Has The Supernatural Hypothesis Failed?

No - the supernatural hypothesis as stated is not necessarily verifiable but the test proposed was for falsification. The test could only verify the observation that goat burnings are linked to crop yield.

We'll need a definition of supernatural that allows (or disallows) verification inherently to be able to say that for sure.In the Mubu example - it might be possible to experiment more precisely with sacrificing to different beings. If the effect only happened when something was sacrificed to Mubu, then that would be fairly decent supportive verifying evidence - for example.Any other questions? Do you still stand by your claim that the supernatural is "inherently untestable"?

Not particularly: it isn't relevant to my point and to do it well would take a lot of effort and would probably have to be honed by lots of minds to be worth discussion. I never said that must be the case. Why would you infer it? It would support them all or a general entity. Yes. How many steps does it take before something is beyond walking distance? The more there are, the stronger the hypothesis of some more general explanation (for instance, a single supernatural entity that has several names).You are still diverting from the original point I raised which was about falsification, not verification. Are you going to get to the point soon so that you can go back to addressing Message 465, as you assured me you would a dozen questions or so ago?If you want a definition, you'll only find me repeating that I think it is an incoherent concept invented to explain why study of the natural world kept not discovering ghosts and demons and angels and fairies and gods and ghouls and goblins and domovoi and leszi, djinn...It's incoherent opportunism is another hallmark of its failure as a theory-class.

By the definition under use, it is. The definition being 'a set of entities included in which are immaterial agents with intentionality that can influence material things'. Not at all. The principle of fallibilism is preventing you from proving Mubu The simple answer is: If a natural event causes another natural event by means of a being classed as supernatural above we can perform a test:If the first natural event is shown to have no causal influence over the second natural event then we have demonstrated that the supernatural claim must be false. This is because the supernatural agent cannot be causing an effect which doesn't actually occur.That's all. How one might provide supporting evidence for such a supernatural entity is not important to the point. It is possible, in all the same ways one provides supporting evidence for any claim - it is only made difficult by the vague nature of the claims so often put forward.

But I'm talking about performing a falsification test on a falsifiable supernatural theory so none of that applies.In case you want to continue down this path I use a falsification test to try to falsify a falsifiable theory. If it is falsified it is not anybody's 'fault' - it just means the theory is false. Give me a complete account of the supernatural and I'll try to figure that out for you.For instance does supernatural materia become visible on the 16th October 2010 (GMT)? If so - I think I can conceive a possible way to provide supporting evidence for the supernatural.Does it exert a natural force under any conditions? Then you've misunderstood the experiment.Here are the possible results and conclusion:pre-sacrificed crops do better - Sacrificing to Mubu increases crop yield.a-sacrificed crops do better - Not sacrificing to Mubu increases crop yield.The crops do equally well - Sacrificing to Mubu has no more effect on crop yield than not sacrificing to Mubu.If the latter result is obtained this then falsifies the theory 'Sacrificing to Mubu will increase crop yields as Mubu will protect your crops from disease spirits' (and the theory 'Not sacrificing to Mubu will increase crop yields')IF you want to balance things out to avoid 'special pleading' you can do it yourself. If the results are the first one then the theory 'Sacrificing to Mubu won't..." is falsified and so on. Now each result has its opposite conclusion accounted for.Edited by Modulous, : No reason given.

Of course it doesn't. But wait - since it is a debate, I'll say more!By "inherently 'unevidencible'" you either mean1) Cannot be falsified: In which case I do not feel the claim is inherently unevidencible.2) Cannot be verified: In which case, I'm not trying to verify it.Either way - it doesn't apply. Perhaps you can try something better than gainsaying if you want to continue down this line of argument? Logic. You are the one that said a supernatural claim could not be tested and failed, not me. I said you could. I gave you a definition that while incomplete was specific enough to allow for a falsifiable supernatural hypothesis. Then I provide you with a test to show that your argument was false. Yes, confused seems to be about right. Let me try again, without assuming you understand the basic setup.Apparatus1,000 fields
Enough seeds to grow 1,000 fields worth of crops
Sacrifices
Common senseExperimentSew seeds in the fields.
Randomly assign each field a letter, A or B.
If a field is part of group A - sacrifice a goat to Mubu.
If part of group B - do not sacrifice a goat to Mubu.
Send postgrads into the fields to record results. Do not tell them which field is in which group.Results and conclusionsIf crops that are grown after a sacrifice to Mubu do better than crops that are grown without a sacrifice to Mubu -
then sacrificing to Mubu increases crop yield.If crops that are grown with no sacrifice to Mubu do better than crops that are grown with the sacrifice to Mubu -
then not sacrificing to Mubu increases crop yield.If crops that are grown after a sacrifice to Mubu do no better than crops that are grown without a sacrifice to Mubu -
then Sacrificing to Mubu has no more effect on crop yield than not sacrificing to Mubu. I think it is adequate, if you want to include other possible results and the conclusions you might draw from them, let me know what they are. Replace Mubu the immaterial agent with Mubu the pesticide (and replace 'sacrifices to' with 'application of' and I'll stick by the reasoning. No special pleading involved there I'm afraid.

Indeed I don't. The experiment cannot tell us why the crops do better when we sacrifice to Mubu, only that this is what happens. Hence that while it could support the supernatural hypothesis, it wouldn't prove it. I think, for the reasons outlined we can agree this only supports the Turdung hypothesis, not proves it. But when we were discussing verification I do remember saying:

quote:

---

In the Mubu example - it might be possible to experiment more precisely with sacrificing to different beings. If the effect only happened when something was sacrificed to Mubu, then that would be fairly decent supportive verifying evidence

---

This, of course is a different experiment. Only one that we really need carry out if the results are positive. And since I was talking about falsification, not verification, it is still not relevant. I think you'll find upon re-examination the test remains unflawed.

---

abe: I think you might be getting confused about evidence here.The experiment gathers evidence so we can establish facts. With Mubu the fact transpires that sacrificing to Mubu does not increase crop yields. With Turdung the fact is that sacrificing to Turdung increases crop yields.The question is, what do these pieces of evidences mean for our theories?For the theory that predicts Mubu will increase crop yields if you sacrifice to him - we have a falsification since no crop yield increase occurs.For the theory that predicts Turdung will increase crop yields - we have gained evidence that supports the theory. An "ah - that's interesting" moment, if you will. If we want to increase our confidence we don't fall prey to confirmation bias, we try to falsify the theory with yet another falsification test. This time involving spreading muck without invoking a being, and spreading muck invoking a different being.When we encounter verification, we run more falsification tests. If a theory survives the initial barrage of obvious tests, it is (tentatively) a success. If it fails, it is (tentatively) a failure. So far, the supernatural, when tested, has fallen into the tentatively failed category. And no supernatural theory has worked its way out of that condition (without ad hoc modifications that render it unfalsifiable).Edited by Modulous, : No reason given.

Jon,
I am confident you have lost at this stage. I strongly suggest you concede the point that you can test the supernatural, before going further down this road. You are reasoning backwards that since you believe I am wrong, there must be a flaw somewhere, and you are confidently attacking every which way and hoping something will stick. Just sit back and reason it through, I'm sure you'll see that I'm right. Really? Very strong words from someone who keeps misunderstanding the basics. My test doesn't allow anything. It's just a test.
Observation and reasoning should lead you to conclude that it is possible to tack any supernatural blip anywhere you want (just look around - as an example a supernatural blip is being proposed by significant numbers of people as an alternative (to science and engineering) explanation for the successful Chilean Mine rescue). That's not a failure of my methodology, it's another failure of the vague piece of crap that is the supernatural. That's not a misuse of the scientific method. You have merely misunderstood the experiment again.Firstly, what you have established (sort of) is that Modus Ponens can be valid but false.

quote:

---

If P, then Q.
P.
Therefore, Q.

---

So yeah - you can put anything you like in the antecedent (P) and then confirm it to 'prove' Q. If you think that you can use this to mislead, then I'd agree, but in this case that would be your fault for misusing logic - not mine for misusing science. Try again.Secondly, This is not related to what I was doing since I wasn't using Modus Ponens but Modus Tollens (denying the consequent rather than affirming the antecedent).Thirdly, the format of your situation is different in a key fashion. In my experiment I was testing the hypothesis that two natural events are causally related as a result of a supernatural intermediary. You have only one natural event.Fourthly, if we assume that Jesus is natural - you have failed to perform a control experiment in which you inhibit the actions of Jesus to see if the sun still rises. This is an example of why we need to avoid confirmation bias. Just the wrong point, at the wrong time, to the wrong person, in the wrong context. I was trying to save on typing because I thought you were capable of being sensible in this discussion. Sorry.The theory 'sacrificing to Turdung increases crop yields' has been confirmed. It is true, from our standards of measurement. As for what happens next - we've been over this.However, this doesn't help our full theory (remember in Mubu's case it was because of Mubu's intercession in driving spirits away). Since the test only weakly confirms the hypothesis we need to embolden the theory if we want to run further tests and get to grips with things.So, with Turdung, being curious and sensible, we might test things such as"Only sacrifices to Turdung result in good crop yield"
"Sacrificing to other beings won't result in good crop yield"
"If one wants better crops, sacrificing to Turdung is the exclusive method."{We do this if we want to have confidence that Turdung has some special or interesting detectable influence}If you want to suggest that we don't ever falsify the theory that"Turdung gives poo magical undetectable powers..."Then that's true, but I'm not interested in falsifying the unfalsifiable.It remains true that it is possible to test supernatural hypotheses, as I have described. Give it up Jon, it's a doomed position: It essentially requires that you demonstrate science can't test hypotheses.

---

Hypothesis: Spreading Mubu, the fertilizer gives better crops than not spreading Mubu, the fertilizer.If my test is crap - why not show me how it is done? Design a test for this natural hypothesis.Edited by Modulous, : No reason given.Edited by Modulous, : No reason given.

Hi Jon
Looks like you've finally conceded defeat. Allow me to be the one that nails the final nail into the coffin lid. Here you admit, what I suspected for a while. You thought I was testing for the existence of a supernatural entity, when I actually wasn't. I was testing a hypothesis that a supernatural entity affects the world in a certain way.Hopefully, if you want to continue along this discussion we could discuss that kind of hypothesis "entity x does y after z" which can certainly be falsified by showing y does not follow z and entity x can be natural or supernatural. If X is supernatural then the hypothesis, for any reasonable understanding I would have thought, is a supernatural hypothesis. Might I suggest you abandon using the word the way you do? It does not gel with common usage and there is a word in common usage that has the exact same meaning:

quote:

---

I draw this from the notion that if your test does test for them, then they are not untestable; if your test cannot test for them, then they are untestable. In either situation, we end up with an untestable 'untestable'

---

I was under the impression you asked me for a working definition so that you might employ it in a discussion with me. But yes, attempting to use shoddy definitions to define yourself as inherently correct does seem silly - I would have hoped that was a hint enough to you to not waste my time with it. Since this is Straggler's thread - I'm pretty sure he gets to define what a 'supernatural hypothesis' counts as. We should be discussing that topic, if you would rather think of it as "Has the immaterial agency causing material or immaterial things to occur hypothesis failed?" then go right ahead. Or make up your own word. But I'm comfortable with a definition that makes gods, spirits and ghosts supernatural. Which means you had no intention of paying attention to me when I gave a definition of the supernatural. I feel vindicated for the rather harsh thing I sad earlier:

quote:

---

You are reasoning backwards that since you believe I am wrong, there must be a flaw somewhere, and you are confidently attacking every which way and hoping something will stick.

---

Indeed - you believe I am wrong because you weren't using the definition I gave, but you were using the definition you had in your head that almost nobody else in the world uses. I'm sure you can see how that might lead to confusion.According to you:
The existence of Mubu, the god, is a supernatural claim (it cannot be tested)
The hypothesis that Mubu, the god helps crop growth (by driving away immaterial disease spirits) if you sacrifice to him is a natural claim (since it can be tested). A hypothesis that is unfalsifiable is a failure by the scientific standard. QED: /threadEdited by Modulous, : No reason given.

Great - so if we're agreeing that my Mubu under the terms supernatural which people actually use is in fact supernatural, and we agree you would not dismiss it on those grounds...would you dismiss it on the grounds the test shows that Mubu sacrifices do not keep away disease or help crops in any tangible way, contrary to the hypothesis. But the first test of a hypothesis that science asks is "Is this hypothesis falsifiable?", supernatural hypotheses can be subjected to this test.It hasn't failed in the sense 'proven wrong', it has failed as a hypothesis to meet the requirements of a hypothesis according to the philosophy of science which is not an unwarranted 'prejudice', but a pragmatic decision which has given us the 20th century. If you want to say that the supernatural hypothesis has not failed according to some other standard, then maybe so. But on scientific terms, it is a failure of a hypothesis. Then we move onto the second step of the process as I actually described earlier (I did not fail to address it):Every time a supernatural hypothesis is falsifiable and has been tested, it is falsified.
Every time a supernatural hypothesis is unfalsifiable it fails to meet the standards of a scientific hypothesis.So, the question then becomes - how many falsifications and unscientific proposals need to be made in the name of the supernatural before we conclude that as an avenue of hypothesis generation it is best considered a dead end. A failed avenue of thought. A failure as a suite of hypotheses.Presumably if luminerferous ether based hypotheses are considered 'failed', by those same standards supernatural based hypotheses can likewise be considered?Or as I succinctly put it some time ago:

quote:

---

Many are testable, have been tested, and have failed.

​

The ones that have not failed have done so by being untestable. Since they are untestable, there is no way anyone can have confidence in them.

​

When 'immaterial causal agent' can be tested, it fails.
When it can't be tested, its useless since we can't hold any confidence in it.

​

How is this not the very definition of a failed hypothesis? Either it consistently fails or gives us nothing.

---

In my first post, that you'd said you'd get back to when you uncovered what I meant by 'supernatural'. I guess now would be a good time to revisit it. Message 465Edited by Modulous, : No reason given.