Thursday, April 04, 2013

John Wilkins Revisits Methodological Naturalism

Thank God John has revived Evolving Thoughts!

As usual, his latest post contains lots of food for thought [God and evolution 2: The problem of creation]. I want to pick out one morsel because it's back in the news recently.

It's the problem of "methodological naturalism" and whether it restricts science. Kairosfocus recently posted an article on methodological naturalism at Uncommon Descent [Optimus, replying to KN on ID as ideology, summarises the case for design in the natural world]. He pointed out, quite correctly, that by restricting science to methodological naturalism it means that Intelligent Design Creationism becomes non-scientific by definition.

I've argued frequently in the recent past that science is not bound by methodological naturalism [Is Science Restricted to Methodologial Naturalism?] [Accommodationism in Dover] [Methodological Naturalism]. My stance has evolved over the past few years. Back in 2007 I was a staunch defender of limiting science to methodological naturalism [Methodological Naturalism].

Jerry Coyne agrees with my current position on methodolocical naturalism. We are both upset by the way it's used to support accommodationism at NCSE [The NCSE Position on Science vs Religion] and at AAAS [AAAS Supports Accommodationism, Illogically]. If you're new to the subject, read Jerry's latest post at: Must we assume naturalism to do science?. That will bring you up to date.

John Wilkins and I have debated this controversy several time [e.g. John Wilkins Defends Methodological Naturalism]. The important point, as far as I'm concerned, is that there are respectable philosophers who disagree with the idea that science can't investigate the supernatural because it is constrained by methodological naturalism.

Here's what John said yesterday (in my time zone).
The term “naturalism”, however, is ambiguous. On the one hand it can mean giving a natural explanation through the use of scientific methods such as the use of human reasoning and observation. Or, it can mean the claim that only “natural” things exist. The first is sometimes called “methodological naturalism”, and it is the underpinning of all science, and indeed all learning about the world. The second is sometimes called “metaphysical naturalism”, although I think it is instead a claim about what exists (which is called “ontology” amongst the philosophical community). God might be natural in that sense. There is no real sharp dividing line between the natural and the supernatural that would satisfy most believers. For example, human nature for some is held to include a soul, which is divine. So let us call the second kind ontological naturalism.
Obviously I don't think that methodological naturalism is the "underpinning of all science." I think science is free to investigate claims of the paranormal (i.e. not naturalism) and can, in principle, discover things that don't meet the definition of naturalism.

What makes me nervous is that this is John's field. Is he saying that among philosophers of science the overwhelming consensus believes that that in science you can only give natural explanations? Or is he simply offering his personal opinion disguised to look authoritative?

Is there a slam-dunk philosophical refutation of the position held by the likes of Yonatan Fishman and Maarten Boudry that Jerry Coyne and I (and many others) are unaware of?


  1. I think science is free to investigate claims of the paranormal (i.e. not naturalism) and can, in principle, discover thinks that don't meet the definition of naturalism.

    I think it would be helpful if you were to give an (imaginary!) example of such a discovery and the logical chain behind it. I think your "in principle possible" is actually "in practice impossible".

    The only way I can think of to prove the existence of a supernatural entity would be to make a list of all "natural" processes, laws and possible consequences thereof, and then observe something happening (booming voice, burning bush etc.) that is impossible according to the "natural" framework.

    I think that in practice, it's impossible to make that list. Science will never be "completed". That means that whatever happens, no matter how bizarre, even if it's inexplicable by all known scientific laws, we can always say "well maybe there's a natural law that allows X to happen, we just don't know it yet".

    Moreover, it may well be impossible even in principle to make such a list. Consider Godel's incompleteness theorem, which proves that in any formal system of logic there will be statements that are both true and unprovable.

  2. Oh, not, not another bogus application of Goedel's theorem.

    I sentence you to read Torkel Franzen's book and not repeat this nonsense until you've read it and understood it.

  3. I always feel stupid in these discussions - it's as if everyone else knows (or think they know) what the word "natural" is supposed to refer to, but no one ever wants to let me in on the sectret. Why does no one ever start by saying what they understand under "natural", so that the rest of the discussion can make sense?

    1. Natural is that which can be investigated by science.*

      The rest is just stories people made up huddling around the campfire and wondering what those strange sounds and glowing red eyes actually were and whatever did happen to cousin joe when he wandered out to take a piss late one evening.

      * Yeah, I know, tautological.

    2. Konrad, here is a good definition of 'supernatural' as provided by Jerry Coyne in the post Larry cites above. It is from the authors of a paper Coyne refers to throughout his post. I think it's an excellent working definition and, one could suppose that 'natural' is pretty much anything and everything beyond this definition

      "(1) ..... operate in ways that fundamentally violate our current understanding of how the world works, (2) ... exist outside the spatiotemporal realm of our universe (though they may still causally interact with our universe), and (3) .... suggest that reality is at bottom purposeful and mind-like, particularly in a sense that implies a central role for humanity and human affairs in the cosmic scheme."

    3. @John: You don't. You point out that the word "natural" is ambiguous, but then proceed to use it anyway without being clear on which meaning you (or the people whose views you are describing) have in mind.

      @Andyboerger: those were three definitions, not one. It's not clear that it makes a lot of sense to use the intersection of the three definitions (you are saying that something satisfying two but not all of the three criteria should be called "natural"? any particular reason for this?), and in fact none of the three definitions are any good:
      (1) means that most (or all?) phenomena in the world start out as supernatural and gradually become natural the more we study them - this doesn't correspond to normal use of the word.
      (2) is problematic given that our understanding of what a concept such as "spatiotemporal realm" means is constantly in flux (at least while we don't have a grand unifying theory of physics).
      (3) many (most?) supposed supernatural phenomena (say, ESP) need not have this property.

    4. No, you are wrong about 1). 1) only means that it is a phenomena which contradicts our empirically established laws of nature. It doesn't necessarily replace the laws of nature, they could remain intact, but the empirical evidence for the supernatural phenomena is a new category of phenomena, it resides outside the laws of nature, and there is nothing here that says that this distinction will gradually disappear, as you are incorrectly claiming.

      Also, regarding 3) and ESP, ESP could definitely be mind-like, or it could be materialistic cause and effect, it could be defined either way, and when it is defined the latter way it is naturalistic. But regardless of which way it is defined, it is counter- evidenced, and therefore must be rejected as fiction.

    5. Re 1): What I'm saying is that our "current understanding of how the world works" is continuously improving, with the effect that things that were previously supernatural under this definition tend to become less so as our understanding advances. E.g. by this definition, all of the early experimental results in quantum physics were supernatural at the time of their discovery, and then became natural once a theoretical framework for their explanation became accepted. That's not a normal usage of the word "supernatural".

      Also, it doesn't seem standard to use a definition where a phenomenon can be supernatural to one group of people (whose understanding of the world doesn't support it) but natural to another group (whose understanding does).

      Re 3): I said "need not" - so I'm referring to non-mind-like definitions. I'm not convinced that most ESP proponents would call such definitions naturalistic.

    6. I agree that the wording for 1) is too loose. It should be possible to rephrase that so that it complies with my description, which I am confident is what they intended. However, it may be difficult to express this clearly, I don't have an alternative phrase to offer that is better. That this distinction can be difficult to clearly delineate actually supports their overall argument that it is a mistake to define science in relation to this murky distinction. Both the methods and the conclusions of science are justified by success, not by the murky natural versus supernatural category distinctions, which are actually superfluous to defining what is good science. Science rejects methods and conclusions that are strongly counter-evidenced, and supernaturalism is a category that covers assertions which are strongly counter-evidenced, which is why it is rejected.

  4. Larry, I think you misread me. I am not defending ontological naturalism (the view that only natural things exist) in this piece. I am defending the idea that any knowledge we have is the result of methodological naturalism (observation, trial and error). Hence, if we can be methodologically natural in our investigations of any phenomena, including putatively supernatural phenomena, we can investigate it.

    1. Yes indeed. The problems come when somebody invokes magic of some flavour. If _the soul_ becomes part of an explanation, without providing some reproducible or testable aspect of a soul we are left at a crossroads of a sort. Either we abandon the rules of naturalism and improvise or we stick to tried and true forms. It's enough to give one the blues.

    2. John, I don't understand why you say we could not do science if occasionalism were true. Surely we could still statistically assess whether the gods are consistent or whimsical, and if they are consistent then we could investigate their attitudes (like whether or not they are always inclined to cause a flame after they cause the striking of a match of a given material)?

      Right here you seem to define "methodological naturalism" as basically the primitive scientific method ("observation, trial and error"). However, this is different from the blog post where (like most people) you seem to define it as a restriction on the kinds of hypotheses which may be permitted (viz. "a natural explanation" where you describe "nature" having to do with secondary causation and the exclusion of otherworldly contributions). Unless you meant something else in that instance of the word "natural"?

      Such a restriction is what the scientists seem to be objecting to, even after you surrender ontological naturalism. We were able to conceive experiments to test whether crop circles were ordinary growth patterns (like mushroom fairy-rings are), or intelligently designed (as proved to be the case), or produced by direct intervention of supreme beings from the heavens (as had been theorised). Quantum physics is another example of scientists being open to explanations which contradict all preconceptions about reality. If you disagree with those who say science cannot test supernatural explanations, but you still consider "methodological naturalism" to be something necessary for any science even in principle, does this mean the definitions of "methodological naturalism" found in most places (like RationalWiki) are wrong (as applied to your writings)?

    3. John Wilkins said,

      I am defending the idea that any knowledge we have is the result of methodological naturalism (observation, trial and error). Hence, if we can be methodologically natural in our investigations of any phenomena, including putatively supernatural phenomena, we can investigate it.

      I'm struggling to understand if your view is any different than the traditional view of Michael Ruse (and others). Do you think it is?

      I don't agree with you that, "any knowledge we have is the result of methodological naturalism." In my opinion, the knowledge gained by science is the result of evidence-based application of sound logic. (Yes, that's a simplification.) It was quite possible that the application of scientific investigation would reveal causes that were not "natural." In fact, many scientists reached that conclusion in the past and some still do today.

      The consensus among knowledge-seekers is that we have no evidence of supernatural beings. Thus the world appears materialistic and naturalism is an appropriate conclusion based on centuries of scientific investigation.

      That's not the same thing as saying that the knowledge we have was due to the assumption of methodological naturalism.

      You suggest that "we can be methodologically natural in our investigations of any phenomena, including putatively supernatural phenomena." How, exactly, does that work? How can we be "methodologically natural? when investigating miracles? Don't we have to at least consider the hypothesis that they are do to the actions of a supernatural being? Why would scientists want to limit themselves to strictly naturalistic explanations rather than go where the evidence might lead them?

    4. Lets say that people rely on trial and error to discover the correct way to worship the divine to acquire divine revelation and they confirm the revelations by observation. Are you saying that these people are practicing methodological naturalism? Are you saying that these people cannot be properly labeled "scientists"? The way I see it, this would be an example of scientists obtaining knowledge due to their assumption of methodological supernaturalism. I see no justification for declaring that this would be, a-priori, by definition, not science or not methodological supernaturalism.

  5. Whatever is real is the object for mans investigation.
    Science is just a more careful investigation of things. It is supposed to be.
    Investigation of nature includes showing its origins and this includes a creator.
    ID says nature has the fingerprints of God all over it.
    YEC and some ID say ideas about origins, evolution, have no evidence and are wrong also.
    Science can't investigate the invisible kingdom of God and so ultimate reality.
    in these matters its all about investigation of nature showing/failing to show God or genesis is true or not.

    1. Robert, if science cannot investigate the invisible kingdom of god, how could you possibly know of its existence? Science isn't restricted to scientists in is just a formalized version of what your mind does to deduce things. If the kingdom of god cannot be accessed by science, then it cannot be accessed by your mind either. Hence, you could not possibly know of its existence, now or in the future.

    2. Amen. Science is just a formula for investigation at best.
      The invisible Kingdom is shown by the visible world having been created in its glory of complicatedness. Complexity.
      The creator is suggestive of his own kingdom.
      Yet its at a spiritual area. so investigating it by us who live in a material universe is a problem.
      Splitting atoms will never reach the true reality. its the other way. Its the spirit behind biology that is in the better direction but also can't reach the spiritual world.

      Natural can't investigate supernatural. Thats why its called super.

  6. Is he saying that among philosophers of science the overwhelming consensus believes that that in science you can only give natural explanations?

    Larry, do you forget that the paper "How not to attack Intelligent Design Creationism: Philosophical misconceptions about Methodological Naturalism" by Maarten Boudry, Stefaan Blancke and Johan Braeckmanyou so much admire said that "the most widespread view ... conceives of MN as an intrinsic or self-imposed limitation of science"?

    Now, of course, consensus in philosophy is no more a slam-dunk refutation of minority views than it is in science. But the confusion is more yours than John's.

    The real problem is, I think, a confusion between science's ability to investigate phenomenon and science's ability to explain phenomena. No adherent of MN that I know says that science cannot investigate the age of the Earth or whether there is some statistical correlation between intercessory prayer and good and/or improving health. The problem is, assuming we found some statistical correlation between prayer and healing, how would science investigate the cause and when, exactly, would we give up the hunt for a "natural" cause?

    For 300+ years we have demonstrated that the phenomena of mass and the phenomenon of gravity are correlated but we still do not have a satisfying explanation of how gravity acts at a distance. Scientists haven't, after all this time, thrown up their collective hands and said "Well it must be because God sucks." In short, scientists act in the real world as if they are applying MN. They look for "natural" causes and are not willing to stop looking for them, no matter how long it takes.

    Now, you may philosophically choose to believe that we have found enough "natural" explanations of phenomenon that we can conclude that those are the only kind there are. But that is not science ... anymore than all the examples of natural selection we have found makes a conclusion that it is the only mechanism of evolution is a scientific result.

    1. Gravity is a geometric property of spacetime, there is no mystery about how it functions over distances. But more to the point, the primary reason that scientists don't look to supernatural explanations is that we know, from historical experience, that they fail. Originally, scientists did not rule out supernatural methodology or ontology. Supernaturalism was considered to be out of bounds for science by the late 19th century for the same reason people rule out walking into walls. The reason is it always fails.

    2. Gravity is a geometric property of spacetime, there is no mystery about how it functions over distances.

      Really? So there are no problems with General Relativity, much less reconciling it with reconciling General Relativity and quantum mechanics? And just how does mass go about affecting the geometric property of spacetime anyway?

      Don't confuse a more or less successful model with an explanation.

      ... the primary reason that scientists don't look to supernatural explanations is that we know, from historical experience, that they fail.

      No, we don't scientifically know that they will fail, (anymore than we "knew" all swans were white ... until we reached Australia ... see Hume's problem of induction). We employ a heuristic that has been wildly successful but does not give us scientific "knowledge" that such explanations are always correct; only that the heuristic is well worth pursuing. But that's the very point of MN. Science will never stop looking for "naturalistic" explanations and, therefore, rules out "supernatural" explanations a priori, because, otherwise, the heuristic doesn't work.

    3. You are pulling a bait and switch, that there is no mystery about how gravity functions over distances doesn't mean that physicists have a complete theory of everything.

      You seem to be saying MS cannot work, but that is nonsense. MS could work, provided that supernaturalism was true. Furthermore, if MS did work then scientists would be relying on MS.

    4. You are pulling a bait and switch, that there is no mystery about how gravity functions over distances doesn't mean that physicists have a complete theory of everything.

      First of all, you didn't answer my question about how mass affects the geometry of space/time. Throwing a bowling ball on to a rubber sheet is a nice metaphor and a good teaching tool for the model but it does not explain.

      In any case, in the absence of a "complete theory of everything" how can you say with scientific confidence that any explanation will "fail." I can say with confidence that the heuristic of science fails once you say "goddidit," simply because such an explanation can never be examined by the methods of science. Therefore, while scientists can and should look for any "naturalistic" explanations for phenomenon, whether or not proposed be religions, the absence of such explanation is not a confirmation of supernatural explanations. The flip side of that coin is that is that no matter how many naturalistic explanations you come up with for phenomena, you cannot rule out supernatural explanations for other phenomena.

      MS could work, provided that supernaturalism was true.

      I'm not sure what you mean by "MS" but, as I indicated before, if we found some statistical correlation between prayer and healing, how would science investigate whether the cause was "supernatural" or some presently unknown "natural" cause, except by trying to find such a natural cause? In other words, the heuristic would be the same even if we had a candidate "supernatural" phenomenon ... we'd assume a priori that we should look for a natural cause.

    5. John, your last point is exactly correct. IF studies showed conclusively, or even gave a strong indication, that prayer DOES 'work', people like Sam Harris would never attribute it to prayer, i.e. a deity listening and responding. They would go straight to what they already know about neuroscience and add a couple layers of just so stories. Richard Dawkins would consider it corroborative evidence of his 'meme' theory, etc.
      It would not be called 'supernatural', it would be labeled as something 'previously held to be attributable to supernatural causes', and the explanations would move forward in such a way that nothing beyond chemistry and already known natural laws were at play.

    6. If, for example, it was demonstrated empirically that revelation following prayer was successful as a method for obtaining knowledge then we would all acknowledge that. The question of whether or not the revealer was a god or advanced technology could remain, but that question is also properly addressed by a best fit with empirical evidences, and would thus depend on the details regarding what prayers are successful and what prayers are not. In that universe people like Sam Harris would either be agnostic (undecided rather than atheist) or theist, depending on he details.

    7. ..but. EA, the investigative work would need to be done according to the scientific methods that the researchers are most familiar with. "Goddidit" would, almost by definition, satisfy believers, so they wouldn't try to examine WHY prayer was working.

      People who were interested in knowing the WHY, wouldn't be able to stop at Goddidit, so they would have to proceed from naturalistic assumptions and methodologies. Nearly all the people doing the work would be people trying to prove that God DIDN'T do it, and the papers would support the preconceived, GodDIDN'Tdo it hypotheses.

    8. No, this is a false caricature of the process of anchoring one's beliefs in the empirical evidences. The issue with belief justification is an issue of best fit with empirical evidence, we are not restricted to only having beliefs that can be quoted from published science textbooks. No one claims otherwise. No science textbook declares there are no gods.

      It would be very difficult (but not impossible) to properly justify belief in god with just one, isolated, area of evidence, such as prayer. It would be easier to justify belief in god with a wide variety of evidences pointing in that direction. But nothing is a-priori ruled out, it is all about best fit with the weight of the overall evidences. Evidence first, evidence is the master, we are slaves who follow the evidences wherever they go.

      A big problem with god beliefs is that gods that merit worship invariably violate the laws of physics. Therefore gods are very strongly counter-evidenced. To justify belief in gods requires overcoming this substantial accumulation of evidence against gods. Again, this is theoretically possible, but it requires a similar accumulation of evidence for gods in quantity and quality and I wouldn't bet on that happening given the current context, where god scores about zero evidences and all of the evidences we have are against.

    9. Sorry for the delay, I was otherwise engaged.

      No, this is a false caricature of the process of anchoring one's beliefs in the empirical evidences.

      We're not talking about anchoring one's beliefs, we're talking about how the process of science works. You are free to construct your metaphysics as you will and I certainly agree that doing so by reference to science is an excellent idea. But your methaphysical beliefs are not coextensive with science.

      ... it is all about best fit with the weight of the overall evidences

      Which, of course, is exactly the argument the IDers make. But what way could we scientifically test the "weight" of evidence? When we restrict ourselves to "natural" causes, that is fairly (but not straightforwardly) easy. When we admit "supernatural" causes, we have abandoned the ability to empirically decide which evidence is better than the other.

      A big problem with god beliefs is that gods that merit worship invariably violate the laws of physics.

      The "laws" of physics are simply regularities we have empirically identified by assuming "natural" lawlike behavior by the universe. If we admit supernatural explanations, there are no such laws.

    10. When we are talking about what is true and false about how the world works we are always talking the evidences. All efforts to change the subject are wrong. This isn't about anyone's metaphysics, it is about the evidences, what the evidences say, who is following the evidences and properly justifying their beliefs in the evidences.

      Yes, the IDers self-claim that they are following the evidences. But mere self- declaration doesn't make it so. I can declare that the evidences show that the universe is 10000 years ago, but I would be wrong. The claims of the IDers have been successfully refuted, as anyone who is honest about following he evidences will know.

      Yes, the laws of physics exhibit many regularities, but that isn't a symptom of a bias of science, it is a symptom of a "bias" in how our universe functions. The methods of observation and trial and error are capable of finding irregularities, and in fact physicists have found some irregularities. Physicists are very interested in irregularities, they don't dismiss or ignore them, they follow the evidences wherever it takes them because they want to know what is real. The notion that naturalism is synonymous with regularities, or supernaturalism is synonymous with irregularities, or that science can only deal with regularities, are all false.

    11. IF Methodological Supernaturalism and/or supernaturalism worked (along the lines of 'hey, this controlled statistical study shows that catholics praying were significantly more likely to be healed/regrow limbs/rise from the dead/etc", and we wanted to investigate further, THEN we'd be obliged to use prayers for revelation, we wouldn't stick with older methods, likesay having experimental groups pray to different sets of saints. Since it doesn't work in the first place, it's not a good tool to investigate even it's own claims. In fact, since it doesn't work in the first place, we can ignore it, since we'd just be able to "pray for all goodly knowledge" or some such, and if we could even do that in some hypothetical universe, we wouldn't even be inclined to, since anything we could get that easily would be just flat out given to us in the first place, no?

    12. MS isn't necessarily easier, it could be very difficult. It could be that only a few people who are unusually righteous receive revelations and only if they worship properly. Worshipping the wrong deity could be unproductive. Correct worship could be time consuming and arduous. There is plenty of room here for trial and error to improve the methods, and a need for observation to confirm the revelations. Since this isn't our universe, it is just endless possibilities. The important point is that following the evidence is not a method that is biased against religion. Following the evidence can either confirm or disconfirm religious beliefs and knowledge could be obtained via MS provided that supernaturalism was true.

  7. What I can't understand is how science could investigate things like the paranormal or supernatural in anything other than a question-begging way. It's not like we have a clear understanding of what a paranormal or supernatural phenomenon is - only what it isn't - so if we did investigate it scientifically, then why would we end up with anything other than an unknown? I just don't know what it would mean not to assume naturalism in order to do science. Perhaps that excludes ID by definition, but doesn't that say something about the vacuity of ID (lack of coherent predictive explanations) over a limitation of the scientific enterprise?

    1. Exactly, in order for science to investigate a "supernatural" claim it has to a priori assume natural causes as only things that can be measured can be used as evidence to support the claim.

      If you can measure it, it is by definition natural.

      You can't disprove a supernatural claim but you can say that there is no evidence to support it and quite often you can present evidence that contradicts the claim, to the point where the probability of the claim being true approaches zero.

      The biggest problems with the science investigating supernatural claims are, in my opinion:

      1) The diversion of finite resources to the investigation specious claims, typically most if not all of these claims should be summarily rejected using the burden of proof criteria.

      2) The veneer of respectability that the claim acquires when it becomes subject to investigation.

    2. When you say "You can't disprove a supernatural claim but you can say that there is no evidence to support it and quite often you can present evidence that contradicts the claim" you simultaneously arguing both sides of the question and thus are contradicting yourself. If by "you cannot disprove" you mean something more than "you cannot contradict with evidence" then your phrase "you cannot disprove" becomes a meaningless truism like saying "water is wet".

  8. It sounds to me like we are arguing over definitions, and perhaps Dr. Wilkins is saying science only works by natural methods because observation, experimentation, logic, peer-review, etc. are all natural methods, not supernatural methods. Which raises the question of what a supernatural method would be - perhaps prayer, meditation, eating sacred mushrooms? To which Dr. Moran might reply, well if any of those worked reliably we would use them. Anyway, I think perhaps somewhere at the bottom of this controversy is the notion that science is the discovery that only natural methods for discovering things about our universe work reliably. Some of us have absorbed that lesson down to our bones to the point where it makes absolute sense to us and we can hardly conceive of anything different (i.e., since it is true it must be necessarily true, like finding out that every prime number of the form 4N+1 that you look at is the sum of two squares), and some are more broad-minded. Or perhaps not.

    Anyway, I also dislike the notion that science can't deal with supernatural claims because of some philosophical demarcation, but would tend to agree that science can't in fact study the supernatural because the supernatural doesn't exist. Because every time I look for it, it isn't there, just like every prime number of the form 4N+1 that I look at is the sum of two squares.

  9. But this is a practical dispute, not just an academic dispute, and saying science cannot investigate things that don't exist isn't resolving the dispute. The bottom line here is that Dr. Moran is correct, if methodological supernaturalism worked than scientists would be people who practiced methodological supernaturalism. Everyone who is insisting otherwise is, as far as I can see, wrong. I keep hearing arguments otherwise, but those arguments invariably are based on adding extra attributes to the definitions of supernatural or natural in a biased way to reach their preferred conclusion. For example, they say natural implies predictability while supernatural implies unpredictability, but that is not true.

  10. For example, they say natural implies predictability while supernatural implies unpredictability, but that is not true.

    Sure it is. You only can reach the opposite conclusion by imagining supernatural explanations unlike those typically proposed (i.e. that act with substantial regularity and predictability, are scrutable, limited in their powers, etc.), and by imagining natural explanations unlike those typically proposed (superpowerful aliens or whatever). Terms should reflect what most people mean most of the time, not the weirdest cases that can be dreamed up by philosophers just trying to poke at conventional wisdom. IMHO.

    1. No, when discussing the alleged limitation of science to only deal with the natural, we cannot confine ourselves only to that which is true about our universe. To address this question objectively and properly, we must take an open minded approach and ask - if our universe were supernatural what would science be like? If you insist on dealing with supernaturalism only in the form it is posited by those people who are deliberately biasing the definition of supernatural to escape detection than you are biasing the discussion. After all, we can play the same game in reverse, and redefine natural to only refer to that which is undetectable, but doing that would be equally arbitrary and biased.

    2. I say it is the theologians, not the atheists,who are dreaming up the weirdest cases that can be dreamed up - like an undetectable god who expects to be worshipped. If you don't see how weird that concept is, if you think that it isn't weird for no other reason than theologians at Notre Dame and other religious universities promote these weird ideas with one success for absolutely no other reason than to uphold their religious beliefs, then you notion of what is weird is not very well grounded.

  11. Perhaps this is just philosophical naivete on my part, but I always thought the correct description of the situation was not "If the supernatural exists, then science should not investigate it." I thought it was "If the supernatural exists, then science cannot investigate it."

    For instance, suppose someone made the claim that gravity does not exist, but that in fact there are squadrons of tiny angels, whose presence cannot be detected by any means humanly possible, whose job is to pull objects together in accordance with what appear to us to be physical laws of the universe. This is not a claim that could be tested in any scientific way, so it cannot be refuted or confirmed by science. Methdological naturalism, then would hold that we simply ignore the question and assume our observation that gravity is a natural process acting according to natural laws to be correct.

    1. But that assertion that science cannot investigate the supernatural Is flat out false. Supernatural is not synonymous with undetectable, natural is not synonymous with detectable. Furthermore, that which intervenes in our universe is detectable indirectly, regardless of whether it is natural or supernatural, and regardless of whether it is detectable directly or not. We reject angels in our description of gravity because we have no need for them, they don't add any explanation that is otherwise missing, which is the same reason we reject adding superfluous naturalistic adornments to our description of gravity.

  12. People who keep defending the reality of the supernatural by claiming that the supernatural is undetectable don't seem to realize that if what they are saying is true then everyone should be an atheist. There can be no proper justification to believe in anything that is undetectable. This is because a necessary condition of a justified belief is that we have reason to consider the belief to be true but if we eliminate the requirement that our beliefs be grounded in supporting evidences then we eliminate our only means to reliabley determine that our beefs are true.

  13. Isn't part of the argument here that while we can attempt to investigate the supernatural, there's always a 'regress' available to the believers, and we can't test that regression very well at all. If a prayer study didn't work, there's a million super-natural reasons why. In science, we don't allow an infinite line ad hoc attempts to 'rescue the hypothesis', but in the super-natural world, it explicitly has to be allowed, no? Perhaps then the super-natural just boils down to that, the 'auxiliary belt' is allowed to be as wide as necessary, while in scientific thinking it's moderately narrow (this would be a methodological distinction between science and the paranormal, no?)

    1. Robert, deciding on a skirt or pants belt is different then deciding what is true or false about how the world works. The latter decision impacts what kind of citizen we are, how we interact other people, what kind of society we build, and the like, while the former decision does not.

      Supernaturalism could be evidenced much like naturalism is evidenced. They could both be evidenced (they are not mutually exclusive, and indeed no one who believes in the supernaturalism completely denies naturalism since we all live in a natural world, which is one reason why when we say something like 'the natural world could be undetectable' people will correctly think that sounds weird because such a universe is unimaginable). So no, I don't agree that supernatural belief requires an infinite line of ad hoc justifications. Supernatural beliefs can be anchored in empirical evidences and be properly justified. That this is not the case in our universe is compelling reason to conclude that our universe is entirely naturalistic. The bottom line is that we cannot do better than follow the evidences wherever they take us, and therefore that is what everyone should be doing.