Book Of Mormon: New Evidence Of 19Th Century Origin

[The Grimace]

You do understand that the underlying assumption in this type of study is that any given text being analyzed is influenced by all the books to which it is compared, right?

 

It is like the stylometry or wordprint studies comparing the text of the Book of Mormon to various individuals, like Joseph Smith, Oliver Cowdery, Sydney Rigon, etc.  These studies comparies the patterns of word usage between the text and writings from these individuals and calculates which person's writing style most closely matches the text of the Book of Mormon.  The whole study is set up to produce a result showing that the Book of Mormon most closely matches one of these individuals' writing style.  It does not allow for the possibility that the Book of Mormon came from an ancient source.  It assumes from the outset that it is the result of one those previously mentioned individuals. 

 

This type of study is the exact same thing- it is designed to rank power of influence independent of whether that influence actually existed.  It is amazing that these types of studies continue to impress people.

 

I think you are mixing up the term assumption with the term hypothesis.

[Calm]

We aren't interested in list size. We are interested in the total score at the bottom.

If you are simply adding up scores, then what is stopping you from just including as many insignificant statistically wise scores until the total looks like it is of significance?  How do you judge each score to be statistically significant?  Wouldn't the significance of the total score be dependent on the number of matches one looked at as well as the total word count...the higher both are, the higher the final score would tend to be?  

 

I don't understand your methodology obviously, but it appears that since there are no negative values, that simply adding as many terms as possible to your list increases the score so you may not be interested in list size but it does affect what you are interested in.

Edited October 21, 2013 by calmoriah

[Calm]

Transparency.

 

 

I assume that there is some value associated with the process that demonstrates statistical significance, error and such.  If one is going for transparency, one should include such information.  It is impossible to understand what the values all mean with just the information given.  Methodology should be explained more.

[Calm]

I googled "size-independent matching scale", it only comes up in reference to this study.  Do you have a link to something that explains the process?

 

As for the reddit post...have you tested the process to see how well it works?

Edited October 21, 2013 by calmoriah

[The Grimace]

One of the interesting things about this study is that it turned up a book that was used in Smith's time, in many schools in New York state. While there is no evidence, there exists a probability that can not be ignored - Joseph Smith Jr. may have had exposure to this book. Combine that with the parallels that are apparent, and the endorsement of the book by one Samuel L. Mitchell, and it's hard to view as mere coincidence.

[rongo]

How much you want to bet that any treatment of this new attack, if any, appears in the Interpreter, and not the Mormon Studies Review?

[rongo]

One of the interesting things about this study is that it turned up a book that was used in Smith's time, in many schools in New York state. While there is no evidence, there exists a probability that can not be ignored - Joseph Smith Jr. may have had exposure to this book. Combine that with the parallels that are apparent, and the endorsement of the book by one Samuel L. Mitchell, and it's hard to view as mere coincidence.

By all accounts, the visit to Mitchill was made well after witnesses had significant exposure to the "voice" and verbiage of the Book of Mormon. 

 

I love the "Holy Grail" mentality of critics of the Restoration. The frantic, never-ending search for the silver bullet is informative and entertaining.

[The Grimace]

I assume that there is some value associated with the process that demonstrates statistical significance, error and such.  If one is going for transparency, one should include such information.  It is impossible to understand what the values all mean with just the information given.  Methodology should be explained more.

Umm...if you read the post a few up, you will see that the authors use a proxy for significance. Significance matters when you are dealing with comparisons and subsets of populations; here, the author is dealing with the entire population.

[Calm]

After reading the reddit thread, I've gotten a better idea of what is involved.  I still would like to see a justification of using low valued 4grams as well as a few other things I need to think about how to phrase...a migraine isn't helping my comprehension at the moment.

[Calm]

Umm...if you read the post a few up, you will see that the authors use a proxy for significance. Significance matters when you are dealing with comparisons and subsets of populations; here, the author is dealing with the entire population.

Meaning?....please be kind and speak simply, my brain hurts today.

Edited October 21, 2013 by calmoriah

[name]

How much you want to bet that any treatment of this new attack

 

Attack it surely is!

Not the first and unfortunately not the last attack we will witness against the BofM!

[canadaduane]

If you are simply adding up scores, then what is stopping you from just including as many insignificant statistically wise scores until the total looks like it is of significance?  How do you judge each score to be statistically significant?  Wouldn't the significance of the total score be dependent on the number of matches one looked at as well as the total word count...the higher both are, the higher the final score would tend to be?  

 

I don't understand your methodology obviously, but it appears that since there are no negative values, that simply adding as many terms as possible to your list increases the score so you may not be interested in list size but it does affect what you are interested in.

 

We're adding up scores using an inverse baseline frequency. So if "Historical Reader: containing the Late War Between the United States and Great Britain" matches a 4-word-sequence (4gram) in the Book of Mormon, then we don't simply tally a "1" and call it done. Instead, we add "1.0/baseline frequency" for that 4gram. The baseline frequency is the frequency fo that 4gram summed from a sample of 5000 randomly chosen books. If the 4gram occurs 3102 times, then we would add "1.0/3102" = 0.000332005. Note that we would have to find 3102 *more* unique 4grams with this relative commonality for it to add up to *one* 4gram that occurs only between The Late War and the Book of Mormon. (Note: it's slightly more complex than that because OCR of text produces occasional errors that throw things off and produce bad (and large) data, but that's the gist of it). See http://www.reddit.com/r/exmormon/comments/1ow284/first_big_data_analysis_completed_on_the_book_of/ccwlzrq for an explanation as to why our methodology shows Mercy Otis Warren's Revolutionary War to be an insignificant match, even though it has many matching (relatively common) 4grams.

 

BTW, thanks for your patience as I try to explain things. It's a challenge to dive in for 3 months and become familiar with your own work, then explain it to others

[rodheadlee]

Gosh, what will the NOMs do now that the BoM has been proven false? Will you pull your children from Church? If it's all a big sham how can it affect their salvation?

[Calm]

Is there any demonstration of scores between books that are known to be related?

[Calm]

We're adding up scores using an inverse baseline frequency. So if "Historical Reader: containing the Late War Between the United States and Great Britain" matches a 4-word-sequence (4gram) in the Book of Mormon, then we don't simply tally a "1" and call it done. Instead, we add "1.0/baseline frequency" for that 4gram. The baseline frequency is the frequency fo that 4gram summed from a sample of 5000 randomly chosen books. If the 4gram occurs 3102 times, then we would add "1.0/3102" = 0.000332005. Note that we would have to find 3102 *more* unique 4grams with this relative commonality for it to add up to *one* 4gram that occurs only between The Late War and the Book of Mormon. (Note: it's slightly more complex than that because OCR of text produces occasional errors that throw things off and produce bad (and large) data, but that's the gist of it). See http://www.reddit.com/r/exmormon/comments/1ow284/first_big_data_analysis_completed_on_the_book_of/ccwlzrq for an explanation as to why our methodology shows Mercy Otis Warren's Revolutionary War to be an insignificant match, even though it has many matching (relatively common) 4grams.

 

BTW, thanks for your patience as I try to explain things. It's a challenge to dive in for 3 months and become familiar with your own work, then explain it to others

Again pardon my slowness today...

 

Do you factor in the total number of 4grams found in some way and if not, why not?  It would seem to me that if one found tons of low value 4grams but only a few high value that could still yield a value that appeared to be high...but why would that be actually significant?  Simply because it was higher than the established baseline?

 

I may have missed this...I get how the baseline was created with 5000 randomly chosen books (what was the randomizing process used, out of curiosity?) ...how were the books then chosen to compare the BoM to? 

[jkwilliams]

By all accounts, the visit to Mitchill was made well after witnesses had significant exposure to the "voice" and verbiage of the Book of Mormon. 

 

I love the "Holy Grail" mentality of critics of the Restoration. The frantic, never-ending search for the silver bullet is informative and entertaining.

Is this the Holy Grail of criticism? Nope, but it does suggest a context within which one can place the Book of Mormon that doesn't involve an ancient setting. The frantic, never-ending search for the silver bullet is directly analogous to the never-ending search for Zarahemla in the Yucatan.

[rongo]

Is this the Holy Grail of criticism?

No, of course not. It's just the latest in the Grail Quest.

 

The frantic, never-ending search for the silver bullet is directly analogous to the never-ending search for Zarahemla in the Yucatan.

 

Sort of. The pro-Mormon side has "chilled out" for the most part, not as a waving of the white flag, but more as an acknowledgment that "Zarahemla" evidence would look largely like run-of-the-mill archaeological/anthropological evidence.

 

I don't see the Grail Quest among critics abating.

[jkwilliams]

No, of course not. It's just the latest in the Grail Quest.

One could say the same about the Heartland Theory, or sheum, or Paanchi, or any number of things.

 

Sort of. The pro-Mormon side has "chilled out" for the most part, not as a waving of the white flag, but more as an acknowledgment that "Zarahemla" evidence would look largely like run-of-the-mill archaeological/anthropological evidence.

 

I don't see the Grail Quest among critics abating.

I don't see it abating among apologists, either. It's basically the same process driven by the same need to be "right." To ascribe bad motives to one side is to ascribe them to the other as well.

Edited October 22, 2013 by jkwilliams

[volgadon]

It was me who said the two struck me as similar, and they do. That doesn't prove anything to me, just that I went to a random page and it sounded familiar.

 

 

At first glance they do seem similar (after all what is a genre if not that), but the similarities are superficial at best. Look, for example, at who is lamenting whom.

[volgadon]

One of the interesting things about this study is that it turned up a book that was used in Smith's time, in many schools in New York state. While there is no evidence, there exists a probability that can not be ignored - Joseph Smith Jr. may have had exposure to this book. Combine that with the parallels that are apparent, and the endorsement of the book by one Samuel L. Mitchell, and it's hard to view as mere coincidence.

 

Perhaps you missed where the textbook version actually shows up lower on the list of possible matches.

[jkwilliams]

At first glance they do seem similar (after all what is a genre if not that), but the similarities are superficial at best. Look, for example, at who is lamenting whom.

Again, I'm not arguing for plagiarism but similarity of style adopted for a similar purpose. That seems pretty obvious to me.

[jkwilliams]

If nothing else, the Johnson brothers have stirred up a lot of interest. Between here and the trailer park, it's in the neighborhood of 8,000 hits.

[canadaduane]

Again pardon my slowness today...

 

Do you factor in the total number of 4grams found in some way and if not, why not?  It would seem to me that if one found tons of low value 4grams but only a few high value that could still yield a value that appeared to be high...but why would that be actually significant?  Simply because it was higher than the established baseline?

 

I may have missed this...I get how the baseline was created with 5000 randomly chosen books (what was the randomizing process used, out of curiosity?) ...how were the books then chosen to compare the BoM to? 

 

We downloaded all of the English language books that have a "text" format on archive.org prior to 1830 (i.e. these have been scanned and OCRed by non-profits, corporations, and universities, so the "hard part" has been done for us). From those 130,000 we chose 5000 using a simple random sample. Then we compared the Book of Mormon against all 130,000 using the ranking algorithm I've explained elsewhere (see reddit thread).

 

Can you help me understand what you mean by "the total number of 4grams"? In the baseline sample?

 

Our algorithm is in essence a search algorithm. I think it's valid criticism that we haven't shown statistical significance yet, since we're basically talking about a "top match". It will be interesting to see if there is statistical significance, for example, by pairing books at random and finding a median score with standard deviation. That takes more processing, however (N books x N books, instead of just 1 book x N books), and we're just doing it on my computer at home at the moment. I think it would be good to find ways of cross validating or invalidating our work. We don't have a complex theory or anything, it's just a search for possible influences on the Book of Mormon.

[rongo]

One could say the same about the Heartland Theory, or sheum, or Paanchi, or any number of things.

 

I don't see it abating among apologists, either

You're participating heavily in the thread here and the one at the 'Park. Tell me: who resembles more the guy in Karate Kid cackling awkwardly and shrieking "Put him in a body bag, Johnny!" the most ---- Mopologists when talking about NHM, Pa-ankh-i, etc., or any number of  people "over there?"  Which one acts most certain that this Grail, finally, will be The One?

 

Honestly, now.

Edited October 22, 2013 by rongo

[jkwilliams]

You're participating heavily in the thread here and the one at the 'Park. Tell me: who resembles more the guy in Karate Kid cackling awkwardly and shrieking "Put him in a body bag, Johnny!" the most ---- Mopologists when talking about NHM, Pa-ankh-i, or any number of those "over there?"  Which one acts ****-sure that this Grail, finally, will be The One?

 

Honestly, now.

Depends on the critic and the apologist. I don't see a lot of difference when either side declares "victory." Reasonable critics, like Chris Smith or Dan Vogel, tend not to do a lot of crowing. Reasonable apologists, like David Bokovoy or Kevin Barney, don't crow much, either, from what I've seen. But your average board poster, critic or apologist, is a different beast altogether, don't you think?