Spot-checking Book of Mormon Word-prints

[wenglund]

Not just to that one, but to the other two Jockers' computerized studies

recently conducted (one of which was read as a paper this year), when

they get to the publication stage.

But that is not all, besides the computerized study of Book of Commandments

authorship, there are forthcoming some similar computerized studies of

authorship for the "Lectures on Faith" and a number of early, unsigned

theological articles in the 1830s LDS newspapers.

In the latter two studies, the Mormon scholars will be hard-pressed to

explain how Jockers' two bogus word-print identification methods

are producing authorship results that historians generally agree with.

It is one thing for the Mormons to argue to Nephites wrote the BoM --

it is quite a different thing for them to argue that Sidney Rigdon

did not write any of the "Lectures on Faith" -- or, that if he did,

Jockers' has again "rigged for Rigdon" his word-print analysis, just

to match what he knows in advance that Mormons will agree with.

In short -- if the dual methodology is accepted as useful in identifying

authors of early 1830s LDS publications, it should also be useful in

separating out different sections of the BoM that came from different

authors.

I have asked the Stanford team to map the "word-print" for a few of

the 1830 Alma's chapters, back against the entire Book of Mormon.

In this case there will be no argument over 19th century authors,

as we are looking at what is ostensibly Helaman edited by Mormon, to

see if that latter part of Alma word-print shows up in Mosiah and

Ether. If it does, then there is a problem with the book and not with

the Stanford University scientists.

UD

Dale,

With all due respect, even a broken clock is right twice a day. That it is right on those two occasions isn't evidence that the clock is not broken. Just saying.

Thanks, -Wade Englund-

[wenglund]

I do science for a living and the original peer reviewed paper, as well as Uncle Dale's independent assessment, look fine to me.

I, for one, am impressed (about you being a scientist). It is just that I am wondering how the peer reviewed paper looks in relation to Stout's article that you have also been tauting. In other words, aren't you fine with two conflicting theories?

Thanks, -Wade Englund--

[Uncle Dale]

...

Just saying.

...

I'm perfectly comfortable with the possibility that the Stanford team

may have mis-identified those parts of the Book of Mormon most like

the writings of Sidney Rigdon.

In fact, if the Mormons will point out to me those parts of that book,

which really DO best compare to Rigdon's word-print, they will be doing

me a great favor, and helping me not to waste my time, trying to

confirm or disconfirm the Jockers' authorship attributions for Rigdon.

So -- which BoM chapters are closest to Rigdon's REAL word-print?

UD

.

[Uncle Dale]

I, for one, am impressed...

Assuming for a moment, that Jockers and associates "hit" upon even ONE

spot in the BoM where the count of frequently occurring non-contextual words

actually corresponds to Solomon Spalding's word-print, is there any good

reason why that same spot should also exhibit a relatively high degree

of overlap (in shared vocabulary and phraseology) with Spalding's language?

I mean, wouldn't we begin by assuming that word-prints do NOT correspond

with changing subject matter in an author's writings, and SHOULD be

fairly consistent across all of his compositions, regardless of what

contextual vocabulary and/or phraseology he/she chooses to make use of?

UD

[wenglund]

I'm perfectly comfortable with the possibility that the Stanford team may have mis-identified those parts of the Book of Mormon most like

the writings of Sidney Rigdon.

In fact, if the Mormons will point out to me those parts of that book,

which really DO best compare to Rigdon's word-print, they will be doing

me a great favor, and helping me not to waste my time, trying to

confirm or disconfirm the Jockers' authorship attributions for Rigdon.

So -- which BoM chapters are clsoest to Rigdon's REAL word-print? UD

I'll be sure and do that once I work up the energy to find those portions of the BoM that most closely reflect Tom Hanks' character in the DaVinci Code.

Thanks, -Wade Englund-

[Uncle Dale]

Have you shared this with Craig? He is very approachable.

Ryan asked Craig to participate in the "Delving Deep" blog's

comments section, but did not get a positive response.

There are two sections in Ryan's blog, relevant to the topic

at hand -- one is his critique of the Stanford team's paper,

and the other is his critique of my own textual studies.

Craig is aware of both blog sections -- and I suppose he

has at least glanced at them. But he will not be adding any

comments of his own.

Craig is also aware of the fact that a few of my results for

Spalding-like language on 1830 BoM pages do not match well

with the Stanford team's 2008 word-print paper's reporting.

We are attempting some other ways of inspecting the texts,

which might help uncover why my results do not always match

well with their results.

One possibility is --- that when Spalding wrote upon subjects

other than the ones prominently featured in his "Roman story,"

that he relied upon rather different vocabulary and phraseology.

I'm not sure how likely that explanation might be -- but we will

try and find a way to test that sub-hypothesis in the near future.

Dale

[Uncle Dale]

I'll be sure and do that once I work up the energy to find those portions of the BoM that most closely reflect Tom Hanks' character in the DaVinci Code.

Thanks, -Wade Englund-

Well, then, let's go back a step and talk about a text Mormons have already

inspected. --- Which parts of the "Lectures on Faith" best resemble the

word-print for Sidney Rigdon? Is there really one chapter in them that

better matches the word-print for Joseph Smith?

I know that the Mormons have been studying such possibilities -- so can't

you share what you know about this sort of computerized authorship attribution?

UD

[wenglund]

Well, then, let's go back a step and talk about a text Mormons have already inspected. --- Which parts of the "Lectures on Faith" best resemble the word-print for Sidney Rigdon? Is there really one chapter in them that

better matches the word-print for Joseph Smith?

I know that the Mormons have been studying such possibilities -- so can't you share what you know about this sort of computerized authorship attribution? UD

Since I am not one of those Mormons, I can't very well share knowledge that I don't have.

But, I understand that Rigdon might have known a cousin of Oliver Cowdrey once, so it is then certain that he authored the "Lectures on Faith".

Thanks, -Wade Englund-

[Uncle Dale]

...

a cousin of Oliver Cowdrey

...

Wouldn't a more proper answer be to say that God authored the "Lectures?"

After all, for many decades they were the "doctrine" part of our

"Doctrine and Covenants." The D&C "covenants" sections are still

honored as the pure and perfect words of Eloheim and Jehovah --

so why dishonor the now deleted "doctrine," by attributing the

counsel to "a cousin of Oliver Cowdrey" ??? Our ancestors accepted

that part of the standard works as God's communication. It was

canonized by the unanimous vote of all the quorums in 1835, after

fasting, prayer and much careful inspection at Kirtland.

All I am saying is that the word-prints for Joseph Smith and Sidney

Rigdon may or may not be discernible in those now deleted pages of

our old D&C. For all I know, it will be the word-print of Jesus Christ

that shows up in any future LDS study.

Probably best not to joke about that sort of study, as the scholars

who undertake it will not appreciate our message board humor.

UD

[LifeOnaPlate]

Wouldn't a more proper answer be to say that God authored the "Lectures?"

After all, for many decades they were the "doctrine" part of our

"Doctrine and Covenants."

Can you find any statement from JS or anyone else calling the lectures a revelation? I think the title "Lectures" speaks for itself. Also, I can't help but think you might be avoiding my question about chapter divisions.

[Uncle Dale]

Can you find any statement from JS or anyone else calling the lectures a revelation?

They were canonized at Kirtland in 1835 and for decades remained

part of our standard works. Eventually (in the 1920s, I think)

both you guys (LDS) and we (RLDS) de-canonized them. After that

they were published separately -- at least I think the Mormons

had a separate edition. The RLDS did something similar with part

of the "covenants" section as well -- which resulted in the

de-canonization of part of our D&C "revelations."

Let's say the Mormons did the same, and next year published their

D&C with Brigham Young's famous "revelation" removed. Would that

demote its message from having been God's Holy Word, down to the

status now held by the de-canonized "Lectures on Faith?"

...I can't help but think you might be avoiding my question about chapter divisions.

That is precisely why I did my own studies based upon the 1830

Book of Mormon textual divisions.

For an even "cleaner" round of examination, I'd like to go back

and make use of the CoC's "Printer's MS" or the recent Skousen

critical BoM text, ignoring chapter divisions altogether.

But Orson Pratt mostly just sub-divided the 1830 chapters into

local components. In most cases a set of his chapter divisions

can be lined up to equal an 1830 chapter, from beginning to end.

Jockers and Criddle used the 1879 chapter divisions because a

single page of the 1830 text is too short for accurate

word-printing (only about 500 words per page) and an average

1830 chapter is too long to provide detailed information.

They discovered that, using the 1830 text, cut into the 1879

Pratt segments, they had just the right mix of page length and

detail for assigning authorship.

I would rather that they stick with the 1830 text and divide it

up into two-page sections -- which would be long enough for

running word-print analysis, but would also logically fit better

with my own breakdown of shared language by 1830 page.

What would you suggest for future studies of this sort?

I hear that Maxewell Institute scholars are composing a

rebuttal to Jockers, based upon their own recent word-printing.

Perhaps they are not using the Pratt chapter divisions - ???

UD

Since you didn't reply, I did not know what else to say about that.

But let's go back to the LDS assessment -- that the Jockers' data

was "rigged for Rigdon," and that the BoM chapters Jockers

attributed to Sidney Rigdon are NOT a true match with Rigdon's

actual word-print.

If that is the case, then you LDS scholars can easily demonstrate

that Jockers, Criddle and Witten have "rigged" their analysis --

how can you demonstrate that? -- by publishing a true report of

which BoM chapters DO match best with Rigdon's REAL word-print.

Using the 1879 Pratt-devised chapters, you can simply list which

ones most resemble Rigdon's word-print, and thus demonstrate that

the Stanford team's report is bogus.

Or, if you do not want to use the 1879 chapter divisions, then I'd

be very happy to see you making use of the 1830 divisions, as I have

done in my "spot-checking" work.

Or, if the 1830 chapter divisions are not very useful, then simply

divide up the text by the CoC Printer's manuscript page divisions,

and run your computerized analysis with those divisions.

Any way you care to run your computerized inspection, you should be

able to show that Jockers and and associates have reported the WRONG

sections of the BoM text as best matching Rigdon's "voice." You can

do that with a page or two of tabulated data, published in some

short paper by the Maxwell Institute.

Or, a statistical whiz like Ben can simply provide us with such a

tabulation here in this forum.

That should decide the "rigged for Rigdon" controversy in your favor,

in a single masterful stroke.

???

UD

.

[cdowis]

Dale,

I asked you several years ago about validation studies on your so-called wordprint studies. You indicated that you will persue that.

Has any research been done to validate your analysis, or is that still a future project.

[Uncle Dale]

Dale,

I asked you several years ago about validation studies on your so-called wordprint studies. You indicated that you will persue that.

Has any research been done to validate your analysis, or is that still a future project.

I must admit that I have not kept up on the recent literature

regarding "word-printing" for authorship identification. I am

not qualified to determine what is a valid "word-print" for a

particular author, or what is not a valid "word-print."

If I understand what I've read in multiple sources, however,

there is not much disagreement over what constitutes a base-text

suitable for determining an author's "word-print." Where the

professional disagreements arise, are in the questions of how

valid the method(s) used in authorship detection might be,

and whether the accompanying statistics, for interpreting the

quantified results of a "word-print" study are used correctly.

I'm happy to look over any of the literature on the subject --

whether for or against the concept itself; or whether for or

against the determination that Book of Mormon word-printing

discerns a Nephite/Jaredite authorship.

It is partly because I am NOT a word-print expert, that I chose

to "spot-check" the recent Jockers' results by a quantifying

method OTHER than measuring frequently occurring non-contextual words.

UD

[LifeOnaPlate]

Since you didn't reply, I did not know what else to say about that.

Thanks, I missed it in the fray. You added: "What would you suggest for future studies of this sort?"

I would suggest not changing the data simply in order to fit what "works" for a wordprint study. (And I would suggest recognizing the relative futility of such studies overall.)

[Uncle Dale]

...the relative futility of such studies overall

...

The world is made up of two different kinds of people:

1. Those who have a testimony of the BoM, who find "such studies" useless

2. The rest of us.

Uncle Dale

ps -- and we outnumber you --

[LifeOnaPlate]

Dale, my opinion on wordprint studies includes the previous studies which have been used to bolster Joseph Smith's claims about the Book of Mormon. My opinion on wordprint studies is not connected in any way to my "testimony" of the Book of Mormon. My opinion on wordprint studies is based on the methodological problems, and in this case, the utterly weak historical foundation upon which it is built.

PS- truth by popularity? No thanks.

[Uncle Dale]

...

PS- truth by popularity? No thanks.

Not "truth," but rather, the Mormons cannot dictate what is allowable

historical/textual study and what is not. They may be 100% against

anybody conducting any further computerized studies, but such studies

will continue to be reported in the scholarly literature, no matter that.

The LDS can either join in that ongoing discussion, or they can opt out.

Or, I suppose they can hide behind Maxwell Institute doors and just

talk among themselves, about how Mormon edited Helaman and how

Moroni edited Ether.

Are you then pretty certain that close examination of the BoM text

and its language, well confirms who the authors actually were?

If so, then a page from Mormon's editing of Helaman should NOT

have the same word-print as does a page from Moroni editing Ether.

Correct?

UD

[Uncle Dale]

...

Are you then pretty certain that close examination of the BoM text

and its language, well confirms who the authors actually were?

If so, then a page from Mormon's editing of Helaman should NOT

have the same word-print as does a page from Moroni editing Ether.

Correct?

...

Can anybody direct me to an on-line article or discussion regarding

LDS computerized authorship studies for the Book of Mormon?

Neither of these two web-pages tell me who the LDS say wrote

which parts of the Book of Mormon --

http://www.lds-mormon.com/wordprin.shtml

http://www2.ida.net/graphics/shirtail/wordprin.htm

???

Uncle Dale

.

[LifeOnaPlate]

Not "truth," but rather, the Mormons cannot dictate what is allowable

historical/textual study and what is not. They may be 100% against

anybody conducting any further computerized studies, but such studies

will continue to be reported in the scholarly literature, no matter that.

Being unaffiliated with this mysterious group of "Mormons" you refer to in no way vindicates your efforts, though. The wordprint stuff is about as intellectually compelling as the "Bible Code" stuff.

[Uncle Dale]

...

The wordprint stuff is about as intellectually compelling as the "Bible Code" stuff.

...

The LDS-conducted word-print studies that "proved" the BoM was written by Nephites?

Or, just the non-LDS studies, which did not find any nephites?

UD

[LifeOnaPlate]

The LDS-conducted word-print studies that "proved" the BoM was written by Nephites?

Or, just the non-LDS studies, which did not find any nephites?

UD

All of them. As I've said before several times.

(None explicitly claimed to "prove" anything was written by Nephites. That is a straw man of your construction.)

[Uncle Dale]

All of them. As I've said before several times.

(None explicitly claimed to "prove" anything was written by Nephites. That is a straw man of your construction.)

So, if I were to take an underlined text from the Book of Mormon to

"The Brethren" in Salt Lake City, and show them the passages where LDS

computer-using scholars have attributed Divine counsel and revelation

directly to the "voice" of Jehovah, those topmost Mormon leaders would

denounce those word-print studies just as strongly as you do?

Can you direct me to any published GA statement, telling us we should

ignore or suppress the previous faith-promoting "word-print" results

reported by LDS researchers?

UD

[LifeOnaPlate]

So, if I were to take an underlined text from the Book of Mormon to

"The Brethren" in Salt Lake City, and show them the passages where LDS

computer-using scholars have attributed Divine counsel and revelation

directly to the "voice" of Jehovah, those topmost Mormon leaders would

denounce those word-print studies just as strongly as you do?

Beats me. They have much bigger concerns right now. And pretty much always.

Can you direct me to any published GA statement, telling us we should

ignore or suppress the previous faith-promoting "word-print" results

reported by LDS researchers?

UD

Nope. I can't point you to an GA statements of endorsement of such studies either. I'm not sure why that would be relevant to the discussion, though.

[Uncle Dale]

...

I'm not sure why that would be relevant to the discussion, though.

I have the feeling, that even in the lower quorums of Seventies they

would have no idea who Jeff Lindsay was -- or what "rigged for Rigdon"

could possibly mean.

A couple of them might have heard something about a "plates" blog though.

UD

[Cold Steel]

Wordprint analysis has always fascinated me, but in this case, who can speak intelligently on it? So far, I've seen the abstract, which notes: "Our results indicate that likely nineteenth century contributors were Solomon Spalding, a writer of historical fantasies; Sidney Rigdon, an eloquent but perhaps unstable preacher; and Oliver Cowdery, a schoolteacher with editing experience. Our findings support the hypothesis that Rigdon was the main architect of the Book of Mormon and are consistent with historical evidence suggesting that he fabricated the book by adding theology to the unpublished writings of Spalding (then deceased)."

Fine, but what about their methodology? I've read extended examples of Solomon Spalding's fictions and I suppose it's not enough to say that there's no outward similarities in style. Before accepting any conclusions for the Book of Mormon, I'd want to see the methodology applied to submitted and blind examples of writings where authors were 1) requested to submit text samples where they were asked to deliberately disguise their writings; and 2) requested to submit text samples where no such attempt had been made.

Initial wordprint studies used the Federalist Papers, which seems to be an excellent control when applying authorship studies. Although statisticians might understand something like "delta and nearest shrunken centroid classification," to me it would seem best remedied by Preparation H. As it is, the methodology isn't to be proven with Mormon texts, but with other, preferably blind, texts. Once it's validity is established, then apply it to the Book of Mormon authors (making sure to delete all "Isaiah" passages beforehand). I don't know, perhaps this has been done already, but if so I cannot find any references. Jeff Lindsey briefly described how it works: "These methods look at the frequency of simple non-contextual words (words like 'of', 'the', and 'and' whose usage typically doesn't vary strongly as a function of the topic or context of writing) and compare stats from chapters in the Book of Mormon to characteristics obtained for text from each of the candidate authors."

Has this actually been shown to work in text that's been deliberately disguised in an attempt to deceive? Or can it reliably pick up on mixed text, a combination of two or more writers or editors? I'm also bothered by Craig Criddle's involvement in the study. Shouldn't those carrying out the study have even a degree of objectivity?

Charging almost thirty dollars just to take a look at their methodology is shabby. It's one thing to establish it for authorship of documents and quite another for judging individual sentences and paragraphs. In short, it's like buying one of those "how to get rid of kidney stones/gall stones" cures that you have to buy before evaluating. Only with those, you at least get your money back if they don't work!

.