The Interpreter; Bayes Theorem; Nephites and Mayans

[_Gadianton]

A whole library of completely different stories about Mesoamerican Israelites would have to be historically authentic as well. But that's absurd.

right, I agree, the question is how might this point be communicated to those with impenetrable skulls. If you were a dale, how would you respond to your suggestion?

[_Gadianton]

Last weekend I looked over the Kass paper, I hadn't really looked at it before and it is a bit math heavy -- I'm sure I'm not the only one who hasn't read it, I know of at least two other people out there who haven't read it (lol) -- but there was one point I've gone back and forth; whether worth bringing up. Maybe it's been discussed before, but here it is.

It's been noted that the "barely worth a mention" has been abused by the authors to mean weak evidence. But to me, the verbal description of 2 as barely anything and 50 as case closed, suggests you'll never be in a position multiplying 12 50's or .02s together along with 37 2's or .5s. If a long string of LR's were a reasonable use case, then the framework would be inadequate, and they'd need to invent a new scale, such that barely worth mentionings can't so easily overpower a "case closed". In other words, the very fact that the Dales got the result they did from Kass suggests that either they abused the theory, or the theory is flawed. It would take 6 fully independent pieces of "barely worth anythings" to over power a "case closed" silver bullet. That doesn't seem right. So something is misunderstood about multiplying LRs, or it's possible independency in the real world is just that hard to come by and the authors assume you'd never get to 6.

Example: The scale Kass introduced seemed to mimic the Richter scale or decibel scale. Both use logs, so a 10 on the Richter scale is 10x the force of a 9. A 9 is 10x an 8. A 10 on the Richter scale is complete destruction while a 1, is "barely worth a mention". The analogy would be, if six tremors on the Richter Scale left greater devastation than a single 10, then the scale would have to be fixed so that doesn't happen -- so that it actually means something.

The same for decibels, if six room-level conversation caused the same hearing loss as one fighter jet passing 20 feet above your head, then they'd need to fix the scale until it meant something we can actually use meaningfully in the real world.

[_Dr Exiled]

i know kirk magleby's blog entry was already posted, but i think it bears a second look:

Posted by Captain Kirk at 3:18 PM , Tuesday, June 4, 2019

Las Vegas Odds

The father/son team of Bruce and Brian Dale, both Ph.D. engineers, published a sensational 110 page article in Interpreter on May 3, 2019 that uses Bayesian statistical analysis to demonstrate a) the Book of Mormon is historical and b) it is set in ancient Mesoamerica.

So, no matter how often the authors of the paper and the editors at the Interpreter assert that this paper is NOT asserting that "a) the Book of Mormon is historical and b) it is set in ancient Mesoamerica," the audience clearly sees it otherwise.

The smart folks at Interpreter (Dan Peterson, Allen Wyatt, Brant Gardner) anticipated a blockbuster, so they kept this article in peer review for over a year where it was polished by both Mesoamericanists and statisticians.

When the provocative piece was finally published a month ago, reactions were fast and furious. Anti-Mormons masquerading behind pen names went ballistic trying to do damage control.

Well. That pretty much sums up the academic side of why I left the Mormon church. Rather than address the critiques or explain the statistical reasoning, the LDS contingent defines objections to the paper as "...Anti-Mormons masquerading behind pen names..."

That's just bad academics. Very, very, BAD academics.

My God .... as long as you believe it, any nonsense is possible with these cheerleaders.

[_honorentheos]

Last weekend I looked over the Kass paper, I hadn't really looked at it before and it is a bit math heavy -- I'm sure I'm not the only one who hasn't read it, I know of at least two other people out there who haven't read it (lol) -- but there was one point I've gone back and forth; whether worth bringing up. Maybe it's been discussed before, but here it is.

... they'd need to fix the scale until it meant something we can actually use meaningfully in the real world.

I don't recall seeing this line of argument but it's compelling.

[_honorentheos]

I'm copying this here so I don't have to retype it if it fails to be posted for the strongly stated view these types of Book of Mormon apologetics are a continuation of old racist views and should not be excused because some people think they are true.

Regarding the latest correspondence on earthen mounds:

The wide spread view of his time was that the mounds left by the mound builders were fortifications built by a lost race overrun by the native American tribes. For example consider this except from the poem The Prairies from 1832:

I think of those
Upon whose rest he tramples. Are they here—
The dead of other days?—and did the dust
Of these fair solitudes once stir with life
And burn with passion? Let the mighty mounds
That overlook the rivers, or that rise
In the dim forest crowded with old oaks,
Answer. A race, that long has passed away,
Built them;—a disciplined and populous race
Heaped, with long toil, the earth, while yet the Greek
Was hewing the Pentelicus to forms
Of symmetry, and rearing on its rock
The glittering Parthenon. These ample fields
Nourished their harvest, here their herds were fed,
When haply by their stalls the bison lowed,
And bowed his maned shoulder to the yoke.
All day this desert murmured with their toils,
Till twilight blushed, and lovers walked, and wooed
In a forgotten language, and old tunes,
From instruments of unremembered form,
Gave the soft winds a voice. The red man came—
The roaming hunter tribes, warlike and fierce,
And the mound-builders vanished from the earth.
The solitude of centuries untold
Has settled where they dwelt. The prairie-wolf
Hunts in their meadows, and his fresh-dug den
Yawns by my path. The gopher mines the ground
Where stood their swarming cities. All is gone;
All—save the piles of earth that hold their bones,
The platforms where they worshipped unknown gods,
The barriers which they builded from the soil
To keep the foe at bay—till o’er the walls
The wild beleaguerers broke, and, one by one,
The strongholds of the plain were forced, and heaped
With corpses.

https://www.poetryfoundation.org/poems/ ... e-prairies

This wasn’t a remarkable guess, and a comprehensive study would have assigned a higher probability this favors the Book of Mormon being fiction rather than fact given its a product of the deeply flawed mound builder mythology that has been refuted by modern archeology. Because it was racist, and still is when is shows up in Book of Mormon apologetics.

[_Physics Guy]

If you were a dale, how would you respond to your suggestion?

Ever seen This Is Spinal Tap?

These go to eleven.

[_Physics Guy]

The same for decibels, if six room-level conversation caused the same hearing loss as one fighter jet passing 20 feet above your head, then they'd need to fix the scale until it meant something we can actually use meaningfully in the real world.

Right. Just because something has numbers doesn't mean it's not stupid, and simple reality checks like this one can reveal the stupidity.

But my take from the Kass-Raftery paper was that it was clearly not telling us to assign an LR of 2 to anything that is barely worth a mention. What it said was nearly the opposite: it was answering the question, "How should I think about LRs in colloquial terms?" You've computed a bunch of LRs, by some more rigorous means than just guessing, and now you're wondering how excited you should be feeling about them.

K-R suggests that an LR of 2 is barely worth mentioning. That is not a prescription for assigning things an LR of 2 just because you feel that they might be barely worth mentioning. Fleabites are also barely worth mentioning, but if everything that was barely worth mentioning was a fleabite we'd all be covered in fleas.

[_Lemmie]

But my take from the Kass-Raftery paper was that it was clearly not telling us to assign an LR of 2 to anything that is barely worth a mention. What it said was nearly the opposite: it was answering the question, "How should I think about LRs in colloquial terms?" You've computed a bunch of LRs, by some more rigorous means than just guessing, and now you're wondering how excited you should be feeling about them.

Agreed re the KR paper, and your last sentence especially sums it up perfectly. The divisions are merely a way to categorize sets of actual data already collected.

The categories in the KR paper cover the full range of LRs that real data could result in, from zero to infinitely large, which the Dales do not incorporate into their analysis, even though they state they do not limit the probabilities in the numerator and denominator of the LRs.

Arbitrarily insisting that data be scored at one of the three ratios allowed by the Dales rules out, for example, the possibility that a data point could have a LR = 1, and thus could be judged so lacking in information as to have no impact on the prior. Billy Shears has discussed his reasons for assigning as LR of 1 to numerous correspondences, but that analysis is artificially and improperly ruled out.

[_honorentheos]

Ironically, Bruce recently described the DNA evidence as qdeserving a 1 in his opinion because he feels the evidence is a mixed bag that could go either way. So he has learned a thing or two from Billy if he's missing the bigger points.

[_Lemmie]

Ironically, Bruce recently described the DNA evidence as qdeserving a 1 in his opinion because he feels the evidence is a mixed bag that could go either way. So he has learned a thing or two from Billy if he's missing the bigger points.

I like that, honor, I'll have to go read again.

DNA is literally the one thing that should have a huge, approaching infinity LR (due to the denominator, since the probability that the DNA evidence exists as it does given the hypothesis that the Book of Mormon is historical is arbitrarily close to zero), but yes, Bruce is growing in his understanding. 1 is undeniably closer on the number line to an extremely large number than is 0.02.