[Merged] Immortality & Bayesian Statistics

[jsfisher]

It seems presumptuous to me as well.

Most serious PhD's aren't all that concerned about the Dr. title. (David Hovid is a good example of a farcical PhD who does flaunt the title.) I wouldn't be surprised if the good doctor's reply email to Jabba was signed simply as "Roger".

That would be a weak invitation to call him Roger.

I am a little surprised, though, that Jabba took the invitation. Someone of Jabba's alleged age tends to be formal addressing people he doesn't actually know. Use of forenames as the default is more of a younger generations phenomenon.

 

[Jabba]

Jabba,

I don't have an objection to your formula or the values you've plugged into it. As long as P(me|A) is infinitely small or giganogargantuanly small, hypothesis A is getting hammered.

Which leaves the alternative(s). That's where you're going to run into a high likelihood of being wrong.
As for your hard-core opponents, they don't care how badly your formula hammers hypothesis A. Probability, particularly conditional probability, means nothing to them. They have no respect for probability. Some of them practically deny the existence of it.* A few tough sessions at a poker table at uncomfortable stakes against competent opponents could teach them some respect, but your formula alone will have no effect on them.

*One just denied the existence of entire pantheons of universes, using their remote locations for an excuse.

Toon,
- I hope you're still listening...
- Why do you think that I have a high probability of being wrong about the alternatives? I'm no longer trying to be specific about what ~A might be...

 

[Jabba]

Most serious PhD's aren't all that concerned about the Dr. title. (David Hovid is a good example of a farcical PhD who does flaunt the title.) I wouldn't be surprised if the good doctor's reply email to Jabba was signed simply as "Roger".

That would be a weak invitation to call him Roger.
I am a little surprised, though, that Jabba took the invitation. Someone of Jabba's alleged age tends to be formal addressing people he doesn't actually know. Use of forenames as the default is more of a younger generations phenomenon.

- Yeah.
- Besides, he's probably 15 years my junior...

 

[Slowvehicle]

Toon,
- I hope you're still listening...
- Why do you think that I have a high probability of being wrong about the alternatives? I'm no longer trying to be specific about what ~A might be...

Good afternoon, Mr. Savage!

I hope you are listening:

Why are you refusing to address your egregious rudeness in telling "Roger" lies about me, and about others on the forum?

 

[Mojo]

Jabba hasn't nailed down a definition for "self"...

I think it's "an everlasting Volkswagen".

 

[Toontown]

The formula itself has never been at issue. Your lack of an objection to it is not surprising. I find your acceptance of Jabba's values a bit strange, though.

Jabba hasn't nailed down a definition for A, but you accept his P(A)?
Jabba hasn't even considered what ~A must therefore be, but you accept his P(~A)?
Jabba hasn't nailed down a definition for "self", but has set a direction incompatible with what's he's implied for A, but you accept his P(me|A) and P(me|~A)?

You must be a very amiable and trusting individual.

I don't need to be amiable or trusting. I have eyes. I can see.

Plug in some values you think are reasonable guesses, and see what difference it makes. Obviously you haven't tried that.

You shouldn't even need to solve it with your selected alternatives. You should be able to see at a glance that as long as P(me|A) = 1/∞ or 1/(some other giganogargantuan finite number), then it doesn't matter what the other values are, unless you deliberately sabotage it. For example, you could set up a division by 0 error, blow the whole thing up, then claim it's invalid because you blew it up.

However, short of overt sabotage, 1/∞ or 1/(some giganogargantuan finite number) bashes hypothesis A with the mathematical equivalent of a sledge hammer.

 

[Toontown]

Toon,
- I hope you're still listening...
- Why do you think that I have a high probability of being wrong about the alternatives? I'm no longer trying to be specific about what ~A might be...

That's the way to avoid being wrong about the alternatives. I frankly don't think there is any chance you'd get it right if you did try to be specific. But of course I can't prove that.

 

[Slowvehicle]

That's the way to avoid being wrong about the alternatives. I frankly don't think there is any chance you'd get it right if you did try to be specific. But of course I can't prove that.

Ah, the flounce it is so hard to stick...

 

[xtifr]

However, short of overt sabotage, 1/∞ or 1/(some giganogargantuan finite number) bashes hypothesis A with a sledge hammer.

Those two cases are completely separate. If you have a die with a hundred billion trillion bogglilion sides, and you roll it, one of those sides (assuming a fair die) will come up.

The fact that you can look at each and every side of that multi-bogglilion-sided die and say "there's effectively no chance this side will come up" doesn't change the fact that if you roll it, one of those sides will come up.

Now, if you say the odds against are 1/inf., that's completely different. Which is why Jabba keeps hoping to get to 1/inf. But he can't, because hypothesis A is that Jabba exists due to the well-understood laws of nature, and those well-understood laws of nature don't allow an infinite number of permutations of the universe-as-we-know-it. (Which is also how modern physics avoids Zeno's paradox.)

 

[John Jones]

- Yeah.
- Besides, he's probably 15 years my junior...

Do you presume that age demands respect for crackpot opinions?

 

[jsfisher]

...deleted...No point feeding the derail.

 

[Toontown]

Those two cases are completely separate. If you have a die with a hundred billion trillion bogglilion sides, and you roll it, one of those sides (assuming a fair die) will come up.

That's irrelevant. No matter how separate they are, Their effect is the same. Either one bashes hypothesis A with the mathematical equivalent of a sledge hammer.

 

[John Jones]

That's irrelevant. No matter how separate they are, Their effect is the same. Either one bashes hypothesis A with the mathematical equivalent of a sledge hammer.

Please stop talking.

 

[Toontown]

It seems presumptuous to me as well.

Do you presume that age demands respect for crackpot opinions?

Please stop talking.

Please curtail the derails. Thank you.

 

[Garrette]

Ah, the flounce it is so hard to stick...

I do not think Too town is flouncing; he may refuse to discuss actual math, but he has been consistent, even going back to his earliest comments near the start of this thread. Recall Jabba's original equation; Toontown agrees that the left side (probability of any particular self given a giganogargantuan number of potential selves is for all practical purposes so close to zero as to make no difference), but he disagrees with Jabba that it will lead to a conclusion of either immortality or reincarnation in the sense Jabba means. He (Toontown) will not say what he thinks the conclusion should be or how he arrived at it, but one can make reasonable guesses.

I think what Toontown believes is some sort of oneness, perhaps like funi in Buddhism or a variation on Emerson's Over soul. Regardless of the particular spin he puts on it, the point is that he removes probability completely from the equation by removing the idea of multiple selves. There is one self; our perception of individuality is the illusion that leads to years long arguments such as this one. I am sure that there are details and nuances I am missing that Toontown will latch on as indicative of my complete misunderstanding; regardless, the point is that he disagrees both with us and with Jabba and has been consistent in that.

As always, I could be very wrong.

 

[Garrette]

More succinctly: Toontown agrees with Jabba's premise but not his conclusion, and he disagrees with both our premise and conclusion.

 

[jsfisher]

Those two cases are completely separate. If you have a die with a hundred billion trillion bogglilion sides, and you roll it, one of those sides (assuming a fair die) will come up.

The fact that you can look at each and every side of that multi-bogglilion-sided die and say "there's effectively no chance this side will come up" doesn't change the fact that if you roll it, one of those sides will come up.

Now, if you say the odds against are 1/inf., that's completely different. Which is why Jabba keeps hoping to get to 1/inf. But he can't, because hypothesis A is that Jabba exists due to the well-understood laws of nature, and those well-understood laws of nature don't allow an infinite number of permutations of the universe-as-we-know-it. (Which is also how modern physics avoids Zeno's paradox.)

Physics avoids Zeno's paradox simply because Zeno's paradox is just a bit of misdirection and sleight of hand. It focuses all attention what what's happening at time, t < 1 -- there's always more to go to catch up. It conveniently avoids asking what happens at t = 1 and t > 1.

In some respects, Jabba is doing the same thing by conflating the probabilities of the existence of an arbitrary individual with that of himself. The former may be 1 / some "giganogargantuan" number, but the latter is certainly 1.

 

[Garrette]

Apologies to Slowvehicle. I read "flounce" as "bounce" meaning inconsistent. I blame it on the fact I am reading and typing on my phone on a business trip in a hotel bar after an airplane flight that cracked my laptop screen. I am grouchy.

 

[Toontown]

In some respects, Jabba is doing the same thing by conflating the probabilities of the existence of an arbitrary individual with that of himself. The former may be 1 / some "giganogargantuan" number, but the latter is certainly 1.

I hope you're not saying what it looks like you're saying, and if you are, I am pretty sure, based on your current existence, that you do not conduct your life accordingly. At any rate, where do you see Jabba conflating those probabilities? I don't see him using p=1. I see him using p=1/infinity.

If you're walking along a sidewalk in Chicagoland and you hear the unmistakable sound of a bullet zipping past your ear, do you tell yourself "Ah, no problem. The probability that some arbitrary individual will hear a bullet zip past his ear is certainly 1."

That would be an incorrect use of probability. That would be you conflating the probability that you heard a bullet zip past your ear with the probability that some arbitrary individual will hear a bullet zip past his ear. That would be a bad mistake. Possibly fatal.

Take evasive action immediately. That's you hearing the bullet. Not some arbitrary individual.

The use of conditional probability is a good thing. And it works.

 

[PixyMisa]

I don't need to be amiable or trusting. I have eyes. I can see.

Plug in some values you think are reasonable guesses, and see what difference it makes. Obviously you haven't tried that.

You shouldn't even need to solve it with your selected alternatives. You should be able to see at a glance that as long as P(me|A) = 1/∞ or 1/(some other giganogargantuan finite number), then it doesn't matter what the other values are, unless you deliberately sabotage it. For example, you could set up a division by 0 error, blow the whole thing up, then claim it's invalid because you blew it up.

Let's give it a try then.

Since I do exist, P(me) = 1.

Since life actually is finite, P(A) = 1.

Therefore, P(me|A) = 1.