The Interpreter; Bayes Theorem; Nephites and Mayans

[_honorentheos]

Hi Res, based on the comments on The Interpreter's site, I think supporters of the paper would take issue with the wording of #2. At least some are determined to argue the paper is a direct response to Coe so the decision to rely on The Maya for comparison is fundamental to the exercise. But even on those grounds, the bias appears to come from the authors chosing which items to compare based on finding any connection, no matter how loose the comparison may be. So perhaps #2 could be reformulated to address the selection process in a way that maintains their stated aim was to compare The Maya with the Book of Mormon?

[_Lemmie]

Hi Res, based on the comments on The Interpreter's site, I think supporters of the paper would take issue with the wording of #2. At least some are determined to argue the paper is a direct response to Coe so the decision to rely on The Maya for comparison is fundamental to the exercise. But even on those grounds, the bias appears to come from the authors chosing which items to compare based on finding any connection, no matter how loose the comparison may be. So perhaps #2 could be reformulated to address the selection process in a way that maintains their stated aim was to compare The Maya with the Book of Mormon?

Good point, but if they maintain their stated aim, it must be noted that they lose the independence, as i noted in my previous post:

Let me emphasize, all 147 statements B, in all 147 likelihood ratios, are coming from a single author, as he is discussing a single topic.

There is no way any reasonable person could conclude that all 147 statements are completely independent from each other, to justify multiplying all 147 LR by each other to change the odds. It is not rational.

by the way, honor, I've been reading your posts on the Interpreter site and it's impressive. You are stating your points extremely well! It's a little disappointing that Brant Gardner sort of concedes your points but can't go so far as to look like he disagrees with the authors.

[_Res Ipsa]

Large as those effects are, I think the biggest problem with the paper is still an enormous bias in favor of weak evidence against fiction, treating all the "hits", however weak, as independent random guesses with moderately low probabilities.

Multiplying so many smallish probabilities together is what makes the final probability come out so low in the end, but it's ridiculously illogical. If the Book of Mormon is fake, it's still trying to portray an ancient society consistently, based on substantial knowledge of at least Biblical ancient societies. It's also trying to portray some kind of American geography (whether North or Meso). So there aren't just a few overlooked correlations among some of the hits: all those 100+ features are massively correlated, such that they really only represent a handful of independent decisions on the part of a fictional writer, at most.

If the Dales are allowed to get away with multiplying 10% a hundred times, then they can afford to laugh off every other criticism, even ones severe enough to boost their final chance of fraud by huge factors. The incredibly tiny product of a hundred small factors will still leave them with a chance so small there's little practical difference between it and zero.

What kicks the legs out from under them is demolishing that big multiplication. Their hits are highly correlated, so their probabilities are not independent. That's what cuts this paper down to the point where the remaining problems finish it off.

I didn’t include that in my list because I have no idea how accounting for the lack of independence affects the bottom line. If I understand Lemmie correctly, it is possible to adjust the math to account for dependence. If that’s the case, even an example of how the improper method could affect the bottom line would be, in my opinion, more persuasive than just saying that the method was improper.

[_Res Ipsa]

Hi Res, based on the comments on The Interpreter's site, I think supporters of the paper would take issue with the wording of #2. At least some are determined to argue the paper is a direct response to Coe so the decision to rely on The Maya for comparison is fundamental to the exercise. But even on those grounds, the bias appears to come from the authors chosing which items to compare based on finding any connection, no matter how loose the comparison may be. So perhaps #2 could be reformulated to address the selection process in a way that maintains their stated aim was to compare The Maya with the Book of Mormon?

Hi Honor,

Number 2 isn’t a criticism of using the two books. The criticism is that the criterion can never affect evidence supporting the Book of Mormon as history, as by definition a hit will always appear in both books. A miss could appear in both books — the horse is a good example. The Maya explicitly states that horses were extinct. But there could be many more missed that are mentioned in only one of the two books. Analytics mentioned jade as an example.

The authors, in my opinion, recognized this bias but failed to fully account for it, by violating their own criterion and including other criticisms by Coe that are not expressly stated in the Maya. Had they not done so, the list of evidence would be so ridiculously short that the bias would have been glaringly apparent.

The authors say the looked at every factual statement in the Book of Mormon and determined whether it was addressed in the Maya. What they omitted was negative evidence in the form of facts stated in one source but not the other. This contributed to them finding many more hits than misses.

[Guest]

Hi Res, based on the comments on The Interpreter's site, I think supporters of the paper would take issue with the wording of #2. At least some are determined to argue the paper is a direct response to Coe so the decision to rely on The Maya for comparison is fundamental to the exercise. But even on those grounds, the bias appears to come from the authors chosing which items to compare based on finding any connection, no matter how loose the comparison may be. So perhaps #2 could be reformulated to address the selection process in a way that maintains their stated aim was to compare The Maya with the Book of Mormon?

Good point, but if they maintain their stated aim, it must be noted that they lose the independence, as i noted in my previous post:

Let me emphasize, all 147 statements B, in all 147 likelihood ratios, are coming from a single author, as he is discussing a single topic.

Hi Lemmie, that makes sense on so far as I Intuit what you are saying and suspect when the dust settles this will be recognized as the fatal flaws in the methodology that leads to the illogical results. You're speaking above my level, though, so I end up taking that more on faith at this point and expect it gets dismissed by the supporters of the paper on the same grounds. Were nothing more said on the subject I think the legacy of this paper will be divided. Which isn't bad per see, and there are many issues with Mormonism where opinions divide based on allegence to the Church so what's one more?

There is no way any reasonable person could conclude that all 147 statements are completely independent from each other, to justify multiplying all 147 LR by each other to change the odds. It is not rational.

While I tend to agree I suspect reality is most of us aren't up to seeing how self evident the issues are but end up viewing the use of Bayes in this case as a battle between experts. So long as those more expert in the tools appear divided it gives the impression to those who hild differing views that this isn't due to objective issues. I think this is a hurdle for the critique of the paper that sounds like it could be overcome.

[/quite]by the way, honor, I've been reading your posts on the Interpreter site and it's impressive. You are stating your points extremely well! It's a little disappointing that Brant Gardner sort of concedes your points but can't go so far as to look like he disagrees with the authors.[/quote]Thanks, Lemmie I appreciate the compliment especially good been your knowledge on the overall topic. I think Brant is in the uncomfortable position of defending see mething which he may agree with in principle but not in detail. My suspicion is he'd rather not be doing so which leads me to wonder I'd he's doing so out of obligation as a moderator for the comments section or had some other role to play regarding which it would be very interesting to know more. Who knows. He certain knows when there is no reasonable, defensible reply and therefore doesn't.

[_Res Ipsa]

Honor, your point on battle of the experts is a good one. I think the independence issue can be explained in a way that avoids that problem.

[Guest]

Hi Res, based on the comments on The Interpreter's site, I think supporters of the paper would take issue with the wording of #2. At least some are determined to argue the paper is a direct response to Coe so the decision to rely on The Maya for comparison is fundamental to the exercise. But even on those grounds, the bias appears to come from the authors chosing which items to compare based on finding any connection, no matter how loose the comparison may be. So perhaps #2 could be reformulated to address the selection process in a way that maintains their stated aim was to compare The Maya with the Book of Mormon?

Hi Honor,

Number 2 isn’t a criticism of using the two books. The criticism is that the criterion can never affect evidence supporting the Book of Mormon as history, as by definition a hit will always appear in both books. A miss could appear in both books — the horse is a good example. The Maya explicitly states that horses were extinct. But there could be many more missed that are mentioned in only one of the two books. Analytics mentioned jade as an example.

The authors, in my opinion, recognized this bias but failed to fully account for it, by violating their own criterion and including other criticisms by Coe that are not expressly stated in the Maya. Had they not done so, the list of evidence would be so ridiculously short that the bias would have been glaringly apparent.

The authors say the looked at every factual statement in the Book of Mormon and determined whether it was addressed in the Maya. What they omitted was negative evidence in the form of facts stated in one source but not the other. This contributed to them finding many more hits than misses.

I'm probably missing something, but it seems to me the issue with this is in how they chose to use The Maya that led to what you noted in two. Had they simply distilled out each point in the book that represented a characteristic of Maya society and culture and then assessed what the Book of Mormon had to say, or didn't say, on the subject then there were be multiple items that appear in one but not the other while preserving the position being used to dismiss this issue that states the intent was a comparison with Coe's own scholarship or statements about the Maya on which the Book of Mormon is being mapped.

[_honorentheos]

Honor, your point on battle of the experts is a good one. I think the independence issue can be explained in a way that avoids that problem.

That's my impression as well. I hope to point out from the layperson's view that this would be worth doing.

[_Lemmie]

I didn’t include that in my list because I have no idea how accounting for the lack of independence affects the bottom line. If I understand Lemmie correctly, it is possible to adjust the math to account for dependence. If that’s the case, even an example of how the improper method could affect the bottom line would be, in my opinion, more persuasive than just saying that the method was improper.

Except that without the independency the entire multiplication collapses, which is the ONLY reason they get their result.

Suppose one element B generates the likelihood ratio .5. if the probability the second element occurs with the first is 1, then the likelihood ratio for elements 1 and 2 is not .5 x .5, it's .5 x 1.

Suppose every additional element is equally likely to occur with the first one, then the odds are not 10 ^ -131, (or whatever they got)

they are 0.5 x (1^146) = 0.5 x 1 = 0.5.

That's the extreme if all items are fully dependent.

Then posterior odds go from 1 billion to one against historicity, to one-half billion to one, still against historicity, instead of the billion billion billion, etc. odds in favor of Book of Mormon TRUE.

It destroys the whole argument to remove dependency.

[_Lemmie]

While I tend to agree I suspect reality is most of us aren't up to seeing how self evident the issues are but end up viewing the use of Bayes in this case as a battle between experts. So long as those more expert in the tools appear divided it gives the impression to those who hild differing views that this isn't due to objective issues. I think this is a hurdle for the critique of the paper that sounds like it could be overcome.

I take your point, but "appearing" to be divided is the real issue.

It gets frustrating when one set of "experts" only posts on the Interpreter with their lack of peer review, and the other "experts" support every theoretical comment with mathematical support that is fully and totally settled as statistical knowledge, as well as comments from actual peer reviewed work.

That's why I don't make math comments on the Interpreter site. What would be the point?