Proof of Immortality, the IV league

[jt512]

It would, indeed, be a valid inference. "Finding water was so unlikely that the intrepid explorer must have been guided to it by God", however, would not.

So, an explorer goes into the desert, gets lost, survives, and claims that he survived only because God told in what direction to walk to find water. If we assume that the probability he would survive if God told him where to find water is 1, then how is this not evidence for the God hypothesis over any other hypothesis? What hypothesis would be better supported by the datum?

What if, out of all the explorers that ever got lost in the desert, the only survivors (say there were 10 of them, each independent of each other) each said (independently, and without knowledge of the others) that he only survived because God showed him which direction to walk to find water? What's going on here? Every time somebody gets lost and survives they insist that God told them where to find water? What non-God hypothesis would be better supported by the data?

 

[Hokulele]

We would have to know how many of the non-survivors were told by God which direction to walk.

 

[jt512]

Beliefs?

Ah, so we aren't talking about science and mathematics after all. Got it.

You don't believe that matter is made of atoms?

 

[Nonpareil]

Actually, if you ignore Jabba, the real meat of the thread is JT512's efforts to school "skeptics" here on what evidence is and how it's used to update beliefs.

What points do you think remain to be addressed?

I am all for continuing the thread, but it does require some input from the other side. With annnnoid suspended and Jabba being even more difficult to engage with than usual, that makes it a little difficult. If there's something that you want to see discussed, toss it out.

 

[Loss Leader]

So, an explorer goes into the desert, gets lost, survives, and claims that he survived only because God told in what direction to walk to find water. If we assume that the probability he would survive if God told him where to find water is 1, then how is this not evidence for the God hypothesis over any other hypothesis? What hypothesis would be better supported by the datum?

What if, out of all the explorers that ever got lost in the desert, the only survivors (say there were 10 of them, each independent of each other) each said (independently, and without knowledge of the others) that he only survived because God showed him which direction to walk to find water? What's going on here? Every time somebody gets lost and survives they insist that God told them where to find water? What non-God hypothesis would be better supported by the data?

That may be the silliest hypothetical I have ever read.

How would this be evidence of God as opposed to, say, some alien pretending to be God?

"God" is a conclusion. Your "evidence" does not make the conclusion necessarily true. It is incompetent.

 

[Nonpareil]

So, an explorer goes into the desert, gets lost, survives, and claims that he survived only because God told in what direction to walk to find water. If we assume that the probability he would survive if God told him where to find water is 1, then how is this not evidence for the God hypothesis over any other hypothesis? What hypothesis would be better supported by the datum?

The probability of his survival given divine intervention is irrelevant.

While we can assume that the probability of me drawing a royal flush is one given if I cheated, the fact that I have a royal flush in my hand is not evidence that I actually cheated.

What if, out of all the explorers that ever got lost in the desert, the only survivors (say there were 10 of them, each independent of each other) each said (independently, and without knowledge of the others) that he only survived because God showed him which direction to walk to find water? What's going on here?

We don't know. That's rather the point.

You won't know until you investigate. Until you do, "we don't know" is not evidence for "it was actually what they say it was". It can be a reason for you to want to investigate, as has already been brought up, but it is not actually evidence of what is going on. It supports the alternative hypothesis "ten independent survivors all had heat-stroke-induced hallucinations of divine intervention" as well as it does "it was actually divine intervention".

And that is entirely without getting into the possibility of surrounding context and evidence. If, say, ten thousand people went into the desert with no water and only ten came out, the fact that those ten all happened to hallucinate similar things in their search for water wouldn't necessarily be all that remarkable. Perhaps there's some cactus juice leaking into the water, or some sort of religious statuary near the oasis. Perhaps most of the surrounding populace is religious and there are stories of dying travelers being saved by angels leading them to water. Maybe there's a college of wizards with a protective illusion of angels around their tower in the middle of the desert. Whatever.

Or maybe they're all just mischievous sorts and like to screw with statisticians by reporting religious experiences that never happened.

We don't know. You don't either. There is no evidence one way or the other (at least, ignoring possible surrounding context, as above).

Of course, you are free to take those ten reports any way you like. If they raise your personal confidence level in the possibility of divine intervention, that's wonderful.

But they don't raise mine, and the fact that they do it for you doesn't make them evidence.

 

[jt512]

That may be the silliest hypothetical I have ever read.

How would this be evidence of Good as opposed to, say, some alien preening to be God?

I suppose it would say it would be about equal evidence for those two hypotheses. What about against other alternative hypotheses, like the 10 surviving explorers found water by wandering around randomly, if say the chance of any one of them finding water that way would be 2%.

"God" is a conclusion. Your "evidence" does not make the conclusion necessarily true.

Evidence ordinarily does not make the conclusion necessarily true. That's why they call it "evidence," not "proof."

 

[Jedi]

Beliefs?

Ah, so we aren't talking about science and mathematics after all. Got it.

EXACTLY!! you have gotten it

 

[marplots]

What if, out of all the explorers that ever got lost in the desert, the only survivors (say there were 10 of them, each independent of each other) each said (independently, and without knowledge of the others) that he only survived because God showed him which direction to walk to find water? What's going on here? Every time somebody gets lost and survives they insist that God told them where to find water? What non-God hypothesis would be better supported by the data?

That theists get lost in the desert a lot.

 

[JimOfAllTrades]

So, an explorer goes into the desert, gets lost, survives, and claims that he survived only because God told in what direction to walk to find water. If we assume that the probability he would survive if God told him where to find water is 1, then how is this not evidence for the God hypothesis over any other hypothesis? What hypothesis would be better supported by the datum?

What if, out of all the explorers that ever got lost in the desert, the only survivors (say there were 10 of them, each independent of each other) each said (independently, and without knowledge of the others) that he only survived because God showed him which direction to walk to find water? What's going on here? Every time somebody gets lost and survives they insist that God told them where to find water? What non-God hypothesis would be better supported by the data?

A guy who hides in the desert, talks to people dying of thirst, tells them he's god, and directs them to water?

 

[jt512]

The probability of his survival given divine intervention is irrelevant.

Bayes' Theorem is correct, so the probability of the obsservation given the hypothesis, matters.

We don't know. That's rather the point.

We know that they said that God led them to the water. That is precisely the point: whether this is evidence that God actually led them to the water.

It supports the alternative hypothesis "ten independent survivors all had heat-stroke-induced hallucinations of divine intervention" as well as it does "it was actually divine intervention".

It does, but if 10 survivors all had delusions, it would seem unlikely that they all had the same delusion, so the data support the God hypothesis over the delusion hypothesis.

And that is entirely without getting into the possibility of surrounding context and evidence. If, say, ten thousand people went into the desert with no water and only ten came out, the fact that those ten all happened to hallucinate similar things in their search for water wouldn't necessarily be all that remarkable.

I agree that if the proportion of lost explorers who survived was approximately equal to the probability of randomly finding water, then that would be strong evidence against the God hypothesis. But by assumption, we don't know that.

 

[jt512]

That theists get lost in the desert a lot.

That, if known, would actually be additional data, or possibly, something to base a prior probability on, similar to the prevalence of a disease in evaluate a the probability of disease given a positive test result.

 

[Nonpareil]

I suppose it would say it would be about equal evidence for those two hypotheses.

<snip>

Evidence ordinarily does not make the conclusion necessarily true.

But it does actually have to support a given hypothesis moreso than it does another.

Anecdotes don't. As you yourself pointed out, the claim of divine intervention is as much compatible with the idea of aliens, hallucinations, wizards, lies, or any other hypothesis as it is with the idea of divine intervention. It is a complete and utter null tell.

It is not evidence.

Bayes' Theorem is correct, so the probability of the obsservation given the hypothesis, matters.

And yet it still doesn't help you to determine whether or not I actually cheated.

We know that they said that God led them to the water. That is precisely the point: whether this is evidence that God actually led them to the water.

And the answer is "no".

This is not complicated.

It does, but if 10 survivors all had delusions, it would seem unlikely that they all had the same delusion, so the data support the God hypothesis over the delusion hypothesis.

Again, no more so than it supports aliens, wizards, or lying reporters. Or, indeed, and despite your objections, even just straight delusions. Your arbitrary ruling of divine intervention being more likely than ten people having similar hallucinations in similar situations, presumably with similar priors, is not particularly convincing.

At most, you could make a case for something happening from this (though even that is stretching it, given the incredible vagueness of the scenario and lack of detail about surrounding conditions). But you have absolutely no actual evidence as to what that something was.

 

[jt512]

But it does actually have to support a given hypothesis moreso than it does another.

Anecdotes don't. As you yourself pointed out, the claim of divine intervention is as much compatible with the idea of aliens, hallucinations, wizards, lies, or any other hypothesis as it is with the idea of divine intervention.

No. Something is evidence for one hypothesis (H1) over another (H2) if and only if it has a higher probability under H1 than under H2. Thus, if an anecdote is more probable under H1 than under H2, it is evidence for H1 over H2; otherwise, it is not.

Something can be equally probable under H1 and H2 (and hence not be evidence for either hypothesis relative to the other), but be more probable under H1 than under a third hypothesis H3. If 10 out of 10 lost explorers finding water by chance (H3) has a probability of 0.02^10, then the observation of 10 out of 10 lost explorers saying God guided them to the water makes the God-guiding hypothesis nearly 10 million times as likely (on the odds scale) as the found-water-by-random-chance hypothesis, assuming that the two hypotheses were equally likely a priori.

 

[theprestige]

Actually, if you ignore Jabba, the real meat of the thread is JT512's efforts to school "skeptics" here on what evidence is and how it's used to update beliefs.

Actually, the meat of the thread is Jabba's attempt to prove mathematically his claim of immortality.

If you ignore Jabba, you might as well start some other thread about whatever it is you'd rather discuss. Everybody here understands how evidence updates beliefs.

What would be nice would be if you or jt could explain how Jabba's unverified, uncontrolled, unfalsifiable narratives should count as scientific evidence in support of his subsidiary claim, that immaterial souls exist.

 

[Loss Leader]

I suppose it would say it would be about equal evidence for those two hypotheses. What about against other alternative hypotheses, like the 10 surviving explorers found water by wandering around randomly, if say the chance of any one of them finding water that way would be 2%.

This makes no sense. It's like asking what the odds of finding a Harvard Law graduate would be at a Harvard Law graduation.

Some unknown number of people went into the desert. The only ones who survived were the ones who found water. Now we're supposed to be amazed that all the survivors found water?

Are you amazed, given that women account for 51% of the population, at the percentage of women found in the women's dressing room at Macy's

 

[theprestige]

I suppose it would say it would be about equal evidence for those two hypotheses. What about against other alternative hypotheses, like the 10 surviving explorers found water by wandering around randomly, if say the chance of any one of them finding water that way would be 2%.





Evidence ordinarily does not make the conclusion necessarily true. That's why they call it "evidence," not "proof."

Oh my god. Instead of arguing by some crappy analogy, why not argue the thing itself?

Jabba's claim is here in this thread for all to see. Jabba's "evidence" is here in this thread for all to see. Why not engage it directly, and judge it as it is?

I mean, take a step back, consider the context, and realize what you are actually arguing in this thread: That Jabba has introduced scientific evidence that supports the claim that immaterial souls exist. Have you bothered to familiarize yourself with the claim you're defending? Do you even know the nature of the evidence you're urging us all to accept? Can you even explain what parts of your desert narrative are analogous to Jabba's claim, and what parts are analogous to his "evidence"?

 

[Nonpareil]

No. Something is evidence for one hypothesis (H1) over another (H2) if and only if it has a higher probability under H1 than under H2.

That is what I said.

Thus, if an anecdote is more probable under H1 than under H2, it is evidence for H1 over H2; otherwise, it is not.

But you cannot establish whether or not a given anecdote is more probable under H1 than H2, because you have no idea if the anecdote is even slightly accurate.

That's the point.

 

[Hokulele]

You don't believe that matter is made of atoms?

Nah, matter is just a bunch of wave functions.

 

[jt512]

Nah, matter is just a bunch of wave functions.

Nice one, that.