[Merged] Immortality & Bayesian Statistics

[wollery]

It's very simple.

Your basic hypothesis is poorly defined in scientific terms. because of this the complement is poorly defined.

However, it's worse than that. In Bayesian terms the complement should have specific measurable qualities, in exactly the same way that the hypothesis should have specific measurable qualities. In your version of the Bayesian equation neither has measurable qualities, let alone specific ones, and there's no measurable data that can be used by which to measure the output.

Your parameters are utterly model dependent, and you are merely making wild guesses at their values with no possible way of knowing what they might actually be.

In summary, your approach is unscientific, your hypothesis is poorly defined, the complement to your hypothesis is almost completely undefined, your assumptions are baseless and your application of Bayes theorem is worthless.

Does that help?

 

[shuttlt]

I've read most of the thread and I've been trying to get my head around Jabba's starting point in all this to unpick his thinking. This is what I'm thinking right now...

His "scientific" universe (universe A) is without purpose or meaning. His complementary universe (universe B) is one where things have meaning and therefore happen for a purpose. He figures that if he lives in universe B then the odds of him being born were always 1, since his life is meaningful and he has a purpose in this universe. In universe A the odds of him having been born, when viewed from even a few hundred years ago, are so small that we'll just pretend they are 0.0000000000000000000000000000000000000001 and not worry to much about the actual odds. Since Jabba has been born he goes off and does Bayes with the above numbers.

Is this correct?

 

[abaddon]

I've read most of the thread and I've been trying to get my head around Jabba's starting point in all this to unpick his thinking. This is what I'm thinking right now...

His "scientific" universe (universe A) is without purpose or meaning. His complementary universe (universe B) is one where things have meaning and therefore happen for a purpose. He figures that if he lives in universe B then the odds of him being born were always 1, since his life is meaningful and he has a purpose in this universe. In universe A the odds of him having been born, when viewed from even a few hundred years ago, are so small that we'll just pretend they are 0.0000000000000000000000000000000000000001 and not worry to much about the actual odds. Since Jabba has been born he goes off and does Bayes with the above numbers.

Is this correct?

Probably.

 

[Jabba]

It's very simple.

Your basic hypothesis is poorly defined in scientific terms. because of this the complement is poorly defined...

Wollery,
- I don't understand your criticism.
- "A" is the hypothesis that each of us has but one, short, life to live -- at most. "Non-A" is simply that "A" is wrong (at least in some respect).
- Specifically, for which term (or terms) do you need a "scientific" definition?
--- Jabba

 

[Hokulele]

Wollery,
- I don't understand your criticism.
- "A" is the hypothesis that each of us has but one, short, life to live -- at most. "Non-A" is imply that "A" is wrong (at least in some respect).
- Specifically, for which term (or terms) do you need a "scientific" definition?
--- Jabba

So do you accept that "Non-A" is "We were never born", as that would be a valid result if "A" were wrong?

Can you see why "Non-A" needs to be defined more strictly?

 

[carlitos]

What the **** do stats have to do with immortality?

 

[godless dave]

Wollery,
- I don't understand your criticism.
- "A" is the hypothesis that each of us has but one, short, life to live -- at most. "Non-A" is simply that "A" is wrong (at least in some respect).
- Specifically, for which term (or terms) do you need a "scientific" definition?
--- Jabba

As Hokule points out, non-A could mean each of us has no life to live.

And A is not specific enough - you have to define "short".

 

[Hokulele]

Yeah, I should have worded the Non-A as "We do not exist", which demonstrates even more clearly why precise definitions are important.

 

[Humots]

Wollery,
- I don't understand your criticism.
- "A" is the hypothesis that each of us has but one, short, life to live -- at most. "Non-A" is simply that "A" is wrong (at least in some respect).
- Specifically, for which term (or terms) do you need a "scientific" definition?
--- Jabba

The negation of a (for the sake of argument) specific statement is not automatically specific.

If Non-A is simply that "A" is wrong in some respect, then it could be that

1. "each of us has but N, short, lives to live (where N is any finite number)"
2. "each of us has but a finite number of short, lives, to live that is a different number for each person"
3. "only one of us is immortal, all the rest have but one, short, life to live"
4. "all but N of us have but one, short, life to live, the other N are immortal (where N is any finite number)"
5. "I am immortal, and all the rest of you have but one, short, life to live"
   .
   .
   .

Non-A is not well-defined or specific.

 

[Hokulele]

The negation of a (for the sake of argument) specific statement is not automatically specific.



If Non-A is simply that "A" is wrong in some respect, then it could be that

1. "each of us has but N, short, lives to live (where N is any finite number)"
2. "each of us has but a finite number of short, lives, to live that is a different number for each person"
3. "only one of us is immortal, all the rest have but one, short, life to live"
4. "all but N of us have but one, short, life to live, the other N are immortal (where N is any finite number)"
5. "I am immortal, and all the rest of you have but one, short, life to live"
   .
   .
   .

Non-A is not well-defined or specific.

Hell, even solipsism is "Non-A".

"There is no 'us'."

 

[Jabba]

I've read most of the thread and I've been trying to get my head around Jabba's starting point in all this to unpick his thinking. This is what I'm thinking right now...

His "scientific" universe (universe A) is without purpose or meaning. His complementary universe (universe B) is one where things have meaning and therefore happen for a purpose. He figures that if he lives in universe B then the odds of him being born were always 1, since his life is meaningful and he has a purpose in this universe. In universe A the odds of him having been born, when viewed from even a few hundred years ago, are so small that we'll just pretend they are 0.0000000000000000000000000000000000000001 and not worry to much about the actual odds. Since Jabba has been born he goes off and does Bayes with the above numbers.

Is this correct?

Shuttlt,

- Interesting analysis. It does seem to me that science cannot (currently) accept any kind of ultimate "meaning" (or purpose) to reality -- that would be "magical."
- But then, I don't think that my particular "scientific" hypothesis ("A") that "we each have but one, short, life to live, at most," necessarily excludes some sort of ultimate meaning...
- I'm just claiming that if that particular hypothesis is true, the probability of any particular person (self) existing right now is esentially zero.
- And since, given that scientific hypothesis, the probability that I would exist right now is essentially zero, I wonder -- does my current existence have something to say about the liklihood of that scientific hypothesis being true? Certainly, if the probability of me existing right now -- given that scientific hypothesis -- was actually zero, the fact that I do exist right now would prove that hypothesis wrong...

- There is much more to discuss about the logic here -- but hopefully, the above will give you and I a good jumping off place.

--- Jabba

 

[Jabba]

The negation of a (for the sake of argument) specific statement is not automatically specific.

If Non-A is simply that "A" is wrong in some respect, then it could be that

1. "each of us has but N, short, lives to live (where N is any finite number)"
2. "each of us has but a finite number of short, lives, to live that is a different number for each person"
3. "only one of us is immortal, all the rest have but one, short, life to live"
4. "all but N of us have but one, short, life to live, the other N are immortal (where N is any finite number)"
5. "I am immortal, and all the rest of you have but one, short, life to live"
   .
   .
   .

Non-A is not well-defined or specific.

Humots,
- Non-A includes all sorts of possibilities, but it's entirely specific about what isn't possible -- and that's all we need in order for the logic to work.
--- Jabba

 

[Mashuna]

Humots,
- Non-A includes all sorts of possibilities, but it's entirely specific about what isn't possible -- and that's all we need in order for the logic to work.
--- Jabba

What does the set, "each of us" include? Everyone alive now, everyone who's ever been alive, every living entity (animals, insects, plants), individual sperm?

 

[godless dave]

Shuttlt,

- Interesting analysis. It does seem to me that science cannot (currently) accept any kind of ultimate "meaning" (or purpose) to reality -- that would be "magical."
- But then, I don't think that my particular "scientific" hypothesis ("A") that "we each have but one, short, life to live, at most," necessarily excludes some sort of ultimate meaning...
- I'm just claiming that if that particular hypothesis is true, the probability of any particular person (self) existing right now is esentially zero.

Well there's your problem. Very improbable is not the same as zero; not even close.

 

[Loss Leader]

- And since, given that scientific hypothesis, the probability that I would exist right now is essentially zero, I wonder -- does my current existence have something to say about the liklihood of that scientific hypothesis being true? Certainly, if the probability of me existing right now -- given that scientific hypothesis -- was actually zero, the fact that I do exist right now would prove that hypothesis wrong..

There was never a zero probability that you would exist. Nor is there even a concept called "essentially zero." Something is either zero or it isn't.

Before your conception, there was a very, very low probability that you would exist. Heck, at the moment of conception, there were some 250,000,000 other sperm vying with the one lone halpoid that would become you.

However, what was the chance that, at the moment of your conception, some sperm would win? It was probably close to 1:1. What was the chance in 1800 that humans would still exist in 2012? Probably very high. What was the chance 2 billion years ago that there would eventually be sentient life on this planet? I don't know, but it wasn't zero.

You are the product of some staggeringly low probability events. But ANYTHING THAT COULD HAVE HAPPENED would also have been the product of staggeringly low probability events.

There was a very, very low probability that I would park next to Elena Haber when I went to the movies on Christmas. But, I did. The amazing thing is that I have no idea who Elena Haber is. I've never met her before, and I didn't meet her at the movies. I had to park next to someone, though. There was a 100% chance I would park next to somebody.

Your mistake, as has been pointed out over and over again, is in thinking that low probability events don't happen. Low probability events happen all the time. They're happening right now. It doesn't make you special and it doesn't require either purpose or design.

 

[Humots]

Humots,
- Non-A includes all sorts of possibilities, but it's entirely specific about what isn't possible -- and that's all we need in order for the logic to work.
--- Jabba

No, that is not all we need in order for the logic to work. Non-A needs to be defined as what it is, not what it isn't.

Since A and non-A are binary events, then P(A) = 1 - P(non-A). But that requires that the probabilities of all the possibilities included in non-A sum to a value that is consistent with P(A).

We cannot simply define an arbitrary probability for A and then just say P(non-A) must be 1 - P(A). That is just pulling values out of the air, which you seem to feel perfectly free to do.

Let A = "I owe you $1.00". Then non-A is "It is not true that I owe you $1.00". That is specific that I don't owe you $1.00, but what does it say that we can define any probability for?

Do I owe you $1.01? Do I owe you $1000? Do you owe me $0.50? Do you owe me $1,000,000,000.00? Do I owe you nothing? Do you owe me nothing? All of these are included in "I don't owe you one dollar".

But it is less likely that you owe me $1,000,000,000.00 than that you owe me nothing, or I owe you nothing. I don't recall having $1,000,000,000.00 to lend to anyone.

 

[Humots]

Hell, even solipsism is "Non-A".

"There is no 'us'."

Indeed. That is covered by
6. "I exist and the rest of you do not."

 

[Brian-M]

Wollery,
- I don't understand your criticism.
- "A" is the hypothesis that each of us has but one, short, life to live -- at most. "Non-A" is simply that "A" is wrong (at least in some respect).
- Specifically, for which term (or terms) do you need a "scientific" definition?
--- Jabba

The criticism being made is that your terms are very vague. For example...

What is meant by "short"? How long would that be? How is this determined? What fundamental criteria do you use to distinguish short from long?

Who is "us"? Those who exist now, those who have or will exist at any point in time, or all those who could have possibly existed, including those who didn't?

By using the word "one" do you use this to mean that we live a life instead of not living at all, or do you use this to mean that we don't live multiple lives?

What function do the seemingly superfluous words "but" and "at most" serve in this statement? Are they intended to clarify your meaning in some fashion, or are they merely decorative?

 

[dlorde]

Statistics requires data, e.g. numbers, not vague hand-waving.

 

[Loss Leader]

The crux of his argument is that "close to zero" means the same thing as "zero."