[Merged] Immortality & Bayesian Statistics

[Garrette]

Jay,

- You are right.

- I thought that I had sent the following already, but apparently not...

- If I ever knew of the "likelihood" terminology, I forgot it.
- But, I would argue that in "everyday" language, the probability of an event occurring, given established beliefs, could reasonably be called the "prior probability."
- Would you agree with that?

Being a moderately intelligent layman, I will dare to step in here:

It doesn't matter if he agrees or not. You are attempting to establish proof; you are not having an everyday discussion down at the pub. If you wish to have your proof treated as proof then you must present it as a proof, not as a conversation. And you must demonstrate that you developed it with the same rigor with which it will be analyzed.

 

[Toontown]

Being a moderately intelligent layman, I will dare to step in here:

It doesn't matter if he agrees or not. You are attempting to establish proof; you are not having an everyday discussion down at the pub. If you wish to have your proof treated as proof then you must present it as a proof, not as a conversation. And you must demonstrate that you developed it with the same rigor with which it will be analyzed.

I think most of us knew at the outset that "prove" was an unfortunate choice of word or intent, as the case may be.

Still, your criticism is essentially semantic.

 

[Garrette]

I think most of us knew at the outset that "prove" was an unfortunate choice of word or intent, as the case may be.

Still, your criticism is essentially semantic.

I can see how it could be seen that way, but it is a mistake. Jabba invoked Bayes as his method; we did not. If he wishes to invoke Bayes, he should be correct in the usage.

I don't mind mistakes or errors; that is to be expected from anyone. What I object to is the dismissal of others' pointing out of mistakes or errors by asking them to agree with a non-Bayesian interpretation/usage.

 

[Toontown]

I can see how it could be seen that way, but it is a mistake. Jabba invoked Bayes as his method; we did not. If he wishes to invoke Bayes, he should be correct in the usage.

There is little of a probabilistic nature that can provide absolute proof of much of anything, by the Bayes method or any other.

So I agree that the use of the word "prove" was a mistake. I just think it's trivial. Jabba has presented his formula. The word "prove" is nowhere in the formula. There is no term for "prove" in the Bayes formula.

I don't mind mistakes or errors; that is to be expected from anyone. What I object to is the dismissal of others' pointing out of mistakes or errors by asking them to agree with a non-Bayesian interpretation/usage.

Criticizers are not above criticism themselves. Bite the bullet.

 

[Garrette]

Toontown, this marks the second time you have been in error in exchange with me. First was when you mistakenly claimed you had said "most" but had not (and it was not a matter of semantics). Now you mistake that to which I was referring. You brought up "proof." I was discussing "prior probabilities." I take my criticism both well and frequently on this forum where I am so often out of my league. The question is whether you can. As you never acknowledged your first error, I am left to wonder.

 

[Toontown]

Toontown, this marks the second time you have been in error in exchange with me. First was when you mistakenly claimed you had said "most" but had not (and it was not a matter of semantics).

I do not recall that. You'll need to quote the post where I allegedly committed the erroneous use of the word "most", so that I may determine my penance.

Now you mistake that to which I was referring. You brought up "proof." I was discussing "prior probabilities." I take my criticism both well and frequently on this forum where I am so often out of my league. The question is whether you can. As you never acknowledged your first error, I am left to wonder.

Below is your lecture to which I responded, with your uses of the word "proof" highlighted. Nowhere in the post does the term "prior probabilities" appear. I am aware that you were lecturing in ostensible response to something Jabba had said about my use of the term "prior probability". However, the substance of your lecture was unrelated thereto.

Being a moderately intelligent layman, I will dare to step in here:

It doesn't matter if he agrees or not. You are attempting to establish proof; you are not having an everyday discussion down at the pub. If you wish to have your proof treated as proof then you must present it as a proof, not as a conversation. And you must demonstrate that you developed it with the same rigor with which it will be analyzed.

We are now well and truly mired in a morass of semantic nitpicking, and you are losing as surely as you would lose a mud wrestling match with a 400 lb hog.

Are we having fun yet?

 

[Garrette]

I do not recall that. You'll need to quote the post where I allegedly committed the erroneous use of the word "most", so that I may determine my penance.



Below is your lecture to which I responded, with your uses of the word "proof" highlighted. Nowhere in the post does the term "prior probabilities" appear. I am aware that you were lecturing in ostensible response to something Jabba had said about my use of the term "prior probability". However, the substance of your lecture was unrelated thereto.



We are now well and truly mired in a morass of semantic nitpicking, and you are losing as surely as you would lose a mud wrestling match with a 400 lb hog.

Are we having fun yet?

Pass the bullet. I was conflating threads in my response. You are right and I am wrong.

In my defense, I am frequently stupid.


ETA: In response to your question about having fun: Yes.

 

[Toontown]

Pass the bullet. I was conflating threads in my response. You are right and I am wrong.

In my defense, I am frequently stupid.


ETA: In response to your question about having fun: Yes.

Ah. At last we agree on something. The semantic log jam has broken.

Just kidding. It has been my experience that most semantic log jams are not so easily dislodged

 

[Jabba]

- I think that I can essentially prove immortality using Bayesian statistics.
- If this belongs in a different thread, or has already been done, please let me know. Otherwise, I'll present my case here.
--- Jabba

- Note that I said "essentially prove." I've never said that I can prove immortality.

 

[Filippo Lippi]

- I think that I can essentially prove immortality using Bayesian statistics.
- If this belongs in a different thread, or has already been done, please let me know. Otherwise, I'll present my case here.
--- Jabba

"Essentially" is not a synonym of "not."

My opinion of you remains unchanged.

 

[Jabba]

You need to define "each of us humans" and "a
life." It is not clear what these mean under the non-scientific
(unscientific?) hypothesis.

Jay,
- Each of us humans” means “each individual human consciousness/awareness.” “A life” means the earthly expression of that consciousness. Hopefully, that communicates and is sufficient.

 

[abaddon]

- Note that I said "essentially prove." I've never said that I can prove immortality.

In other words, you can't prove it at all.

 

[MRC_Hans]

- Note that I said "essentially prove." I've never said that I can prove immortality.

However, you couldn't even essentially prove it. In fact you couldn't even make it probable.

Hans

 

[Toontown]

The following are hills I am not willing to die on:

Here's the best you could have hoped for, Jabba.

The Four Stages of Paradigm Acceptance

Decade 1: Hostile opposition

Decade 2: Denial

Decade 3: Indifference

Decade 4: Waved off as self-evident

There is no "Decade 5: Profit"

There is no upside here. If your proposition is true, you're casting pearls before swine. If false, you're throwing up airballs.

It is unlikely that anything truly paradigm-shifting has ever been presented in this forum, or ever will be. In fact, it would be a mistake to attempt it.

Your idea is not new. I first encountered the gist of the idea, presented differently, in a philosophy book published @1933 (no, I'm not that old). The author devoted a page to the idea, found it fascinating, and correctly deemed it unprovable. I remember because my interest was piqued, having already had some thoughts along similar lines.

I do agree that the finite uniqueness assumption is vulnerable. I disagree with your linking the finite uniqueness assumption to "the scientific model", whatever that is. If such linkage truly exists, and you disprove the finite uniqueness assumption, then "the scientific model" would presumably go down with it. That won't happen, in part because no such absolute linkage exists. Nor will centuries of scientific research and observation be so easily overthrown.

If there is a 'scientific' way to unstack the apparent infinite odds a finitely unique "you" must have beaten, then it lies on a different path from the one you chose. For starters you need to get science on your side.

 

[Jabba]

- So for now, I'll focus on what I think are your three major objections:
1. That my current existence is so improbable -- given hypothesis "A" -- is NOT evidence against that hypothesis.
2. My hypotheses are too poorly defined to be dealt with by Bayesian inference, taken seriously or, understood.
3. The probability numbers that I propose are “off the wall” at best…

...Yes. Your priors are off the wall. The prior probability of a hypothesis is the probability of the hypothesis given all the available background information, not two numbers picked out of thin air...

Jay,
- “Out of thin air" also needs defining. Here, I’m taking you to mean that I have no RATIONAL basis for my probability choices. If that’s the case, I’m happy to provide the rationals for my choices.

 

[Jabba]

...Furthermore, if your hypotheses concern the lives of all humans, then how can you justify only considering as data the likelihood of only your life?

Jay,
- It seems to me that this is the only significant objection to my argument. I THINK that I have an effective answer to this objection, but like before, it will take me a while to compose.

 

[Filippo Lippi]

Ask the Whangers what they think

 

[Agatha]

Jay,
- “Out of thin air" also needs defining. Here, I’m taking you to mean that I have no RATIONAL basis for my probability choices. If that’s the case, I’m happy to provide the rationals for my choices.

It's almost as if I hadn't asked you for your rationale - to justify your numbers on both sides of your equation - several times already.

No, wait, I did. So it's good to hear you're happy to provide these justifications, I hope you don't delay too long.

 

[Akhenaten]

Jay,
- “Out of thin air" also needs defining. Here, I’m taking you to mean that I have no RATIONAL basis for my probability choices. If that’s the case, I’m happy to provide the rationals for my choices.

This century?

 

[Akhenaten]

- Note that I said "essentially prove." I've never said that I can prove immortality.