Laws of Supernature

On his Dangerous Idea blog Victor Reppert reprints an old exchange between us (we have been going at it for over 35 years now):

An Old Exchange between Keith Parsons and myself over theistic explanations

This is one that appears in my paper on Hume on Miracles, Frequencies and Prior Probabilities.


Science is unavoidably naturalistic, or atheistic if you prefer. Science operates in terms of scrutable, independently testable entities that operate in accordance with knowable regularities. Supernatural beings, on the other hand, are essentially mysterious; claims made on their behalf are not independently checkable, and there are no “laws of supernature” governing their behavior. Furthermore, “explanations” in terms of supernatural entities are inevitably post hoc and untestable. In other words, proponents of supernaturalistic theories can glibly account for things we already know, but become strangely silent when asked to predict something new, something that would allow their theory to be tested.[18]


Even though the locus of discussions of miracles is historical rather than scientific, if it is the case that supernaturalist hypotheses are inevitably untestable, this would mean that supernaturalist claims cannot be genuinely supported by evidence. But some points can be made in response to this position. First of all, I see no in principle impossibility in “laws of supernature.” One cannot, of course, generate deterministic laws governing divine conduct, but one cannot generate such laws concerning the behavior of subatomic particles, either. One can, of course, form probabilistic expectations concerning the conduct of subatomic particles, but, as we have noted, one can generate probabilistic expectations concerning divine conduct as well. It would disconfirm belief in the Christian God if Jim Bakker were to die and rise again on the third day, ascending into heaven a few weeks later. The “laws” of supernature that Christians or other theists are inclined to postulate may not be as detailed as the laws scientists hope to discover in nature, but they leave theistic claims open to confirmation and disconfirmation.

[18] Keith Parsons, “Is there a Case for Christian Theism?” in J. P. Moreland and Kai Nielsen, Does God Exist: The Great Debate (Nashville: Thomas Nelson, 1990) p. 189.

I basically agree with what I wrote twenty one years ago, but I think I should clarify and expand some things. I think that hypotheses postulating powerful supernatural entities are in principle testable. A template for such a test is found in I Kings Chapter 18 where you have the famous experimentum crucis between Elijah and the priests of Baal. You recall: Elijah makes a sacrifice for Yahweh and the priests make one for Baal. The priests call all day for Baal to send fire and burn the sacrifice, to no avail. Then Elijah steps up and calls upon The Lord, and fire falls from heaven to consume the sacrifice. Elijah then leads the people in a massacre of the priests of Baal (hey, every good OT story has to have at least one massacre).

Now, of course there would be many intractable practical problems with setting up such a test. For instance, you would probably need to bring in James Randi to make sure that no trickery was involved. However, I see no reason why, in principle, such a test could not be done nor why a supernatural hypothesis could not be confirmed in this manner. The “inevitability” I mention in the above quote I would now take as referring not to the in-principle impossibility of such a test, but to the predictable refusal of religious apologists to submit their hypotheses to such testing. We could have a test like in I Kings now live on TV and broadcast worldwide. Fire falling from the sky would pretty much shut up pesky atheists like me (if, again, I could be sure that it was the real deal and not that somebody had dropped napalm). Why, then, instead of resorting to recondite philosophical arguments, do religious apologists not avail themselves of the obviously superior choice of performing a public experiment? Nullius in verba! Do the experiment! Give atheist smartasses their comeuppance! (No massacres, OK?)

Well, of course, no modern day apologist, however brash and confident his claims, will conduct such an experiment. He will probably decline mumbling something about how you should not put The Lord to the test. No such scruples bothered Elijah, and I hope that it is not too cynical to assert that were they confident of the sorts of results Elijah got, today’s theists could set aside their qualms also. Why don’t they have such confidence? This gets back to what I said about “laws of supernature.” The biggest practical problem facing a latter-day, would-be conductor of an Elijah-type demonstration is that there do not seem to be any detectable regularities governing miracle events that would provide a basis for predicting experimental results. God will send fire from heaven when and only when he damn well pleases, and when he will please seems completely unpredictable. No detectable laws appear to govern the performance of miracles. Technically, we know of no true nomological conditionals of the form “If a condition of type C were to obtain, then God would perform a token of miracle-type M.” Of course, God could tell us when to expect a miracle, but that does not seem to happen either.

Now Victor thinks that there might nevertheless be knowable (non-deterministic) laws of supernature that permit the confirmation or disconfirmation of theism. Unfortunately, the example he gives is completely unhelpful. I cannot at all see how Jim Bakker rising from the dead and ascending into heaven would disconfirm Christianity. Rather than quibble, though, I can just say “Ok, then, put up or shut up!” Less truculently: If there are laws of supernature that could support something like a scientific experiment, then name the laws and provide at least a sketch of the experimental design. Let’s do it! Enough of the endless logic chopping; let’s do the experiment.

Of course, as he indicates, Victor will reply that he did not mean that there were discoverable laws of supernature of sufficient detail and specificity to provide the basis for an actual experiment. We cannot say when and where God will perform his next miracle (e.g., the Houston Astros winning next year’s World Series). We can only make much vaguer predictions, maybe like this: Any world that God creates will have sentient creatures to be in communion with God. Obviously, there is no way to test such a “prediction” scientifically, but it could be (and is) used as the basis of a philosophical argument for theism. Philosophers can argue out whether the existence of sentient creatures in this world confirms the existence of God.

However, such a reply concedes my original point in the passage quoted, which concerned science, not philosophy. My argument was that trying to support theism with science is dicey, in part, because theistic hypotheses just do not seem amenable (in practice if not in principle) to confirmation or disconfirmation with the sorts of experiments and tests regularly employed in the natural sciences. As far as I can tell, Victor here has said nothing to challenge that claim.

About Keith Parsons
  • Victor Reppert

    Keith: Is the following claim untestable?

    Reducing the budget deficit only by cutting domestic programs will harm the economy and undercut the economic health of the middle class.

    Now it looks to me as if

    a) the claim is testable.
    b) there is all sorts of relevant evidence
    c) you aren't going to be able to perform some experimentum crucis that will settle the question once and for all.

  • Keith Parsons


    I think we may be talking past each other. All sorts of issues are arguable, that is, evidence may be adduced for one side or the other without the question being testable in anything like a rigorous scientific sense. Questions of, for instance, law, economics, politics, religion, philosophy, history, art, or literature may be discussed rationally in the light of evidence. I think the existence of God is one such question that admits of rational discussion but eludes rigorous testing. In the physical sciences (historical sciences are somewhat different) the gold standard of evidence is the experimental test. What often earns a field the honorific label "science" is precisely that its claims are amenable to rigorous testing with precise predictions, narrow margins of error, and exact measurements. The sorts of much looser plausibility considerations that have to do in other fields are not enough for the physical sciences. For instance, solid state physics has a higher standard of evidence than, say, Roman history. I take this as obvious. The passage you quoted from my essay from 1990 was simply making the also obvious point that God's existence cannot be confirmed (or disconfirmed) by rigorous experiment. Rather, discussions about God's existence must employ the less epistemically compelling sorts of arguments and evidence more typical of debates in history or the humanities. The upshot is that zealots like Moreland, who seem to think that the evidence for theism is coercive (or nearly so)are engaging in wishful thinking. Yes, you may rationally argue for or against the existence of God. I think these debates should take the form of inferences to the best explanation, and all parties should admit ahead of time that, in the end, we may have to agree to disagree. No, you cannot establish God's existence in the same way or with the same degree of evidence that we can establish the existence of protons (and antirealists like Van Fraassen would say that there is room for rational doubt even with all unobservables).

  • Alex Dalton

    Very interesting exchange, folks. I really like these discussions of the nature of theistic explanation.

  • Keith Parsons


    Long time, no hear from you! Welcome back!

  • Havok

    I found Theism and Explanation By Greg Dawes to be very informative about the form and content of what might make up a successful explanation of this sort

  • Keith Parsons


    The Dawes book is quite good. I plan to review it in this space in the not-too-distant future.

  • Havok

    Keith Parsons: The Dawes book is quite good. I plan to review it in this space in the not-too-distant future.
    That sounds great Keith – look forward to read it.
    Also, if you know of any other reviews of the work, I'd be interested in reading them (especially from those who otherwise argue for theistic/supernatural explanations) :-)

  • Richard Wein


    I note that your original quoted paragraph made a claim about the supernatural broadly, but here you've narrowed this to just the question of God's existence. Perhaps that was indeed the context in which you wrote the original paragraph, but nevertheless you made the broader claim, and that was the one Victor originally addressed. (It was also suggested by the title of your OP.) I think for the sake of clarity it would help if you acknowledged you are now limiting yourself to defending the narrow claim.

    "In the physical sciences (historical sciences are somewhat different) the gold standard of evidence is the experimental test."

    I don't want to argue that the existence of God is a matter for "science", but I do want to argue that "science" is a broad and fuzzy concept, and it's a mistake to impose hard demarcation criteria on it. I would reject your demarcation criterion, even for the physical sciences. It's quite possible for a scientist to make a new inference based on pre-existing theory and facts, and I don't think we have to wait until it's been tested before we can call it a scientific inference, at least a tentative one.

    I think that those who assert that the existence of God is a matter of science are making a certain useful point: a good judgement about whether God exists will be one that's informed by scientific knowledge and by scientific inference in the broadest sense. (The progress of science has made God much harder to believe in.) Nevertheless, I think to call the inference to the non-existence of God "scientific" is going too far. This is not based on any one demarcation criterion, but on the judgement that, all things considered, the factors which lead us to call something "science" (as opposed to some other area of empirical enquiry) are not present in sufficient degree in this case.

    I think that far too much time is spent arguing over whether a question or inference is a matter for "science" or some other field of empirical enquiry. The judgement is of little practical value. Let experts in a field address any questions to which they think their interests and skills are matched. The perceived importance of the judgement arises primarily from the positive connotations of the word "science", which some people are keen to bestow, and others to deny.

  • Keith Parsons


    You raise a number of interesting and insightful points, and these are rather complex issues, so let me see if I can be as clear as possible on my views. What I say about the problems with testing theistic hypotheses applies to supernatural hypotheses in general–equally if not a fortiori. Again, it is not a matter of in-principle untestablitily. We can imagine ideal tests that would establish to any rational person's satisfaction the existence of ghosts, ESP or other paranormal powers, miraculous healings, the efficacy of prayer, etc. The problem is that attempts to verify such claims by rigorous tests always meet with failure or frustration. Two typical outcomes are these: (1) The "results" of tests of claims in parapsychology are so marginal that they cannot meet reasonable burdens of proof. (2)The proponents of such hypotheses succeed in surrounding them with ad hoc insulators making tests a practical impossibility. E.g., when skeptical magicians were present, Israeli "psychic" Uri Geller would claim that their skepticism created "negative energy" that interfered with his miraculous telekinetic powers. These are just two instances of a bottomless bag of tricks and excuses whereby proponents of supernatural hypotheses shield their treasured claims from rigorous testing.

    With theistic hypotheses we are generally told (from Deuteronomy, I think) "Thou shalt not put the Lord to the test." In short, we cannot test because that would be sacrilege! Neat trick, huh? Reminds me of a character I saw in a bad movie one time who claimed the power to make himself invisible, but only if no one was watching!

    BTW, I was not offering testability, or anything else as a demarcation criterion. I do not think that any such criterion exists. The difference between those fields we denominate with the honorific label "science" and other areas of inquiry is chiefly the degree of stringency and rigor of the sorts of evidence we can adduce within those fields. In the "hard" sciences evidence constrains hypotheses much more severely than in, say, cultural anthropology. Informally, hypotheses in "softer" areas are generally left with far more "wiggle room," i.e., interpretive latitude vis-a-vis the evidence. I take this observation to be an epistemic commonplace. This does not mean that those "softer" fields are illegitimate or even that they do not deserve the label "science." When the issue is a metaphysical claim like the existence of God, the bonds of evidence are much looser still, and interpretive latitude is even greater. This is why I now feel great ennui with arguments in the philosophy of religion.

  • Richard Wein


    Thanks for your reply, and for clarifying that you are still making a claim about the supernatural generally. However, it remains unclear just what you are claiming. Perhaps you could help by stating your central claim explicitly.

    I think the critical issue is whether you're making a claim about what could be expected if supernatural phenomena actually existed, or just making a claim about what can be expected given the way things are. The actual failure of supernatural claims to be supported by rigorous evidence could be explained by the fact that these claims are all untrue. It doesn't tell us whether a _true_ supernatural claim could be rigorously supported by evidence. To draw that sort of epistemic conclusion we need to consider what characteristics make a phenomenon "supernatural" and what are the epistemic consequences of those characteristics.

    Your original paragraph seemed to be making the first (stronger) sort of claim. It _did_ attempt to argue from alleged characteristics of supernatural phenomena, e.g. "Supernatural beings…are essentially mysterious." Your latest comment, on the other hand, only argues from the lack of success of supernatural claims to date (and from psychological facts about the minds of believers). It only seems to be supporting the second (weaker) sort of claim.

  • Keith Parsons


    You are quite right that in the original paragraph I made a stronger claim than I would make now. When that paragraph was written over twenty years ago I thought that supernatural claims were in-principle untestable. I no longer think that, but I do think that the history of attempts to put supernatural claims to rigorous testing (as long-time readers of The Skeptical Inquirer are aware), always meet with grief. When predictions ARE made, they turn out false or insignificant. Other hypotheses are hedged about with ad hoc insulators that stymie testing. Indeed, proponents of supernatural hypotheses have many excuses for not conducting rigorous tests of pet claims.

    With the God hypothesis, I certainly can imagine demonstrations that would put me on the front pew of the church or synagogue of my choice. Reading Norwood Russell Hanson's essay "What I do not Believe" convinced me that there could be spectacular, Spielbergian demonstrations that would convince any rational person of the existence of the Biblical God. Yet such decisive demonstrations are never offered. Why? Don't ask me; ask the defenders of theistic hypotheses. Personally, I am all for it. I would love to see an attempted experimental demonstration of the existence of God (and, naturally, I am pretty confident how it would turn out).

    My central claim in a nutshell is this: Supernatural hypotheses are generally untestable, not in principle, but because their proponents will not allow them to be tested. For instance, defenders of theistic hypotheses hold that God COULD establish his existence beyond reasonable doubt at any time by performing spectacular miracles. However, defenders of the God hypothesis add an auxiliary hypothesis to the effect that God will not permit his existence to be demonstrated because it is essential that he maintain a degree of "hiddnness." The effect of this auxiliary hypothesis (whether we regard it as ad hoc or not) is to stymie experimental demonstrations.

    So, again, when Victor tells me that claims about God are testable, then, if he means anything like the rigorous tests in the physical sciences, my reply is: "Great! What kind of test did you have in mind?" I then inevitably find that the notion of "testing" has been watered down to mean something like "rationally discussed."

  • Richard Wein


    Thanks for the reply. I'm afraid it seems to me that you still haven't clarified your claim, and that you're still unwittingly equivocating between weaker and stronger claims.

    You wrote: "My central claim in a nutshell is this: Supernatural hypotheses are generally untestable, not in principle, but because their proponents will not allow them to be tested."

    Is your "because" an indication of causal consequence or semantic consequence (a consequence by virtue of the meaning)?

    If you intend it as a causal consequence, then you're making a claim of how it comes about that "supernatural hypotheses are generally untestable", but you still haven't explained what you mean by "supernatural hypotheses are generally untestable".

    If you intend it as a semantic consequence, then I say that you are not using words in a reasonable sense. To say that "supernatural hypotheses are generally untestable" is–by any reasonable usage–to say something stronger than "the proponents of supernatural hypotheses will not allow them to be tested".

    Either way, you haven't addressed what I called "the critical issue" in my last comment.

  • Richard Wein

    P.S. Let me put it a bit differently. Whether the consequence was supposed to be a causal one or a semantic one, the conclusion ("supernatural hypotheses are generally untestable") does not appear to follow from the premise ("the proponents of supernatural hypotheses will not allow them to be tested"). If you think you can find a reasonable interpretation of "untestable" which makes the consequence work, then I think the onus is on you to give it. It's not sufficient merely to make the negative stipulation that you're not talking about "in-principle" untestability.

    BTW I'm saying "consequence" rather than "entailment" because I accept we're not talking about strict logical entailment. Perhaps philosophers have some other word to use in such cases.

  • Keith Parsons


    Thanks for your patience! I AM trying to clarify and not equivocate! The "because" is meant in a causal sense, and I do not think there is any semantic malfeasance. "Untestable" IS ambiguous. It can mean "untestable in principle" or "untestable in practice." Generally, when we say that something is so or not so "in principle" we are referring to matters of logical possibility. Hence, if I were to say that supernatural theories were in-principle untestable, I would be asserting that it is impossible in the strongest sense,i.e., logically impossible, that they be tested. In-practice untestability, therefore, would be any situation where testing is logically possible, but is contingently rendered impracticable by any of many possible circumstances. In-practice untestability is hard to characterize in general terms because there are so many different kinds of impediments that can interfere with robust testing. In astrophysics, for instance, hypotheses are often proposed that, though perfectly testable in principle, cannot yet be tested because instruments of sufficient accuracy have not yet been deployed. On other occasions, as I indicated, a hypothesis that is in-principle testable may be conjoined with an auxiliary hypothesis that has the effect of obviating stringent testing.

    Let me emphasize again that the charge of untestability is not my STIPULATION, it is my CONCLUSION based on many years of observation of the behavior of apologists for supernatural hypotheses, from gods to ghosts. A proponent of such a hypothesis has a very simple and effective way of disputing my conclusion: Call my bluff. Propose the test. If my bluff is persistently not called, but, on the contrary, the hypotheses at issue are surrounded by more and more test-killing qualifications, what am I to conclude? Is it then unreasonable to conclude that, as a practical matter, the hypotheses appear untestable? How is articulating this conclusion a violation of reasonable usage? Indeed, when such experience is repeated many times, what is the problem with concluding, as a matter of inductive inference, that in general and practically speaking such hypotheses appear untestable?

    Concerning the "critical issue" you mention, do I think that a TRUE supernatural hypothesis could be experimentally demonstrated? Yes, I do. Do I think that there is a snowball's chance in hell that we can ever come up with an actual such test? No, I don't, and centuries of experience with such hypotheses provide strong inductive support for that inference.

  • Keith Parsons


    BTW, if the word "untestable" is the stumbling block, I would settle for asserting the following mild and, I hope, unobjectionable statement:

    "In general it does not appear practicable to evaluate supernatural hypotheses, including theistic hypotheses, using the sorts of stringent experimental tests typically employed to confirm hypotheses in the physical sciences. Typically with such supernatural hypotheses,including theistic hypotheses, it is impossible to draw clear, precise, and unambiguous predictions that would permit strict testing. Even when such hypotheses are in-principle testable (as many are), they are generally conjoined with auxiliary hypotheses that have the effect of preventing rigorous testing. The upshot is that supernatural hypotheses, including theistic hypotheses, generally must be evaluated with somewhat looser and less stringent criteria than those serving to confirm hypotheses in the physical sciences."

    Really, these quite weak and unremarkable observations are all that I am saying. I really do not see that it would be much of a bone of contention between me and someone like Victor. We both agree that theistic hypotheses may be rationally discussed and evaluated vis-a-vis the evidence, and, as I indicated in my response to him, I think that our apparent disagreement over whether such hypotheses should be called "testable" is really more verbal than substantial.

  • Richard Wein


    Thanks for your clarification, and for _your_ patience. I was afraid my last couple of posts may have been rather pedantic or unclear.

    I think there's still some lack of clarity in your claim. But the essential point seems reasonably clear now. You're not talking about something that's in the nature of supernatural hypotheses themselves. You're saying something about the typical behaviour of the proponents of supernatural hypotheses, e.g. they typically save their hypotheses from falsification by adding or changing auxiliary hypotheses. My concern was that your wording makes it seem that you're doing the former as well as the latter.

    Your use of the terms "in principle" and "in practice" hasn't helped matters. The claim that supernatural hypotheses can be tested in principle but not in practice can be interpreted in at least two ways (probably more):

    1. It's in the nature of supernatural phenomena that, while supernatural hypotheses are testable in principle, such tests are impossible to conduct in practice.

    2. The typical behaviour of proponents of supernatural hypotheses is to vitiate any attempt at testing by adding/changing auxiliary hypotheses.

    I would say that #1 is the more natural interpretation. It's also the one suggested by your recent attempt to explicate "in principle" in terms of logical possibility. But you seem actually to mean something like #2. (You may have in mind more types of behaviour than just the one I've mentioned.)

    As you say, this interpretation makes your claim quite weak and unremarkable, and unlikely to be a bone of contention with Victor. But your OP seemed to be saying something significantly stronger, something which you clearly did consider to be a bone of contention with Victor.

    I think your OP might have been a lot clearer if you'd omitted your original paragraph, and described your current position from scratch. But it seems as if you wanted to defend your original paragraph from Victor's criticism while actually arguing something very different.

  • Keith Parsons


    No patience required on my part. Your remarks have been VERY helpful in getting my thoughts focused. As to the choice between your #1 interpretation and #2 interpretation, I think it is actually some of both–hence my appearance of hedging (A problem here is the medium. Expressing a complex and nuanced position in a few short paragraphs is impossible). Supernatural hypotheses often posit conscious, indeed capricious, entities who are not bound by impersonal laws (no "laws of supernature" as I say) so, their behavior is harder to predict than natural entities. How do you test for God or ghosts if they do not want to be tested and simply refuse to cooperate? So, yes, supernatural hypotheses are, in practice, often harder to test because the tests require the cooperation of the posited entities, and their cooperation cannot be forced.

    However, "harder" does not mean "impossible," and sometimes tests can be proposed because cooperation can be presumed. An instance is the one I mentioned where a "psychic" like Uri Geller claims to have telekinetic powers. Skeptical magicians such as James Randi challenged him to demonstrate those powers in their presence. Geller refused claiming that the skeptics created "bad vibes" that interfered with his paranormal powers, thus rendering the claim untestable.

    Similarly with God we can assume that there will be times and places where he will want to demonstrate his existence and powers to an unbelieving world. Indeed, scripture records many such instances when God made his powers abundantly clear, often when called upon by a holy prophet like Elijah. Why then do we not have such unambiguous demonstrations of divine power in our day? Surely, we are just as needful of divine guidance now as then. It is when such questions are asked that the excuses start rolling in. E.g., if God were to demonstrate his existence to all, there would be no need for faith, or people would be good merely out of prudence rather than for moral reasons. And so forth. Thus, as I say, there are MANY factors that make supernatural hypotheses hard to test, but I think the MAIN reason is simply that the proponents of those hypotheses do not want them tested.

  • Richard Wein


    Geller's case is a good example. I think there are two separable issues there:

    G1. Geller's participation is (arguably) essential for any test of his powers, in which case it is impossible to test his hypothesis given his refusal to participate. I think it's reasonable, therefore, to say that his refusal to participate makes the hypothesis untestable in practice. It's not specifically the supernatural aspect of the hypothesis that makes it susceptible to this problem. It's the fact that the hypothesis relates to the abilities of a specific person. It would also be susceptible to his refusal to participate if the hypothesis was that Geller could achieve the same effects through natural prestidigitation. That said, supernatural hypotheses much more often relate to the abilities of uncooperative people than do the hypotheses typical of science, and more readily provide excuses for refusal (like "bad vibes"). In a broad sense I would agree that they tend to be "harder to test".

    G2. Geller has added an auxiliary hypothesis: skeptics create "bad vibes" that interfere with his powers. The involvement of a skeptic is (arguably) essential for any test of his powers, in which case it is impossible to carry out a test of the original hypothesis in conjunction with the auxiliary hypothesis. But I'm unwilling to say that the addition of the auxiliary hypothesis makes the original hypothesis untestable in practice (or harder to test). We can still test the original hypothesis to the satisfaction of reasonable obervers, even though Geller and his supporters will use his auxiliary hypothesis to claim that the test is useless. (Of course, we still need Geller's participation, and it would be very odd for him to participate after announcing his auxiliary hypothesis. But it's not impossible, and we can separate the two issues.) When speaking casually it seems like a reasonable shorthand to say that the addition of the auxiliary hypothesis "makes the original hypothesis untestable". But here I want to be as clear as possible, and I find that way of speaking to be misleading.

    Geller's refusal to participate is a physical constraint on our testing. His auxiliary hypothesis is merely an assertion that we don't have to take seriously in the context of testing the original hypothesis.

    I tried to draw the distinction between issues of type #1 and issues of type #2 by associating the latter with the "behaviour of proponents". The example of Geller shows that this was a poor way of drawing the line, since both G1 and G2 involve Geller's behaviour.

    Anyway, I won't go any further in trying to distinguish between different types of problems that arise in connection with supernatural hypotheses. You only seem to be claiming that some such problems are typically the case when supernatural hypotheses are actually put forward. That's still a weak and uncontroversial claim, at least among skeptics.


  • Richard Wein


    There's a reason I've included the qualifier "when supernatural hypotheses are actually put forward". If we simply generalise about "supernatural hypotheses" it can be unclear whether we're generalising about the set of supernatural hypotheses that are actually put forward, or the set of all possible supernatural hypotheses. When you claimed that supernatural hypotheses are generally untestable "in practice", your addition of "in practice" didn't prevent this potential ambiguity, because the term "in practice" appeared to be qualifying "untestable" and was not necessarily qualifying "supernatural hypotheses".

    One further point I'd like to add is that we've been concentrating here on tests for the presence of claimed supernatural effects. Some different issues come into play when we talk about testing proposed supernatural explanations of these effects, or about whether it's appropriate to apply the word "supernatural" to any effect or explanation that we may discover. I mention this only for the sake of clarity. I don't wish to discuss those issues at present.

  • Richard Wein

    P.S. I wrote: "(Of course, we still need Geller's participation, and it would be very odd for him to participate after announcing his auxiliary hypothesis. But it's not impossible, and we can separate the two issues.)"

    There are other cases where the claimant doesn't announce an auxiliary hypothesis (or in common parlance an "excuse") until after he or she has participated in a test with negative results. Randi has reported such cases involving the preliminary testing of applicants for his £1,000,000 challenge. Needless to say, Randi doesn't dismiss the results of the test in response to the assertion of an auxiliary hypothesis. The test remains a valid test of the original claim.

  • Keith Parsons


    Thanks again for the extensive and perceptive comments. I'll be making a lengthy post next week to set out my views more clearly and address some of the points you make.