Is the Book of Mormon Homogeneous?

[Benjamin McGuire]

Like Holley, I find these parallels striking on their own merit. Before computers came along, Dale pointed out the vocabulary overlap that you now characterize as "misleading." But we have here a convergence of evidence for a story that is also (independently) attributed to Spalding by Jockers method and now also clusters with other Spalding attributed BOM chapters on Bruce's PCA chart, all of which support the original assertions of the witnesses who, in 1833, started this Spalding ball rolling.

It isn't just me that labels it misleading. It is a century of discussion on literary criticism. Shall I provide a couple more? One of the favorites is from a letter written by Lord Tennyson in response to a Canadian edition of his poem []The Princess. There, among other things, he makes this comment:

But there is, I fear, a prosaic set growing up among us, editors of booklets, book-worms, index-hunters, or men of great memories and no imagination, who impute themselves to the poet, and so believe that he, too, has no imagination, but is for ever poking his nose between the pages of some old volume in order to see what he can appropriate. They will not allow one to say "Ring the bell" without finding that we have taken it from Sir P. Sidney, or even to use such a simple expression as the ocean "roars," without finding out the precise verse in Homer or Horace from which we have plagiarised it.

Brander Matthews in 1902 wrote this (referring to the letter above):

A pleasant coincidence of thought is to be noted between these words of Lord Tennyson and the remarks of Sir Walter Scott about 'Gil Blas.' Both poets think ill of the laborious dulness of the literary detective, and suggests that he is actuated by malice in judging others by himself. The police detective is akin to the spy, and although his calling is often useful, and perhaps even necessary, we are not wont to choose him as our bosom friend; the amateur literary detective is an almost useless person, who does for pleasure the dirty work by which the real detective gets his bread.

By the way, these are some of the more kind remarks. In 1929 E. H. C. Oliphant commented on this topic in his essay titled

[Benjamin McGuire]

Dale writes

So then, we have come full circle, to what my priesthood superiors told me thirty years ago: "There is nothing here to explain."

You can put this however you like, but yes, in essence, you haven't presented me with anything that requires an explanation - but I have told you how you can correct this, and produce something interesting that could use an explanation.

I asked them two simple questions: #1. "Was Joseph a polygamist," and #2. "Was there something in the Book of Mormon that made early witnesses assert it was not written by Nephites?"

In both cases, I was told that there was "nothing to explain." Of course, Joseph Smith was never a polygamist, and so the RLDS Church was indeed the one true church. And the early claims against the Book of Mormon were created by a certain D. P. Hurlbut, who made the whole thing up and found liars who would sign false statements, to back up his absurd fabrications.

The problem isn't these questions Dale. The problem is that what you have produced isn't capable of providing an answer. You will at some point have to deal with that issue if you want to carry this forward. The concerns about Hurlbut seem to me to be quite legitimate. But we have hashed out some of that already without actually coming to much common ground. I have real problems conceptually with this though. On the one hand, you seem to be suggesting that there is absolutely no reason for Hurlbut and those associated with him to lie (regardless of any evidence here for the moment) but it seems that implicit in your argument is the assumption that Joseph Smith and those associated with him must be lying. This is a real inconsistency, and its why - at least from my perspective - you have this bias that you are ignoring right from the beginning. You want to choose who is lying and who is telling the truth. In attempting to remake my position into a theological discussion, you have one real problem - and that problem simply put is this - attribution studies of the sort you are carrying out are not new. They have been going on for a very long time. There is a lot of discussion (some of it quite empirically based) on methodology, on criteria, on how evidence can be produced and used. In your search, you haven't gotten into any of this. Your proposals stand in isolation - outside of what is accepted in normal scholarly practice. This doesn't mean that your familiarity with the texts can't be useful still in doing what you set out to do - it just means that you need to find a way to deal with the gaping holes in your process. And if you choose to do something like this, you will either find a way to express what your intuition tells you in a way that has real merit, or you may find yourself moving across the aisle to agree with me.

And do you know what, Ben? -- For years those priestly answers were good enough for me. They made perfect sense. I felt better about paying my tithing and donating countless hours to church work.

Then I met Vernal Holley, and you know all that followed.

But changing one authority for another uncritically is never productive and never useful in a quest for knowledge. I am not asking you to take my position uncritically. I am trying to provide you with some tools to further the dialogue. But this kind of response tells me effectively that you too believe that "there is nothing here to explain" - just from a different side of the coin. But throwing the religious aspect of it towards me isn't going to help in this discussion. It really doesn't matter how you got to where you are, the question is where are you going to go with it. I believe that you can apply this vast knowledge that you have (and it is vast) and your impressive familiarity with early 19th century texts and literature (which is truly impressive) to the question at hand, and look at it in completely different ways, you may well come up with something that excites your interest and curiosity again. But "I know it when I see it" (not quoting you - but others as you know) isn't much of an argument to make.

Ben M.

[Uncle Dale]

...

But this kind of response tells me effectively that you too

believe that "there is nothing here to explain"

...

So far as Nephites are concerned -- yes, that has become my position.

Unless you can present evidence that convinces the non-LDS experts,

academics, scholars, scientists, etc., I have given up on "Nephites."

Nevertheless, I find their names convenient for talking about the

Book of Mormon narrative. We have something that the book itself

calls the "Record of Helaman," in the latter part of a text that

the book itself calls "Alma," which the book itself says was recorded

on the "large plates of Mormon."

All of that terminology I still find useful. But when I speak of

Mormon redacting Helaman, I am not supposing those were actual

Americans living somewhere in the jungles of Mexico, etc.

Why should I have to "explain" the case for historical Nephites?

I do not have to do that, and so I move on to what I can discuss.

That leaves the possibility that either Joseph Smith wrote the

book himself -- or that he edited/dictated material not fully

composed by himself.

However, for purposes of our discussion here, we can still speak

of Mormon having edited Helaman. Shall we agree that is what the

book itself purports for Alma 46-53?

And, if we can find at least that much common ground, can we move

on to discussing how we might determine whether or not the same

writer composed Alma 45 as composed some (or all) of Alma 46-53?

This is the sort of thing I had hoped some LDS scriptural expert

would contribute to this thread long before now.

Obviously the Church teaches certain things about the "prophets"

of ancient America, and what they wrote, and who wrote what.

Is there some reason we are not discussing such stuff?

Bruce produced a specific pca chart, by which he seemed to say

that the authorship controversy was over -- settled. If his chart

carries that much weight, in terms of scientific authority, then

my pointing out the cluster of Alma 46-53, in that very same chart,

should be worthy of at least a superficial examination.

You and Bruce may try to convince me that I have used inappropriate

means to identify "Spaldingish" chapters in the book --- that there

are far better, rigorous and exhaustive methodologies, by which to

determine whether or not the dots for Alma 46-53 should be colored

in red on that chart.

That's OK. If I have colored in the wrong dots, then simply show

me, on Bruce's chart, what scientific, non-subjective methodology

DOES identify as mathematically "similar" to Spalding's fiction.

Surely somebody has an hour or two of free time, by which they

can settle the "end of the controversy" in support of Bruce's discovery.

???

UD

.

[Uncle Dale]

...

Surely somebody has an hour or two of free time

...

Suppose that a Mormon student of the book DID take the time to

answer my question -- and suppose that this was his answer:

"Dale, I ran the two texts through several computerized comparison

methodologies, and have thus determined the ten most 'Spaldingish'

Book of Mormon chapters -- they are:

Mosiah 14, Moroni 3, Ether 11, 2Nephi 14, Alma 8,

Mormon 6, 3Nephi 5, Alma 45, Mosiah 25, and Enos,

in descending order. Comparing Spalding's own personal

letters #1, 2 & 3, as receiving the highest match

with his Oberlin MS (99.7 to 100%), I rate all of the above

Book of Mormon chapters as 2% or lower in their

relative (comparative) degree of 'Spaldingness.'"

And my theoretical LDS correspondent might add:

"Also, Dale, you will notice that these 10 chapters

are not contiguous and do not fit together well, in

telling any one story, or relating anything in particular.

The only possible conclusion, is that your Conneaut

witnesses (and subsequent witnesses) were NOT gleaning

their so-called recollections from the parts of the

book most resembling Spalding's writings (ranging from

0.19% up to 1.87%, respectively).

You have had your fun, and made a useful contribution

to on-line resources, but you can now retire in peace."

????

UD

.

[4truth]

Ben wrote:

You see, the elements that you present aren't really that disparate. They occur fairly regularly in texts - particularly those about warfare. Rivers make a natural barrier to army movements. Hills (or mountains) create geographical barriers that can conceal armies. When you want to have an ambush, these kinds of features make sure it will work. You don't have an ambush in the middle of a plain with nothing to hide behind. It just doesn't work well ...

In other words... coincidence.

You're right, we've been down this road before. Holley notes something quite unusual (actually this is just one example of several unusual somethings between the two works) and you pull up your favorite quotes critical of parallel hunters in renaissance plays. You highlight the (expected and explainable) differences as though they somehow totally negate the similarities. You seem to want the same story reprinted word for word before you will accept it as something unusual.

If it were simply this one set of parallels, your case against the broader comparison of Spalding to the BOM would be reasonable. If it were simply a few obscure observations of similar shared phrases, we could all agree that there's not much here to go on. Of course, it's merely my subjective opinion, but, IMHO your best attempts to paint each independent component as "bad evidence" fail which, in turn, dooms your larger point that the resulting combination is therefore worthless.

Regardless, it is apparent that no matter how detailed the parallels get, you will focus on the differences and then claim those differences cancel out the similarities. Of course, that is your prerogative. I simply think you're being unreasonable in the process. Any truly objective observer can see the similarities exist. Heck, even biased observers can see them. So they either exist because of coincidence or dependence.

But the larger point is that this is merely one set of parallels within a larger group that Holley and Broadhurst have identified--none of which had to be there, and, indeed(!) were originally alleged by critics of the S/R hypothesis (like yourself) to have not existed at all.

And yet, now we have critics on the other side (like Chris Smith) who suggests it was these and other similarities that led the Conneaut witnesses to conclude there was indeed a connection between the two works! Even though they claim to have read about Nephi and Lehi (etc.), we know they were actually exposed to the Oberlin Manuscript (because no other manuscript exists and it's irrational to think that if it ever did it wouldn't now) and came to their false memories because, by george, there are some parallels there after all!

It's not just a detailed set of parallels in Alma 37 as this thread illustrates. I merely brought that one example up in response to your challenge to Dale to come up with something uniquely "Spaldingish."

Of course it's merely my opinion, but to blatantly plagiarize, I'm beginning to conclude that S/R critics "do not, as a rule, set out to be truthful and impartial. They are hell-bent on proving a point."

All the best.

[Uncle Dale]

...

I'm beginning to conclude that S/R critics "do not, as a rule,

set out to be truthful and impartial...

As we all know, there are two categories for such "critics;"

those who are ostensibly "objective," and those who have a

God-given testimony that the book is truly an ancient text.

In the Church tradition I come out of, the latter sort of

"critics" do not articulate their testimonies so much along

the lines of the book being "true," as the gospel being "true."

That has allowed them to at least tolerate researchers like

myself. They may be my "critics," but they still sit next to

me on the bench on Sunday morning to take the sacrament.

I'm not sure that sort of toleration exists within the LDS Church.

I doubt that a member could function (for very long) as a Seventy

(like B. H. Roberts) or an Apostle, who advocated a 19th century

origin for the Book of Mormon. We have a few such leaders in the

Reorganization, but they tend to be "Smith-alone" advocates, and

thus remain my critics.

I have encountered a few LDS who were willing to "leave their

testimonies on the doorstep" and proceed in discussion and

discovery, with an open mind as to whether Nephites are "true."

And -- without exception -- all such folks I've dealt with have

eventually left the Church. You can still find a handful of those

people over on the "Message Board That Shall Not Be Named," but

most of them have simply dropped out of sight in recent years.

Sooner or later we shall see another Palmer, Vogel or Quinn, who

departs the LDS over matters such as BoM authorship. And it is

just a matter of time before one of those departing scholars

follows after Craig Criddle and adopts some (or all) of the S-R

arguments as his/her own.

But that is not what I'm trying to accomplish here. I'd be mortified

if Bruce or Ben (or even Glenn) agreed with me to the point of fully

abandoning his testimony.

I'm more hopeful that one of them will supply useful information in

a discussion such as this one. Bruce did so with his pc1/pc2 chart --

providing me with an excellent opening graphic for my "Book of Solomon"

pages. I mean, come on! -- where else would I have found three dozen

of Jockers' "Spalding chapters" all lumped together in a single picture.

Probably Ben will provide something similarly useful regarding Rigdon's

language, and affinities for it within the Nephite Record. I'm patient.

No need to re-invent the wheel, time after time. But it is a bit

tiresome --- waiting for the Mormons to do my research for me.

UD

.

[Benjamin McGuire]

Roger writes:

You're right, we've been down this road before. Holley notes something quite unusual (actually this is just one example of several unusual somethings between the two works) and you pull up your favorite quotes critical of parallel hunters in renaissance plays. You highlight the (expected and explainable) differences as though they somehow totally negate the similarities. You seem to want the same story reprinted word for word before you will accept it as something unusual.

Not at all. I want something reasonable. The problem is that for you, what you have is reasonable. It cannot be coincidence you say - but, you have nothing to base it on. Merely your own intuition. What you have is a methodological house of cards.

If it were simply this one set of parallels, your case against the broader comparison of Spalding to the BOM would be reasonable. If it were simply a few obscure observations of similar shared phrases, we could all agree that there's not much here to go on. Of course, it's merely my subjective opinion, but, IMHO your best attempts to paint each independent component as "bad evidence" fail which, in turn, dooms your larger point that the resulting combination is therefore worthless.

But all it is is a set of parallels. There is no other evidence Roger. You have built a monument of speculation without a shred of evidence. Every time I have asked you to present some formal methodology (and there are certainly several out there), you simply opt out. You aren't interested in making a reasonable argument - you merely want to continue asserting the same thing over and over again. And, to be honest, it simply doesn't work.

Regardless, it is apparent that no matter how detailed the parallels get, you will focus on the differences and then claim those differences cancel out the similarities.

This is easy for you to say, but, you have NEVER given me any significant details. This list of eleven parallels is not detailed. It seems to be attempting (as I point out) to reduce the two narratives to very common denominators just to make the theory stick. And never have you attempted to deal with the actual differences that occur - difference which are plainly obvious to anyone who reads the texts in context. The stories are not the same - only in creating soundbites of them can you even begin to argue that they are like each other - only in the most general terms. We try to get more detailed and we end up with what I posted. What I see here is denial - you suggest that I won't care how detailed you get - but lets try it once, shall we? Why don't you try dealing with my concerns in a detailed fashion. Use the texts. Show me how my analysis of the parallels is inappropriate. Because personally, I think that you are simply blowing smoke here.

Any truly objective observer can see the similarities exist. Heck, even biased observers can see them. So they either exist because of coincidence or dependence.

Any objective observer can find similarities between any two texts. It doesn't matter what the texts are. Give me two texts and I can tabulate similarities between them. I have already demonstrated this. That isn't the issue here. The issue is whether those similarities come even close to establishing a claim that one text is reliant on another, or that one text was at least in part authored by a particular individual. That is the question here.

But the larger point is that this is merely one set of parallels within a larger group that Holley and Broadhurst have identified--none of which had to be there, and, indeed(!) were originally alleged by critics of the S/R hypothesis (like yourself) to have not existed at all.

I have never alleged that parallels didn't exist. I have consistently alleged that they are contrived. Holley and Broadhurst both present nothing more than parallelomania. I have yet to see a defense against that charge. There is no question that the kind of approach involving two columns displaying parallels is easy to make, is convenient - it even looks good. There is also no doubt that in the academy, such presentations are nearly universally considered deceptive - and consequently not useful as evidence for the claims that are being made.

And yet, now we have critics on the other side (like Chris Smith) who suggests it was these and other similarities that led the Conneaut witnesses to conclude there was indeed a connection between the two works!

Yes, funny that. You insist that these parallels imply some kind of biological connection in the case of the Book of Mormon and Spalding, but you insist that there is no reason to make that same assertion for the much better parallels in the letters that Hurlbut collected. Why is this? Was the Book of Mormon a source used by the translator of Jules Verne's Around the World in 80 Days? If not, then why can I find similar connections between it and the Book of Mormon as Dale finds between the Spalding text and the Book of Mormon? Its not that the connection between the Verne book and the Book of Mormon is very strong, its that the connection between the Spalding text and the Book of Mormon is quite weak. The problem is that you have never looked at any other texts - you have never compared them (Dale hasn't either). If you compared the Book of Mormon in such a detailed fashion to the rest of early 19th century literature, you would find this to be the case. At best, the Spalding connection is typical.

Even though they claim to have read about Nephi and Lehi (etc.), we know they were actually exposed to the Oberlin Manuscript (because no other manuscript exists and it's irrational to think that if it ever did it wouldn't now) and came to their false memories because, by george, there are some parallels there after all!

The way that you would come up with a good argument for the Hurlbut texts is to show details in the Hurlbut documents that aren't attested elsewhere. There isn't a single one of those documents that actually requires exposure to any Spalding manuscript. Every one of the details in them can be accounted for in published material about the Book of Mormon. They all recognize Nephi and Lehi (but where is Zerahemnah? Or even Lobaska or Hadokam?) We get this wonderful geographic connection where Zarahemla is placed relative to real world locations - only to discover that quite some time earlier, Pratt had already been teaching this geography model (and we find it published in local newspapers). Its not that your theory isn't demonstrable - its that you simply haven't done it. You keep repeating this litany of how no one will accept your position because of a prior bias. My rejection is the fault of a bias - its a rejection of your presentation of the theory. There is no evidence for it.

It's not just a detailed set of parallels in Alma 37 as this thread illustrates. I merely brought that one example up in response to your challenge to Dale to come up with something uniquely "Spaldingish."

Then you completely misunderstood my use of the phrase. What makes something Spaldingish isn't that Spalding uses it. We expect that Spalding himself plagiarized as much as anyone else does (as William Inge once wrote: "What is originality? Undetected plagiarism. This is probably itself a plagiarism, but I cannot remember who said it before me.") Spaldingish language is the things that occur in Spalding that don't occur elsewhere. Your list of 11 parallels doesn't even begin to approach something Spaldingish.

If this was just an example, then bring me more. I can tear them all apart - its not hard, because its not detailed. I am certainly willing to look at them - but perhaps instead you would be willing to respond to the detailed analysis I provided above. I think it would be a good place for you to start.

Of course it's merely my opinion, but to blatantly plagiarize, I'm beginning to conclude that S/R critics "do not, as a rule, set out to be truthful and impartial. They are hell-bent on proving a point."

Ah yes. Well let me leave you with the final remarks that that particular author made at the conclusion of his nine vices (and before I do, I want to point out that I am in fact using recognized scholarship on the subject - and you are using .... what?):

Whatever his vices or virtues, the parallel-hunter is a hardy species. He is destined, as someone had said, to persist until Judgment Day, when he will doubtless find resemblances in the very warrant that consigns him to the nether regions.

Ben M.

[Uncle Dale]

...

My original PCA graph showed that the actual Spalding writings were

completely separate from all Book of Mormon chapters

...

I think that this is a dangerous conclusion, presented without

reference to the scientific evidence showing that a pc1/pc2 chart

is more reliable for authorship attribution, than are other charts

in the pca series.

For example, I will paste in below my copy of the pc2/pc3 chart,

in which Spalding's chapters DO overlap Book of Mormon chapters.

I have centered a pink disk upon Jockers' Spalding #17 -- which

most readers will agree contains a good deal of religious subject

matter, with words found in the Book of Mormon -- and have chosen

a radius that extends out to the most distant Oberlin MS chapter

from that center point. To the right of center ("east," plus x-axis)

the number of red dots diminish -- the same may be said for the

"southeast (x/plus, y/minus). To the left of center ("west,"

minus x-axis) the frequency of red dots/squares increases -- the same

may be said for the "northwest" (x/minus, y/plus). These are easily

visible distribution patterns for the red (Spalding-attributed

BoM chapters) dots and red squares (Oberlin MS sections).

As can be seen by a very brief inspection, more red dots fall within

the pink disk than fall outside of it. There is definitely some sort

of overlap going on here. It is probably not all based upon linguistics,

or else the two Biblical texts would fall outside of the pink bounds.

Various factors (such as text length, inter-spacing of repetition, etc.

may also contribute to this pattern of dispersion).

This is not the best pca chart I could have selected for presentation

here -- but it is the next one in the series I have had created --

which extends upwards to pc9/pc10. Other charts in the series show

a better overlap of Spalding's Oberlin MS chapters and BoM chapters.

Unless some mention is made of such facts, I strongly suggest that

pca charts NOT be used to show "that the actual Spalding writings were

completely separate from all Book of Mormon chapters..."

Uncle Dale

-------

added note: chart should read "PC2 is the X-axis; Pc3 is the Y-axis."

[Uncle Dale]

...more red dots fall within

the pink disk than fall outside of it....

The astute observer will also notice that most of the

Record of Helaman sequence I've mentioned here so

often falls within the bounds of the pink disk --

that includes Alma 46-47-48-49-50-51.

In other pca charts in the series, several of these

Record of Helaman chapters also become red dots,

positioned relatively close to one or more of the

red squares (representing the Oberlin MS chapters).

Call it coincidence, if you wish -- but do not suppress

such discoveries merely for pro-LDS polemical purposes.

UD

[Chris Smith]

And yet, now we have critics on the other side (like Chris Smith) who suggests it was these and other similarities that led the Conneaut witnesses to conclude there was indeed a connection between the two works! Even though they claim to have read about Nephi and Lehi (etc.), we know they were actually exposed to the Oberlin Manuscript (because no other manuscript exists and it's irrational to think that if it ever did it wouldn't now) and came to their false memories because, by george, there are some parallels there after all!

I recommend you pick up a copy of the Cowdery/Davis/Vanick book and read through the first few chapters. Although their intention was to persuade readers of the theory, I found it to be a very effective antidote. Reading about the theory in bits and pieces on the boards makes it seem persuasive, but once all the material is gathered together in one place and treated comprehensively, the weaknesses of the evidence become glaringly obvious.

[4truth]

Ben & Chris:

I will create another thread to continue the discussion there so as not to derail this one.

Uncle Dale:

For example, I will paste in below my copy of the pc2/pc3 chart,

in which Spalding's chapters DO overlap Book of Mormon chapters.

I have centered a pink disk upon Spalding's chapter #17 -- which

most readers agree contains the most linguistic and thematic

parallels with the Book of Mormon -- and have chosen a radius

that extends out to the most distant Oberlin MS chapter from that

center point.

As can be seen by a very brief inspection, more red dots fall within

the pink disk than fall outside of it. There is definitely some sort

of overlap going on here. It is probably not all based upon linguistics,

or else the two Biblical texts would fall outside of the pink bounds.

Various factors (such as text length, phrase repetition, etc. may

also contribute to this pattern of dispersion.

I know I keep asking this over and over... sorry. It still has not quite sunk in. Can you briefly explain why there is a difference between PC1 and PC2 and PC3, etc? Or in other words, why did the dots move? You seem to be saying, on Bruce's chart the dots are here, now on this chart they've moved. If the answer is "we don't know" then I'm really confused... since you produced the charts, how could you not know what causes them to change?

Thanks.

[Uncle Dale]

Ben & Chris:

I will create another thread to continue the discussion there so as not to derail this one.

Uncle Dale:

I know I keep asking this over and over... sorry. It still has not quite sunk in. Can you briefly explain why there is a difference between PC1 and PC2 and PC3, etc? Or in other words, why did the dots move? You seem to be saying, on Bruce's chart the dots are here, now on this chart they've moved. If the answer is "we don't know" then I'm really confused... since you produced the charts, how could you not know what causes them to change?

Thanks.

The pc1/pc2 chart exhibits those characteristics of the texts

MOST deviant from the mean, and SECOND most deviant from the mean.

The next logical chart in the series would be pc3/pc4, which exhibits

those characteristics of the texts THIRD most deviant from the mean,

and FOURTH most deviant from the mean.

But, to fill in a sequencing gap between pc1/pc2 and pc3/pc4,

I created a special chart -- pc2/pc3. It thus includes some of

the texts' characteristics which we viewed (expressed as the y axis)

in pc1/pc2. I need to clarify that my new chart uses pc2 as its

x-axis and pc3 as its y axis.

At any rate, my new chart preserves some of the relationships seen

in the first chart, where the BoM texts were mostly widely separated

from the 19th century authors.

When we re-cast the data as a pc2/pc3 chart, that separation disappears

(since it was mostly caused by the pc1 x axis).

When we remove the pc1 separation, the various texts just naturally

begin to overlap in their distribution patterns.

This is NOT the best illustration I could have shown. According to

Bruce's previous "heat map," some of the principle components after

pc3 do a better job of bringing individual 19th century authors into

juxtaposition with BoM chapters' "clusters."

Bruce's own categorization places Alma 46-53 in a common cluster.

Right now I'm trying to generate a BoM "chapter tree" in which the

most textually similar chapters are visually brought together, I know

from the raw data that Alma 43, 44, and 46-53 will be adjacent on that

"tree," but am having a heck of a time making the chart.

When I get that finished, even Bruce and Ben will have to admit that

those parts of the Record of Helaman are more similar to each other,

than they are similar to most of the rest of the book.

That will be a small (but necessary beginning) for my intents here.

UD

[Uncle Dale]

I recommend you pick up a copy of the Cowdery/Davis/Vanick book and read through the first few chapters. Although their intention was to persuade readers of the theory, I found it to be a very effective antidote. Reading about the theory in bits and pieces on the boards makes it seem persuasive, but once all the material is gathered together in one place and treated comprehensively, the weaknesses of the evidence become glaringly obvious.

Criddle has a new (and very long) PowerPoint demonstration

on the web, that is 100 times better than Art's book.

Ask him for the URL, and take a look for yourself.

UD

[Uncle Dale]

...

I'm really confused...

Then relax your mind with a little parallelomania -- it will drive

Ben crazy. On this new chart we can see that Alma 2 and Alma 3

plot out close to Sp10 and Sp17. Those are Oberlin MS chapters

8 and the end matter (because Jockers used different numbering):

http://solomonspalding.com/SRP/ColorSMS.htm#chap08

http://solomonspalding.com/SRP/ColorSMS.htm#endmatter

Read through Alma 2-3, and see whether or not the language

seems at all similar to Spalding:

http://solomonspalding.com/SRP/MEDIA/BookAlm1.htm#224a

UD

[Chris Smith]

Criddle has a new (and very long) PowerPoint demonstration on the web, that is 100 times better than Art's book.

Ask him for the URL, and take a look for yourself.

I don't have Craig's contact info. And in any event, last time I heard from him he was intimating that my rejection of the S-R theory was motivated by selfish careerism. I'm not exactly looking for an excuse to make his acquaintance. However, I'll probably make time to take a look at the link, should you or another of Craig's friends feel led to post it.

[Benjamin McGuire]

Dale writes:

I think that this is a dangerous conclusion, presented without reference to the scientific evidence showing that a pc1/pc2 chart is more reliable for authorship attribution, than are other charts in the pca series.

Dale, this is part of where you don't understand the data.

Take the data to be represented by a pencil. We hold the pencil up in front of a projector, and the light puts a shadow on the screen. Now, if we hold the pencil just right, we get a long cylindrical shape. If we hold it another way, we get a small dot. This is kind of like your different PCA plots. The moment we start considering PC2, PC3, and so on - as you do, all you are doing is rotating the pencil - away from the view that is most descriptive of the data. Eventually, you will get a tight point. In effect, by using these other plots you are taking out of consideration the features of the text which are the most different. If we only use the 3 or 4 most common features, all you will end up with is a very tight point of data - but, taken this way, any set of texts by any set of authors will also produce exactly the same thing. The thing about the various PCx plots is that they aren't independent but cumulative. That is, the the most descriptive plot shows the data most effectively. After that, the adding additional plots increases the distance - they never narrow it. The only way you could use PC2 or PC3 as original coordinates is to suggest that PC1 simply does not exist. I am not sure its necessary for Bruce to have to explain the basic statistical principles of his analysis. It is discussed here a bit:

http://en.wikipedia.org/wiki/Principal_component_analysis

Where we read:

"PCA essentially rotates the set of points around their mean in order to align with the first few principal components. This moves as much of the variance as possible (using a linear transformation) into the first few dimensions. The values in the remaining dimensions, therefore, tend to be highly correlated and may be dropped with minimal loss of information."

The further down the series you go, there is increasingly less value in the results in a non-linear fashion. And, the results as you go further down need to be added to the first sets of numbers. You don't get any value from PC3 unless you make a 3-Dimensional rendering involving PC1, PC2 and PC3. PC4 would be adding in a 4th dimension.

Ben M.

[Uncle Dale]

...

The further down the series you go, there is increasingly less value in

the results in a non-linear fashion.

...

Doubtlessly correct.

Even evolving from pc1/pc2 over to pc2/pc3 obviously leads to a

less specific (or less constrained) view of spatial relationships

between the plot points -- and thus, probably, of actual relationships

between the texts themselves.

But ---- I still say that using pc1/pc2 as an authority on authorship

attribution is a dangerous conclusion. We do not know what factors are

involved in producing all of those plot locations. It appears to be

a relatively smaller set of factors/characteristics than we employ

"further down the series."

I assume that there are better ways of creating scatter diagrams of

related (or non-related) texts, than the method which progressively

narrows down the number of characteristics being employed in charting.

I suspect that what we are left with, in pc1/pc2, is only that set

of data which furthermost removes all plotted points from the mean.

The factors/characteristics may thus be a relatively small sub-set,

as well as comprising relatively small differences.

Once we label the identities of plots inside of the Book of Mormon

"cloud" in pc1/pc2, we see that the expected internal clustering

is "off" -- something is wrong -- Moroni's chapters do not cluster;

Ether's chapters do not cluster. Other supposed writers of the

Nephite Record have their respective chapters scattered over

relatively great distances -- including the Isaiah sub-set.

If Bruce wishes to say that HE HIMSELF has concluded that the

groupings in pc1/pc2 put an end to the Spalding-Rigdon claims;

then he certainly can publish that opinion. He can offer up

his hypothesis for inspection and testing.

On the other hand, if he makes that pronouncement as an axiom --

as a maxim we all must accept with the same assurance as we

accept that "2+2=4," then I assert that he must also provide

some context, by admitting overlap of authorship attribution

"clouds" in other charts in the pca series.

My great suspicion is --- that many of the Book of Mormon chapters

are complex texts which will not plot out neatly within the

confines of any particular known author's "cloud" -- and that

the unusual, quasi-biblical format of the Nephite Record causes

its chapters to roughly cluster around Isaiah, Malachi & Matthew.

On the other hand, I have identified a "resistant cluster" in

the sequence of Alma 48-51 (or 46-51, or 46-53) which remain

closely grouped in all sorts of analytical depictions.

Suppose that we "word-print" Alma 46-53 and map its possible

distribution (as a "voice") across the entire set of BoM chapters.

If NSC methodology requires some "actual contributors" among its

author-candidate list, then we have one here, already.

If we go back to the 1830 edition, and retain its "Preface," we

have another identified author-candidate, in the person of

Joseph Smith. We can also add in Isaiah, Malachi & Matthew as

known contributors. Possibly two or three other BoM "known"

contributors could also be isolated and word-printed.

Do that sort of additional homework -- and then submit the

resultant NSC classification, accompanied by its own pca chart.

At that point we would have some context for Bruce's implied

assertion, that 19th century BoM authorship claims are effectively

ended for Spalding, Rigdon, Pratt and Cowdery.

As Bruce's contention now stands, I find it as suspect and as

unconvincing as his other conclusion -- that Spalding and the

BoM share absolutely no stylistic similarities.

But, don't take my word for it --- publish the "science"

behind such reliance upon that specific pc1/pc2 chart, and

sit back and wait for the professional response.

UD

.

[4truth]

UD wrote:

The pc1/pc2 chart exhibits those characteristics of the texts

MOST deviant from the mean, and SECOND most deviant from the mean.

The next logical chart in the series would be pc3/pc4, which exhibits

those characteristics of the texts THIRD most deviant from the mean,

and FOURTH most deviant from the mean.

...then it does seem to be the case that we can't (at least at this point) put our finger on exactly what it is that causes "those characteristics of the texts MOST deviant from the mean, and SECOND most deviant from the mean," etc. Hence the speculation that it is probably King James English. I think I am catching on.

But, to fill in a sequencing gap between pc1/pc2 and pc3/pc4,

I created a special chart -- pc2/pc3. It thus includes some of

the texts' characteristics which we viewed (expressed as the y axis)

in pc1/pc2. I need to clarify that my new chart uses pc2 as its

x-axis and pc3 as its y axis.

At any rate, my new chart preserves some of the relationships seen

in the first chart, where the BoM texts were mostly widely separated

from the 19th century authors.

When we re-cast the data as a pc2/pc3 chart, that separation disappears

(since it was mostly caused by the pc1 x axis).

Okay so let me really go out on a limb.... could that mean that the PC1 factor that was causing the separation in Bruce's chart was indeed King James English? If so, then you're saying when that factor is removed, so is the separation?! Hmmm! That seems to be what you're saying here:

When we remove the pc1 separation, the various texts just naturally

begin to overlap in their distribution patterns.

Which can't be good if you're trying to argue that the separation caused by PC1 means there can't be any connection between the two clouds. On the contrary, there can be a connection if PC1 is simply King James English.

This is NOT the best illustration I could have shown. According to

Bruce's previous "heat map," some of the principle components after

pc3 do a better job of bringing individual 19th century authors into

juxtaposition with BoM chapters' "clusters."

Bruce's own categorization places Alma 46-53 in a common cluster.

Right now I'm trying to generate a BoM "chapter tree" in which the

most textually similar chapters are visually brought together, I know

from the raw data that Alma 43, 44, and 46-53 will be adjacent on that

"tree," but am having a heck of a time making the chart.

When I get that finished, even Bruce and Ben will have to admit that

those parts of the Record of Helaman are more similar to each other,

than they are similar to most of the rest of the book.

That will be a small (but necessary beginning) for my intents here.

Which means what? That Alma and Helaman were the same dude?

All the best.

[Uncle Dale]

...Okay so let me really go out on a limb.... could that mean that

the PC1 factor that was causing the separation in Bruce's chart

was indeed King James English?

...

I'm not sure I would state it in those words, that way.

Remember, the only data our computer program has to work

with is our compilation of frequently occurring non-contextual

words -- differentiated into sub-sets, according to the various

author-candidates' "word-prints."

With that massive amount (but very limited scope) of data to

sort through, our computer spits out the pc1/pc2 chart --- in

which the most EXTREME deviations from the mean plot out.

The first set of principal components realized in that chart

includes some mathematically small sub-set of relationships

which cause the texts to "cluster" as we see them do there.

I suppose that the operative force is one of repulsion --

pushing everything away from the x=0/y=0 center of the chart.

As a result of that process certain sets of dots group together.

Not because our computer was attempting to isolate those "clusters"

of author-candidates' texts ---- but because our computer was

trying to show us a very small sub-set of language patterns which

pushed all the dots outward from the 0,0 point.

Now you ask why the BoM dots cluster as they do.

Half of the answer is that the pc1 effect pushes them to the right.

The other half of the answer is that the pc2 effect pushes them up.

Why?

Because there is something about the content of the frequently

occurring non-contextual words' relationships in those texts which

causes those effects. Since we are only measuring the most extreme

cases here, it must be a very small and minor set of differences.

Say, maybe, the ratio of "beholds" to "wherefores." --- But, since

those words are not included in our tabulations, the effect must

be caused by some less evident surrogate for my example. Say, maybe,

the number of "thes" divided by the number of "ands", plus the

square roots of the "buts" occurring in those texts.

Don't try to comprehend the exact reason.

Just look at the most distant-from-the-mean BoM chapters and

compare your reading experience with that occasioned by browsing

through the Oberlin manuscript. What languages "jumps out" at you

as being remarkably "different?" --- That will be the "tip of the

iceberg" for the hidden surrogate.

Consult my previously-posted blow-up of pc1/pc2, in order to

construct your reading list of the most oddball BoM chapters,

up in the "northeast" of the BoM "cloud."

UD

.

[Uncle Dale]

...

Which means what? That Alma and Helaman were the same dude?

...

More like, Alma the younger was a Spaldingesque pastiche,

created by Oliver Cowdery and Parley P. Pratt.

And, when they made up their Spalding emulation, Helaman

was the next real Spalding character in the narrative --

so the story continues with his "record," while our hopeful

emulators get some much need rest from their writer's cramp...

UD

[Uncle Dale]

...

Consult my previously-posted blow-up of pc1/pc2, in order to

construct your reading list of the most oddball BoM chapters,

up in the "northeast" of the BoM "cloud."

...

Another useful exercise would be to look among Jockers'

totals for all the frequently used non-contextual words,

and determine which of those words the Book of Mormon chapters

USE THE MOST FREQUENTLY, but the 19th Century authors use

the LEAST FREQUENTLY -- and the other way around.

What I'm trying to say, is that you could identify those words

which do not match up very well in the frequency of their occurrence

among the several 19th Century writers and in the Book of Mormon.

The factor pushing the Book of Mormon chapters to the right hand

side (and top) of the pca chart may be something as simple as

there being a couple of words appearing very often in ALL the

Book of Mormon chapters, but not very often among the other texts.

The difference in frequency may be relatively small, and thus not

immediately discernible to the reader's eyes.

However, once such a culprit is suspected, it could be temporarily

removed from the tabulations and the pca charting re-compiled. By

trial and error the "culprit" set of differing words could probably

eventually be determined.

But there is something other than that sort of "culprit" at work

in the distribution of plots on the chart -- because, besides just

being separated from the BoM "cloud," the various 19th Century writers'

texts also form rough clusters themselves -- one for each author.

We shall probably never know exactly what the determining factors

are in this oddity --- but I would again warn Bruce not to place

much reliance upon a single pca chart's plot distributions. That

alone will not negate the evidence for a 19th Century BoM authorship.

UD

[Glenn101]

Another useful exercise would be to look among Jockers'

totals for all the frequently used non-contextual words,

and determine which of those words the Book of Mormon chapters

USE THE MOST FREQUENTLY, but the 19th Century authors use

the LEAST FREQUENTLY -- and the other way around.

What I'm trying to say, is that you could identify those words

which do not match up very well in the frequency of their occurrence

among the several 19th Century writers and in the Book of Mormon.

UD

Well, you could do that, and keep doing it until you have a nice, tight fit, but the results would be meaningless.

Glenn

[Uncle Dale]

Well, you could do that, and keep doing it until you have a nice,

tight fit, but the results would be meaningless.

Glenn

If the "results are meaningless," then Bruce's argument

from a pca chart is equally meaningless -- as professional

statisticians will agree, once he publishes that chart.

If the entire controversy over the possibility of 19th century

authorship claims dies with the publication of Bruce's chart,

I'll consent to a Mormon baptism (and pay 30 years back-tithing

to those saintly fellows in Salt Lake City).

Uncle Dale

[Glenn101]

If the "results are meaningless," then Bruce's argument

from a pca chart is equally meaningless -- as professional

statisticians will agree, once he publishes that chart.

Two different things. You are suggesting manipulating the data. Throwing out things that make the Book of Mormon different from the nineteenth century authors. Once you start doing that you are biasing the results.

But are you saying that Bruce does not know what he is talking about? He is a professional statistician you know.

If the entire controversy over the possibility of 19th century

authorship claims dies with the publication of Bruce's chart,

I'll consent to a Mormon baptism (and pay 30 years back-tithing

to those saintly fellows in Salt Lake City).

Uncle Dale

It would not die. But I do not think that Bruce is publishing a pca chart. He is publishing a critique of the Jockers study in the Literary and Linguistic Computing Magazine. Just what issue it will come out in I do not know. It should be fairly soon, I would think.

Glenn

[Uncle Dale]

...

I do not think that Bruce is publishing a pca chart.

...

A crude version of it was featured in his pre-pub of that paper;

and he later offered us a more refined version here at this

message board. If he has decided to drop the chart, that will

be helpful to his hopes of getting the thing published.

A different, more mathematical paper, is scheduled for publication

during the first half of 2011 in a Stats Journal; but I had not

heard that the LLC reviewers had accepted the initial anti-Jockers

study paper.

UD