Cont: Proof of Immortality, V for Very long discussion

[Waterman]

Waterman,
- It looks backwards to establish the probability of an hypothesis, given that a certain event has occurred. We're not trying to establish the probability of an event; we're trying to establish the probability of the hypothesis, given new information.

Waterman,
- It looks backwards to establish the probability of an hypothesis, given that a certain event has occurred. We're not trying to establish the probability of an event; we're trying to establish the probability of the hypothesis, given new information.

Aah, I have learned something new. So to simplify an example. I know that Bob is rolling either a die with 6 sides (d6) or a die with 20 sides (d20).

Bob rolls a 5. This has a probability of 1:6 on a d6 and 1:20 on a d20

It is more likely that to be a d6 than a d20. The second roll is a 4. This increases the likelihood that it is a d6 but a d20 is still possible. The 3rd roll is a 7 that eliminates the d6 option it is 100% the d20 regardless of what the odds were after roll two.

I am sure there are formula for this but that is not really relevant yet. Before this can be applied a clear understanding of the relative probabilities if the two hypothesis is required.

What you have appear to have done is to declare that one of the hypotheses is impossible prior to putting it into the equation. Therefore stacking the deck in favor of the perferred hypothesis regardless of how unlikely.

So Bayes cannot be applied until a reasonable probability can be established for BOTH hypotheses. Current scientific theory is that a normally functioning human brain will have the property of self awareness and sense of continuity and growth.

If a healthy human brain exists the odds that it will have a property of self is 1:1. Do you disagree?

 

[Waterman]

<snip>
So Bayes cannot be applied until a reasonable probability can be established for BOTH hypotheses. <snip>

Ok in reviewing the information again I see that you are claiming that you are using the ~ notation to indicate that something other than the base claim. This would mean that for it to be meaningful that the probability for the base case MUST be SOLID as it drives everything.

You've stacked the deck in the ~ favor by declaring the base claim as impossible. But you have not demonstrated this.

May I count potential meals of some items that do not exist in the same place and time? If I can there must be a near infinity of them. The odds that I would have a particular meal this evening was astronomical. It must have a non-mundane explanation. Or am I abusing the logic in some way.

If I am wrong, how can you claim:

To count potential people from some eggs and sperm that do not exist in the same place and time?

 

[Filippo Lippi]

By the way, I ran across this by accident.

It's Jabba giving his "virtually proving immortality through Bayesian statistics" spiel at a statistics forum in early 2015, and people, it ain't pretty.

Among other things, they point out a mistake I don't recall Jabba making in this thread. Possibly he did take away one lesson.

Has anyone seen and/or linked to this before?

Contains the suggestion that Jabba's a 'bot... could be

 

[Jabba]

Ok in reviewing the information again I see that you are claiming that you are using the ~ notation to indicate that something other than the base claim. This would mean that for it to be meaningful that the probability for the base case MUST be SOLID as it drives everything.

You've stacked the deck in the ~ favor by declaring the base claim as impossible. But you have not demonstrated this.

May I count potential meals of some items that do not exist in the same place and time? If I can there must be a near infinity of them. The odds that I would have a particular meal this evening was astronomical. It must have a non-mundane explanation. Or am I abusing the logic in some way.

If I am wrong, how can you claim:

To count potential people from some eggs and sperm that do not exist in the same place and time?

Waterman,
- If one were frozen and later combined with the other, wouldn't this produce a new, different, original person?

 

[Filippo Lippi]

Waterman,
- If one were frozen and later combined with the other, wouldn't this produce a new, different, original person?

Why are you pretending that the statisticians didn't hand you your arse on a plate? Why didn't you have the intellectual honesty to tell us that you had engaged in that debate?

 

[zooterkin]

Waterman,
- If one were frozen and later combined with the other,

In which case, they would be existing in the same time and place.

wouldn't this produce a new, different, original person?

Who's saying they wouldn't?

 

[jond]

Waterman,
- If one were frozen and later combined with the other, wouldn't this produce a new, different, original person?

Could you, today, freeze your father's sperm and cleopatra's eggs? Were their sperm and eggs frozen before they died? No, and no. Which means that a union of the two is impossible.

Oh, and: why do you continue to insist that the likelihood of your brain is different from the likelihood of your self under H?

 

[Belz...]

Bob rolls a 5.

AC is -8 so it's a miss.

 

[JayUtah]

If one were frozen and later combined with the other, wouldn't this produce a new, different, original person?

Yes, any time a human is produced, by whatever means, using constituents however preserved or handled, you would get a new organism. In materialism, the organism is synonymous with person. How that person's sense of self developed would depend on his environment over time. But he would have one, simply by virtue of being alive.

It's apparent where you're going, but under H most of your trumped up magic simply isn't operative. Under H the sense of self is an emergent property. It isn't anything like a soul that can be said to have any sort of static or independent existence. Words like "new, different, original person" seem to be aimed at expressing materialism in soulish language.

 

[JayUtah]

Why are you pretending that the statisticians didn't hand you your arse on a plate? Why didn't you have the intellectual honesty to tell us that you had engaged in that debate?

And this isn't trivial. Many times in both this and the Shroud thread, Jabba assured us that his argument would get a toehold anywhere but on JREF/ISF. Turns out that's a lie. Turns out it's not just those annoying skeptics who can't see Jabba's brilliance. Turns out a lot of people apparently can't. What about that, Jabba? Since part of your argument claims your critics here are extraordinarily stubborn, can you explain how others elsewhere seem to arrive at the same conclusions as ISF regarding the validity of your proof? Is the whole world against you?

 

[Jabba]

Ok in reviewing the information again I see that you are claiming that you are using the ~ notation to indicate that something other than the base claim. This would mean that for it to be meaningful that the probability for the base case MUST be SOLID as it drives everything.

You've stacked the deck in the ~ favor by declaring the base claim as impossible. But you have not demonstrated this.

May I count potential meals of some items that do not exist in the same place and time? If I can there must be a near infinity of them. The odds that I would have a particular meal this evening was astronomical. It must have a non-mundane explanation. Or am I abusing the logic in some way.

If I am wrong, how can you claim:

To count potential people from some eggs and sperm that do not exist in the same place and time?

- Yes. But, there are numerous situations in which the possible outcomes are so numerous (say, a million) that any specific outcome will be EXTREMELY unlikely. In such a case, in order for its unlikelihood to 'count,' there must be something about this particular outcome that 'sets it apart' from the remaining million (or at least, most of the remaining million).

 

[Hokulele]

- Yes. But, there are numerous situations in which the possible outcomes are so numerous (say, a million) that any specific outcome will be EXTREMELY unlikely. In such a case, in order for its unlikelihood to 'count,' there must be something about this particular outcome that 'sets it apart' from the remaining million (or at least, most of the remaining million).

No.

 

[abaddon]

- Yes. But, there are numerous situations in which the possible outcomes are so numerous (say, a million) that any specific outcome will be EXTREMELY unlikely. In such a case, in order for its unlikelihood to 'count,' there must be something about this particular outcome that 'sets it apart' from the remaining million (or at least, most of the remaining million).

But what is the probability of an outcome? 1. It doesn't matter how many possible outcomes you can imagine.

 

[JayUtah]

- Yes. But, there are numerous situations in which the possible outcomes are so numerous (say, a million) that any specific outcome will be EXTREMELY unlikely.

Texas sharpshooter fallacy.

In such a case, in order for its unlikelihood to 'count,' there must be something about this particular outcome that 'sets it apart' from the remaining million (or at least, most of the remaining million).

There has to be something ahead of the data collection that sets it apart. You collect the data, announce that it's what you were looking for, and claim it to be improbable.

 

[Mojo]

- Yes. But, there are numerous situations in which the possible outcomes are so numerous (say, a million) that any specific outcome will be EXTREMELY unlikely. In such a case, in order for its unlikelihood to 'count,' there must be something about this particular outcome that 'sets it apart' from the remaining million (or at least, most of the remaining million).

Jabba, if we were to take H as being the hypothesis that you have a body and an independently existing soul, then your existence is at least as unlikely as it is under the hypothesis that you have a body and a consciousness that is produced by that body. Your argument is designed to disprove whatever hypothesis it is used to attack in favour of whatever this leaves as ~H.

ETA: But if it is used to compare the two hypotheses I have just mentioned, it cannot possibly favour the former over the latter.

 

[Mojo]

- Yes. But, there are numerous situations in which the possible outcomes are so numerous (say, a million) that any specific outcome will be EXTREMELY unlikely. In such a case, in order for its unlikelihood to 'count,' there must be something about this particular outcome that 'sets it apart' from the remaining million (or at least, most of the remaining million).

The only thing that sets the observed outcome apart from all the other possible outcomes is that it is the one that has happened.

 

[Monza]

Jabba agreed to withdraw his map? Shucks, I missed that. Can anyone point me to it, if it's not too much trouble?

By the way, I ran across this by accident.

It's Jabba giving his "virtually proving immortality through Bayesian statistics" spiel at a statistics forum in early 2015, and people, it ain't pretty.

Among other things, they point out a mistake I don't recall Jabba making in this thread. Possibly he did take away one lesson.

Has anyone seen and/or linked to this before?

It's interesting that this was the assessment from one of the posters over there.

You posted here presumably because you have some awareness that you don't know much about probability. But you seem to be unwilling to listen to anything that any of us are saying about probability. Instead you just keep reiterating the same, obviously flawed, arguments. This has become a waste of time.

 

[zooterkin]

- Yes. But, there are numerous situations in which the possible outcomes are so numerous (say, a million) that any specific outcome will be EXTREMELY unlikely. In such a case, in order for its unlikelihood to 'count,' there must be something about this particular outcome that 'sets it apart' from the remaining million (or at least, most of the remaining million).

You mean like any regular conception?

 

[RoboTimbo]

- Yes. But, there are numerous situations in which the possible outcomes are so numerous (say, a million) that any specific outcome will be EXTREMELY unlikely. In such a case, in order for its unlikelihood to 'count,' there must be something about this particular outcome that 'sets it apart' from the remaining million (or at least, most of the remaining million).

Can you explain the Texas Sharpshooter fallacy to me? Maybe give some examples?

 

[Jabba]

Aah, I have learned something new. So to simplify an example. I know that Bob is rolling either a die with 6 sides (d6) or a die with 20 sides (d20).

Bob rolls a 5. This has a probability of 1:6 on a d6 and 1:20 on a d20

It is more likely that to be a d6 than a d20. The second roll is a 4. This increases the likelihood that it is a d6 but a d20 is still possible. The 3rd roll is a 7 that eliminates the d6 option it is 100% the d20 regardless of what the odds were after roll two.

I am sure there are formula for this but that is not really relevant yet. Before this can be applied a clear understanding of the relative probabilities if the two hypothesis is required.

1)What you have appear to have done is to declare that one of the hypotheses is impossible prior to putting it into the equation. Therefore stacking the deck in favor of the perferred hypothesis regardless of how unlikely.
- I didn't mean to "declare" that one was impossible; I meant to "argue" that one was (just about) impossible.


2)So Bayes cannot be applied until a reasonable probability can be established for BOTH hypotheses. Current scientific theory is that a normally functioning human brain will have the property of self awareness and sense of continuity and growth.
- In Bayes these are "prior probabilities," and I'm suggesting .99 for P(H), and .01 for P(~H). I accept that most of the scientific evidence supports the one lifetime hypothesis. I'm claiming, however, that such has not been proven, and that there is significant relevant information that has not -- yet -- been taken into account. That's where Bayes comes in.


3)If a healthy human brain exists the odds that it will have a property of self is 1:1. Do you disagree?
- No -- I agree.

Waterman,
#1. I didn't mean to "declare" that one was impossible; I meant to "argue" that one was (just about) impossible.
#2. In Bayes these are "prior probabilities," and I'm suggesting .99 for P(H), and .01 for P(~H). I accept that most of the scientific evidence supports the one lifetime hypothesis. I'm claiming, however, that such has not been proven, and that there is significant relevant information that has not -- yet -- been taken into account. That's where Bayes comes in.
#3. No -- I agree.