Interpretering Bayesian Analysis

[Philo Sofee]

kyzabee
You'll see me conclude at the end of this not that the Book of Mormon must be authentic, but only that my read of the evidence gives me reason for hope.

Only if you have used Bayes properly I would suggest. It's not about hope, it's about what the probabilities tell us our reaction needs to be concerning the theory that the Book of Mormon is authentic. Don't lose sight of the fact that it has nothing to do at arriving at either a positive or a negative, but arrive at what is justified, based on the proper use of Bayes Theorem. That is all that matters.
In Bayes, you have no right to arrive at a conclusion you wish is true. It must come out that way only after an accurate assessment of all the evidences and all the background of what we know... no preordained conclusions are allowed. This is not about faith, it is about probabilities.

[kyzabee]

Hi Philo!

"In the above case, quit worrying about Joseph Smith."

You'll have at least one more post that takes Joseph (and the witnesses) into account. The episodes follow a rough progression through the Book of Mormon, so it made sense to get Joseph and the introductory material out of the way first.

"And you can't ignore the former attempts of apologists to present authenticating evidences which have now been refuted either."

We'll see if we agree that they've been refuted.

"you cannot ignore that you honestly have no first ground for even supposing Nephites and Lamanites ever existed"

That's good, because I don't ignore the lack of potential artifacts, though I'd disagree that there's no ground for supposing Nephites and Lamanties, mostly because I also don't ignore everything that faithful scholars have had to say on the subject.

We can discuss that further once we actually get to that post, though, which will be Episode 14.

"The DNA is gonna be a lulu as well since that too is also part of our background knowledge which must be acknowledged and calculated, as well a the various General Authorities teachings of who the Lamanites were, where they originated, etc."

DNA will be coming next week, and learning about it was a fun process. And the potential errors of prophets, seers, and revelators will show up in Episode 18.

"...background knowledge..."

Since my estimated background knowledge (i.e., my initial prior) does no favors to the faithful position, my analysis agrees that the Book of Mormon has a pretty high hill to climb.

We could spend years talking about the content, though. I'm assuming you agree then that my intended approach is a decent starting place, even if others might instantiate it differently?

[Doctor CamNC4Me]

*a dramatic puff of smoke appears*

Wonderful! I look forward to you collaborating with the scholars here who’ve shown a keen interest in your work. Thank you for taking the time to set the table, as it were.

- Doc

[Dr Moore]

Welcome to the board, Dr. Rasmussen. It’s encouraging to see you here, and I do have some basic questions.

First though, I must confess to being caught surprised Dr. P would ever encourage any sort of engagement with me, as his final email signoff with me some months ago amounted to rather aggressively calling me a hypocrite because I chose to publicly hold him to his word (which by the way, I’ve essentially given up on, realizing that he actually does truly enjoy hearing his friends assign all sorts of negative names and labels to this message board, more so than he enjoys being a man of honor as the moderator and host of his blog. You can ask others for details…). But with a cash wager on the line, and knowing I’m good for the honor of a deal, I suppose Dr. P will stoop to my level - just this once - for a chance to get dat money. And as always, should the retired Dr. P choose to use some his newly discovered free time to
actually moderate his blog per the terms of our generous Interpreter donation deal, then I will be delighted to hold him up as a man of honor on that matter.

Basic questions:

1) How much of the negative evidence do you intend to treat with the creative flair and reaching as your positive evidentiary episodes? All of it? Really all of it? Will you take peer review on negative evidence missing from the analysis? Or do you prefer to pick which “ones” through “fives” you will consider?

2) Why have you chosen to poke fun and construct straw man arguments about something so elementary as establishing a baseline for statistical independence in an analytical body of work that depends wholly on statistical independence?

Second Amendment) If time travel for more than a few seconds is an astronomically unlikely scenario, then what does a single unresolved anachronism in the Book of Mormon do to the analysis?

3) What, if any, constructive criticism would cause you to hold back publication? As in, is there any amount of logical identification of garbage-in/garbage-out in your process that would matter enough to pull up and reconsider the entire approach in the name of rigorous scholarship, a la normal peer reviewing processes, as opposed to just putting stuff out there for people to chew on?

3a) If I offered to hire a current BYU professor
to publicly or privately provide a proper peer review of your applied statistics in advance of further publication, would you accept that assistance? (this would void other offers, but I believe would have tremendous benefit for lay readers of your work, and the added benefit of removing me - a pseudonymous stranger who apparently you mistrust implicitly - from the equation)

4) Why, if you believe in the ultimate merit of your approach, publish in Interpreter and not seek for a venue with wider distribution?

And to answer the question above, I do not believe there is merit in this course of study. While it may be fun to look at, and I know plenty of folks who have tried variations on this theme, proper application of the tools you’ve chosen cannot be made within the rules of the art. You can’t de-correlate Joseph from himself, neither can you control for the unknowable contents of Joseph’s mind, capacities, experiences, neuro-psychological makeup, his entire bricolage etc etc. At a macro level it’s the sharp shooter fallacy. Joseph started something that was unique, almost died but then became huge. So what? So did L. Ron Hubbard. This line is ultimately doomed by abundance of assumptions, biases and logical fallacy if performed without open, transparent and significant peer input by people on both sides of the underlying subjectives: assessment of historical ideas and facts, selection of control sets, relevant evidence to include and exclude, and what questions lead to meaningful vs irrelevant answers. But that’s just my opinion.

[kyzabee]

Greetings, good Dr. And really, I'm fine with just Kyler if people prefer.

There are fair questions underlying the ones you pose here, and I'll do my best to ignore the "why don't you stop beating your wife" framing to some of them.

"1) How much of the negative evidence do you intend to treat?"

All the major classes of evidence that I could think of I tried to work into the analysis, generally focusing on those that seemed to have gained traction in the internet age. I'm bound to have missed a bunch (Treasure Digging didn't make the list, for instance, as I realized when Billy brought it up this week, though that probabiy isn't independent with my analysis on prophetic imperfections), and I'm happy to take feedback on missing ones you feel would have some weight to them. Here's the full list of negative posts, in case you're curious:

2 - First Vision
4 - DNA
6 - Political Incorrectness
8 - Transoceanic Voyages
10 - King James Plagiarism/Errors
12 - Late War/View of the Hebrews Plagiarism
14 - Archaeological Anachronisms (collated a bunch of anachronisms here)
16 - Thematic Anachronisms (also another collation)
18 - Prophetic Imperfections
20 - Place-Name Plagiarism (i.e. Holley's map)
22 - Book of Abraham (collation)

It'll be interesting to see how people here feel about my treatment of the negative evidence. The goal was to try to treat it in a fairer and more complete fashion than critical treatments generally have for the positive evidence.

"2) Why have you chosen to poke fun and construct straw man arguments about something so elementary as establishing a baseline for statistical independence in an analytical body of work that depends wholly on statistical independence?"

If you're talking about my discussion of independence on Dan's blog, it's because 1) I'm a man who enjoys fun, and finds it an effective way to communicate, and 2) when it comes to determining the independence of DNA and Chiasmus, that's no straw-man--that's exactly how I'd approach that problem, and it would be no fun for anyone involved. And then I'd need to put forward a similar effort at least 230 more times to account for the various combinations of independent pieces of evidence.

"Second Amendment) If time travel for more than a few seconds is an astronomically unlikely scenario, then what does a single unresolved anachronism in the Book of Mormon do to the analysis?"

Keeping in mind that Book of Mormon anachronisms and alleged errors have had a tricky habit of being overturned as time passes, a single anachronism should probably be weighted quite a bit lower than 10^-20. As providing detailed estimates for every anachronism (or positive correspondence) is probably beyond mortal capacity, I generally followed the same method as the Dales for applying Bayes factors on either side (though I was a lot pickier about independence and other relevant matters). Despite the criticisms the Dales' received (including from me), their choice and method for applying Bayes factors wasn't one of them, and has its basis in secular Bayesian analysis. In that particular analysis, I also essentially neuter all of the positive evidence in an attempt to apply a fortiori reasoning. When I talked to Bruce Dale about it, he's like, "you can't do that, that's not fair to the Book of Mormon", and my reply was "it's totally not fair, and that's the point".

"3) What, if any, constructive criticism would cause you to hold back publication?"

At this point I'm committed to having my thoughts see the light of day, and I'm confident enough in their utility and willing enough to be wrong that I have no qualms in doing so. These are thought experiments, but I also learned a lot of potentially useful things from each look at the raw data, and those lessons learned wouldn't do anyone any good sitting around in my head.

"3a) If I offered to hire a current BYU professor to publicly or privately provide a proper peer review of your applied statistics in advance of further publication, would you accept that assistance?"

These have received a friendly review from people with the appropriate credentials, and I made quite a few changes on the basis of their feedback. At this point because all of the analyses are intertwined pretty heavily, they're essentially locked down beyond minor adjustments, so the time for additional peer review has unfortunately passed. If these were going in the journal proper I would've sought out a more hostile review, which gets us to your last question.

"4) Why, if you believe in the ultimate merit of your approach, publish in Interpreter and not seek for a venue with wider distribution?"

In the end, these are on the Interpreter blog because: 1) these are, in the end, amateur analyses, conducted with limited time and resources--I'm much more comfortable sharing these as blog episodes, and would've needed to put in a lot of refinement and collaboration with subject-matter experts to make them a matter of academic record; 2) I have no desire for fame. I would've been perfectly happy to just have these on my personal blog to chat about with my friends, but throwing them on Interpreter felt like a decent compromise between a public profile and complete obscurity; and 3) who knows, maybe they will reach primetime someday, in which case I'll have been glad to have them tested by you good folks before they hit the big tent.

"While it may be fun to look at..."

And can't that be reason enough for them to exist? Potentially interesting ideas packaged in a relatively novel approach, and one that's seemed to have generated some interesting dialogue?

"proper application of the tools you’ve chosen cannot be made within the rules of the art"

People like Carrier would seem to disagree, though I seem to be more interested in the collection and analysis of raw data than he's been to this point.

"You can’t de-correlate Joseph from himself, neither can you control for the unknowable contents of Joseph’s mind, capacities, experiences, neuro-psychological makeup, his entire bricolage"

I can put forward an argument that some of those faculties should be uncorrelated, the same way your nose length and your blood pressure would probably be uncorrelated. Again, all you need is an argument otherwise and I'll be happy to reconsider (and yes, if absolutely necessary, make numerous changes to 20+ blog posts while attempting to balance the rest of my life).

"At a macro level it’s the sharp shooter fallacy. Joseph started something that was unique, almost died but then became huge. So what? So did L. Ron Hubbard."

L. Ron was definitely huge. But in what way is he unexpected? I expect that his religious movement falls pretty well as expected on the distribution of other religious movements, and his books on the distribution of other fictional works. As you say, it's not enough for me to argue that the book's unique, I have to argue that it's different from what we would've had good reason to expect (and aligns in more ways than it should with what we'd expect from the ancient world). At least I'll sleep well knowing that if I fail in that, you'll be sure to let me know.

[drumdude]

I would hope you have listened to Mormon Expression’s podcast on how to build a trans-oceanic vessel. The details are important and I hope you don’t just gloss over them.

Basically, the likelihood that Nephi and his family accomplished what is said in the Book of Mormon is on the same order of magnitude as me building a Mars colony spaceship in my back yard.

I would give it a probability so low that it *alone* would easily counter all of your arguments in favor of the Book of Mormon being true.

[kyzabee]

"I would hope you have listened to Mormon Expression’s podcast on how to build a trans-oceanic vessel."

I haven't, though I worked off of Zelph on the Shelf's treatment, which I assume was similar. My guess is that Mormon Expressions similarly glossed over the wide variety of ways ancient societies engaged in trans-oceanic travel.

[drumdude]

"I would hope you have listened to Mormon Expression’s podcast on how to build a trans-oceanic vessel."

I haven't, though I worked off of Zelph on the Shelf's treatment, which I assume was similar. My guess is that Mormon Expressions similarly glossed over the wide variety of ways ancient societies engaged in trans-oceanic travel.

This suggests to me you haven’t read the relevant section of the Book of Mormon, if you’re comparing it to what we know about ancient seafaring peoples.

The Book of Mormon is very specific about building bellows, mining ore, smelting ore, and crafting tools, just as a means to begin to construct the ship. All which was accomplished in a matter of months by a couple men.

I am glad that your instinct on this one is to go with the “Nephi didn’t do anything particularly miraculous” apologetic approach rather than “God can make anything happen no matter how impossible.” At least we can discuss the details and compare them to actual ancient societies to see if they do indeed match up.

[Dr Moore]

5) Would you please explain why rare = true? I get that rare = rare, but how do you justify rare equals true?

[Dr Moore]

Greetings, good Dr. And really, I'm fine with just Kyler if people prefer.

There are fair questions underlying the ones you pose here, and I'll do my best to ignore the "why don't you stop beating your wife" framing to some of them.

"1) How much of the negative evidence do you intend to treat?"

All the major classes of evidence that I could think of I tried to work into the analysis, generally focusing on those that seemed to have gained traction in the internet age. I'm bound to have missed a bunch (Treasure Digging didn't make the list, for instance, as I realized when Billy brought it up this week, though that probabiy isn't independent with my analysis on prophetic imperfections), and I'm happy to take feedback on missing ones you feel would have some weight to them. Here's the full list of negative posts, in case you're curious:

2 - First Vision
4 - DNA
6 - Political Incorrectness
8 - Transoceanic Voyages
10 - King James Plagiarism/Errors
12 - Late War/View of the Hebrews Plagiarism
14 - Archaeological Anachronisms (collated a bunch of anachronisms here)
16 - Thematic Anachronisms (also another collation)
18 - Prophetic Imperfections
20 - Place-Name Plagiarism (i.e. Holley's map)
22 - Book of Abraham (collation)

It'll be interesting to see how people here feel about my treatment of the negative evidence. The goal was to try to treat it in a fairer and more complete fashion than critical treatments generally have for the positive evidence.

"2) Why have you chosen to poke fun and construct straw man arguments about something so elementary as establishing a baseline for statistical independence in an analytical body of work that depends wholly on statistical independence?"

If you're talking about my discussion of independence on Dan's blog, it's because 1) I'm a man who enjoys fun, and finds it an effective way to communicate, and 2) when it comes to determining the independence of DNA and Chiasmus, that's no straw-man--that's exactly how I'd approach that problem, and it would be no fun for anyone involved. And then I'd need to put forward a similar effort at least 230 more times to account for the various combinations of independent pieces of evidence.

"Second Amendment) If time travel for more than a few seconds is an astronomically unlikely scenario, then what does a single unresolved anachronism in the Book of Mormon do to the analysis?"

Keeping in mind that Book of Mormon anachronisms and alleged errors have had a tricky habit of being overturned as time passes, a single anachronism should probably be weighted quite a bit lower than 10^-20. As providing detailed estimates for every anachronism (or positive correspondence) is probably beyond mortal capacity, I generally followed the same method as the Dales for applying Bayes factors on either side (though I was a lot pickier about independence and other relevant matters). Despite the criticisms the Dales' received (including from me), their choice and method for applying Bayes factors wasn't one of them, and has its basis in secular Bayesian analysis. In that particular analysis, I also essentially neuter all of the positive evidence in an attempt to apply a fortiori reasoning. When I talked to Bruce Dale about it, he's like, "you can't do that, that's not fair to the Book of Mormon", and my reply was "it's totally not fair, and that's the point".

"3) What, if any, constructive criticism would cause you to hold back publication?"

At this point I'm committed to having my thoughts see the light of day, and I'm confident enough in their utility and willing enough to be wrong that I have no qualms in doing so. These are thought experiments, but I also learned a lot of potentially useful things from each look at the raw data, and those lessons learned wouldn't do anyone any good sitting around in my head.

"3a) If I offered to hire a current BYU professor to publicly or privately provide a proper peer review of your applied statistics in advance of further publication, would you accept that assistance?"

These have received a friendly review from people with the appropriate credentials, and I made quite a few changes on the basis of their feedback. At this point because all of the analyses are intertwined pretty heavily, they're essentially locked down beyond minor adjustments, so the time for additional peer review has unfortunately passed. If these were going in the journal proper I would've sought out a more hostile review, which gets us to your last question.

"4) Why, if you believe in the ultimate merit of your approach, publish in Interpreter and not seek for a venue with wider distribution?"

In the end, these are on the Interpreter blog because: 1) these are, in the end, amateur analyses, conducted with limited time and resources--I'm much more comfortable sharing these as blog episodes, and would've needed to put in a lot of refinement and collaboration with subject-matter experts to make them a matter of academic record; 2) I have no desire for fame. I would've been perfectly happy to just have these on my personal blog to chat about with my friends, but throwing them on Interpreter felt like a decent compromise between a public profile and complete obscurity; and 3) who knows, maybe they will reach primetime someday, in which case I'll have been glad to have them tested by you good folks before they hit the big tent.

"While it may be fun to look at..."

And can't that be reason enough for them to exist? Potentially interesting ideas packaged in a relatively novel approach, and one that's seemed to have generated some interesting dialogue?

"proper application of the tools you’ve chosen cannot be made within the rules of the art"

People like Carrier would seem to disagree, though I seem to be more interested in the collection and analysis of raw data than he's been to this point.

"You can’t de-correlate Joseph from himself, neither can you control for the unknowable contents of Joseph’s mind, capacities, experiences, neuro-psychological makeup, his entire bricolage"

I can put forward an argument that some of those faculties should be uncorrelated, the same way your nose length and your blood pressure would probably be uncorrelated. Again, all you need is an argument otherwise and I'll be happy to reconsider (and yes, if absolutely necessary, make numerous changes to 20+ blog posts while attempting to balance the rest of my life).

"At a macro level it’s the sharp shooter fallacy. Joseph started something that was unique, almost died but then became huge. So what? So did L. Ron Hubbard."

L. Ron was definitely huge. But in what way is he unexpected? I expect that his religious movement falls pretty well as expected on the distribution of other religious movements, and his books on the distribution of other fictional works. As you say, it's not enough for me to argue that the book's unique, I have to argue that it's different from what we would've had good reason to expect (and aligns in more ways than it should with what we'd expect from the ancient world). At least I'll sleep well knowing that if I fail in that, you'll be sure to let me know.

Okay. Asked and answered then. I won’t fault you for going out to market with a half baked cakr. Not sure how many customers there are for that product, but you are clear about the goal at any rate. Best of luck with the fun.