Another Wordprint-type Study

[why me]

It is certainly hard to thus account for Parley P. Pratt's tenacious

defense of Rigdon, the BoM, and the associated history, if Pratt

were indeed one of the writers.

He specifically testifies that the first time Sidney Rigdon ever saw

"Elder Joseph Smith" was when Rigdon went to visit Smith in 1830.

At Kirtland Pratt felt very much cheated by Smith over certain

real estate and financial dealings. He was probably just about ready

to leave the Church -- but said nothing about a BoM conspiracy.

In 1839 Pratt evidently led the effort to get Elder Jesse Haven into

the home of Spalding's widow, to interview her and (according to

the family) obtain some of Spalding's writings. Again, at that time,

Pratt was a very vocal defender of the BoM in NYC newspapers.

So, based upon his own post-baptismal experiences. perhaps we

should eliminate Pratt as a possible contributer. He certainly never

admitted to any such role.

On the other hand, Pratt produced some entertaining fiction, including

the only known Mormon literary creation in which Joseph Smith

refutes the Spalding authorship claims. Pratt was capable of writing

something like the BoM (or parts of it). But, since he never admitted

to any such thing, I guess we can say he was not involved.

UD'

Thanks for the information. There is also the account of Emma Smith and David Whitmer of Joseph putting his head in a hat and translating the book of mormon. The antimormons have a field day with this but would this also discount Rigdon, Spaulding and Cowdery as authors of the book of mormon? Rather difficult to be a front person and do such a stunt. And when one looks at it objectively, why would a small group of men bother to write a book and have their front man stick his head in a hat to 'translate' the book?? It sounds just a little strange.

[Uncle Dale]

Thanks for the information. There is also the account of Emma Smith and David Whitmer of Joseph putting his head in a hat and translating the book of mormon. The antimormons have a field day with this but would this also discount Rigdon, Spaulding and Cowdery as authors of the book of mormon?

A good question, Why-Me.

And we could add to that the fact that at least some pages

of the preserved "Dictated Manuscript" of the BoM show signs

in the handwriting, that those pages were probably written

down by Oliver Cowdery from Joseph's dictation -- presumably

whilst Joseph was gazing into a seer-stone (perhaps in a hat).

Can we use these bits of evidence to rule out examining such

writers as Solomon Spalding and Sidney Rigdon as having been

BoM textual contributors?

If THAT is how we set up our writer selection criteria, for conducting

word-print testing, I think that we run into immediate problems and

pretty much are forced to limit such testing to the word-print of

Joseph Smith, Jr. and "Nephites."

Rather difficult to be a front person and do such a stunt. And when one looks at it

objectively, why would a small group of men bother to write a book and have their

front man stick his head in a hat to 'translate' the book?? It sounds just a little strange.

Yes, it does -- unless we first of all consider that this was reportedly

Joseph's method, as a village seer, well before he claimed to do any

BoM translating. If this was his method, then the method must be

accepted as an historical peculiarity, and we are left with your more

important question -- was Joseph a "front man" doing "stunts?"

Mormons and anti-Mormons can debate this sort of stuff all day long;

but it has no place in our performing and reviewing word-print studies.

I realize that you, and "Cold Steel" and Daniel Peterson are very much

used to the back-and-forth banter that goes on within the sundry postings

on a Message Board such as this one ---------- but we are entering a

different sort of discussion territory when we stop to consider the recent

Stanford team's BoM word-print reporting. That sort of topic, published

in the peer-reviewed scholarly press, requires a different sort of response

than our simply stating our opinions on a Message Board like this one.

Probably we are doomed to carry on our endless banter here; but while

we do so, serious scholars/scientists/academicians/historians/statisticians

can (and will) begin to examine this recent word-print study in a more

serious manner. In that scholarly discussion that is forthcoming, folks like

you, and Dan and Cold Steel cannot simply submit your opinions regarding

Nephites and head-in-the-hat recollections, to serve as the faithful LDS

response to what purports to be authorship testing, utilizing scientific methods.

If Mormons offer no better responses than these, they risk having the

authorship discussion pass entirely out of their control -- entirely out of

their ability to shape and direct the discussion and associated research.

As I see it, THAT would be a very serious mistake.

UD

[John Williams]

I think you are missing the point here, Dan. We each have our private opinions

about what could or could not have been sought out and compiled by an early

18th century writer. But our private opinions/testimonies should not get in the

way of our conducting such word-print studies (or, at least allowing others who

are interested in the subject to pursue it).

Hopefully our private opinions will be kept to a minimum in scholarly reporting

on these matters. Then again, perhaps it is unavoidable that we will sort ourselves

out into two opposing groups; each deriding the efforts and conclusions of the other.

But I hope not.... I am so t-i-r-e-d of replaying that old game.

Dale

Excellent point. Saying "he could not possibly have known" or "he obvious did know" is an ideological position, not a statement of fact, and can get in the way of a careful and objective examination of the text.

BTW, it's nice to see you around again. I hope all is well for you out in paradise.

[Vex]

Rather than using non-contextual words (the, it, and, but, etc.), these researchers determined

a number of the most frequent words used by seven different writers, and then used that

information in order to determine which potential author out of that group could have

written each chapter in the modern LDS edition of the Book of Mormon.

If the LDS Church officials/scholars can come up with a lengthy text sample that they

can reliably assign to the young Joseph Smith, Jr., it sounds like the writers of this new

paper would be pleased to map out Smith's word-print in the Book of Mormon. When

they conducted their study they could not obtain a lengthy, reliably "Smith" text sample.

Probably other researchers will conduct more studies, testing a wider variety of old authors,

to determine whether or not this latest word-print analysis has any validity.

At least that is what I hope will happen in the coming year.

UD

What my concern would be however would be large amounts of false positives. Conducting this word-print study on one book is interesting, but doesn't hold merit unless those findings can be compared to others (contemporary or not). For instance would we see the same distribution of "hits" in works of other contemporaries, what about ancient works (Illiad, etc)? To simply put out findings like this with nothing else to substantiate the claim (or compare to) seems a little misleading. I believe you've raised the point of is it mere coincidence or something more.

There's just too many questions concerning the methodology and distribution of affirmatives for this to actually go anywhere right now. Though the speculation is enjoyable.

To add fuel to the fire I am not asking for anything that hasn't already been implemented in other similar studies:

http://llc.oxfordjournals.org/cgi/content/abstract/2/3/153

[John Williams]

What my concern would be however would be large amounts of false positives. Conducting this word-print study on one book is interesting, but doesn't hold merit unless those findings can be compared to others (contemporary or not). For instance would we see the same distribution of "hits" in works of other contemporaries, what about ancient works (Illiad, etc)? To simply put out findings like this with nothing else to substantiate the claim (or compare to) seems a little misleading. I believe you've raised the point of is it mere coincidence or something more.

There's just too many questions concerning the methodology and distribution of affirmatives for this to actually go anywhere right now. Though the speculation is enjoyable.

To add fuel to the fire I am not asking for anything that hasn't already been implemented in other similar studies:

http://llc.oxfordjournals.org/cgi/content/abstract/2/3/153

Agreed. There are a lot of questions remaining about the efficacy of wordprint studies.

[Uncle Dale]

What my concern would be however would be large amounts of false positives. Conducting this word-print study on one book is interesting, but doesn't hold merit unless those findings can be compared to others (contemporary or not). For instance would we see the same distribution of "hits" in works of other contemporaries, what about ancient works (Illiad, etc)?

Yes, I agree -- if this approach is to ever become meaningful to

the general public (non-specialists in statistics, etc.), then the study

results must consist of more than a handful of miscellaneous writers'

word-prints, mapped out across the Book of Mormon.

Two additional steps should eventually be added to what we have

seen so far --- but it will take time.

1. A much larger number of potential candidates for BoM text contributor

must be selected, tested and the results for those writers reported. It

would also be useful if those results could be rank ordered, from #1

probability, down to, say, #500 or so.

2. Once that has been done, each of those many writers should be

tested as the potential author of numerous other texts, than just

the Book of Mormon.

Until all of that has been done, we can only speak of the RELATIVE

standings of a very small number of writers tested. Those results

can serve as an incentive for additional research, and the results

can be examined to discern errors in methodologies. etc., but so

long as our reported data is limited to RELATIVE quantification of

a very small number of potential writers, the results are problematic.

For example, let's say we chose three more contemporary writers

whose word-prints could be fairly easily established, and we then

mapped out the word-prints for those three potential contributors

to the text. Say we chose Lucy Smith, W. W. Phelps and Wm. Morgan.

What sort of results might we expect to see in their mapped-out wordprints?

Probabably Lucy Smith would stand out as the most likely author of

certain parts of the BoM, while Phelps and Morgan would show up as

the better candidates for having written other parts of the text.

Which proves? ---- nothing, other than the relative results for the three.

However, what would we make of the findings, if Morgan's word-print

was relatively "flat," and Phelps only showed up as probable writer of

one or two chapters, and Mother Smith was also "flat" throughout all

the text, except for 1st Nephi ------ where her word-print scored at

90% probability for all the chapters in that part of the BoM?

That is the sort of oddity we might keep a lookout for -- if such things

ever develope in future word-print analysis.

To simply put out findings like this with nothing else to substantiate the claim (or compare to)

seems a little misleading. I believe you've raised the point of is it mere coincidence or something more.

There's just too many questions concerning the methodology and distribution of affirmatives for this to actually go anywhere right now. Though the speculation is enjoyable.

To add fuel to the fire I am not asking for anything that hasn't already been implemented in other similar studies:

http://llc.oxfordjournals.org/cgi/content/abstract/2/3/153

We shall see --- I'm told that another team of word-print researchers

have expressed an interest in re-running the Stanford team's testing,

but this time with the inclusion of the word-print for Joseph Smith, Jr.

UD

.

[Uncle Dale]

Excellent point. Saying "he could not possibly have known" or "he obvious did know"

is an ideological position, not a statement of fact, and can get in the way of a careful and objective examination of the text.

BTW, it's nice to see you around again. I hope all is well for you out in paradise.

I'm trying to get through various health problem constraints, but will

come back to contribute what I can, now and then.

We really cannot expect too much from posters on a Message Board,

Sometimes they surprise us, by communicating great insights and

suggestions --- but a great deal of the discussion here is just banter.

The new word-print study results deserve a better, more informed discussion

than most of us here can provide. So, we will knock the topic around a little,

but probably will not come up with a suitable response for publication in the

journal that featured the new paper.

Given that admitted limitation, I still invite interested readers to offer their

ideas on what the LDS response (if any) should be to the Stanford researchers.

Either this matter will fade and be forgotten in a few weeks, or it will grow

to become a major new explanation for Book of Mormon orgins, among the

non-LDS folks. Are we prepared for the latter possible development?

Uncle Dale

.

[wenglund]

I have read most, but not all, of the thread, so I don't know if this has been asked and responsed to or not. If so, please forgive the redundancy.

Regardless of how confident or skeptical we may each be in the authorship attribution techniques employed in the study at issue, can we all agree that at the very least the results of the techniques may only have some evidentiary value if the actual author or authors of the text in question are included in the study?

In other words, suppose author "A" wrote a section of the Book of Mormon being tested, and only authors B, C, D, E, F, G, and H were included in the "training" portion of the study, then even were author "B" to end up with the highest relative probability for the section being tested, all this would really tell us is how closely "B"'s writings matched the text in question. It doesn't really tell us that "B" or "A" wrote the text. Right?

Thanks, -Wade Englund-

[Cold Steel]

...we are entering a different sort of discussion territory when we stop to consider the recent Stanford team's BoM word-print reporting. That sort of topic, published in the peer-reviewed scholarly press, requires a different sort of response than our simply stating our opinions on a Message Board like this one.

And I believe we should consider to look at such evidence. Since many of us aren't into wordprint studies, we simply raise other issues that would detract from the conclusions, and content is as valid a judgment as any other.

The Book of Mormon opens up the door for ever more revelation and is, itself, an incomplete translated volume. Thus, whoever wrote it was careful to keep that door open. There are already two different versions of the "sealed portion" of the plates and both read about like what I would expect a fiction writer to come up with. Okay, perhaps Tolstoy could do better, but certainly no one has yet produced anything to match (or even come close) to the Book of Mormon.

I believe John Welch was the first to discover chiasmus in the Book of Mormon. Having learned it was a hallmark of Semetic writing, he began searching the Book of Mormon and found complex chiasms throughout. For some reason this has never really impressed the book's critics, but it's not something that should be overlooked when trying to determine authorship.

I understand why some want to simply investigate Book of Mormon authorship without considering these and other elements, but ultimately they must be considered and the question asked, could these have been known at the time?

I'm one of those who are still hoping that wordprint analysis can be used to substantiate the Book of Mormon, but the very content of the Book of Mormon is a major impediment to wordprint studies being used against it.

On the surface, mathematical probability studies sound impressive. Who hasn't heard of the famous Bible Code? Entire documentaries have been done on it and its advocates are difficult to refute; however, it presupposes that the Torah and other books are word-for-word correct transliterations (something pointed out by numerous LDS scholars). If even a few words are off, the Bible Code theory would fail, and yet we know that errors have crept in. Still, there are many who believe in it, despite the statistical irregularities.

[why me]

If Mormons offer no better responses than these, they risk having the

authorship discussion pass entirely out of their control -- entirely out of

their ability to shape and direct the discussion and associated research.

As I see it, THAT would be a very serious mistake.

UD

I am sure that mormons who have a grasp of the issue will begin to respond. However, in academic discussion there is much bantering back and forth with the use of sourced evidence in academic journals. However, no clear cut winner is often to be shown from the academic bantering. But academics do love to print articles in academic journals to impress their peer group. And so, here we may have chitchat among nonprofessionals but in academia we have sourced chitchat between academics.

And at times, in the end, it is usually the simple explanation as evidenced in human nature that rings true.

[why me]

A good question, Why-Me.

And we could add to that the fact that at least some pages

of the preserved "Dictated Manuscript" of the BoM show signs

in the handwriting, that those pages were probably written

down by Oliver Cowdery from Joseph's dictation -- presumably

whilst Joseph was gazing into a seer-stone (perhaps in a hat).

Can we use these bits of evidence to rule out examining such

writers as Solomon Spalding and Sidney Rigdon as having been

BoM textual contributors?

UD

When one looks at the evidence just how the book of mormon came about and compare it with the word study that recently was published, and if such a study is found to be true, one can say that the people around the bringing forth of the book of mormon were all non-violent psychopaths or sociopaths as they all showed a total disregard of human emotion and conscience. And yet, it would be difficult to believe that any of these people showed such a tendency in their life.

From a humanist point of view, if I comtemplate Sidney or Oliver in on a hoax and how they made up such a story complete with angel visitations etc, I can only conclude that they would have been sociopaths. And to find a front guy who would eventually give his life for the hoax was a piece of luck. And to find all those witnesses to go along with it and stick to their story was quite a piece of luck too. And to find Jospeh Smith and to have him stick his head in a hat to translate the book from a memorized manuscript was a total piece of luck on sidney's and oliver's part.

But with so much luck, I can begin to assume that the luck was not real, that there needs to be another explanation for the book of mormon than the statement that Sidney and Oliver wrote it. Or that it was written by early 19th century men looking to implament the hoax of the century.

[Danna]

Spalding and the Book of Mormon have a very high degree of vocabulary overlap (about 40 percent in either direction). While the BoM shares about 40% of the BoMs vocabulary, the Longfellow text shares only about 30 percent of the BoMs vocabulary. The Longfellow text and the Spalding text are pretty close to the same length, and yet the vocabulary of the Longfellow text is quite small by comparison (the Book of Mormon also has a relatively small vocabulary for such a lengthy work).

There are some notable differences. Among them is the fact that the Spalding text shares far more locutions with the Book of Mormon than does the Longfellow text (nearly twice as many 3-word locutions for example). And if the Book of Mormon is taken apart a chapter at a time, this will dramatically affect the clustering and likewise the predictions of the study. If the Book of Mormon were to be cut into larger chunks, I am not sure this distinction would be quite so pronounced. The shared locutions is quite an interesting thing though. Spalding, for example, duplicates about 1.5 % of all unique 3 word locutions in the Book of Mormon (this is, by the way, quite low by comparison for other contemporary works that I have looked at - Warren's _The Rise, Progress and Termination of the American Revolution_ by comparion duplicates just over 5% of all the unique locutions in the Book of Mormon: a number which is higher than the average I would also expect to see of about 3%). By comparison, the Longfellow text duplicates about .8% of the unique 3 word locutions of the Book of Mormon - significantly less than any other text of I had previously analyzed (Spalding had previously had one of the lower values - and Longfellow is about half of Spalding). As a comparison, perhaps, Jules Verne's _Around the World in 80 Days_ yielded nearly as many unique 3-word locutions in common with the Book of Mormon as did Spalding and Longfellow combined.

Delta and NSC specifically do not analyse locutions, and by restricting their analysis to words common to all authors with a cutoff determined by frequency, breadth of vocabulary would NOT impact on the analysis. I can see your point though, if you are establishing authorial similarity based on locutions and vocabulary. Do you have similar data for Borrows?

I think you would have to show a correlation between overlap in unique locutions/vocabulary, and similarity in the frequency of common words. On what theoretical basis would you expect a correlation?

edit to add 'NOT'

[Uncle Dale]

...

there needs to be another explanation for the book of mormon than the statement that Sidney and Oliver wrote it. Or that it was written by early 19th century men looking to implament the hoax of the century.

That raises an interesting question.

IS there an explanation we have not yet considered (other than the following list)?

1. Authentic ancient record of authentic persons and events

2. Authentic ancient record, but contains errors, myths, mis-statements, etc.

3. Not an authentic record, but a fairly true reconstruction of PreColumbian history/religion

4. Partly authentic and partly fabricated -- contains some historical truth

5. Not an authentic record; nor authentic history; but still contains a divine revelation

6. The product of a 19th century deluded mind; perhaps that of Smith

7. A 19th century fabrication, but Smith thought it was an authentic record

8. A 19th century fabrication, made with good intentions by one or more writers

9. A 19th century fabrication, made for base purposes by one or more writers

10. ......... ???

Finally -- can word-printing studies in any way help to solve the issue (for us non-LDS, at least)?

UD

[Cold Steel]

The new word-print study results deserve a better, more informed discussion than most of us here can provide.

Exactly. I haven't been able to read anything more than the outline because the study is only on a subscription site.

In a review of a second wordprint analysis done in the 90s by Presbyterian minister convert Roger Keller, FARMS writer John Tvedtnes took issue with much of its methodology. It was an expansion on the first study, conducted in 1980, but like any study involving wordprint analysis, care must be taken to sort out the translation process, editing process (by both Joseph Smith and Moroni) and inclusion of various authors. There are so many input/output possibilities that it would be nice to see the methodology used in these more recent tests.

The abstract states, in part: "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)."

I was unaware of any historical evidence that Rigdon added theology or anything else to either the Book of Mormon or Spalding's missing transcript. Speculation and "historical evidence" are hardly the same, yet if there is evidence, I'd like to see it. Criddle himself calls the Rigdon-Spalding connection a "theory" and has little but conjecture to back up his claims. This theory "proposes that the unpublished writings of Solomon Spalding (a Dartmouth-educated man who died in 1816) came into the possession of Sidney Rigdon, a reformed Baptist (Campbellite) preacher, and that Rigdon inserted into those writings his literalistic, pre-millennial, Restorationist Christian theology, along with other information useful for his purposes, to create the 1830 version of The Book of Mormon." His motive, "to become the leader of a new Christian religious movement and to gather Israel to a 'promised land' in North America, where he would usher in the Christian millennium."

Criddle concedes the weakest parts of his theory include:

1) The Spalding manuscript from which the Book of Mormon was said to be taken is missing.

2) How and when Rigdon first came into contact with a member of the Smith family is unknown.

3) Previous wordprint studies have not connected Spalding or Rigdon to The Book of Mormon.

4) Some of the testimony cited in support of the Spalding-Rigdon Theory came many years after the alleged incidents and can be questioned on the grounds of memory fallibility.

Of course the weakest part of the theory is content, itself, in the Book of Mormonâ??content that cannot be explained in the context of what was known by anyone living when it was written. Still, proponents of the latest wordprint study want us to put this point aside, and so I assume it will for the point of arguing and evaluating methodology.

John Hilton, in his paper, On Verifying Wordprint Studies: Book of Mormon Authorship, noted: "Statistical reliability is rigorously demonstrated by using the technique with a large number of control-author texts for the purpose of verifying the authorship of known texts." The Criddle/Jocker wordprint study, however, uses only two controls and texts from five suspect writers. Joseph Smith's texts was not one of them, a point I find peculiar since he's the central figure in this whole thing.

[wenglund]

I have read most, but not all, of the thread, so I don't know if this has been asked and responsed to or not. If so, please forgive the redundancy.

Regardless of how confident or skeptical we may each be in the authorship attribution techniques employed in the study at issue, can we all agree that at the very least the results of the techniques may only have some evidentiary value if the actual author or authors of the text in question are included in the study?

In other words, suppose author "A" wrote a section of the Book of Mormon being tested, and only authors B, C, D, E, F, G, and H were included in the "training" portion of the study, then even were author "B" to end up with the highest relative probability for the section being tested, all this would really tell us is how closely "B"'s writings matched the text in question. It doesn't really tell us that "B" or "A" wrote the text. Right?

Thanks, -Wade Englund-

What I am asking is, can the computerized author attribution techniques (CAAT) used in the so-called Stanford study produce a false positive--particularly if the actual author is not included in the study?

If so, then what is the probative value in using such techniques in cases where the actual author is unknown and in question?

I mean word-prints studies are not like fingerprinting, where attribution only occurs when there is a perfect match.

Thanks, -Wade Englund-

[Uncle Dale]

What I am asking is, can the computerized author attribution techniques (CAAT) used in the so-called Stanford study produce a false positive--particularly if the actual author is not included in the study?

If so, then what is the probative value in using such techniques in cases where the actual author is unknown and in question?

I mean word-prints studies are not like fingerprinting, where attribution only occurs when there is a perfect match.

Thanks, -Wade Englund-

I paraphrased your question, in a query to one of the Stanford team members,

and the answer came back that in any such computerized results, using their

methodologies, there is one chance in ten of a false authorship % rating for

a particular section of a lengthy text.

If I understand correctly, that answer does not mean that one out of every

ten chapters in the BoM that they have reported probable authorship for

is a "false positive;" but it does mean that, generally speaking, they expect

their methodology to turn up the wrong "first choice" for authorship, about

that frequently.

Again, this is just my understanding of what I was told, and I'm probably

oversimplifying things -- but, as I understand it, if the Stanford researchers

rate Oliver Cowery as the first choice for authorship of nine chapters, and

Parley P. Pratt for second choice in those chapters, but then also rate a

tenth chapter with Pratt in first place and Cowdery in second place, then

there is a significant possibility that Pratt's position is a "false positive,"

and that, perhaps, that tenth chapter should also be rated with Cowdery

as first choice and Pratt as second choice for authorship.

If any of what I just said seems possible, then we should be able to discern

probable anomalies in their authorship ratings, by looking at patterns in

authorship distribution, matched against how the various sections of text

tested seem to "look and feel."

Let me give you an example. If we have two chapters that purport to be

Nephite sermons, by the same speaker, and the writing style, etc., for

those two BoM chapters appears to us to be very, very similar, we can

mistrust the Stanford researchers' opinion on there having been two different

writers, about one time out of ten.

Does that make any sense?

UD

.

[wenglund]

I paraphrased your question, in a query to one of the Stanford team members,

and the answer came back that in any such computerized results, using their

methodologies, there is one chance in ten of a false authorship % rating for

a particular section of a lengthy text.

If I understand correctly, that answer does not mean that one out of every

ten chapters in the BoM that they have reported probable authorship for

is a "false positive;" but it does mean that, generally speaking, they expect

their methodology to turn up the wrong "first choice" for authorship, about

that frequently.

Again, this is just my understanding of what I was told, and I'm probably

oversimplifying things -- but, as I understand it, if the Stanford researchers

rate Oliver Cowery as the first choice for authorship of nine chapters, and

Parley P. Pratt for second choice in those chapters, but then also rate a

tenth chapter with Pratt in first place and Cowdery in second place, then

there is a significant possibility that Pratt's position is a "false positive,"

and that, perhaps, that tenth chapter should also be rated with Cowdery

as first choice and Pratt as second choice for authorship.

If any of what I just said seems possible, then we should be able to discern

probable anomalies in their authorship ratings, by looking at patterns in

authorship distribution, matched against how the various sections of text

tested seem to "look and feel."

Let me give you an example. If we have two chapters that purport to be

Nephite sermons, by the same speaker, and the writing style, etc., for

those two BoM chapters appears to us to be very, very similar, we can

mistrust the Stanford researchers' opinion on there having been two different

writers, about one time out of ten.

Does that make any sense? UD

That makes sense were one to presume that the author of the text in question is included in the study. If the author of the text happens not to be included in the study, then there is a 100% chance of a false positive. Right?

Thanks, -Wade Englund-

[Uncle Dale]

That makes sense were one to presume that the author of the text in question is included in the study. If the author of the text happens not to be included in the study, then there is a 100% chance of a false positive. Right?

Thanks, -Wade Englund-

99.99999% of the time, that is indeed correct.

But there may be exceptions. Let's say you have a twin brother who

went to the same schools, read the same books, learned the same

language and idioms as you have. It is possible that such a person's

word-print would be so similar to your own, that it would score a

very high probability for authorship of a tested lengthy text from you.

The word-print determination in such a case would be a "false positive,"

but it would be an expected/understandable one.

But, let's take that scenario a bit farther -- Let's say that out of every

million people, speaking the same idiomatic language, that two will

have such similar preferred vocabulary, so as to produce practically

identical word-prints. In such an instance the "false positive" would be

expected/understandable.

Now, back to the Stanford researcher's recent paper -- in the word-print

mapping of Longfellow, he came out as the first choice, probable author

in a chapter or two, out of all the chapters in the Book of Mormon. Even

if the study were conducted again, and you, me and Longfellow were the

authorship candidates, I suspect that Longfellow would still emerge as the

favored writer of those one or two BoM chapters.

It is that merely a statistical fluke, generated by the computerized process?

Or, is there perhaps something peculiar about those BoM chapters that

causes their wording to correspond at a high level with Longfellow's?

Whatever the cause, I think that we can both agree that Longfellow did

NOT write those chapters -- and neither did we. Whoever it was who wrote

them, those person(s) cannot possibly be included in a study where only

you, me and Longfellow are thus examined for our word-prints in the BoM

Clear as mud?

UD

.

[wenglund]

99.99999% of the time, that is indeed correct.

But there may be exceptions. Let's say you have a twin brother who

went to the same schools, read the same books, learned the same

language and idioms as you have. It is possible that such a person's

word-print would be so similar to your own, that it would score a

very high probability for authorship of a tested lengthy text from you.

The word-print determination in such a case would be a "false positive,"

but it would be an expected/understandable one.

But, let's take that scenario a bit farther -- Let's say that out of every

million people, speaking the same idiomatic language, that two will

have such similar preferred vocabulary, so as to produce practically

identical word-prints. In such an instance the "false positive" would be

expected/understandable.

Now, back to the Stanford researcher's recent paper -- in the word-print

mapping of Longfellow, he came out as the first choice, probable author

in a chapter or two, out of all the chapters in the Book of Mormon. Even

if the study were conducted again, and you, me and Longfellow were the

authorship candidates, I suspect that Longfellow would still emerge as the

favored writer of those one or two BoM chapters.

It is that merely a statistical fluke, generated by the computerized process?

Or, is there perhaps something peculiar about those BoM chapters that

causes their wording to correspond at a high level with Longfellow's?

Whatever the cause, I think that we can both agree that Longfellow did

NOT write those chapters -- and neither did we. Whoever it was who wrote

them, those person(s) cannot possibly be included in a study where only

you, me and Longfellow are thus examined for our word-prints in the BoM

Clear as mud? UD

Simply stated, and if I understand things correctly, the word-print techniques are not designed to determine whether the actual author(s) of the text in question is/are included in the study or not. And, if the actual author(s) of the text happen to be included in the study, then there is a 20% chance of a false positive; whereas if the actual author(s) aren't included in the study, it is expected/understandable that the study will invariably (100%) return a false positive. Therefore, there is between 20% and 100% chance that the results of the study will returned a false positive, and this without the ability to know which percent applies.

Should that instill confidence?

Thanks, -Wade Englund-

[Uncle Dale]

Simply stated, and if I understand things correctly, the word-print techniques are not designed to determine whether the actual author(s) of the text in question is/are included in the study or not. And, if the actual author(s) of the text happen to be included in the study, then there is a 20% chance of a false positive; whereas if the actual author(s) aren't included in the study, it is expected/understandable that the study will invariably (100%) return a false positive. Therefore, there is between 20% and 100% chance that the results of the study will returned a false positive, and this without the ability to know which percent applies.

Should that instill confidence?

Thanks, -Wade Englund-

I think that the degree of confidence in the reader depends very much on

what questions are asked before the study is ever conducted.

If the primary question we ask is, "Does word-print testing/mapping help us

determine whether there is more than one author for a text?" -- then I think

that the process can eventually instill in us confidence that the BoM was not

the product of a single writer, who just happened to insert some quotations

from Isaiah, etc., into his generally homogeneous literary production.

If the primary question is, "Which of these handful of names is the author?"

-- then the results can only instill confidence, if we are certain that one of

the named persons must be the author (or perhaps one of the authors).

Say we have a prison cell with four prisoners. One of them writes a lengthy

note, threatening the life of the Warden -- but all four deny the act. Can

word-print testing/mapping help us determine who wrote the nasty note?

If the note HAD TO come from one (or more) of those men; then yes.

But, if we expect word-printing to reliably tell us who wrote an unsigned

editorial in the Deseret News, then we will probably only discover the

true author, if we happen to have his/her word-print on file.

Lastly, there are side-effects from this sort of word-print testing/mapping,

which can begin to help us know which parts of a lengthy, complex text are

related to other, separated sections in the same text.

If a man writes his life story, and his friend takes that story and secretly

inserts lengthy blocks of additional material, we may eventually be able to

discern that pattern of insertions, even if our word-print analysis never

includes word-prints from those two men.

In other words, as a side-effect of the computerized process, we may be

able to determine where the insertions begin and end, even though we

still do not know who the two writers were, for certain.

UD

.

[wenglund]

I think that the degree of confidence in the reader depends very much on

what questions are asked before the study is ever conducted.

If the primary question we ask is, "Does word-print testing/mapping help us

determine whether there is more than one author for a text?" -- then I think

that the process can eventually instill in us confidence that the BoM was not

the product of a single writer, who just happened to insert some quotations

from Isaiah, etc., into his generally homogeneous literary production.

If the primary question is, "Which of these handful of names is the author?"

-- then the results can only instill confidence, if we are certain that one of

the named persons must be the author (or perhaps one of the authors).

Say we have a prison cell with four prisoners. One of them writes a lengthy

note, threatening the life of the Warden -- but all four deny the act. Can

word-print testing/mapping help us determine who wrote the nasty note?

If the note HAD TO come from one (or more) of those men; then yes.

But, if we expect word-printing to reliably tell us who wrote an unsigned

editorial in the Deseret News, then we will probably only discover the

true author, if we happen to have his/her word-print on file.

Lastly, there are side-effects from this sort of word-print testing/mapping,

which can begin to help us know which parts of a lengthy, complex text are

related to other, separated sections in the same text.

If a man writes his life story, and his friend takes that story and secretly

inserts lengthy blocks of additional material, we may eventually be able to

discern that pattern of insertions, even if our word-print analysis never

includes word-prints from those two men.

In other words, as a side-effect of the computerized process, we may be

able to determine where the insertions begin and end, even though we

still do not know who the two writers were, for certain.

UD

Those are all good questions and wonderful points to consider.

However, they don't surmount the critical problem I mentioned above. The word-print techniques are not designed to determine if the actual author is included in the study or not. There is no way to tell if the results of the study are a false positive or not. There isn't a way to determine if the chances of a false positive are 20% or 100%. As such, I personally don't see any reason to have confidence in the techniques--at least as far as author attribution is concerned.

As you and others have pointed out, though, the techniques may be somewhat useful in determining whether the text in question has one or multiple authors. If nothing else, the Stanford study may undermine the Joseph Only authorship theories that have gained prominance of late in certain circles. ;-)

Thanks, -Wade Englund-

[4truth]

UD:

A good question, Why-Me.

And we could add to that the fact that at least some pages

of the preserved "Dictated Manuscript" of the BoM show signs

in the handwriting, that those pages were probably written

down by Oliver Cowdery from Joseph's dictation -- presumably

whilst Joseph was gazing into a seer-stone (perhaps in a hat).

Can we use these bits of evidence to rule out examining such

writers as Solomon Spalding and Sidney Rigdon as having been

BoM textual contributors?

In answer to your question... I don't see how that would rule it out. Wouldn't that simply mean that that portion was likely dictated? It wouldn't have to have been dictated from words appearing in a stone. Joseph could either have been cheating in some as yet unknown way, or others were in on the hoax and Joseph was actually dictating & embellishing from a manuscript.

Do you know aproximately what percentage of the extant manuscript pages appear to be thusly dictated? What are the specific "signs" that indicate transcription from oral dictation (as opposed to textual plagiarism & embellishment, for example)? Also, since you say "at least some" ...does that mean there are some that don't contain signs of oral dictation?

Thanks.

[why me]

UD:

In answer to your question... I don't see how that would rule it out. Wouldn't that simply mean that that portion was likely dictated? It wouldn't have to have been dictated from words appearing in a stone. Joseph could either have been cheating in some as yet unknown way, or others were in on the hoax and Joseph was actually dictating & embellishing from a manuscript.

Do you know aproximately what percentage of the extant manuscript pages appear to be thusly dictated? What are the specific "signs" that indicate transcription from oral dictation (as opposed to textual plagiarism & embellishment, for example)? Also, since you say "at least some" ...does that mean there are some that don't contain signs of oral dictation?

Thanks.

That still begs the question: why did JS stick his head in a hat to do the translating from a book written by Spaulding, Rigdon, and Cowdery? The process was witnessed by David Whitmer and Emma Smith. JS must have had a great mind to do such a feat as to memorize a prepared manuscript and then stick his head in a hat and read off the manuscript. It is quite amazing, isn't it?

And then to find all those individuals who would be willing to be accomplices to the fraud and stick to their story for the rest of their lives is also quite amazing. Not to mention the falling out that Rigdon and Cowdery had with Joseph Smith over polygamy. Did JS try to get Nancy Rigdon as a plural wife? All good reasons to expose the fraud. But...all involved never jumped ship.

[why me]

That raises an interesting question.

IS there an explanation we have not yet considered (other than the following list)?

1. Authentic ancient record of authentic persons and events

2. Authentic ancient record, but contains errors, myths, mis-statements, etc.

3. Not an authentic record, but a fairly true reconstruction of PreColumbian history/religion

4. Partly authentic and partly fabricated -- contains some historical truth

5. Not an authentic record; nor authentic history; but still contains a divine revelation

6. The product of a 19th century deluded mind; perhaps that of Smith

7. A 19th century fabrication, but Smith thought it was an authentic record

8. A 19th century fabrication, made with good intentions by one or more writers

9. A 19th century fabrication, made for base purposes by one or more writers

10. ......... ???

Finally -- can word-printing studies in any way help to solve the issue (for us non-LDS, at least)?

UD

Well, I have said it before: if the book of mormon is a fraud, I can only wish those attempting to establish that fact good luck with best intentions. No one should be living a fraud. But I have to say that the early beginnings of the book do seem to rule it out. Too much flys against human nature that no one involved exposed the fraud even after falling out with the lds church and with Joseph Smith. And certainly some financial gain could have been made from those involved in exposing the fraud. They would have certainly have found some 'love' with the protestant denominations that existed at that time.

But it certainly is all a mystery...maybe too mysterious and maybe that is the way it needed to be.

[Uncle Dale]

...

As you and others have pointed out, though, the techniques may be somewhat useful in determining whether the text in question has one or multiple authors. If nothing else, the Stanford study may undermine the Joseph Only authorship theories that have gained prominance of late in certain circles. ;-)

Thanks, -Wade Englund-

Allow me to give yet another example, which may be helpful in our discussion here.

Suppose you are a professor in the Art Dept. of a university and you administer an

end-of-the-term "blue book" exam to your students in an art history class. I am one

of your students -- but when you read my submitted "blue book" you suspect cheating.

I am a Freshman in a class of 21 art history students (10 Juniors and 10 Seniors). In

reading my two-part essay in my "blue-book," you suspect that I received help in

writing the second part of the essay. Since all the students submitted their tests

under controlled conditions, confined for an hour's period to a testing room, you are

fairly certain that one of the other students helped me write my essay's part two.

Can word-print testing discern my probable accomplice?

Let's say that the "blue-books" were submitted on controlled, limited-access laptop

computers attached to the desks of the testing room. So, you have before you the

two-part essays of all 21 students. You are thus able to generate word-prints for

all 21 students, for each of their two parts of the "blue-book" essay.

You test the Juniors' and Seniors' two-part essays separately -- and determine that there

is a very high probability that the students thus tested each wrote both parts of their

respective essays. However, when you compare my two-part essay with the word-prints

of the Juniors and of the Seniors, I score a very high probability as author of part one,

but a low probability as author of part two.

The monitor at the time of the hour's period of testing tells you that she thought she saw

another student talking to me briefly during the testing. That student may have secretly

passed me some written notes --- she thinks it was a Junior, but is not certain.

From all of that information before you, neatly available on your computer, can you

use word-printing to "prove" that I was not the writer of the second part of my essay?

Can you determine probabilities, as to whether it was one of the Juniors, or one of

the Seniors, who was my helpful ghost-writer?

Can you rank-order the probable candidates, as to the probability of who was my helper?

My guess is, that with such information before you, through word-printing, you could rule

me out as the author of the essay's part two, based upon two or three statistical criteria.

1. My first part of the essay does not match up well with the second part. The probability

of their having the same author (based upon examining the word-prints of two separate

batches of 10 Seniors and 10 Juniors) places my ranking as part two's possible author at the

bottom of the computerized output list for all possible authorship candidates.

2. However, my first part of the essay does not match up well with the word-prints of any

of the other 20 students. You suppose that it is indeed a product of my writing, and you

can later compare it with other writing samples you obtain from me, to be sure.

3. Having decided that I did not write the second part of my submitted essay, you have

compared it (in two separate instances of computerized testing) with the word-prints of

the Juniors and Seniors. With the Juniors I rank at the bottom, with a 5% chance of

having been the writer. Compared to the Seniors, I rank at the bottom with a 6%

probability of having been the writer. Is that 1% difference relevant/significant?

4. Looking at the two sets of data, you see that among the Seniors, the highest ranking

candidate for author ranks at 15%. Among the Juniors, the highest ranking candidate

for author ranks at 90%, with the next highest writer scoring only 16%.

Have you found my secret accomplice?

Answer: No -- you have only determined a relative probability, based upon examining

21 word-prints. The answer you have thus obtained may be good enough for you to

decide to boot me and the other student out of the class for cheating --- but it is NOT

absolute proof that I cheated.

If there is a "reasonable doubt," in your conclusions, I may be able to take you to

court and get a jury to award me damages for your actions against me.

This has a direct analogy in our Book of Mormon authorship investigations. So long as

the Mormons can cite a "reasonable doubt" in considering word-print findings, they can

state that the supposed 19th authorship has NOT been demonstrated "beyond a doubt."

UD

.