Has The Supernatural Hypothesis Failed?

Hi, Straggler.I'm a little uncomfortable with the idea of a Supernatural Hypothesis, for two reasons:

1. How can it be a hypothesis?
   ​
2. Can all supernatural concepts be combined under a single, unified hypothesis?

I’m a little leery of the notion that any given set of supernatural ideas is an appropriate sample for testing the merits of all supernatural ideas in general, especially given the wide disparity in the concepts of different supernatural ideas. I think, at the very least, you need to break it up into a few distinct hypotheses: maybe supernatural phenomena, and supernatural beings, for starters.On another note, how do you feel about natural processes that are entirely stochastic (assuming such processes exist)? We can’t really offer much in the way of a naturalistic mechanism for such things, so calling them natural is a little off. But, calling them supernatural is probably also not warranted.

---

-Bluejay (a.k.a. Mantis, Thylacosmilus)Darwin loves you.

Hi, Straggler. This is not true.I have only invalidated hypotheses that explicitly include the claim that untestable alternatives are categorically false.Real scientific theories do not include this claim. Bluegenes’ "theory" does.-----I will explain this in detail, using a different approach from one I've used before. This is just for thoroughness.Any data set of notable size will include some data points that do not fit well with the rest of the data: outliers. Ecological data sets, such as I am accustomed to working with, often contain very many outliers.When creating scientific theories, outliers are ruled out as aberrant data points, and no attempt is made to explain them scientifically.This is perfectly normal and statistically defensible, because if only a small percentage of data points defy the model, the model is still a good and useful approximation of the data set as a whole.Scientific theories do not claim to explain outlying data points (in fact, the tests used clearly conclude that the theory cannot explain these points: that’s what makes them outliers). We generally just chalk these up to stochastic factors, and leave it at that.If these outliers are due to supernatural causes, it doesn’t matter, for two reasons: (1) because the model still explains most of the data, which means we can be confident that the model is pretty damn good; and (2) because no scientific theory includes the clause that supernatural factors cannot explain the few outliers in the data set.That is, of course, except for Bluegenes’ theory. Bluegenes’ theory is explicitly a statement that there will be no outliers to the model. It claims that something untestable is never correct.This problem is compounded by the fact that any data supporting the alternative view cannot be confidently determined as such, which means that they cannot be included from the data set. The aggregate result of this is that the only data points that can be included in analyses are data points that support the theory that is purportedly being testing. This obviously means that the test can only return an answer that is consistent with the theory it is supposed to by testing.So, in summary, we can't have confidence in any theory that claims that something untestable does not exist.But, we can have confidence in mainstream scientific theories, because they don’t claim that untestable things do not exist.

---

-Bluejay (a.k.a. Mantis, Thylacosmilus)Darwin loves you.

Hi, Straggler. Nonsense. I’m not talking about the tentativity of his theory: I’m talking about the claimed scope of his theory. A universal claim---even a tentative universal claim---is not a scientific theory.If, by, All gods* are made up by humans, what Bluegenes really means is just, Humans make up gods, then I suppose you’d be right that his theory is logically identical to all other scientific theories, and that we can have confidence in it. Clearly, though, in this non-universal form, his theory is entirely different from it’s universal form. *I’m going to use the word god from here on, instead of supernatural being, because it’s easier and less awkward to write. Please note that I am not trying to change the substance of the theory by altering the wording.This is not the case for real scientific theories.Look at natural selection. Any data set collected for natural selection has random noise, outliers and/or aberrant data points. But, the fact that natural selection can run into data that it could theoretically explain, but fails to explain in reality, does not utterly destroy the theory of natural selection. Natural selection coexists perfectly fine with genetic drift and sexual selection, and would still coexist perfectly fine with, But God put a rabbit fossil in the Cambrian.Look at the Biogenetic Law. All life from an egg is obviously not entirely true, but Biogenesis still coexists just fine with Abiogenesis, and would also coexist just fine with human-designed organisms or with, But God created Opabinia.Look at relativity. We’ve run into a number of data points that have unequivocally demonstrated themselves to be unexplainable by the current model, but relativity still coexists just fine with these singularities, and would also coexist just fine with, And God is responsible for these singularities.Now, look at Bluegenes’ theory. If there were one aberrant data point, one outlier, then the theory could not be recovered in any recognizable form. It will have completely lost its power. It cannot coexist just fine with, But Zeus is real.This is the difference between Blugenes’ theory and real scientific theories. It is wholly dependent on its own universality. Proposing universality, even proposing it tentatively, is not what scientific theories do, and is not what is required of scientific theories to do. Scientific theories are not about accounting for all the data: they are about identifying real mechanisms and processes. But, Bluegenes’ theory is explicitly about accounting for all the data.If you can’t see this, then I don’t know what else I can say to you.Edited by Bluejay, : "and/or" instead of "and"Edited by Bluejay, : added "though," in the second paragraph

---

-Bluejay (a.k.a. Mantis, Thylacosmilus)Darwin loves you.

Hi, Straggler. But, the theory of natural selection does not say this.And, in fact, evolutionary biology does not conclude this, either. Not all of the data is explained by natural selection, and there is no theory that claims that all the data is explained by natural selection.The actual claims of the theory of natural selection are not parallel to the universal claims of Bluegenes’ imaginary-gods theory or your light bulb theory.Why is it that you can’t grasp this?----- Do you understand what the word confidence means in relation to hypotheses and statistics?
You are using it in a much more general and non-rigorous way than it is actually used in relation to scientific hypotheses and theories.The confidence you would have in the theory that supernatural things are invented by humans is not the same thing as the confidence you would have that that theory can account for all the data.I’m going to call them -confidence and -confidence. -confidence:Your confidence in the human invention theory is based on the statistical methodology called hypothesis testing: evidence is gathered, and the probability that patterns in the evidence actually represent genuine deviations from a null model is calculated. -confidence can be quantified and expressed as a percentage (95% is the generally-accepted cut-off point).Hypotheses that pass this kind of testing can eventually grow up to be theories. -confidence:Your confidence in the full accountability claim is not based on hypothesis testing (which also means the claim is not a theory): it is based on a zero-one rule. This means the claim either meets the criteria for -confidence, or it doesn’t: it cannot be quantified. Either an alternative explanation has be found for some of the data, or it hasn’t.This means that -confidence is entirely independent of the strength of the evidence for the one existing theory, because it isn’t based on that evidence, but on the lack of evidence for something else. So, a theory based on 8 data points can be assigned just as much -confidence as a theory with 17,843 data points, so long as no alternatives explain any of the data.This means that -confidence in a brand-new theory, for which no alternatives have yet been conceived, is exactly identical to -confidence in an old theory for which several alternatives have been proposed, but discredited.So, contrary to your claims, adding evidence to the human invention theory does not improve confidence in the full accountability of the human invention theory. Thus, the confidence you can reasonable have in this claim is not the same as the confidence required for a scientific theory.So, I still reject your claim that you can have confidence in the ability of a theory to account for all the data. Any confidence that you can have is not quantifiable and has nothing to do with scientific theories.----- I have provided at least three different explanations for this now. Your response so far has been to pretty much ignore them all and explain for me the same process of reasoning that you posted in your very first post to RAZD about the subject of the supernatural.The reasoning you are using is not quantifiable, not rigorous, and not scientific, so I still reject your contention that it counts as anything more than a practical heuristic.

---

-Bluejay (a.k.a. Mantis, Thylacosmilus)Darwin loves you.

Hi, Straggler.You have an unusual gift for making your debate opponents feel like responding to you is of the utmost urgency. Regrettably, however, I’m going to have to back down after this post: I’ve only got two weeks left until my candidacy exams, and this topic is cutting into my ability to focus on my study.If you still want to talk about this in three weeks (which, obviously, you will, because you’re Straggler), feel free to invite me back in. But, for now, I need to not be involved in any of these deeper, time-consuming debates.So, here is my final response before respite:----- This is subtly inaccurate. I am under the apprehension that theories that explicitly state an ability to comment on the supernatural are not defensible logically or empirically.The theory is invalid because it explicitly cites the demonstration of the inherently indemonstrable as the only way to falsify it.----- I need to respond to two different parts of the above quote. Displaying primitive theistic tendencies.I’m quite frustrated by this, because I am still of the opinion that purportedly supernatural beings cannot actually be demonstrated to be supernatural, and I seem to be the only participant in this debate who regards this as a particularly relevant point.Your examples seem to completely neglect the investigatory methods required to verify the supernatural quality of something. You and Bluegenes seem to both assume that witnessing something like a mermaid or a chimpanzee who can walk on water would demonstrate the existence of the supernatural. But, surely you can recognize this as a God of the gaps argument? The only way to verify that some given phenomenon is supernatural is to demonstrate that the gap into which you want to insert a god is infinitely large: otherwise, it’s just another singularity in the theory that may or may not be explained by some future development.If God of the gaps cannot be used to support the existence of deity, then it also cannot be used to falsify a naturalistic hypothesis, which means that Bluegenes’ hypothesis needs a non-supernatural falsification criterion, which means that his theory must surrender its claim to commentary on the supernatural in order to be retained as a valid, testable theory. All species on Earth are the result of... natural selection...There are two main problems I have with this.First, the scope can be altered at whim to make it true even in cases when it isn’t true. If one species became two by means of isolation and genetic drift, natural selection need not be invoked to explain this data. But, you can slide the rule and say that natural selection was surely involved at some point in the species’ evolutionary history, and thereby claim that the theory is still true. So, unless you are making an already-falsified claim (such as, all speciation events are the direct result of natural selection"), I don’t think this theory is actually saying anything substantive.Second, it can be falsified by an observation with a naturalistic explanation. Naturalistic explanations can be verified empirically and statistically. On the other hand, supernaturalistic explanations cannot be verified empirically or statistically. So, once again, because Bluegenes requires an observation with a supernaturalistic explanation to falsify his theory, while natural selection does not, these examples simply are not logically analogous to one another.----- I don’t recall having claimed that the Big Bang hypothesis has rightfully achieved the status of high confidence, and I lack the qualifications to satisfactorily answer your questions about it (I don’t even know what Cosmic Background Radiation is), so I feel no shame in declining this challenge.-----See you on the other side, Straggler.Edited by Bluejay, : added the farewell.

---

-Bluejay (a.k.a. Mantis, Thylacosmilus)Darwin loves you.

Hi, Straggler.I'm back. I think my exams went well: I suppose I'll find out when I get them back.I have not read anything that has been written on this thread since I left a few weeks back, so I don’t know where the debate has gone in my absence.I really only have one thing to say here: This really doesn’t make as much sense as you think it does.This statement is simply not an explicit or implicit part of any scientific theory. Rather, it’s pretty much the most basic heuristic on which all of scientific reasoning is based: i.e., that the natural world can be comprehended and explained by deterministic rules.You can’t test a heuristic. You can’t have confidence in a heuristic. The ability to extrapolate a theory through heuristic reasoning does not count as confidence in the heuristic. A heuristic is just a rule that is followed because the alternative approach to reasoning (in this case, pooling all the possible explanations and testing them all together) is unmanageable.I honestly can’t think of any better way to say it. And, I honestly don’t think I’ll ever understand why you think I’m wrong about this.

---

-Bluejay (a.k.a. Mantis, Thylacosmilus)Darwin loves you.

Hi, Straggler. How can this possibly be the point, Straggler?We weren’t talking about people who irrationally claim supernatural causes for things: we were talking about people who irrationally claim that naturalistic methodologies are set up to test a theory that can only be falsified by alternatives that cannot be demonstrated naturalistically.----- I have already done this, Straggler. I have done it multiple times, from multiple different angles. I have explained it from the standpoint of statistics and from the standpoint of scientific philosophy.I can’t figure out why you don’t find these arguments convincing, because all you ever do is say random crap about gravity and evolution, and ask me to explain (again) why I think these are different from Bluegenes’ theory.This will be my final attempt to explain why I think these theories are different from Bluegenes’ theory, and I won’t continue this discussion if I don’t get some sort of indication that you have actually assimilated them.Review: Statistical Point of View
Confidence is a statistical term that describes the suitability of a theory for explaining a given data set. It results from showing that one naturalistic hypothesis is better at explaining a given data set than another naturalistic hypothesis. It cannot be formatted to provide commentary on supernatural hypotheses.So, for example, confidence in ToE arises from showing that ToE is better than any naturalistic alternative, not from knowing that ToE is better than any supernaturalistic alternative. So, even if we can’t show evidence that Jesus’s magical hand is not guiding evolution, we can still have confidence in ToE, because confidence in ToE has nothing to do with what Jesus’ magical hand may or may not be doing, or even with whether or not Jesus’ magical hand even exists.In contrast, confidence in Bluegenes’ theory can only arise from showing that it is better than a supernaturalistic alternative. This confidence has everything to do with showing what Jesus’ magical hand may or may not be doing (or something conceptually similar). This is not a mode of analysis that statistics or logic can provide tests for. Thus, there can be no statistical confidence in Bluegenes’ theory, and Bluegenes’ theory cannot actually be a theory.This is why your continual allusions to ToE, to gravitation and to electrical resistance are not applicable to this scenario. So, before you rattle them all off again, please address the actual point that I'm making, rather than repeat the stupid point you've been repeating since we (actually since you and RAZD) started this discussion. Philosophical Point of View
Scientific reasoning begins with the assumption that data is explainable by deterministic, naturalistic hypotheses. But, there is no reason to believe that any genuine supernatural cause would have to leave a mark that naturalistic reasoning would pick up on, so there is no reason to think that naturalistic reasoning is going to get us to a position from which we can comment on supernaturalism. Scientific reasoning---including all the theories that you keep tossing at me---is therefore fundamentally based on the heuristic rejection of supernatural explanations.Heuristics are rules of reasoning that are followed when practicality is more important than accuracy. We rule out supernatural, not because we can actually reasonably comment on it and have shown it to be false, but because we can’t actually reasonably comment on it.Naturalism has not shown itself to be more accurate than supernaturalism: it has only shown itself to be more useful in organizing information than supernaturalism. I don’t see any reason to suspect that supernatural processes could even be modeled with a logical formula, nor why they would even have to be exclusive of natural processes.-----I await your explanations for why these two points are false. If your reply is just another assertion that my reasoning says we can't have confidence in ToE or electrical resistance, don't expect any further participation from me. I'm confident that the majority of readers will be able to see the logic in what I'm saying, even if I don't get in the last word.

---

-Bluejay (a.k.a. Mantis, Thylacosmilus)Darwin loves you.

Hi, Onifre.I was using "heuristic" as you say: rules of thumb. Well, the assumption behind all scientific reasoning is that, if we have sufficient information about the factors that influence a certain process, we can reliably predict the outcome.If our predictions aren't reliable, we generally conclude that our information about the influencing factors was insufficient, and must be refined.I admit that I don't know a lot about the history of sciences other than biology, so I’ll introduce the caveat that I didn’t mean my statement to refer to the thought processes of all scientists throughout history. But, I think the examples you listed can be explained more by the slowness of the historical paradigm shift away from dogmatic religious thinking than by the actual fundamentals of scientific thinking.

---

-Bluejay (a.k.a. Mantis, Thylacosmilus)Darwin loves you.

Hi, Straggler.Sorry about the delay: both my laptop battery and my PSU burnt out on Monday (probably all that twelveplus-hours-a-day online studying I was doing), and I haven’t had a computer all week.And, as an aside, I didn’t realize how complicated science is until I tried to do a week’s worth of it without a computer. No, not really. We’re talking about whether or not it is possible to determine this question.----- All except for Bluegenes’ theory, that is; Blugenes’ theory can only be falsified by a supernatural alternative.In my book, this means that it is not a scientific theory.----- I can’t figure out why you are incapable of seeing that empiricism would not be a valid or meaningful way to study genuine supernature.You’re so focused on naturalism, empiricism, and the scientific method, that you’ve completely failed to account for the fact that supernature doesn’t have to behave in any way that is meaningful to you, or that is meaningful in a naturalistic, empirical or scientific sense. It doesn’t have to leave interpretable patterns in data sets, and it doesn’t have to not leave patterns that would support alternative explanations.----- Why would the existence of supernature preclude the ability of naturalistic explanations to make successful predictions?You are a fool if you think successful predictions mean anything at all when the alternative to the prediction cannot be demonstrated.You are a fool if you think Bluegenes’ theory really successfully predicts anything, anyway.Given any set of similar ideas or things (even naturalistic things), it is a sure bet that at least some of them came from human imagination. So, I say that it is completely impossible for human imagination to not successfully predict at least some observations in any data set. Examples: So, given any data set, all you have to do is look hard enough, and you can validate your prediction in at least some circumstances.In the case of the supernatural, the genuine articles legitimately cannot be demonstrated to be genuine, so all you can work with is the subset that can be demonstrated to be imaginary, and you are claiming that this is somehow meaningful, and should be extrapolated to the entire collection of analogous things, simply because genuineness has not yet be demonstrated (even though it cannot be demonstrated).----- Straggler's entire argument amounts to nothing more than an over elaborate attempt to get around the very real impotence of empirical reasoning in testing supernatural concepts by pretending that he can instead have "confidence" in an alternative theory that can only be falsified by demonstrating the supernatural, apparently oblivious to the fact that this is exactly the same thing.He also has an annoying habit of attributing peoples’ tendency to disagree with him to the irrationality of their theistic background, even when he has been told on multiple occasions that his opponent considers himself agnostic, and has, in fact, been arguing for an agnostic, rather than a theistic, conclusion.I will not be participating further in this debate.

---

-Bluejay (a.k.a. Mantis, Thylacosmilus)Darwin loves you.

Hi, Straggler.All right: you talked me back into it again. This is a reply to Message 399 in the Peanut Gallery. Why is it that people always interpret frustration as a symptom of a greater psychological syndrome or disorder? Isn’t it equally likely that there are things out there that are legitimately frustrating? After all, I’m not the only one who found this frustrating.I feel like I work damn hard on each post, but don’t really get much in response except the same assertions as before I put all that effort into it. Somehow, everything I say gets interpreted as a demand to disprove the supernatural.I say, The assessment of confidence in science isn’t about how well it compares to supernatural alternatives: it’s about how it compares to naturalistic alternatives, and your very next response is, but, based on your reasoning, we can’t have confidence if there are supernatural alternatives.I don’t get it: I can’t figure out why you can’t accept this point. You don’t ever explain it: you just keep saying that my argument leads to the exact opposite conclusion of this point, even though this point is my argument.Here’s the point of it: confidence, in the sense used by science, comes from comparisons. This is why the unfalsifiability of supernature is a big deal. You can’t say you have confidence in one theory just because you have evidence that supports it: you also need a way to compare it to the alternatives. Positive evidence for one idea doesn’t mean much when evidence leaning the other way cannot exist. And, since the alternative is immune to testing by its very definition, such comparisons cannot be performed, and, consequently, confidence cannot be rightfully claimed.The implication of your argument is that everything that science can’t study is imaginary. By default, this precludes the supernatural from being anything but imaginary.----- First, I don't think I have any supernatural beliefs.Second, why do you think I have to believe in something to believe that it can't be handled with deductive logic?-----You also kept trying to get me to respond to something you said in Message 184. I have no idea what about that particular post you were referring to, but there is at least one point in it that I have not addressed, so I’ll address it here: Confidence comes from comparisons.Do you have confidence, right now, that, even if alien intelligence was discovered, your human imagination theory would still explain all supernatural beings?If so, then you are pretty indiscriminate about what you place confidence in.If not, then I have to ask you the follow-up question: Edited by Bluejay, : No reason given.

---

-Bluejay (a.k.a. Mantis, Thylacosmilus)Darwin loves you.

Hi, Straggler.This whole time, you’ve been arguing from the position that you think I’m reasoning inconsistently. So, you keep demanding that I respond to various examples, thinking that, when I see what conclusions my arguments come to, I will suddenly decide that I don’t agree with it.Your reasoning for this is apparently that you personally find the conclusions nonsensical.
But, I ask you, what reason do you have to believes that I will also find the conclusions nonsensical?I am aware of what conclusions my argument leads to, and I don’t have a problem with them. Good grief, drop the gravity crap already!As far as I’m aware, there isn’t a clear naturalistic explanation for gravity yet, so this is hardly as analogous to the situation at hand as you seem to think it is.As far as can be told with present knowledge, gravity is supernatural.It is possible to have confidence that a pen will fall when I drop it.
But, it is not possible to have confidence that the explanation for why the pen will fall when I drop it is naturalistic, because naturalism cannot be used to demonstrate its own veracity.This is the logical conclusion of my argument.
I do not have any problems with this conclusion, and I don’t find it to be nonsensical. Furthermore, it doesn’t have the catastrophic effects on my ability to reason that you seem to think it would.----- I have made no such claim, and thus, do not have to provide the comparison on which I have made the claim.And, once again, I have already told you that scientific confidence is not robust to supernatural explanations.
And, I have no problem with the ramifications of this position, which are neither as severe nor as melodramatic as you seem to think they are.Why do I have to keep saying this?
Will you finally accept that this is, indeed, my position?
Or, will you demand that I hold your hand and walk you through yet another example?If you’d like, we can go through every theory that currently exists in science, one at a time, and I can show you how to apply my reasoning to each of them, and describe for you what the conclusions will be. However, I think you are fully capable of doing that on your own if you actually apply my reasoning the way I’m explaining it to you.I am not afraid of the conclusions of my own reasoning, Straggler. So, please stop throwing out examples of the ramifications of my reasoning, because I am aware of them already, and you keep messing it up anyway.In particular, my argument never has the ramification that supernatural alternatives are confidently rejected. Stop presenting examples in which you think it does.

---

-Bluejay (a.k.a. Mantis, Thylacosmilus)Darwin loves you.

Hi, Straggler.That was about the worst job I’ve done explaining myself in a long time. It doesn’t even make sense the way I said it. My apologies. My position is not that the mere existence of a supernatural alternative makes confidence invalid, but that comparisons with a supernatural alternative make confidence invalid.In your gravity example, you’re not presenting a natural and a supernatural alternative: you are just presenting two alternatives, the natural/supernatural quality of which are not known. I know of at least two natural situations in which your pen-drop scenario will not result in falling: (1) a tornado; (2) an electromagnet. If the pen were to fly out the window, I would be inclined to first analyze the probabilities of these (or similar) naturalistic alternatives before resorting to supernature, and I think you would too.I can dismiss these two naturalistic alternatives based on the following reasoning and evidence:

1. You weren’t intentionally including tornadoes or electromagnets, so any interference by these things would have to be coincidental.
2. The chances of a tornado coincidentally interfering are pretty low, even in Tornado Alley.
3. Tornadoes are predictable, so the trial would likely be cancelled long before a tornado interfered.
4. The chances of a magnet coincidentally interfering are also pretty low.
5. Not all pens respond to magnetism, anyway.

All of these things give me the means to quantify probabilities of different naturalistic alternatives, and allow me to say that I have confidence in one of them.However, I cannot say that this confidence is robust to the supernatural alternative you describe above (again, my apologies for implying that I could in my last post). But, that’s okay, because I’m not comparing my expected result with the supernatural alternative when I calculate my confidence, anyway.In the case of the human imagination theory, only two alternatives have been presented for use in comparisons so far: (1) genuine supernature; (2) alien imagination. Since I am pretty sure that you don’t actually have the means to legitimately perform either of these comparisons, and that, even if you did, you haven’t actually performed either one, I am equally sure that your confidence is not statistically valid.-----My second observation on the matter is that gravity isn’t really a naturalistic explanation: it’s just a pattern. Remember that only nature has to obey patterns: supernature may or may not obey patterns. Until there is a naturalistic explanation for the pattern, this gravity example is pretty disingenuous.

---

-Bluejay (a.k.a. Mantis, Thylacosmilus)Darwin loves you.

Hi, Straggler. No, you can’t. But, this has no bearing on confidence.No, you can’t. But, this has no bearing on confidence.No, you can’t. But, this has no bearing on confidence. That’s for the next two times you ask me, as well, because past experience has shown me that my answering it has no effect on whether or not you’re going to ask it again, so I decided to stockpile them now.I have answered this question five times now (including the three above). I would like it to go into the record this time.-----Straggler, I really need to stop participating in this debate: it’s fruitless, pointless and, most of all, not fun. There’s a personality profile that spells a recipe for major disaster whenever they get into discussions with Straggler: obsessive-compulsive, meticulous, overly analytical, pedantic and not an atheist. See Rob, Iano and RAZD for examples.And, unfortunately, I fit the profile too. But, I don’t intend to let my participation in one stupid, pedantic topic define the rest of my participation at EvC, as Rob, Iano and RAZD have. I’m done with it: you’ll never hear (see, actually) another word about the supernatural from me.

---

-Bluejay (a.k.a. Mantis, Thylacosmilus)Darwin loves you.