural” or not. On that, I think that you, Sean and PZ all agree that science can and indeed is the only appropriate tool for investigation.

    He says that scientists would support “god” if it was the best explanation. Where you and PZ differ (and where I don’t see Sean taking a clear position) is what it would take for that to happen. You’ve given examples of how a being could show that it was very powerful but what does it take to demonstrate omnipotence? It may be intelligent, but what does it take to demonstrate omniscience? And what do we do about all of our past, failed attempts to investigate a god?

    I don’t think I’ve seen a good answer to these questions from you or your supporters and that’s why I’m going to side with PZ for the moment.

    But yes, if there was something which appeared to be supernatural, I’m entirely in agreement that it will be science (and scientists) which gives us answers.

    • whyevolutionistrue
      Posted November 2, 2010 at 9:48 am | Permalink

      Well, I’m backing off on testing God’s “omnipotence” and “omniscience”; the arguments have convinced me that those traits, at least, can’t be demonstrated scientifically. But there can be degrees of knowing and power that are beyond anything we know of or could conceive of. I’m still formulation my thoughts on this.

      • Tulse
        Posted November 2, 2010 at 10:01 am | Permalink

        there can be degrees of knowing and power that are beyond anything we know of or could conceive of

        And we have ways of knowing and power that are beyond anything spiders could conceive of — does that mean we’re spider gods? Or does that instead mean that spiders have certain fundamental limitations to their understanding of the completely natural world, and that we might have analogous limitations?

        • gillt
          Posted November 2, 2010 at 10:51 am | Permalink

          Tulse: “Or does that instead mean that spiders have certain fundamental limitations to their understanding of the completely natural world, and that we might have analogous limitations?”

          I don’t see a meaningful comparison between a spider’s lack of knowledge about the universe and our own, and it’s because, Peter Parker aside, spiders aren’t scientists.

          • Tulse
            Posted November 2, 2010 at 11:40 am | Permalink

            spiders aren’t scientists

            But that’s irrelevant to the question of cognitive limitations. It is possible in principle that there are things about the natural world that humans are simply not cognitively equipped to understand, and I find that possibility far more likely than the notion that the supernatural exists.

      • articulett
        Posted November 2, 2010 at 12:52 pm | Permalink

        If a being were omnipotent, then they could use their “all powerfulness” to demonstrate their omnipotence even if we don’t know how they’d do such a thing, right? They would also have the power to make any non-believer into a believer –even those who say nothing could convince them (otherwise they would not be omnipotent.)

        I’m not sure the atheist has to know what would change their mind about gods since the god most theists believe in WOULD know and has failed to provide that evidence.

        When it comes to any supernatural claim, the question seems to be, how did the believer come to “know” of such things when science cannot? Belief in the supernatural requires a belief in “other ways of knowing”. I think a believer would have to demonstrate that “other ways of knowing” are valid before a skeptic would conclude they know anything about the supernatural at all.

        We can, at least, devise experiments like: “if X is true we should expect to see y”– and start an investigation in that manner, but, as James Randi’s challenge shows, we never seem to get beyond the basics.

        (Thinking about it makes my head hurt.)

  4. Jack van Beverningk
    Posted November 2, 2010 at 9:51 am | Permalink

    “if the best explanation scientists could come up with for some set of observations necessarily involved a lawless supernatural component, that’s what they would do”

    I have a HUGE problem with that statement.

    It’s eerily close to “if we can’t find any other explanation then it’s okay to invoke a supernatural one”

    How can we even determine that an observation “necessarily involves a lawless supernatural component”?

    This all sounds to me as opening the door widely for the old ‘God of the gaps’: I can’t think of any natural explanation, so god-did-it.

    • whyevolutionistrue
      Posted November 2, 2010 at 9:56 am | Permalink

      What about if Jewish prayers are answered 10% of the time, or that healing invoking Jewish prayer sometimes works, but Christian prayers are never answered and invocations of Jesus never heal anyone? I think you’d be stretching it to think that, if these tests were done with sufficiently rigorous controls, there would be a “natural” explanation.

      • Sigmund
        Posted November 2, 2010 at 10:03 am | Permalink

        The ‘living in the matrix’ or ‘living in a computer simulation’ scenarios easily cover that problem. Alternatively it could be a powerful alien who has a sense of humor and is playing a joke on us! These are all hypothetical ‘natural’ solutions to the problem that would be pretty difficult to separate out from a verdict that the God of one particular religion was responsible.

      • Tulse
        Posted November 2, 2010 at 10:04 am | Permalink

        Perhaps the super-aliens are Jewish.

        Seriously, what is a more likely explanation: that some natural beings operating under physical laws we don’t understand cause these effects, or that the universe is so vastly different from our current understanding that it encompasses violations of natural law? It seems to me the latter requires a far more radical rejection of practically all of what we understand science to be.

      • Posted November 2, 2010 at 10:44 am | Permalink

        What about if Jewish prayers are answered 10% of the time, or that healing invoking Jewish prayer sometimes works, but Christian prayers are never answered and invocations of Jesus never heal anyone?

        But if it were this regular, it wouldn’t be “lawless”, right?

        • whyevolutionistrue
          Posted November 2, 2010 at 10:47 am | Permalink

          Yeah, and I say that in the main post. I don’t think “lawlessness” is required for a demonstration of the supernatural.

          • Tulse
            Posted November 2, 2010 at 11:42 am | Permalink

            But if it’s lawful but inexplicable, how is that different from, say, the statistical nature of quantum mechanics?

      • gillt
        Posted November 2, 2010 at 12:32 pm | Permalink

        Coyne: “I think you’d be stretching it to think that, if these tests were done with sufficiently rigorous controls, there would be a “natural” explanation.”

        What you’re describing sounds like an association study. How can you be so willing to rule out natural causes before you’ve done any functional investigation? When was healing triggered in relation to cerebral cortex activity? We’d want to observe the miracle at the cellular or molecular level before we could rule out natural mechanisms.

      • Thanny
        Posted November 2, 2010 at 12:47 pm | Permalink

        All that would mean is that there is apparently an agent capable of producing those effects who likes Jews and doesn’t like Christians.

        Such evidence would in no way justify leaping all the way to “Yahweh is real”.

        Likewise, if there ever were a body of evidence in support of lawless phenomena, that in no way supports supernatural agency. Assuming that lawlessness equals deity is the same broken logic that leads creationists to thinking that discrediting evolution leaves their view victorious be default.

      • AdamK
        Posted November 3, 2010 at 6:33 am | Permalink

        So NOW He starts with the miracles.

      • Jack van Beverningk
        Posted November 3, 2010 at 9:21 am | Permalink

        “What about if Jewish prayers are answered 10% of the time”

        Once we have scientifically established that, then:

        a. Get Jewish prayer covered by the health insurance industry.
        b. Keep working on scientifically figuring out WHY it works.
        c. Don’t give up on b.

    • Sigmund
      Posted November 2, 2010 at 9:57 am | Permalink

      That’s the reason I suspect we will be unable to determine whether such evidence indicates a ‘God’ or a ‘God-like alien’.

      • Posted November 2, 2010 at 1:22 pm | Permalink

        Well and what’s the difference anyway, and how do we know, and who determines? Will there be a panel of judges to vote “God” or “god-like alien”? If there is, what will the criteria be?

        • Dominic
          Posted November 2, 2010 at 3:05 pm | Permalink

          I vote you onto the panel!

    • Kevin
      Posted November 2, 2010 at 10:10 am | Permalink

      Seems to me that it’s not a matter of “we can’t think of an explanation, therefore supernatural.”

      Instead, it’s more along the lines of “we know that this phenomenon violates all of our well-understood laws of physics, chemistry, and biology; therefore supernatural cannot be ruled out.”

      It’s not lack of knowledge that’s the issue. It’s phenomena that run counter to our knowledge. And despite what someone is bound to say about the conditional nature of our understanding of the universe, I think we know plenty enough to know when the laws of physics, chemistry, and biology are being broken.

      • Posted November 2, 2010 at 10:34 am | Permalink

        And despite what someone is bound to say about the conditional nature of our understanding of the universe, I think we know plenty enough to know when the laws of physics, chemistry, and biology are being broken.

        But are nature’s laws broken, or is the laws in our latest model of nature that are broken? How would you know?

        • gillt
          Posted November 2, 2010 at 10:58 am | Permalink

          Sean actually answers this here and in a follow-up post:

          “The view of electrons and protons and neutrons interacting through the Standard Model and gravity will stay with us forever — added to and better understood, but never replaced or drastically modified.”

          • Posted November 2, 2010 at 12:24 pm | Permalink

            I’ve read that post before, but I fail to see how it answers my question. It just says that at this point, our models have proven to be so useful, that they are more likely to be extended, rather than discarded. If we find a phenomenon that doesn’t fit our current model, we will likely still be able to use it for the vast majority of conditions we use it for today – just like you can still use Newton’s laws in non-relativistic conditions.

            But this doesn’t tell you if this new phenomenon was merely a consequence of an unknown natural law, which was not accounted for in our current model, or whether it breaks the laws of nature itself.

            • gillt
              Posted November 2, 2010 at 1:10 pm | Permalink

              He does answer your query because you were responding to the miraculous, which by most any definition is something directly visible that requires a violation of the laws underlying everyday phenomenon.

              We would easily detect such a thing because our current version of general relatively and the Standard Model is complete in this context. That’s what I believe Carroll is saying.

            • Kevin
              Posted November 2, 2010 at 2:22 pm | Permalink

              Basically, you’re offering the argument that we’re no further along in our understanding of physics (primarily) than we were 100+ years ago.

              I’ll respectfully disagree.

              Any phenomenon that violates our known models (read that carefully, I said violates) would not at this point lead us down a “whole new physics” pathway.

              Any new physics will not violate the old physics.

            • Posted November 3, 2010 at 5:20 am | Permalink

              @gillt and Kevin: I don’t think I’m being clear enough. In that article, Carroll is arguing that for everyday phenomena, physics is pretty much complete. This I would agree with. But this is not relevant for this discussion, because a miracle is almost by definition not an everyday phenomenon.

              Furthermore, suppose we would have evidence that miracles happen. Suppose we had evidence for what we could tentatively call “divine intervention”. In that case, just like Newton’s laws are still valid in absense of high velocities etc, the Standard Model would still be valid in absense of “divine intervention” – which would be the vast majority of the time. So in that sense, I fully agree with Carroll.

              However,we can’t say a priori that we don’t need new physics to describe what happens during those rare cases of divine intervention.

              And again, what happens during “divine intervention” might be a rare, not-yet-understood phenomenon that follows currently unknown natural laws of its own, or it could somehow really break the laws of nature itself. How would you know which is the case? That is the question that needed answering.

              By the way, whether it’s even possible to break the laws of natura is another discussion, which hinges on your definition of “nature”. If the divine existed, and acted in our world, would it be part of nature?

            • Posted November 3, 2010 at 8:07 am | Permalink


              Any phenomenon that violates our known models (read that carefully, I said violates) would not at this point lead us down a “whole new physics” pathway.

              But that was entirely my point: it may violate our models of nature, but would it violate nature itself?

  5. Posted November 2, 2010 at 9:58 am | Permalink

    Ha! I was about to email you the link, but as usual you’re way ahead of me.

  6. Kevin
    Posted November 2, 2010 at 10:05 am | Permalink

    So, I think that the issue here is whether we can describe the supernatural versus whether we expect that a supernatural “anything” will ever be demonstrated.

    My personal opinion two those separate questions is yes and no.

    PZ seems to be on the side of no and no. He’s considered the evidence and come to the conclusion that supernatural = imaginary. And I agree with him on that point. But as a result, he’s taken the next logical step and refuses to engage in the thought experiment by which one might prove an imaginary creature actually existed. And I respect him for that.

    I just prefer to at least play that game, lest I be tainted with a “close-minded” label. I think that’s the risk inherent in PZ’s position. But I can’t criticize him for arriving at that conclusion.

    • Posted November 2, 2010 at 10:41 am | Permalink

      Indeed. Saying that there can’t be any evidence that will convince you of the supernatural may not be the best thing to do from a strategic standpoint. By the time you have explained the reasoning behind it (“supernatural” is badly defined, can’t tell whether something is supernatural or just unknown natural, etc), most people will already have their prejudices confirmed and won’t be listening anymore.

  7. MadScientist
    Posted November 2, 2010 at 10:16 am | Permalink

    I think people are being drawn in too much by the “sophisticated theology”. The most basic question is:

    Why do people claim they have the True God(s) when they have no credible evidence whatsoever for their claims.

    Once that question is answered, *then* it makes sense to proceed to questions of why it may not be possible for science to study a god. However, theologians immediately jump into discussions about things such as the color of the CeilingCat without first bothering to establish the existence of the CeilingCat. It is plain stupid at best.

  8. Posted November 2, 2010 at 10:32 am | Permalink

    This concept of “lawlessness” raises all sorts of questions: What is the explanatory power of a “lawless component”? What predictions follow from accepting a “lawless component”?

    • Gregory Kusnick
      Posted November 2, 2010 at 1:24 pm | Permalink

      Exactly. How would you recognize a lawless phenomenon when you saw it? By what criteria could you conclude that two apparently separate phenomena were in fact the effects of a common, lawless cause? Wouldn’t such criteria imply a regularity to the underlying cause, thereby promoting it to lawfulness? Given these difficulties, how could you ever know that the lawless category is non-empty?

  9. TheBear
    Posted November 2, 2010 at 10:34 am | Permalink

    Here we go again…

    A couple of objections:

    1: Carrols piece hasn’t got a workable, rigorous definition of “supernatural”

    2: There’s still no adress to the fact that claims of the supernatural has been falsified to the point that talking about evidence for them is akin to talking about evidence for a flat earth.

    Yes, of course there *could* in they very very theoretical principle be evidence that we’ve all been fooled by a complex and all-encompassing conspiracy to think that the earth is round, but the odds are so remote that treating the flat-earth society like they might have a point is simply lunacy.

    Playing “what-if”-games with the theists is simply giving them a level of respect and credibility their “theories” doesn’t warrant.

  10. Rob
    Posted November 2, 2010 at 10:36 am | Permalink

    I’m not so sure that I agree that the deist god is category one. Since it did, at one point, interact with the universe, that’s category 2.

  11. Tim Martin
    Posted November 2, 2010 at 10:49 am | Permalink

    There is a perfectly good question of whether science could ever conclude that the best explanation was one that involved fundamentally lawless behavior.

    This is exactly why I don’t understand how some scientists are willing to say that there are events at the quantum level that are undetermined. How could one ever conclude, logically, that there are effects without causes?

    • Dan L.
      Posted November 2, 2010 at 11:25 am | Permalink

      How could one ever conclude, logically, that there are effects without causes?

      How did you conclude that all effects should have causes? What reasoning led you to that conclusion? Finally, what reason is there to suspect that nature should behave the way you “logically conclude” that it should?

      I think there’s still some contention regarding the philosophical interpretation of QM which might be part of the confusion here, but I suggest you not lean too hard on traditional conceptions of causality in any event.

      Take radioactive decay. The time of decay of any particular nucleus is undetermined, but the half-life of the whole sample can be determined from the structure of the nucleus (with some difficulty). Sure that seems weird — that a low-level unpredictable behavior would become predictable at a high-level — but why shouldn’t things be weird?

      • Tim Martin
        Posted November 2, 2010 at 2:18 pm | Permalink

        Dan, I make no claims about what should or shouldn’t be. The universe can be however it wants. I’m merely asking how one can make a logical argument that there are things which do not conform to logic.

        It’s similar to the old conundrum about using science to prove that our sensory perceptions are entirely inaccurate. Such a conclusion would invalidate itself, since the science we used to make the conclusion is something that we built and confirmed using those inaccurate sensory perceptions. Therefore the science is inaccurate, therefore the conclusion is invalid and we’re back to the drawing board.

        Similarly, all of human reasoning is based on certain logical principles (principles which cannot be proved because… what would we use to prove them? Logic?) Our system of logic does not allow effects without causes. Consider the following argument:
        1. A
        2. If A, then B
        3. C

        Well you certainly could have used this argument to derive B, but C? Where did C come from? It is not an acceptable answer to say that it came from nowhere whatsoever, an effect without a cause. The premises must logically entail the conclusion, else we have no reason to believe it is true.

        And yet this is what quantum indeterminacy is apparently about. The answer to “why did this particle move in this direction” is… “for no reason whatsoever.” I personally cannot conceive of their not being some reason. Maybe that’s just a limitation of the human mind. I don’t know, and I’m not saying the universe has to be limited just because I am.

        But I am asking, since we are humans using a human system of logic, how can one ever make an argument that ends, “therefore, this particle moved this way for no reason whatsoever”? What premises could logically entail a conclusion that, essentially, destroys logical entailment?

        • Paul W.
          Posted November 2, 2010 at 2:34 pm | Permalink

          Our system of logic does not allow effects without causes.

          That’s not true.

          It’s not logic itself that doesn’t allow that, it’s certain ingrained assumptions about the nature of physical reality.

          Some of our natural intuitions about how causation works just turn out to be wrong.

          • Tim Martin
            Posted November 2, 2010 at 2:49 pm | Permalink

            Why make such an assertion without giving any citation or even argument to back it up? Your position implies that arguments such as these should be valid:
            1. A
            2. If A, then B
            3. C

            They are not. And even if they were sometimes valid, how would we differentiate between the valid ones and the invalid? If I’m completing my math homework and I derive a contradiction, am I wrong, or have I just shown another assumption of logic to be wrong?

            Your statements bring no understanding to the issue, Paul.

            • Paul W.
              Posted November 2, 2010 at 4:55 pm | Permalink

              If your 2nd premise is supposed to resemble nondeterministic quantum effects, I think it should be more like

              if A then maybe B

              where “maybe” simplistically stands for a modal logic operator (sort of like the diamond “possibly” operator in normal modal logic) with an associated probability distribution.

              There’s no problem representing such things logically. You just have to not get it wrong.

              So, for example, with the “paradox” of wave/particle duality, there’s no real paradox. There’s just something wrong with the particular overly-concrete ideas of waves and particles that people started out with—e.g., the waves are not waves of motion in a medium, but waves of probability of spontanteous particle interactions, and particular particles don’t exist until wavefunction collapse or decoherence, or or something like that.

              (And really, the right deep representation would depend on what the correct interpretation of quantum mechanics actually is.)

            • Tim Martin
              Posted November 3, 2010 at 8:51 am | Permalink

              Actually, it was the conclusion, C, that was meant to resemble nondeterministic action. If logic is a way of taking what you know is true and deriving further truths from that, then a truly random effect would not be derivable from any premises, and would thus “appear” on your piece of paper as a fact from nowhere. That is what C represents.

    • Tyro
      Posted November 2, 2010 at 11:41 am | Permalink

      I’m not a physicist so this is based purely on reading pop-level physics books and while they do tackle these questions, I’m sure I’m outclassed here. That said…

      The question of causality was a sticking point for many scientists including Einstein. This lead to many hearty debates which lasted for decades which have resulted in tests to see whether the components of the system could, in theory, possess the necessary values to cause the observed effects. It looks like no, they cannot.

      Dan mentioned atomic decay but there are many other quantum events which are strictly probabilistic. One of the most famous experiments which could show whether the system results were determined by some properties and thus was “caused” by the initial values was created by John Bell, called “Bell’s Inequality”. The tests have since grown better and the data more conclusive.

      There are some good books which cover this (“The Elegant Universe” by Brian Greene is a good all-around popular science book and should be accessible to anyone with high-school math and physics).

      • Tim Martin
        Posted November 2, 2010 at 2:25 pm | Permalink

        See, this is what I don’t understand. With any other subject, if you come up with a result that you can’t explain, or if you come up with a contradiction, you conclude that you did something wrong or there’s something about the system you have yet to understand. This is a preferable alternative to concluding that logic is broken and effects don’t need causes.

        However, in quantum mechanics, you get a contradictory result and conclude that logic is broken, rather than your theory?

        Pardon me if I have this completely wrong, as I don’t understand the experiments involved very well at all. But when people try to explain this to me, the gist of what they’re saying always seems to be the description I just gave above.

        If Greene covers this well in his book, I’ll be happy to read it!

        • Posted November 2, 2010 at 4:00 pm | Permalink

          Check out Jim Al-Khalili’s Quantum: A Guide for the Perplexed.

          Short version: the waveform equations work perfectly well in describing and predicting quantum behaviour. Describing quantum behaviour accurately in other ways — e.g. by analogy — is the hard part.

        • Rob
          Posted November 2, 2010 at 4:50 pm | Permalink

          Other than observation, where do you get effects have causes?

        • Tyro
          Posted November 2, 2010 at 6:14 pm | Permalink

          See, this is what I don’t understand. With any other subject, if you come up with a result that you can’t explain, or if you come up with a contradiction, you conclude that you did something wrong or there’s something about the system you have yet to understand. This is a preferable alternative to concluding that logic is broken and effects don’t need causes.

          What you’re saying is almost exactly what Einstein thought so you’re hardly on the lunatic fringe here. In many ways I’m sure I’d be tempted to agree with you as would most 20th century physicists. Indeed many physicists did think that the theories must be wrong and fought hard to prove it. But in the end, evidence matters more than intuition, no matter how deeply ingrained.

          The classical view didn’t go down without a fight but as the data came in, it kept siding with quantum mechanics and not with “logic”. Indeed, it turns out that we aren’t really talking about logic but our intuitive, classical understanding of the world, something that is so deeply ingrained that contradictions are almost inconceivable.

          So here’s the thing. QM isn’t just some wild idea scrawled on a napkin but is the best tested, most accurate theory (or set of theories) that we’ve ever come up with. Nothing, not even gravity, comes close to the precision, repeatability, and predictive value of these theories. It’s not that they don’t require a cause but they actually preclude a cause. In order to settle these questions, Bell (and others after him) designed experiments which could empirically determine whether quantum events were determined by some hidden variables which we didn’t know about or whether they were independent of any external agent and were truly “random”, uncaused. His tests have been performed and others have refined them to be even more vivid and each time the evidence continues to support a lack of any causal agents. Being good scientists, physicists accepted the evidence and agree that Einstein was wrong in this case.

          It’s strange, startling, totally counter-intuitive and violates everything we understand in our day-to-day world. I think the positive sign here is that it’s good reason to think that if genuine “supernatural” events were discovered (whatever that means), scientists would follow the evidence even if it means discarding beliefs they thought were fundamental and unshakeable.

          (And to throw in an extra confusion, subatomic physicists rarely talk about terms like “cause”, even back in Einstein’s day. The notion of causation in our macro world is almost recognizable in the quantum world so even these fights were about the last vestiges and remnants of a badly shredded concept, reduced to “hidden variables” and “locality”. Long before this, events like a high-energy electron emitting a photon and falling to a lower orbit were known and accepted to happen without any trigger or agent, without any “cause”. Same with atomic decay and other events.)

          • Tim Martin
            Posted November 3, 2010 at 8:53 am | Permalink

            Woot, at least I’m in good company!

            Apparently I’m going to have to do some reading on this. Thanks, everyone, for your explanations and book recommendations.

        • Posted November 3, 2010 at 5:43 am | Permalink

          However, in quantum mechanics, you get a contradictory result and conclude that logic is broken, rather than your theory?

          Logic isn’t broken, it works just fine on our macro scale. On our scale, the indeterminism of quantum mechanics are usually nicely averaged out to conform to our sense of cause and effect. Our common sense just turns out to not be appliccable at the quantum scale.

  12. Dominic
    Posted November 2, 2010 at 11:09 am | Permalink

    It is the whole idea of the soul that really gets me. I am 100% convinced that it does not exist, whereas I admit to the possibility of the possibility of god/s. But frankly, the more I consider these things the closer I come to considering the possibility of god/s absurd.

  13. john
    Posted November 2, 2010 at 11:37 am | Permalink

    All religions claim to offer some form of immortality to their adherents. Most religions claim their god is immortal. Buddhism might be an exception. The claim of “immortality”, as apposed to the ambiguous “supernatural”, is directly opposed to natural law. And much easier to highlight the differences between Science and religion. It’s too easy to get sidetracked by defining “god”.

    • Tulse
      Posted November 2, 2010 at 11:48 am | Permalink

      The claim of “immortality”, as apposed to the ambiguous “supernatural”, is directly opposed to natural law

      There are organisms that are effectively immortal (HeLa cells, for example, or indeed most single-celled creatures that reproduce via mitosis). And certainly one can imagine even humans attaining effective immortality through a number of means (biological, cybernetic, even digital). Immortality is not a sufficient criterion, unless one is going to define “immortality” in such as way that it requires supernatural abilities, in which case we’re back to the drawing board.

      • Posted November 2, 2010 at 12:34 pm | Permalink

        I do believe John is right: “The claim of “immortality”, as apposed to the ambiguous “supernatural”, is directly opposed to natural law. [based on our current models]”

        Negligible senescence is not true immortality.
        A. as time progresses the chance of dying through external causes approaches 100% for any given non-aging organism (theoretically, this can be prevented w/o violating our current models; however, considering that some cosmic events can annihilate planets or solar systems quite quickly and without much warning this may be rather difficult)

        B. to me as a layperson it seems almost every prediction of the universe’ ultimate fate involves dramatic events that would destroy any life forms (or even any information we may try to preserve). Heat death, neutron decay, big crunch, etc.


        sorry if that sounds pedantic; but I do agree with the epistemological objections you raised. If the universe does not end, our models may have been simply wrong – there is probably no way to tell.

        • Tulse
          Posted November 2, 2010 at 12:43 pm | Permalink

          theoretically, this can be prevented w/o violating our current models; however, considering that some cosmic events can annihilate planets or solar systems quite quickly and without much warning this may be rather difficult

          But then the key feature is not “lives forever”, but “has the power to prevent the annihilation of planets and solar systems”, which just gets us back to “supernaturally powerful”.

    • Paul W.
      Posted November 2, 2010 at 1:29 pm | Permalink

      Not all religions promise immortality, or even think that the gods are immortal. That’s just not a constitutive principle of the supernatural or of religion. It’s a common one that admits of a degrees, not necessary or absolute. (E.g., some religions posit several generations of gods, with the later gods having replaced the earlier ones, e.g., after killing them.)

      As I understand it, the ancient Hebrews mostly didn’t believe in an afterlife. (They believed in a breath of life that animated bodies, but that breath of life was not itself life—once it’s gone and your body falls apart, you’re dead.) The afterlife only became a popular Jewish belief a few hundred years BCE.

      Even today, lots of Jews don’t particularly believe in an afterlife, and promises of immortality or otherworldly reward are not a big part of the religion.

      Religions mostly aren’t really about immortality; they’re about this-worldly concerns. Once you have dualism, though, it’s pretty easy to promise people pie in the sky when they die.

      As for gods, they don’t really need to be immortal. It’s just really handy for storytelling if the current ones aren’t going to die in the foreseeable future, so that you can have a stable belief system about them.

    • john
      Posted November 2, 2010 at 2:02 pm | Permalink

      On second reading, I think I made a jump from the attempt to define supernatural to assuming god is the cause of all supernatural occurrences.

    • Tim Harris
      Posted November 2, 2010 at 5:03 pm | Permalink

      Paul W (below, or will he be above after I post this – this getting rather theological already)- has rightly said that all religions do not offer immortality. What anthropogists-cum-cognitive scientists like Boyer and Pyysaiainen think defines religion more or less is belief in the existence of agents who act or exist in counter-intuitive ways (in ways that are beyond what human beings are capable of: they don’t need to eat, they don’t age; in Pyysiainen’s words, they violate ‘panhuman intuitive expectations’).

  14. stvs
    Posted November 2, 2010 at 12:00 pm | Permalink

    Carroll: What if there truly are lawless supernatural actions in the world, but they appear only very rarely? …  How can we guard against that error? We can’t, with complete confidence.

    I’m disappointed that Sean doesn’t address Einstein’s question on this point, “What really interests me is whether God had any choice in the creation of the world.” It may be that there is only ONE logically consistent set of laws that make the universe possible. If we know these laws and can prove uniqueness in this sense, then that kills God forever. This is not idle barroom speculation from Einstein—he asked this question because this is the conclusion to which all science points. IF such a Theory of Everything is established, then “lawlessness” becomes an impossibility, not merely an improbability.

  15. Anonym
    Posted November 2, 2010 at 1:00 pm | Permalink

    Were there a ‘god’, human beings and all other living things would need be nothing more than magically animated specifically formed clumps of dirt (read ‘golem’) — there would be no need for bones, muscles, veins, blood, … or any other such ‘nasty bits’ suggesting biological relatedness; … and evolution would be patently absurd.

  16. Posted November 2, 2010 at 1:02 pm | Permalink

    I have posted a response to Jerry on my blog. As a quick summary, I disagree with Jerry. I think that examples of the third category will really be seen as in the second category, and I suggest this is an illustration of the Quine-Duhem thesis.

  17. Don
    Posted November 2, 2010 at 1:04 pm | Permalink

    This is an interesting conversation, but I will just never understand anybody of a scientific bent saying there will never be enough evidence to prove X. Only if X is logically inconsistent can this be true. I know a number of posters argue that god is an inconsistent idea. I tend to agree actually, but I have never seen a proof that any and all possible gods are logically inconsistent. Instead I see that the god concept is most likely inconsistent, that it is most likely that all or any evidence for god will always turn out false, and most likely that no god or gods exists. Still, I, and I believe science, is obliged to examine the evidence when it appears at least superficially credible and follow where that evidence leads.

    • Posted November 2, 2010 at 1:29 pm | Permalink

      But if it’s a matter of any and all possible gods, why use the word “gods” at all? Why not use a more neutral, technical, matter of fact word? Using “gods” just seems to confuse the issue.

      If we drop the word “gods” the question just becomes “could there be a force or a conscious agent in the universe that we don’t know about?” – and then the question seems much less loaded.

      • Don
        Posted November 2, 2010 at 2:20 pm | Permalink

        I have no problem with dropping the word gods. In fact I think the whole definition question is a non-problem. If there is evidence to investigate then it should be investigated. The appropriate definitions for the concepts will evolve and become more rigorous with the findings of the investigation. They could conceivably match at some level our current fuzzy definition of god at which time it would be appropriate to call it that. I find it highly unlikely, but I don’t dismiss the possibility.

        • Tim Harris
          Posted November 2, 2010 at 5:07 pm | Permalink

          Try the anthropologistIlkka Pyysiainen’s ‘counter-intuitive agents’ – as he suggests, that gets us away from the Judaeo-Christian bias of the word ‘god’.

          • Posted November 2, 2010 at 6:23 pm | Permalink

            Or just agent – that by itself would help. It’s still surprising etc – but without the associations of divine, worthy of worship, etc etc.

            I have zero problem accepting that there could be unknown agents in the universe. I have much more problem thinking they know anything about us, or have any effect on our planet and us.

            • Tim Harris
              Posted November 2, 2010 at 6:58 pm | Permalink

              No, not just agents: they are a specific kind of agent. But these are agents that, as I think Boyer and others have shown pretty conclusively (and it will get more conclusive as more evidence comes in and more serious thinking is done) are basically functions of how our brains work. You could say, I suppose, that the question of the existence of counter-intuitive agents is bracketed, in some Husserlian way, by these scientists, and Tremlin becomes a little too generous, I feel, in the final parts of his book, but I don’t really think that anyone can come away from a reading and a genuine understanding of the work of these thinkers with their religious faith intact.

            • Dominic
              Posted November 3, 2010 at 4:50 am | Permalink

              We should call them ‘secret agents’ then! 😉

            • Paul W.
              Posted November 3, 2010 at 11:54 am | Permalink

              I agree with Tim. (Above, I think.)

              What people are actually interested in when discussing “naturalism” and “supernaturalism” isn’t just alien agents. It’s supernatural agents.

              (A god isn’t just a powerful alien.)

              We need to address that squarely, by clarifying what “supernatural” means, and why people tend to find irreducibly spooky things more worship-worthy.

              Pyysiainen’s term “counter-intuitive agents” doesn’t ring true for me, though.

              The distinctive thing about supernatural agents is that they are quite intuitively understandable, at the usual level we normally think about things. It’s only if you know more and think hard that they don’t make sense.

              (E.g., the idea of a disembodied mind is a simple comprehensible idea, and intuitively credible, until you learn that real minds are doing a whole of lot information processing, requiring something like a computer made out of something very like matter.)

              The problem with supernatural concepts isn’t that they’re counterintuitive; it’s that they’re all too intuitive,but wrong—they’re what you know that ain’t so.

            • Posted November 3, 2010 at 1:47 pm | Permalink

              But why isn’t a god just a powerful alien? Besides the fact that fans of the idea just don’t want it to be that.

            • Paul W.
              Posted November 3, 2010 at 2:30 pm | Permalink

              But why isn’t a god just a powerful alien? Besides the fact that fans of the idea just don’t want it to be that.

              Because there’s a particular psychology and particular basic categories involved.

              If the spookiness of supernatural beings is reducible, it stops being so spooky.

              (As opposed to, say, creepy or amazing or just unknown and frightening, like a powerful alien would be.)

              For example, suppose you had a powerful alien who could make you fall in love by rewiring your brain at the synaptic level, to have the appropriate propositional and affective attitudes toward somebody. (Either using advanced technology or maybe a freakishly but naturalistically evolved gift for mind control.)

              If you understood what they were doing, it’d mainly be amazing and scary. It’s not that different from controlling somebody’s mind with date rape drugs and hypnosis.

              That’s just not how anybody conceives of a Love God or Goddess. (If they use such terms in earnest.)

              They conceive of those things using naive, essentialist categories.

              Love is an essence that doesn’t reduce to anything like information processing in a computer. It’s just love, which is a kind of substance or invisible aura or something.

              A Love Goddess like Aphrodite is a special kind of being with a special kind of essence related to the essence of Love—she can create and control Love fairly directly because that’s the kind of being she is. Her essence and the essence of love are profoundly related, if not identical, so she can envelop people in the right kind of metaphysical glow or something, and that just is being in love.

              She’s not somebody who uses technology that’s indistinguishable from magic, in practice. She’s different, in principle—she’s magical in a way that doesn’t reduce to nonmagic, or it just doesn’t count.

              I think Boyer and Carrier are right that there’s a similar thing going on with essentially all supernatural concepts, including anything that anybody calls God in earnest—e.g., anything that they might actually worship and have specifically religious attitudes toward.

              The common New Age impression of quantum mechanics is like that—people think that quantum physics is irreducibly bound up with consciousness, so that you get mind over matter.

              New Agers who believe that see quantum mechanics as magical and even godlike, because it’s omnipresent and magical and directly related to minds. (Cue the wooooooo music…) Reality is just a dream in the Universal Consciousness, or is the Universe Creating Itself with mind over matter, and we are the Universe evolving teleological so that it Know Itself, or some irreducibly spooky crap like that.

              Physicists these days generally can’t go there. Quantum mechanics is interesting and bizarre and important and pervasive, but it’s just not anything remotely like God, because it’s dumb and meaningless. It’s like electricity, which turns out to be boring in the sense religious folks are interested in. It has no mind, and no directly interesting mind-related or teleological properties. It doesn’t understand or care about anything at all. It’s still just stuff that’s interestingly bizarre in a geeky way, but not a spiritually meaningful way.

        • Tim Harris
          Posted November 3, 2010 at 7:20 pm | Permalink

          I don’t know if anybody’s still reading (I live in Japan so there’s a large time-difference), but I just want to pick up on what Pyysiainen means by ‘counter-intuitive’ agents’; Boyer also, by the way, speaks of agents with ‘counter-intuitive’ properties, and in a series of ‘brilliantly designed experiments’ (I quote Harvey Whitehouse)he and his collaborators have shown that what he calls ‘minimally counterintuitive (MCI) representations’ are ‘easier to recall than intuitive, strange, or profoundly counterintuitive concepts.’ That is to say, agents who violate ordinary expectations in a moderate way are especially easy to notice, recall and transmit. It is such agents that are ‘gods’ – including Jehovah, concerning whom no believer believes in a deeply emotional way that he is omnipotent, omniscient, omnipresent etc (except when they are trotting out the theological niceties they have been told are ‘correct’); they in fact believe (and this has been demonstrated in experiments)that Jehovah acts much like a human agent and can’t do more than one thing at once, so that after unleashing a tsunami in Indonesia and Thailand or floods in Northern England to deal with all those gays and show his disapproval of gay marriage, he will come to Arizona to save a little girl from being mowed down by a car, and then will offer Ray Comfort, if our little ray of sunshine is feeling gloomy, some comfort… ‘Counter-intuitive’ is not meant by Boyer or Pyysiainen or Whitehouse to mean ‘unnatural’ or ‘hard to envisage’ but is rather a technical term to describe the properties of the kind of agent human beings in fact regard as gods. (Whitehouse goes out of his way to entitle a section of his ‘Modes of Religiosity’ ‘The Naturalness of Gods’.) Which is why it seems to me that conducting the debate on a high theological, or a-theological, level is, though not entirely pointless, not very helpful, since it is not on that level that people actually entertain the concept of ‘counter-intuitive agents’. And it leads to the sort of constant exasperation with which some atheists regard the believer, as though the latter were merely being perverse in not accepting the atheists’ arguments. Anyway, enough of that: the famous Calvinist preacher in Scotland had the right idea about ‘God’ to my mind. In a Sunday sermon, he told the story of Jock who drank on Sundays. After his death, Jock found himself lying on a lake of fire burning from head to toe. He looked up and there was God glowering down at him from Heaven. ‘Lord, Lord,’ wailed Jocky, ‘I didna ken, I didna ken…’ ‘Well, ye ken noo!’ said God.
          Now that’s a proper god.

    • Posted November 3, 2010 at 6:10 am | Permalink

      I tend to agree actually, but I have never seen a proof that any and all possible gods are logically inconsistent.

      Then again, a single counterexample could show that consistent definitions can exist. All I can say is that so far, I have yet to see a definition of “God” or “supernatural” that is coherent enough to allow evidence for it to exist, while not reducing it to a natural phenomenon at the same time.

  18. Paul W.
    Posted November 2, 2010 at 1:05 pm | Permalink

    I haven’t read Carroll’s original argument, but this discussion here seems to be mostly missing the point of the supernatural.

    The supernatural isn’t lawless. That’s just not what the supernatural is about, never has been, and never will be.

    Supernatural entities obey a slightly different kind of law than natural ones. If that weren’t true, people could not understand stories about supernatural entities and repeat them. There’d be no way to schematize them, and no interesting plots. They’d be so weird and confusing that the stories would die.

    For example, ghosts act just like people in most basic ways—people who happen not to have normal material bodies, and who can walk through walls. (And who don’t die, because they’re already dead. An afterlife isn’t lawless—it’s like life only different in certain ways.)

    Supernatural entities obey laws that are basically laws of intuitive psychology. (Or teleology, which is similar.) That is how people understand them—they’re quite comprehensible and regular, but they’re regular in ways that are like minds—they have goals like humans, or make distinctions relevant to human goals.

    For example, superstitious Luck is either for you if you have good luck, or against you if you have bad luck. It’s quite lawful and understandable in its own terms, which is why you can make sense of stories about people who are “lucky” or “unlucky.” Behaviorally, Luck is like having a friend or enemy who sees all and knows all, and can manipulate things in subtle ways so that they work out well or badly for you.

    Too often, we get sidetracked talking about apologetic theology, and fail to talk about the supernatural and religion as people actually practice it.

    Religious people don’t really care whether God is literally omnipotent and omniscient. What matters is that he’s a whole lot more powerful than you and can catch you doing stuff you’d otherwise get away with. “Almighty” doesn’t mean omnipotent—it means he “has all the power” in any relationship. That’s what makes him God, not just a god.

    Likewise, religious people don’t really think God is utterly mysterious. They think he’s a person with beliefs and desires that they can make deals with.

    That’s almost the furthest thing from lawless.

    There’s an element of unpredictability there, because you can’t always predict what somebody who’s smarter, more knowledgeable, and more powerful than you will do.

    That’s true of people in general, though, even if they’re not smarter than you; they’re complicated and you have to make guesses based on general tendencies.

    What makes supernatural concepts work is that supernatural entities are lawful and predictable in general, high-level ways, like people.

    Belief in the supernatural generally involves applying to intuitive (“folk”) psychology inappropriately. We see agency where we shouldn’t, and we psychologize about the non-psychological.

    • Posted November 2, 2010 at 1:30 pm | Permalink

      H/t Pascal Boyer. :- )

      • Paul W.
        Posted November 2, 2010 at 1:35 pm | Permalink

        Absolutely. As I’ve said before, everybody interested in discussions of the “supernatural” should read Boyer’s Religion Explained.

        (Richard Carrier’s good, too, but Boyer does some fairly broad and deep cross-cultural cognitive anthropology of religion.)

        • Tim Harris
          Posted November 2, 2010 at 5:14 pm | Permalink

          YES: Boyer, Boyer & Boyer; also Harvey Whitehouse, Ilkka Pyysaiainen, Todd Tremlin, Robert McCauley, E.T. Lawson, David Lewis-Williams, James Laidlaw, Stuart Guthrie and many other anthropologists and cognitive scientists. There is a huge and fascinating scientific debate about the nature of religion going on, and yet the quarrels between atheists and theists seem to revolve around parochially Judaeo-Christian conceptions and to be couched in terms that haven’t changed much in 200 years or more.

          • Tim Harris
            Posted November 2, 2010 at 7:04 pm | Permalink

            A little Milton to cheer us all up:
            ‘In discourse more sweet/ (For eloquence the soul, song charms the sense,)/ Others apart sat on a hill retired,/ In thoughts more elevate, and reasoned high/ Of providence, foreknowledge, will and fate,/ Fixed fate, free will, foreknowledge absolute,/ And found no end, in wandering mazes lost./ Of good and evil much they argued then,/ Of happiness and final misery,/ Passion and apathy, and glory and shame, Vain wisdom all, and false philosophy…’

  19. Kirth Gersen
    Posted November 2, 2010 at 1:36 pm | Permalink

    I have to admit that I’m at a loss to come up with a mechanical way in which the “supernatural” can ever be distinguished from “the natural.” If ghosts are “souls” made of “ectoplasm” that can “manifest” on this plane from another one, then aren’t we merely saying that nature includes such things as souls, ectoplasm, and coterminous phases of existence?

    Even a seemingly omnipotent God who violates what we see as “natural laws” at whim would have some means of doing so — a relativity effect, as it were.

    So, to my mind, the strictest definition of “supernatural” is as follows:

    Supernatural (Adj.) a word used colloquially to denote topics into which the speaker, because of his pre-existing bias or world-view, does not wish for any rigorous investigation to be conducted.

    • Paul W.
      Posted November 2, 2010 at 2:00 pm | Permalink

      If ghosts are “souls” made of “ectoplasm” that can “manifest” on this plane from another one, then aren’t we merely saying that nature includes such things as souls, ectoplasm, and coterminous phases of existence?


      The “supernatural” is just NOT the complement of the “natural” in the sense that science looks for natural laws to explain whatever is observable.

      The word “natural” has several related senses, and these are very different senses.

      Defining the supernatural to be not natural isn’t a good definition—it’s a bad pun.

      Think about the distinction between “natural” and “artificial”. Artificial things are natural in the broader sense relevant to whether science can study them. There’s no rule of “naturalism” that says we can’t study artifacts.

      So too with the supernatural.

      The conflatiion of different senses of “natural” is the problem here. There’s nothing particularly lawless about the supernatural, and no good reason that not to regard it as part of “nature,” when determining whether it’s within the scope of science.

      Like “artificial,” “supernatural” is a cluster category with several related senses, and there’s a particular pattern to those senses.

      What we call “natural” is context dependent—it’s what we’re thinking of as the normal operation of the system in question, according to certain principles, without interference from something we think of as above or “beyond” that.

      So, for example, we might study predator/prey population cycles, with booms and crashes that occur “naturally,” just given the cycles of availble food. (E.g., food for the prey is plentiful in certain seasons, predator population booms lag prey population booms, etc.)

      We would say that regular cycles in the population of predators are a “natural consequence” of those underlying facts, and what we mean by that is that we don’t need other entities principles to explain such observations. (E.g., superpredators from another region wandering into our system periodically and eating our predators, or humans coming in and shooting predators.)

    • Posted November 2, 2010 at 6:26 pm | Permalink

      Yes. That’s much what I was saying a week or so ago about nature being just a word for whatever there is. It’s not a special walled-off category different from the supernatural, it’s just what there is. If it is, it’s part of nature.

    • Myron
      Posted November 3, 2010 at 12:30 am | Permalink

      “If ghosts are ‘souls’ made of ‘ectoplasm’…”

      What is “ectoplasm” supposed to be? Some exotic kind of matter or some immaterial stuff? If the former, ectoplasmic souls are material, and if the latter, ectoplasmic souls are nothing, because the concept of immaterial stuff is self-contradictory.

      • Kirth Gersen
        Posted November 3, 2010 at 12:52 pm | Permalink

        “the concept of immaterial stuff is self-contradictory.”

        Does light have mass?

        • Gregory Kusnick
          Posted November 3, 2010 at 1:44 pm | Permalink

          Yes, and momentum. “Matter” includes energy in all its forms, because (as Einstein showed) energy is mass.

        • Myron
          Posted November 3, 2010 at 2:41 pm | Permalink

          Light is (visible) electromagnetic energy travelling through space; and energy is not a kind of stuff but a quantitative property of physical things. Thus, energy isn’t a kind of immaterial stuff either.

          • Gregory Kusnick
            Posted November 3, 2010 at 3:16 pm | Permalink

            Energy certainly is a kind of stuff in the sense that you can weigh it and measure its impact when it hits things, just like “matter” particles.

        • Tyro
          Posted November 3, 2010 at 2:53 pm | Permalink

          Photons do not have rest mass but they have “relativistic mass” which might be what Gregory is getting at, I’m not sure.

          Two photons can occupy the same space at the same time and have other properties which “regular” matter does not. Does this mean they’re immaterial? Who the hell knows – tell me what the word means and then we can talk.

          You should also know that particles like photons don’t really play well together nor form stable structures so if this is meant to salvage “god” from the scrap heap, you’ll have to think a little harder.

          • Myron
            Posted November 3, 2010 at 3:37 pm | Permalink

            In the broadest sense, “immaterial” means “lacking intrinsic physical properties”.

            • Tyro
              Posted November 3, 2010 at 4:00 pm | Permalink

              Then photons and all other particles are definitely material.

            • Posted November 4, 2010 at 4:49 am | Permalink

              And so is space-time, which has at least the intrinsic property of curvature.

          • Kirth Gersen
            Posted November 4, 2010 at 8:50 am | Permalink

            “… if this is meant to salvage “god” from the scrap heap, you’ll have to think a little harder.”

            Dunno who this question is aimed at; hopefully not me — that’s the LAST thing I’m trying to do. One semi-snarky comment (a play on the vagueness of the term “immaterial”) does not a god-bot make.

  20. john
    Posted November 2, 2010 at 2:14 pm | Permalink

    The definition of supernatural may be insufficient. Under “3. The lawless: things that affect the world in ways that are observable (directly or otherwise), but not subject to the regularities of natural law”, the double slit experiment in physics would fall under the definition of supernatural because the possibility of a single proton being in two places at the same time is “not subject to the regularities of natural law.”

  21. Nick B.
    Posted November 2, 2010 at 3:10 pm | Permalink

    The first category is not, of course, subject to scientific investigation, and if that’s your kind of deistic or apophatic god, then of course science has nothing to say about it.

    I think science does have something to say about it. Science broadly construed, that is. Just because we can’t run an experiment doesn’t mean we can’t bring scientific knowledge and, perhaps more importantly, scientific thinking to bear on the claims/beliefs. A deistic god, it would seem, would be susceptible to the Ultimate Boeing 747 Gambit, for example. And of course a good deal of science underlies that argument (evolution, the relationship of mind to matter). And what about religion as natural phenomenon? How humans create gods in their own image, how we are predisposed to see agency in the world where there is none, etc. (I realize a deistic god isn’t much of an agent but it clearly stands on a continuum with all other gods.) Throw in a little Occams Razor and the door is pretty much closed to such a supernatural being by science.

    I make this point because one move religious apologists often make is to claim that God is not measurable or testable by scientific tools and then implicitly conclude that reason has no applicability to the question of God at all.

    • MadScientist
      Posted November 2, 2010 at 4:38 pm | Permalink

      I would go a bit further and say that current scientific understanding in fact rules out the existence of deities. A god would violate the laws of thermodynamics if that god is composed of any of the particles (and subatomic particles) which we know of. If any god at all is to exist, we need new classes of matter which follow very different laws from our own natural laws and which can somehow influence our class of matter. The god hypothesis fails at every point. Ironically, to believe that some god created the universe and created humans in a single day is to believe that it is in fact possible for a 747 to materialize with no human intervention. Contrary to what creationists believe the sudden appearance of a 747 would be a pretty strong indicator that their deity may in fact exist – and since 747s don’t magically materialize like that, odds are (and this is consistent with absolutely all evidence to date) humans didn’t magically poof into existence either.

  22. Alex SL
    Posted November 2, 2010 at 7:14 pm | Permalink

    Always late to the party – damn time zones.

    Again, the discussion seems all bloody confused to me because hardly anybody seems to agree on definitions and what question is actually being discussed.

    Definitions: while JC has now made a point out of addressing the definition of supernatural – and my view is always that the actual believers do not really, if we are honest, behave as if they expect what they call supernatural to be unreliable and capricious – the other end of the question (whatever it is, see below) requires a definition of science. People like MP define science in such a narrow way that, once again, it cannot reject supernatural claims by definition – that rejection is then considered metaphysics and not science.

    Questions: Many commenters seem to discuss different ones. Which of the following is the focus here?

    1. Is science allowed to reject the existence of supernatural phenomena if no evidence is found for them (or would that be arrogant scientism)?

    2. If there were evidence for the supernatural, would it be scientific to accept supernatural phenomena, or is science by definition required to search for “natural” (meaning here gradualistic and non-intelligent) phenomena as alternative explanations forever and ever, even if they never turn up?

    3. Can supernatural be defined in a coherent way at all? (And if not, if nature is by definition all that exists, could a hypothetical superhuman intelligence that is part of nature be called a god? Cue superadvanced aliens argument.)

    4. Would it now, with all we already know about the world, still be theoretically possible to find evidence for gods etc.?

    And perhaps some more that I missed.

    Thing is, JC says yes to #2, and then comes PZ and says “I disagree” and to give his reasons, he weirdly argues for no to #4; and then comes MP and says, “I also disagree, and PZ is right”, but what he actually means is no to #1.

    It gets weirder in the comments here. Everybody is talking past each other, and it will be a long way to have even just that sink in…

    • Tim Harris
      Posted November 2, 2010 at 7:29 pm | Permalink

      Regarding your point 3, read anthropologists and cognitive scientists like Boyer, Whitehouse, Pyysiainen… I know I (and, I’m glad to see, Paul W) harp on about this, but a great reason for the unsatisfactoriness of these arguments is that people haven’t bothered to acquaint themselves with an important body of scientific work (I don’t want to sound snide, but science does extend beyond physics, chemistry and biology) that simply allows these unanswerable questions, about which people circle like Milton’s devils in wand’ring mazes lost’, to be side-stepped…

    • Myron
      Posted November 3, 2010 at 12:08 am | Permalink

      “…if nature is by definition all that exists…”

      Nature is not by definition all that exists. My simplest definition is:

      Nature = the MEST complex
      (MEST = Matter-Energy-Space-Time)

      [I know that “MEST” is an acronym coined by Ron Hubbard and used by Scientology. I use it too, but I have absolutely nothing to do with Scientology.]

      • Alex SL
        Posted November 3, 2010 at 3:41 am | Permalink

        Sigh. The thing is not that you or me decide the answer by fiat, but becoming aware of the fact that it is one of several questions that people disagree about without realizing that there are, in fact, several different questions.

        That being said, my personal opinion is indeed that there are only two sensible definitions of nature: either “all that can be inferred to exist” or “the opposite of culture”, i.e. everything not created or changed my humanity.

        • Myron
          Posted November 3, 2010 at 10:35 am | Permalink

          From the point of view of philosophical naturalism, culture is part of nature. Cultural products are in a sense non-natural, but this sense is irrelevant with regard to philosophical naturalism.

      • Posted November 3, 2010 at 7:18 am | Permalink

        Nature = the MEST complex
        (MEST = Matter-Energy-Space-Time)

        So “nature” got a whole new definition in the early 20th century, after Einstein showed that these four were linked? What if M-Theory ever gets accepted, will you then redefine nature again so that it also includes branes? Or will branes continue to be supernatural or extranatural?

        • Myron
          Posted November 3, 2010 at 10:36 am | Permalink

          M-theory is a theory of the nature of the MEST complex.

          • Paul W.
            Posted November 3, 2010 at 11:38 am | Permalink

            But what if all the stuff in M-theory turns out to be a dream in the mind of an irreducibly supernatural God?

            That is, what if everything reduces to the nonmental and nonteleological at one level, but that in turn reduces to an irreducibly mental/teleological level.

            In that case, I’d say that naturalism appears true down to that level, but is actually false.

            What makes a theory “naturalistic” isn’t its subject matter, (e.g., matter-energy and spacetime, or minds and societies) but its own structure.

            In particular, the crucial thing is whether it posits that some things don’t reduce to nonteleological and nonmental brute matter and energy and so on.

            For example, what made electricity turn out to be obviously natural—despite seeming spooky and magical at first in a variety of ways—is that it turned out to be utterly impersonal and mindless, and not especially directly to anything of intrinsic human interest, like Truth or Justice, or Life, or the Wrath of God.

            It’s important how things like lightning (and x-rays nuclear energy and noneuclidean space and so on) were new, and didn’t fit into our particular scientific theories at the time, but did turn out to fit into the basic materialist paradigm, which is reductionist in a broad sense. (E.g., including functional reduction, not just naive greedy “reductionism.”)

            Whether a theory is naturalistic or supernaturalistic is a matter of what reduces to what, and what’s irreducible.

            • Myron
              Posted November 3, 2010 at 12:36 pm | Permalink

              Paul Draper uses the following definition:

              “x is supernatural =df. x is not part of nature and x can affect nature.

              Nature =df. the spatiotemporal universe of physical entities together with any entities that are ontologically or causally reducible to those entities.”

              (Draper, Paul. “God, Science, and Naturalism.” In /The Oxford Handbook of Philosophy of Religion/, edited by William J. Wainwright, 272-303. Oxford: Oxford University Press, 2005. p. 277+)

            • Posted November 4, 2010 at 3:59 am | Permalink

              @Myron: Paul Drapers definition only shifts the problem of defining “natural” to the problem of defining “physical”.

          • Posted November 4, 2010 at 3:57 am | Permalink

            You miss my point. You define nature as a list of interacting entities, but at least two of them (space and time) were not recognized as interacting entities in their own right until Einstein proposed his theory of relativity. If M-theory introduces more new entities, like branes, would they become part of nature or not?

            • Myron
              Posted November 4, 2010 at 2:31 pm | Permalink

              – Yes, the proper definition of the concept of physicality is a nontrivial problem.

              “Is there any good way to delimit the realm of the material that does not preclude further discoveries in physics, but also does not trivialize the category by allowing it to include anything that people in departments labeled ‘Physics’ might eventually come to study? This is anything but a trivial problem[.]”

              (BonJour, Laurence. “Against Materialism.” In /The Waning of Materialism/, edited by Robert C. Koons and George Bealer, 3-23. Oxford: Oxford University Press, 2010. p. 7, n. 4)

              – To define “nature” as “the MEST system” is not to decide the questions as to what its fundamental structure is and how many dimensions there are.

  23. Neil
    Posted November 2, 2010 at 7:45 pm | Permalink

    I hate the term “supernatural” because the prefix “super” suggests “superior to”. (Forgive me if someone else has already made this point.)

    I suggest the term “infranatural” for any phenomenon not yet explained by science because it simply means, we are not just there yet, explananation-wise.

    • Dominic
      Posted November 3, 2010 at 4:48 am | Permalink

      What about ‘differently-natural’?!

      • Sigmund
        Posted November 3, 2010 at 6:20 am | Permalink


  24. Myron
    Posted November 2, 2010 at 10:51 pm | Permalink

    In my opinion, if there were any supernatural agents, i.e. active (self-)conscious immaterial beings, then their behaviour wouldn’t be totally chaotic. They would presumably have a certain character and certain habits, and so display certain regular, i.e. repeatedly occurring, patterns of behaviour.

  25. Myron
    Posted November 2, 2010 at 10:55 pm | Permalink

    Methodological naturalism (as metascientific nonsupernaturalism, not as metaphilosophical scientism) seems to boil down to this simple motto:

    “Naturalistic explanations first—supernaturalistic explanations last!”

  26. Myron
    Posted November 2, 2010 at 11:07 pm | Permalink

    “I like the term ‘lawless’ instead of ‘supernatural’.” (J. Coyne)

    What about possible laws of supernature (governing hyperphysical mental-to-mental causations), or possible bridge laws causally connecting (physical) nature and (hyperphysical) supernature?

  27. Myron
    Posted November 2, 2010 at 11:28 pm | Permalink

    A “violatable” or “breakable” law of nature is but a pseudo-law, i.e. a nonlaw, strictly speaking. Miracles would simply not be subject to any law of nature. So, here, the term “circumvention” is more adequate than “violation”.

  28. Myron
    Posted November 2, 2010 at 11:37 pm | Permalink

    “The first category is not, of course, subject to scientific investigation, and if that’s your kind of deistic or apophatic god, then of course science has nothing to say about it.” (J. Coyne)

    Absolutely inactive immaterial concrete (spiritual) entities belong to “the silent”, and so do abstract entities, which are causally powerless by definition; and so do other worlds which are spatiotemporally and causally isolated from our world.

  29. Myron
    Posted November 2, 2010 at 11:47 pm | Permalink

    By the way, it is helpful to keep in mind that “supernatural” and “hyperphysical” are virtually synonymous words. Hyperphysical is what is neither physical nor ontically and genetically dependent (i.e. dependent for its being and becoming) on anything physical.

  30. Notagod
    Posted November 3, 2010 at 4:40 pm | Permalink

    We might have ten arms including eight that don’t do anything. Everyone knows that is possible but we don’t speculate about what would or could happen if one of the eight started flailing about suddenly. Remove all the emotional attachment to gods and extra arms have the same observable properties. Isn’t it inconsistent to say gods are possible but not give equal time and consideration to all the other possibilities that don’t do anything?

    If one of our eight invisible arms ever starts doing anything then we can acknowledge it, until then the correct answer to how many arms a human has is two, correct? And we say we have zero wings even though wings could be there? Why should gods be treated specially without regard for all other things that have no evidence, form or function?

    If I state I have two arms and no wings or beak would you be just as inclined to tell me that there is the slightest minuscule possibility that I do and that I am wrong for not acknowledging It? Does consistency matter?

  31. abb3w
    Posted November 4, 2010 at 12:32 pm | Permalink

    The problem I see is at the term “fundamentally lawless”. This is in part due to what seems an ambiguity in the term “law”, which refers to either the actual rules the universe uses, and to the rules which humans infer. Finding evidence to violate the latter is philosophically trivial; for example, finding objects whose inertial and gravitational masses were non-equivalent, or finding a type-2 perpetual motion machine. The universe has no obligation to conform to our expectations.

    However, finding evidence that violates the actual rules is more problematic; because any rule that is contradicted by evidence is by definition not the “actual” rule, but merely some approximation. And, contrariwise, any finite set of evidence can be given by at least one rule: simply saying “the rule is that (whatever-it-is) happens”, enumerating the evidence. (Science is about the search for WHICH rule should be used; something better than mere enumeration is usually possible.)

    Working with infinite sets (to anticipate the possibilities of future evidence) makes this (unsurprisingly) more difficult; and gets into type-0 grammars, Recursively Enumerable languages, Church-Turing automata, the Halting Problem, ordinal degree Turing hypercomputers, a mathematically rigorous form of Occam’s Razor, and something very kin to science that incorporates both Falsification and Parsimony. Remaining “fundamentally lawless” becomes pathologically more and more difficult – though not philosophically impossible.

    One can get more “lawless” still (at least, if you take the axiom of the Power Set)… but on taking “fundamentally lawless” beyond any ordinally recognized law, you seem to lose all hope to resolve Hume’s problem of induction, even enough to tell a hawk from a handsaw when the wind is southerly. Since deities are not considered utterly random to that degree, we’re left with the deities being talked about being ones that science can be used to consider – and perhaps dismiss, either as being incoherent or as being probably wrong.

2 Trackbacks/Pingbacks

  1. […] First some links.  Chris Schoen has posted Is “Dark Matter” Supernatural? Sean Carrol has further commented, with Is Dark Matter Supernatural? And now Jerry Coyne has commented on that, with Sean Carroll on the “supernatural”. […]

  2. […] baggage, and warps thinking about it from the outset. I kept stumbling on that in a discussion on a post of Jerry’s yesterday. Sigmund saidI suspect we will be unable to determine whether such evidence indicates a ‘God’ […]

%d bloggers like this: