“Lies, [Expletive Deleted] Lies, and Statistics”

Posted on by

I recently perfunctorily read a new study written by three Stanford University professors, titled “Reassessing authorship of the Book of Mormon using delta and nearest shrunken centroid classification,” and published in the latest issue of the journal, Literary and Linguistic Computing.   While I admit to being disturbed by the article, it is not for the reason they would like it to be.  I teach statistics at America’s largest privately-owned university, and, frankly, I see so many problems, I’m not sure where to begin.

To start, they were not exhaustive in alleged authors. In this regard, the “study” is not unlike choosing five National Football League (NFL) teams at random (and two Canadian Football League (CFL) teams as a “control”), and stating that the one with the best won-lost percentage won the Super Bowl. Choosing the best of an woefully incomplete list tells the reader nothing about “Who really wrote the Book of Mormon”–the Stanford authors’ clucking agreement with the authors of the book by that name notwithstanding–just as not checking out ALL NFL teams says nothing about the winner of the Super Bowl.

For one thing, Joseph Smith’s writings were missing from this study. I can imagine that even many anti-Mormons would be upset at crowning Solomon Spaulding’s Manuscript Story as the true Book of Mormon, if it were demonstrated that Joseph Smith’s letters or journal was a better match!

Moreover, the authors alleged by Latter-day Saints are VERY conspicuous by their absence. While I admit that there are no other samples of Mormon’s or Alma’s or Nephi’s writing than what is found in the Book of Mormon, the Book does extensively quote Jesus Christ–and not just the rerun of the Sermon on the Mount!

Why is the New Testament missing from this study? Surely, if Solomon Spaulding were the author of III Nephi 9-12; 15-21; 23; 26-30, rather than the Saviour, wouldn’t testing His words in the New Testament put the final nail in the coffin burying Mormonism? However,whether or not Jesus’ New Testament statements are a match with His alleged words in the Book of Mormon, these authors have made sure that the reader would never know it. What are they afraid of–that the Gospels would be a match?

It gets worse. As I read the results, not one of the works studied had more similarities with the Book of Mormon than dissimilarities. Their choice of Spaulding’s manuscript as the true source of the Book of Mormon, then, is like taking the last place team in each division of the NFL and congratulating the best of them for winning the Super Bowl. Just as every last place team is disqualified from the playoffs, let alone winning the Super Bowl, a manuscript with more dissimilarities than likenesses to the the Book of Mormon should also be disqualified from consideration.

In short, this game is rigged, and such a misuse of scholarship offends me as a teacher of statistics, as a Latter-day Saint, as a Christian, and as a fair-minded person.  The fact that this pseudo-scholarship is in reality a poorly-reasoned anti-Mormon bromide makes it worse.

51 thoughts on ““Lies, [Expletive Deleted] Lies, and Statistics”

  1. Patrick

    While I agree that this is a prime example of scholars misusing scholarship, I’m curious as to how it’s different than much of what FAIR does—that is, examining questions with a predetermined answer/bias, selectively using “evidence” than confirms those predetermined conclusions, utilizing scholarly methods to debate faith claims of a group, etc.

  2. Phouchg

    I would expect that sort of view from somebody who can not even bring himself to use the word “damned”.

  3. NOYDMB

    Steven,
    I’ve also begun perusing the article. I guess if english as a major can’t feed a family of four at least their going to start trying to tear down religions. Note the anti-ex-Mo listed as the third author from the a non-mathematics, non english department.

    The cost of submitting an article to a pseudo-science journal~500$
    Using ones “engineering skills” to do a haphazard job of proving the BoM isn’t genuine, Priceless.

    I sorry, but if I claimed to translate the book “Why all liberals are going to hell” and three stanford professors tested Steven, Phough, and Nick Literski for wordprint authorship evidence, and didn’t test me, there’d be something wrong, and in that case, even apostates like Phough would admit it. But as long as it appears to put Mormons in a bad light, Phough turns off his reasoning ability.

  4. Nick Literski

    I agree that the lack of effort to compare against known writings of Joseph Smith was pretty much a tipoff here that the outcome was predetermined. It would appear that the “scholars” made the age-old assumption that Joseph was simply too stupid to have written The Book of Mormon, therefore, they needed to look to Rigdon, Spaulding, etc.

    On the other hand, it’s fair to say that Joseph’s literacy improved dramatically from the 1820s to 1844. Emma was quite emphatic that in his younger years, he couldn’t even manage a coherent letter. Perhaps “wordprints” are more stable than I expect, but it seems to me that the only fair sample these “scholars” could have used from Joseph would be holographs from around 1828 to 1832. Based on Dean Jesse’s compilation, that would be a very small sample size.

  5. Steven Danderson Post author

    Hi Patrick!

    While I would like to see instances where FAIR personnel cherry picked evidences, that IS a good question.

    In theory, peer review is supposed to ensure that scholars use proper research procedures. For example, my study measuring the Temple’s impact on local housing prices was reviewed by a statistician at one university, and the dean of another college’s Business Department. Before submitting that study for publication, I asked the then-Chairs of the another university’s Finance and Economics Departments to check my work. Moreover, the data was included in an appendix so that others with knowledge of regression analyses could replicate it.

    In short, FAIR wanted to make sure that my research would not draw laughs from the professional community, and that the secular methods that I used in a religious subject proved applicable in both religious and secular contexts (I understand that the Economics Chair used my study in his own research.).

    Unfortunately, though, peer review is not foolproof, especially when editors let biases run rampant. If, for example, an Evangelical-based journal published a stylometric study showing that the Book of Isaiah had only one author, I doubt that they would publish a study by the same author that showed that the Jesus in the Book of Mormon was the same as the One in the New Testament.

    Sometimes that bias can be quite subtle. A journal’s editors may not be biased against Latter-day Saints relative to, say, Evangelicals, but if they have an anti-religion bias, they might publish an anti-Mormon paper, even if it makes Evangelicals look good in comparison, because the opportunity to tear down one religion is too good to pass up.

    To sum up, while peer review is useful, it is not foolproof. The best way to minimize bias is to use multiple referees–both officially and unofficially–who are competent in the relevant field(s), and to show one’s work, so other competent people in the relevant field(s) could duplicate the tests to check for accuracy and relevance.

    Do I make sense?

    Unfortunately, peer review broke down in this case–possibly because the reviewers were ignorant of the relevant background.

  6. Steven Danderson Post author

    Floyd:

    The short answer to your question is, “Yes.” ;)

    See my answer to Patrick for more comments along that line.

  7. Steven Danderson Post author

    Phouchg:

    I was trying to maintain a sense of humour. There are WAY too many people who are WAY too serious to know their limitations! ;)

  8. Steven Danderson Post author

    Hi NOYDMB!

    I’m not sure that Phouchg is apostate. Frankly, I don’t know enough about him (her?) to make that call! ;)

    Unfortunately, this isn’t a case of fly-by-nighters. Rather what we have are one reputable, high-quality journal, two reputable, high-quality universities (Stanford and Oxford), and two reputable professors who were snookered by one anti-Mormon who knew that NONE of them had any expertise in the relevant history or religious study required to know that Spaulding’s Manuscript Story is utterly different from the Book of Mormon, and that Spaulding only wrote one book.

    If any of them had that expertise, I suspect that they would have looked more closely at the OTHER errors, and the study might never have seen the academic light of day.

  9. Confutus

    To debunk the conclusion of this study only requires a comparative reading of Solomon Spaulding’s manuscript and the Book of Mormon. To get to the Book of Mormon from Spaulding’s manuscript would require such a drastic and thorough rewrite that it would look identical to an entirely different work in a different genre by a different author with a different mindset and a different literary style.

    Traditionally it has taken ignorance, desperation, or both to reach any other conclusion. Unfortunately, no matter how advanced computer-aided statistical analysis gets, the conclusions it spits out still depend on what assumptions were fed in.

  10. Steven Danderson Post author

    Hi Nick!

    That is an EXCELLENT question!

    Yes, wordprints are remarkably stable throughout one’s life. As Larson Rencher, and Layton’s 1978 BYU Studies article informs us, Sir Walter Scott’s wordprint pattern proved quite consistent over several decades–and five strokes!

    Moreover, Joseph Smith’s 1832 journal entries alone provide enough data to test against the Book of Mormon, so I don’t think we have much to worry on that score. Further, even with blocks of less than 300 words, I was able to distinguish between the Saviour and Luke in chapter 12 of his Gospel! ;)

  11. Steven Danderson Post author

    Hi Confutus!

    You have it right, in short form.

    GIGO: Garbage In, Garbage Out.

    :)

  12. Matthew

    When will these people ever learn, you cannot prove the reality of a spiritual entity via physical means. All the back and forth rhetoric, word play, scholarship and science in the world cannot authenticate the Book of Mormon. It is a stand alone position that each person takes, between him and His God to KNOW, of him self that the record is true, there is no other way, nor means that this can be produced, in order that we may have sufficiant faith in Christ, than to have personal, direct communication from the heavens that this is, what it says that it is.
    Here is a word study for you, there is no other book( to my knowledge) that has the audacity to say in and of it self, to ask God if it is NOT true, such a third party verification was out of the hands of whatever original author that it was. If one trully wants to test this autheniticity of the record, that is the only way. By responding to such a thread we acknowledge not only the existence of such a God but also of His ability to communicate to his children. But let one be warned, that response that one recieves (as per D&C 20) will either save you or to us the threads author…damn you if you do not abide it.

  13. Floyd the Wonderdog

    I asked whether the journal was peer reviewed because from experience I know that many non-peer reviewed journals are nearly worthless.

    Not that the peer reviewed journals are much better, mind you. It is too easy to manipulate editors into excluding reviewers that might raise significant objections to methodology and inferences. e.g. “We can’t have anyone actually familiar with Mormon studies review this because they would be biased.”

    Remember that some editors spend much of their time actively seeking article submissions. They have space to fill in the journal and sometimes not enough submissions.

  14. frank

    wow- i( MS operations resarch/statistics naval post graduate school monterey ) plowed thru the article and must say i agreed with everything even their comment that JS “writings” hopefully will be available soon(thru lds JS project) and the study can be run again.
    in email conversation with one author matt jockers? there seemed to be no attempt to take the BOM “down”.
    so i hope byu/maxwell will run the numbers and show off their credentials in the wordprint analysis area. unfortunately i see an apologetic train coming down the tracks to derail this study and “preserve LDS truth”.
    this study will gain momentum in non-lds circles as it comes out of stanford and has almost instant credibility in the non-lds intellectual world.
    i also see the midnight oil being burned at byu/FARMS!

    so many apologetic issues-so little time!!!

  15. Dr Everyday

    Steven,
    I’ve taken a fair amount of statistics classes (into the graduate level). But, considering my line of work and experience, I would feel underqualified to write a response. For the sake of academic integrity, is anyone planning on writing a response, pointing out these serious flaws?

    Even if this were a paper on comma use in the OED, if the stats are bad, these “professors” should be discouraged from mediocre statistics. (It sounds like a student with a project due, who was too lazy to try to do it right, so turned in what they got around to.)

  16. cinepro

    FYI, Craig Criddle (the anti-Mormon member of the team) has written about the process through which this study come about. Here are some of his comments:

    “With the help of friends, I was able to obtain most of the texts we needed for analysis, but I was not successful in obtaining reliable text for Joseph Smith (which we are still hoping to obtain). Matt took the texts I provided, segmented and encoded them into the xml that his tools require. He did the same with two control texts that he obtained.

    Initially, Matt provided me some lists of frequently used words, bigrams, and phrases in the Book of Mormon and in the other texts. I ran some tests using my amateur methods, and Matt ran some tests of his own using his methods. We had some similar outcomes, with Rigdon appearing as a likely major contributor. This led Matt to believe that the theory was worth rigorous testing using more sophisticated methodologies. He decided to organize and lead a team effort with an eye toward publication of the research. We both understood the need for a bona fide statistician, and Matt recruited Daniela Witten, a doctoral student in Statistics with expertise in machine learning and classification.”

    “On April 5, 2008, Matt submitted the manuscript to the Journal of Literary and Linguistic Computing. The anonymous peer review process took six months. On October 7, 2008, we finally received notification that the paper was accepted pending an adequate response to the reviewer comments. We completed our response to the reviewer comments and submitted the corrected manuscript on November 6. On November 24, we received word that the manuscript was accepted. At the same time we received page proofs. We corrected them and returned the article on November 27. It was published electronically on December 6.”

    “as many apologists have already (correctly) pointed out, I am the team member with bias. Matt and Daniela were unbiased, and had very little knowledge of Mormonism before they became involved with this project.

    While I contributed expertise as a former Mormon, Matt and Daniela carried out the data analysis and the results are what they are, independent of my participation in the research.”

  17. NorthboundZax

    Steven, having read a number of your other posts, I know you are much better than this. Your dismissal of the paper seems to be derived more from emotion than from real scrutiny of what information one can glean from this study. Information (not proof) about authorship can be gained from this study, and we shouldn’t be so rash to toss it away without seeing what we can learn from it.

    I posted this starting point on MAD:
    First of all, what should we expect the results of this study to be assuming the BoM is of Nephite origin? The answer depends strongly on whether the translation was ‘loose’ or ‘tight’. A loose translation should have definite imprints of a strong Joseph Smith signal across the entire work. A tight translation should result in no overwhelming JS signal or any overwhelming signal correlating with any of the selected author pool, giving a statistically similar (flat) probability distribution function across the board (not counting Isaiah-Malachi which will clearly show up in either as a special case).

    Now, Joseph Smith wasn’t included, so that should leave us to consider what a JS signal should look like. My supposition is that a JS signal for this group of authors would either be flat (i.e., little correlation with any and mapping everywhere simultaneously) or would preferentially map on Cowdery as I would guess he would be more similar to Smith than any of the others. Another necessary and important note to mention is that errors in data selection, poorly constructed centroids, or whatever should again bias the results to a flat distribution (more controls relative to actual signal).

    So what does it mean that the results are not flat? For starters, that seems to me to rule out a tight translation. So what to do with the strong Rigdon-Spalding pairing? Maybe one could suppose that the unaccounted for JS signal preferentially mapped there, indicating a loose translation. Personally, that particular pairing is a little too striking for me to chalk up that way, and seems that I should take the Rigdon-Spalding hypothesis more seriously than I have in the past. Clearly, not the only way they must be interpreted (especially in the absence of a Smith centroid), and it will be interesting to see how robust the signal remains in future studies, particularly when Joseph Smith can be reliably included. But at the moment, we have a quantitative result that points to a surprising Rigdon-Spalding connection of BoM origins. For anyone interested in the question of BoM origins, it would be far more useful to penetrate what that signal likely means than to simply dismiss the study as trash.

  18. cinepro

    NorthboundZax raises two interesting questions:

    1. Since Joseph Smith’s writings weren’t included in this initial test, it allows apologists (and others) to think scientifically and propose an expected outcome for such a test based on their theories of Book of Mormon origin. So how about it? What would be expect from a well designed study that includes Joseph Smith’s writings as a sample?

    2. While it’s possible (but not logical) for different portions of the Book of Mormon to have been translated with different methodologies (“tight” vs. “loose”), it is impossible for any specific portion of the Book of Mormon to have been simultaneously translated with both tight and loose techniques. So studies such as this might be applied to the question of whether it was all loose, all tight, or a mixture of both. Again, here is an opportunity for apologists (and critics) to suggest a falsifiable theory, and test it. Sounds like fun!

  19. Steven Danderson Post author

    Hi cinepro!

    Here were some of my impressions of Professor Criddle’s post:

    You quote Professor Criddle:
    “I was not successful in obtaining reliable text for Joseph Smith (which we are still hoping to obtain).”

    Wouldn’t it be prudent to not publish at all; or at the very least, to reach only a tentative conclusion, seeing that they omitted several alleged authors–not just Joseph Smith?

    At the very least, couldn’t Professor Criddle have used Joseph Smith’s journals from the Book of Mormon publication period? They are found in Professor Dean Jessee’s The Personal Writings of Joseph Smith, and were found to be quite genuine.

    You quote Professor Criddle:
    “I am the team member with bias. Matt and Daniela were unbiased, and had very little knowledge of Mormonism before they became involved with this project.”

    Wouldn’t that indicate that Professors Joskers and Witten were extremely vulnerable to biased information provided by Professor Criddle?

    Part of my bachellor’s thesis was a stylometric analysis of A Course in Miracles. While I could have dug up information from bitter ex-New Agers, I chose not to. Instead, I chose the charitable approach and got background data from contemporary New Agers.

    You quote Professor Criddle:
    “While I contributed expertise as a former Mormon, Matt and Daniela carried out the data analysis and the results are what they are, independent of my participation in the research.”

    The problem is, the biased “expertise as a former Mormon”–coupled with the omission of several alleged authors–made their conclusion, at best, highly suspect, and, at worst, completely useless, and therefore, unconscionably misleading.

  20. Steven Danderson Post author

    Hello Frank!

    The Naval Postgraduate School is an excellent institution.

    It appears that we have deuling quant grads! ;)

    Here are some of my credentials:

    My graduate training was at the University of South Florida. My concentrations were in finance, economics, and international business. My undergrad college mathematics classes included algebra, trigonometry, pre-calculus, and calculus. I have taken statistics courses at both the graduate and undergraduate levels, and my other graduate quantitative courses include econometrics, economic policy analysis, and mathematical economics.

    And, as I mentioned in the blog entry itself, I teach statistics at America’s largest privately-owned university (It is NOT BYU! BYU is a distant second.).

    Moreover, my bachellor’s thesis was a stylometric analysis of some alleged extended quotes of Jesus Christ from non-New Testament sources, using a modified version of that used by Professors Larson, Rencher, and Layton.

    I think I know about the quantitative subjects! ;)

  21. Steven Danderson Post author

    Hi Everyday!

    You ask:
    “I’ve taken a fair amount of statistics classes (into the graduate level). But, considering my line of work and experience, I would feel underqualified to write a response. For the sake of academic integrity, is anyone planning on writing a response, pointing out these serious flaws?”

    Well, yes. I think there are about three of us at FAIR working on individual responses. Look for something from FAIR. perhaps in late January or early February.

    You said:
    “Even if this were a paper on comma use in the OED, if the stats are bad, these “professors” should be discouraged from mediocre statistics. (It sounds like a student with a project due, who was too lazy to try to do it right, so turned in what they got around to.)”

    Yeah, that, and their foisting on us this “work” as the final word on the subject, really bothered me. :(

    Professor Criddle’s sin was pretty obvious, and it showed in that article. Where Professors Joskers and Witten went wrong was in not following through using the Russian proverb, “Trust, but verify.” They should have held Professor Criddle’s feet to the fire until he provided enough documentation; they should have done some homework themselves, so they could have avoided Professor Criddle’s bias, and, when samples from Joseph Smith and others were not forthcoming, they should have made their conclusions tentative, and obviously so–in the article.

    To have acted as they did is to be unconscionably misleading.

  22. Steven Danderson Post author

    Hello, NorthboundZax!

    You say:
    “Steven, having read a number of your other posts, I know you are much better than this. Your dismissal of the paper seems to be derived more from emotion than from real scrutiny of what information one can glean from this study.”

    First of all, I admit that my initial reading was cursory. At the time, I had neither the time or the space for a full-scale review. The weaknesses I pointed out were, as some in the academic business community call it, “blinding flashes of the obvious”! ;)

    Second, Since that was only a first impression, you should expect that I would look more closely later on. Indeed, I am trying to work that in with Christmas and vacation plans! ;)

    Third, I probably would have been more restrained in my post if Professors Joskers, Witten, and Criddle had been more restrained in the conclusions they reached in their study. I find no justification to reach their conclusion when their data seem to have (at best) barely cleared 0.9 significance, let alone the standard, more restrictive 0.1 significance.

    Moreover, in my FAIR study of the Temple’s impact on property values, I ran tests to a even-more restrictive 0.05 significance–and an R squared exceeding .8 made sure of the study’s relevance.

    You posted this on MADB:
    “First of all, what should we expect the results of this study to be assuming the BoM is of Nephite origin? The answer depends strongly on whether the translation was ‘loose’ or ‘tight’. A loose translation should have definite imprints of a strong Joseph Smith signal across the entire work. A tight translation should result in no overwhelming JS signal or any overwhelming signal correlating with any of the selected author pool, giving a statistically similar (flat) probability distribution function across the board (not counting Isaiah-Malachi which will clearly show up in either as a special case).”

    This is true. Thus, at best, a test that excludes the two most prominent possibilities tells us very little, if anything at all.

    This is like trying to find out who played in the 2006 Super Bowl, while omitting the Indianapolis Colts and the Chicago Bears. ;)

    Though the authors didn’t actually pair Isaiah or Malachi with the non-Old Testament parts of the Book of Mormon, my cursory examination shows both to be somewhat of a match. Thus, even without Joseph Smith, the decidedly non-flat Isaiah-Malachi results should have told the study’s authors that–maybe–the Book of Mormon might not have been of so modern an origin, after all.

    You say:
    “So what does it mean that the results are not flat? For starters, that seems to me to rule out a tight translation. So what to do with the strong Rigdon-Spalding pairing? Maybe one could suppose that the unaccounted for JS signal preferentially mapped there, indicating a loose translation.”

    Possibly–depending on how non-flat the results are.

    You say:
    “Personally, that particular pairing is a little too striking for me to chalk up that way, and seems that I should take the Rigdon-Spalding hypothesis more seriously than I have in the past. Clearly, not the only way they must be interpreted (especially in the absence of a Smith centroid), and it will be interesting to see how robust the signal remains in future studies, particularly when Joseph Smith can be reliably included. But at the moment, we have a quantitative result that points to a surprising Rigdon-Spalding connection of BoM origins.”

    Again, possibly. There is, however, an historical roadblock: Sidney Rigdon claimed–to the end of his life–that he never met Joseph Smith until 1832–two years after the Book of Mormon’s publication. While I am not claiming that Rigdon couldn’t have been lying–or that Solomon Spaulding’s Manuscript Story couldn’t have reached Joseph Smith some other way before the Book of Mormon was published, but Professor Criddle, et al., would have to demonstrate that beyond at least some reasonable doubt–especially since the data itself seems insufficient to prove their conclusion beyond a 0.1 significance–and the Manuscript Story itself (which appears to be the only novel written by Reverend Spaulding!) is measurably different from the Book of Mormon.

    You conclude:
    “For anyone interested in the question of BoM origins, it would be far more useful to penetrate what that signal likely means than to simply dismiss the study as trash.”

    I haven’t dismissed it–yet.

    In my more thorough review, I expect to find things that were better than I thought. While I would have liked for Professor Joskers, et al., to have included other tests to have demonstrated the validity of their model (My study used a standard regression analysis.), it seems to be promising, especially since they apply it as Don Foster did his model when he dicovered that Joe Klein was the Anonymous who wrote Primary Colors.

    ALAS! Others will probably prove worse!

  23. Steven Danderson Post author

    Hi again Floyd!

    You make a good point. Peer-reviewed articles, while they are NOT infallible, are set up to minimize amateurism and bias.

    This is why I like FAIR: the editorial staff tends–as a rule–to err on the cautious side. My non-LDS colleagues in grad school were pleasantly surprised that my conclusions from my study weren’t “shouted from the housetops”–especially since the R squared was about double “yelling and screaming and shouting territory.”

  24. Steven Danderson Post author

    Hello Matthew!

    While it is true that temporal methods alone can neither prove nor disprove the Book of Mormon, they can present evidence.

    In Moroni 10:3, we are told to “ponder”–meaning to weigh carefully. What is it we need to ponder? The evidence.

    What FAIR does is to try to give you tools to properly weigh that evidence.

    God won’t punish us for having honest intellectual problems with the Book of Mormon–but He WILL damn us if we refuse to do our homework!

    Do I make sense?

  25. Greg Smith

    Point #1:

    You quote Professor Criddle:

    “I am the team member with bias. Matt and Daniela were unbiased, and had very little knowledge of Mormonism before they became involved with this project.”

    This little gem shows where the whole problem is.

    Repeat after me: Everybody. Has. Bias.

    How can anyone trained in the sciences not get this? We have biases we don’t know about. We have biases we do know about. We have biases we don’t know about that we think we know about. And so on.

    (After all, they had Bob McCue, bastion of bias-free Mormon scholarship to thank in the credits! How ”could” they have possibly gone wrong? Bob apparently thinks this is some type of watershed in “Mormon Studies.” The mind boggles.) :-)

    Everybody has bias. Even if “Matt and Daniela” don’t know much about Mormonism, they probably have some thoughts or biases on whether God tends to give gold plates to farmboys who translate them with seer stones. :-)

    #2: I’ll bet the stats were peer-reviewed, and they were probably done properly; I’d bet money that no one peer reviewed the polemic masquerading as LDS “history” that informed the stats.

    Greg

  26. NorthboundZax

    SD: I teach statistics at America’s largest privately-owned university (It is NOT BYU! BYU is a distant second.).

    University of Phoenix?

    SD: First of all, I admit that my initial reading was cursory. At the time, I had neither the time or the space for a full-scale review. The weaknesses I pointed out were, as some in the academic business community call it, “blinding flashes of the obvious”! ;)

    Unfortunately, the “blinding flashes of the obvious” seem to have put you at less than your best. You really would have done yourself and the rest of us a favor to wait until your vision had cleared a bit to post your review.

    SD: Third, I probably would have been more restrained in my post if Professors Joskers, Witten, and Criddle had been more restrained in the conclusions they reached in their study. I find no justification to reach their conclusion when their data seem to have (at best) barely cleared 0.9 significance, let alone the standard, more restrictive 0.1 significance.

    Their conclusion was actually quite restrained and lands right where the results take them: “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).” IOW, they had a hypothesis that could be falsified by their method – and wasn’t. That does indeed lend support to such a hypothesis (no claim of proof – which would have been unrestrained). It is also worth keeping in mind that the 90% confidence threshold is for chapter by chapter. When there is a consistent signal that keeps appearing at a given confidence, I’m sure you will agree that the aggregate will have an overall higher confidence level. I agree that the significance angle is a good angle to pursue, though – and you have lots of room for exploration. But simply waving it away as not as good as your property value study is a bit shallow.

    NZ: “First of all, what should we expect the results of this study to be assuming the BoM is of Nephite origin? The answer depends strongly on whether the translation was ‘loose’ or ‘tight’. A loose translation should have definite imprints of a strong Joseph Smith signal across the entire work. A tight translation should result in no overwhelming JS signal or any overwhelming signal correlating with any of the selected author pool, giving a statistically similar (flat) probability distribution function across the board (not counting Isaiah-Malachi which will clearly show up in either as a special case).”

    SD: This is true. Thus, at best, a test that excludes the two most prominent possibilities tells us very little, if anything at all.

    This is where I would expect a teacher of statistics to show a deeper understanding of what the tests mean. Excluding possibilities is by no means valueless (a common application of statistics!). Even more valuable in this case is the prominent peaks in the signal. Even if you prefer a different explanation to that stated in the conclusion, these results should inform that explanation.

    SD: This is like trying to find out who played in the 2006 Super Bowl, while omitting the Indianapolis Colts and the Chicago Bears. ;)

    Yes, I saw this analogy in your OP, but it’s not a very insightful one so I didn’t intend to bring it up. If you really want to pursue it, at least acknowledge that the method would determine which team in your tested group is ‘most like’ the Superbowl winner in whatever parameter space is investigated. If the teams are all equally good (bad) fits, then the results would be pretty flat – no team significantly more like the Superbowl winner than any other team. If one team is significantly more like the Superbowl winner, that tells you something too – not necessarily the team that won, but the team that was more likely to have to have won the Superbowl than the others. Interpreting the why that team is actually pretty easy in this case (even if ‘wrong’). In the Jockers et al. study we have the Rigdon-Spalding pair of voices that are ‘most like’ BoM authors, whether Nephite or 19th century. Deducing why that would be the case is where the real interest lies.

    SD: Though the authors didn’t actually pair Isaiah or Malachi with the non-Old Testament parts of the Book of Mormon, my cursory examination shows both to be somewhat of a match. Thus, even without Joseph Smith, the decidedly non-flat Isaiah-Malachi results should have told the study’s authors that–maybe–the Book of Mormon might not have been of so modern an origin, after all.

    It would help move things along if you would elaborate on your cursory examination here and explain why your cursory methodology for similarity is superior to Jockers et al.’s Nearest Shrunken Centroid method.

    NZ: “So what does it mean that the results are not flat? For starters, that seems to me to rule out a tight translation. So what to do with the strong Rigdon-Spalding pairing? Maybe one could suppose that the unaccounted for JS signal preferentially mapped there, indicating a loose translation.”

    SD: Possibly–depending on how non-flat the results are.

    ??? We know how non-flat the results are. Please elaborate.

    NZ: “Personally, that particular pairing is a little too striking for me to chalk up that way, and seems that I should take the Rigdon-Spalding hypothesis more seriously than I have in the past. Clearly, not the only way they must be interpreted (especially in the absence of a Smith centroid), and it will be interesting to see how robust the signal remains in future studies, particularly when Joseph Smith can be reliably included. But at the moment, we have a quantitative result that points to a surprising Rigdon-Spalding connection of BoM origins.”

    SD: Again, possibly. There is, however, an historical roadblock: Sidney Rigdon claimed–to the end of his life–that he never met Joseph Smith until 1832–two years after the Book of Mormon’s publication. While I am not claiming that Rigdon couldn’t have been lying–or that Solomon Spaulding’s Manuscript Story couldn’t have reached Joseph Smith some other way before the Book of Mormon was published, but Professor Criddle, et al., would have to demonstrate that beyond at least some reasonable doubt–especially since the data itself seems insufficient to prove their conclusion beyond a 0.1 significance–and the Manuscript Story itself (which appears to be the only novel written by Reverend Spaulding!) is measurably different from the Book of Mormon.

    You are talking ‘proof’ rather than ‘support’ here, and maybe the historical roadblock will prove insurmountable. In any case, I don’t think the historical burden is on Criddle’s shoulders – the applied method has nothing to do with history other than what can be derived specifically from the text. Also, I admit to being puzzled by your “measurably different from the Book of Mormon” – you are in essence arguing that Spalding’s work cannot be connected with BoM origins as the two are measurably different as a way to discount a study demonstrating quantitative similarity between the two. That argument is unlikely to get much traction.

    SD: I haven’t dismissed it–yet.

    Well, words like “meaningless”, “rigged”, and “anti-Mormon bromide” sounded like a pretty round dismissal to me. I look forward to seeing your adjectives if you do actually dismiss it!

    SD: In my more thorough review, I expect to find things that were better than I thought. While I would have liked for Professor Joskers, et al., to have included other tests to have demonstrated the validity of their model (My study used a standard regression analysis.), it seems to be promising, especially since they apply it as Don Foster did his model when he dicovered that Joe Klein was the Anonymous who wrote Primary Colors.

    In your cursory reading, you seem to have missed the part that discusses tests of the validity. It’s not a great discussion, but they are there.

    SD: ALAS! Others will probably prove worse!

    Worse than “anti-Mormon bromide”?!?

  27. cinepro

    While I find the criticisms of the study to be strong and persuading, I do not see them as dead-ending this line of investigation.

    Instead, it would be cool if apologists and “believing” scholars used this is as an opportunity to formulate strong, valid tests. And before performing the tests, make prediction about what the outcome would be to support different theories of the Book of Mormon translation/dictation. If we acknowledge the possibility (however slight) of an authentically loose translation, and authentically tight translation, or a 19th century origin, what kind of testing would delineate between those three? What kind of results would we expect?

    I suspect the only safe harbor for apologists will be to argue that there is no way for this type of testing to inform on any of those possibilities, or that the 19th century origin theory isn’t worthy of honest consideration. But I am anxious to be proven wrong.

    So how about it, Steven? How about a blog post suggesting future lines of investigation that could shed light on these theories, specifically stating what the next test should be, and what the different results could indicate?

  28. Ryan

    Again, we behold the tragic fruits of so much effort being poured into deconstructing the Book of Mormon, and so very little being put into actually READING it.

  29. Steven Danderson Post author

    Actually, cinepro, I am working on a scholarly article about it! Unfortunately, it’s going to be a little slow-going; Christmas, you know! ;)

    Really, there are many promising paths for this method–if done right! I had earlier commented that Professors Joskers, Witten, and Criddle applied their method in a similar manner that Professor Foster applied his. That is NOT a bad way to proceed–provided you are exhaustive in testing alleged authors. If the test is done right and shows that the Book of Mormon has a 19th century author (specifically, Solomon Spaulding and/or Sidney Rigdon) beyond a 0.1 significance (and a high, say, 0.5 R squared), then I would have to seriously look at their theory of how the book got from Spaulding to Rigdon to Joseph Smith. As of now, all their “background” (which is merely warmed-over from Who Really Wrote the Book of Mormon, by Howard Davis, Warren Cowdrey, and Arthur Vanick, which is only the second edition of a book of the same name, but Don Scales, rather than Vanick being the third author) is mere supposition–with precious little (if any) supporting evidence.

    Moreover, this can be used to study similarities in style among contemporaries–even if they are definitely separate authors. I believe that this would play well with Professor Joskers’ specialty, and indeed, Professors Larson, Rencher, and Layton found that, though Mormon and Moroni proved entirely separate people, their wordprints more closely resembled each others’ than those of say, Nephi ben Lehi.

    I would like to see these:

    1. I would like to see them use the forthcoming volumes of Joseph Smith’s words, or, at the very least, using the authentic parts of Professor Jessee’s The Personal Writings of Joseph Smith or Professor Ehat’s The Words of Joseph Smith. If it were demonstrated to a 0.1 (or better) significance that Joseph Smith wrote the Book of Mormon, then it would be good (though not conclusive) evidence that the Book of Mormon is the brainchild from the 19th century, rather than an ancient document.

    2. Since we have no reliable alternate (non-Book of Mormon writings of Nephi, et al., I would like to compare non-Sermon quotes of Jesus Christ from the Book of Mormon (Larson, et al., tell us there are about 9000 words.) compared with quotes of the Saviour in the New Testament (The bulk would comprise a book of about seven chapters, consisting of the Sermon on the Mount [Matthew 5-7], the Olivet Discourse [Matthew 23-24], and the epistles that He dictated to the Apostle John [Revelation 2-3].). If the “Biblical” Jesus is closer to the “Book of Mormon” Jesus than the 19th century authors, then that is eloquent evidence of the Book of Mormon’s ancient, divine orgin.

    3. I would like to see more quantification of their results. It doesn’t matter if one NFL team has a better record than the rest of the sample. What matters is if any of the teams have a good enough record to get into the playoffs. In a similar manner, having a better match than others in a sample to the Book of Mormon tells us next to nothing; what matters is whether that closest match is deonstrated to a 0.1 significance, or better.

    4. I would like to see tests of authors from places outside the USA–besides Jules Verne–and I would like to see comparisons with non-19th century authors. It would be interesting to see if people from the same culture and/or time periods have similar wordprints. ANOVA comparisons seem apt here.

    Do I make sense?

  30. Steven Danderson Post author

    SD: I teach statistics at America’s largest privately-owned university (It is NOT BYU! BYU is a distant second.).

    NZ: **University of Phoenix?**

    Yes! :)

    NZ: **Unfortunately, the “blinding flashes of the obvious” seem to have put you at less than your best. You really would have done yourself and the rest of us a favor to wait until your vision had cleared a bit to post your review.**

    I did. The problem is, the more I looked, the more problems I saw. ;)

    NZ: **Their conclusion was actually quite restrained and lands right where the results take them: “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).” IOW, they had a hypothesis that could be falsified by their method – and wasn’t. That does indeed lend support to such a hypothesis (no claim of proof – which would have been unrestrained).**

    Except that the word, “support” implies that the study is strong evidence in favour of the Spaulding/Rigdon hypothesis. As you admit above, the study merely failed to eliminate that theory. However, since failing to eliminate is considerably short of “supporting,” Moreover, we don’t know to what confidence level the match is. Given these and the fact that their data set is admittedly incomplete, “support” is too strong a word.

    NZ: **We know how non-flat the results are. Please elaborate.**

    That is exactly the problem: We don’t know. For example, is the match closer to zero or to 100%? Are the results closer to an 100% match than those of ALL alleged authors? Are the results close enough to an 100% match to give up the null hypothesis (Ho) that the Book of Mormon was NOT written by Spaulding and/or Rigdon in favour of the alternate hypothesis (Ha) that it was?

    NZ: **You are talking ‘proof’ rather than ‘support’ here, and maybe the historical roadblock will prove insurmountable.**

    There is a reason I talk “proof.” ;)

    Professors Joskers, Witten, and Criddle asserted an hypothesis here: That Solomon Spaulding and/or Sidney Rigdon wrote the Book of Mormon. By definition, that is the Ha, and its opposite, that they didn’t, is the Ho–the “default” assumption. Under the rules of the statistics “game,” until there is enough evidence to demonstrate the actuality of the Ha beyond the requisite level of significance (usually 0.1), we continue with the Ho.

    NZ: **In any case, I don’t think the historical burden is on Criddle’s shoulders – the applied method has nothing to do with history other than what can be derived specifically from the text.**

    Except that, since Professor Criddle and his associates, Professors Joskers and Witten, made the assertion, by the rules of the statistics “game,” they bear that burden.

    One thing that irks me is when atheists demand that, unless we can prove God’s existence to THEIR satisfaction, WE must adopt their atheism. The problem is that atheism is a relatively NEW philosophy, and God’s existence has been historically presumed. Hence, atheism is the Ha and theism the Ho; and the burden of proof rests with the atheists.

    It is my sad experience and observation that anti-Mormons of both the atheist and Evangelical stripes share the demand that the Latter-day Saints bear the burden of proof.

    NZ: **Well, words like “meaningless”, “rigged”, and “anti-Mormon bromide” sounded like a pretty round dismissal to me. I look forward to seeing your adjectives if you do actually dismiss it!**

    I start using terms like “fraud,” “criminal,” and “immoral”! ;)

    More seriously, I used the first term because Professors Joskers, Witten, and Criddle, far from “supporting their thesis, merely failed to eliminate that thesis. I used the word, “rigged” because, by artfully inverting the Ha and Ho, they “begged the question” of Book of Mormon authorship. Moreover, by not testing either Joseph Smith or the one Book of Mormon Person Whose words are extent (that is, Jesus Christ in the New Testament), the Latter-day Saints are declared the “loser” without even being allowed to compete.

    I used “anti-Mormon” because this study is being presented as an argument AGAINST Momrmonism.

    NZ: **In your cursory reading, you seem to have missed the part that discusses tests of the validity. It’s not a great discussion, but they are there.**

    I’ll reread it. Were the test that you saw applied in a non-religious context?

    NZ: **Worse than “anti-Mormon bromide”?!?**

    Oh, yes! There are MANY things worse than an anti-Mormon bromide! ;)

  31. cinepro

    I’m curious how #2 (comparing the words of Jesus) above would be accomplished? Would you use the KJV, or a more reliable translation? How would you control for the Book of Mormon’s use of olde English? Are the New Testament records of Jesus’s sayings reliable transcriptions, or were they influenced by other factors when they were recorded?

    While the New Testament certainly claims to be a record of what Jesus said, it’s not exactly as if Jesus sat himself down and put pen to paper to record his thoughts, or even as if an apostle sat nearby with a scroll to transcribe the sermons. There may be someway to account for all of this, and they may not be important considerations, but I’ll wait for you to say so.

  32. Steven Danderson Post author

    Hi cinepro!

    Those are good questions!

    I would probably use the KJV, since that was standard in Joseph Smith’s day. However, I’m not averse to using more modern translations. It would be interesting to see what effect different translations have on the tests. I suppose that the paraphrasing translations, like The Living Bible, would have quite different results from either the Book of Mormon or more literal translations like the King James Version.

    The 1978 BYU Studies article by Professor’s Larson, Rencher, and Layton, indicated that they had previously used their method on works from different authors translated from the German, correctly identifying the samples.

    I’m not sure, on the other hand, if Professors Joskers, Witten, and Criddle’s method would work as well.

    You may be right that Jesus’ words in the Gospels may not be verbatim, and that could render such a test impossible. In that case, since Professors Joskers, Witten, and Criddle cannot eliminate the Spaulding/Rigdon hypothesis (I assume that they were honest and accurate in their data and results), it all boils down to: Faith.

    Do I make sense?

  33. NOYDMB

    While we’re going to start introducing translators into the mix, is it possible for wordprint to distinguish between author and translator? Has any work distinguished whether they can be isolated?

  34. Steven Danderson Post author

    Actually, NOYDMB, I believe the 1978 Larson-Rencher-Layton article in BYU Studies does what you ask. It appears that they compared the wordprints of several authors that somebody translated from German to English with samples of the translators own English-language words. Each author differed from each other–and from the translator.

  35. Jeff Lindsay

    I discussed this study in a couple of posts at Mormanity (First Post and Second Post). In addition to at least some minor errors in the data, the whole foundation for their approach is incredibly flawed, as has been noted above and elsewhere. Surely the Standford folks and probably Mr. Criddle understood that the test guarantees that each chapter of the Book of Mormon will be assigned to a “winner” from the tiny set of candidates provided, regardless of how dissimilar the style of that chapter is to the “winner.” Thus, they might be able to conclude that having chapters assigned to someone on their “dream team” of Spaulding, Rigdon, and Cowdery rather than the highly questionable controls of poetry from Longfellow and Barlow or to Orson Pratt might be helpful in eliminating Orson Pratt as a potential author, but hardly confirms any hypothesis of plagiarism. Incredibly, amazingly, this allegedly peer-reviewed publication with the help of a Stanford professor of humanities (not a statistician) and a Stanford student (not a professor!) of statistics actually makes utterly unfounded claims on the basis of such weak evidence. From their abstract:

    We use both methods to determine, on a chapter-by-chapter basis, the probability that each of seven potential authors wrote or contributed to the Book of Mormon. Five of the seven have known or alleged connections to the Book of Mormon, two do not, and were added as controls based on their thematic, linguistic, and historical similarity to the Book of Mormon.

    Our results indicate that likely nineteenth century contributors were Solomon Spalding, a writer of historical fantasies; Sidney Rigdon, an eloquent but perhaps unstable preacher; and Oliver Cowdery, a schoolteacher with editing experience. Our findings support the hypothesis that Rigdon was the main architect of the Book of Mormon and are consistent with historical evidence suggesting that he fabricated the book by adding theology to the unpublished writings of Spalding (then deceased).

    What?? They determined the probability that someone wrote a chapter of the Book of Mormon?! Not even close. They determined probabilities that one candidate was closer than another candidate to the style of a chapter, but that says nothing about the probability that any of those candidates wrote the chapter. How absurd, naive, and irresponsible. I feel very sorry for the Stanford folks who let this kind of garbage be attributed to their names. They were duped – but how did they miss the grotesque error in the conclusory statements made on their behalf?

    I also think its interesting how they see the data supporting Cowdery’s involvement, when it shows almost as strong a signal for Orson Pratt as Cowdery, who is quickly dismissed, along with Longfellow, who shows up as the lead candidate for several chapters. The delta method makes Isaiah/Malachi the top voice for the Book of Mormon, and both methods assign far more chapters than expected to Isaiah/Malachi – almost as if the Book of Mormon has more of the voice of an ancient Hebrew writer than it does of Rigdon or Spaulding – an observation that is not brought out with the quick dismissal of the “non-flat” Isaiah/Malachi wins.

    The study is rigged and the conclusions are unjustified.

    Run this again with some of my writings as a control instead of Joel Barlow’s poetry, and I bet you’ll see I’m one of the top contributors to the Book of Mormon. Maybe I could even become #1 if you let me pick the other control besides Longfellow, for the choice of controls can have a dramatic effect on the nature of the study. Their approach uses words that appear at least 0.1% of the time in ALL the candidates. The choice of a control can shift which words are considered and thus dramatically change the results.

  36. Gotcha

    This is amazing. Criddle and his colleagues have been caught with their hands in the cookie jar. I’m shocked at lack of quality their advocates are demonstrating on this board – especially you NorthboundZax. Do you really want to continue supporting their conclusion?

    “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.”

    This is nuts – bizarre even. I feel like we’re talking at a wall – where is the quest for truth in all of this? Why don’t people like you admit that this is a study that does not support the above statement?

    What are the odds that Criddle’s peers were duped or are now feeling duped (if their even aware of what they’ve done)? How would we craft a reasonable study that could rationally be said to “support the hypothesis that Rigdon was the main architect of the Book of Mormon”. Let’s construct reasonable criteria that we would need to see and go to these guys with a sponsored research agreement. My guess is they are the typical starving academics that depend upon government dollars for their research. Seriously – let’s actually design a study that a reasonable person could look at and say “Yes, I now agree that Sidney Rigdon was the main architect of the Book of Mormon”. It would have to be very large, inclusive etc. Let’s create a near perfect world scenario, a budget and hire a capable group of people that HATE Mormons or are unbiased, we certify their efforts and see what this yields?

    I would think we could round up money for something like this.

  37. NorthboundZax

    Ignoring Gotcha’s weirdness for a moment, I think it is worth pointing out a few things about Jeff Lindsay’s criticisms above.

    Beyond the hyperbole of duping and rigging, most of Jeff’s criticism above lies in the statement, “the test guarantees that each chapter of the Book of Mormon will be assigned to a ‘winner’ from the tiny set of candidates provided”, which is turning a blind eye to what the statistics are really saying. While it is true that one author will necessarily get the top spot for any particular chapter, the statistics themselves guide us in how much stock to put in that ‘winner’, particularly in how far the probabilities deviate from a flat spectrum. IOW, there will always be some variation across the spectrum, but as long as all authors are listed at probabilities of less than ~2/N (where N is the number of authors in the target list), they can be considered statistically flat.

    Picking probabilities as done in the Jockers study is not rolling the dice to see who the ‘winner’ is, but rather should be thought of as a method of constructing meaningful dice for such a determination. A flat spectrum means constructing a die that assigns all target authors as equally [im]probable. We do in fact see that for many chapters. These ‘flat’ probabilities could be interpreted in a number of ways. 1) all authors contributed equally to the chapter (no one will actually take this position – but it is allowed by the results) 2) the actual author is not present in the target author list, so the method can do no better than rolling a equally weighted die in determining authorship for that chapter. 3) The correct author is present, but the signal is too weak to detect because of poorly constructed centroids, improper chapter cuts, or anything that will degrade the signal or sharpness of the centroid, so again the signal is not much stronger for any particular one over the others.

    Note that the majority of the most prescient criticisms (Joseph Smith missing, bad chapter cuts, centroids that should have had a different word in its list, etc), to the extent they are valid mean that the results have been biased towards flat – i.e., far easier to have false negatives, that to have false positives at statistically significant levels (not just ‘winning’ by a small amount).

    The argument that the probabilities are ‘just relative’ also seems to have far more traction than it deserves. Relative probabilities are scientifically stronger in a study like this than absolute probabilities, because we know exactly what they are normalized against (the other authors in the target list). That Rigdon and/or Spalding have such huge signals across the board relative to the other authors is saying something very strongly about the level of input they had on BoM origins. If Rigdon had little input, his centroid would fare statistically no better (or worse) than the others.

    It would be interesting to see Jeff’s writings included as a control, though. Given the statistically low level of Barlow (even for a control), Jeff probably would have been a better control centroid.

  38. Steve Bigler

    OK. I’m a newcomer here, and I have a question/comment or two.

    First I’ll disclose my bias upfront. I’m a faithful Mormon. I’m not a professor of statistics or linguistics, I’m just a simple Ob-Gyn doc. So bear with me and my admittedly rudimentary understanding of statistics and linguistics.

    However, with my medical background, I do have a pretty decent understanding of the requirements necessary to construct and conduct a quality study.

    So, here’s my first comment:

    1. The authors glibly state:

    “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).”

    Unfortunately their study simply can’t say that. Their sample size is far too small to confer that much conclusive power.

    Even if, for the sake of argument, we presume that Joseph Smith’s story is false, how do the Stanford authors know that some unknown person whom they DIDN’T test wasn’t really the author?

    For example, if they had tested writing from, say, E. B. Grandin how do they know they wouldn’t have found an even better match than what they found with Rigdon’s writing?

    Perhaps Martin Harris would have been a better match than Rigdon – but they didn’t include him in the study. How about Professor Anthon? But again, they didn’t include him in the study either.

    In fact, in their study they didn’t include literally thousands of people in the Eastern United States at Joseph Smith’s time who, for all we know, might have revealed a better match than Rigdon did!

    And since there is a virtually limitless pool of people in Joseph Smith’s time and vicinity that they DIDN’T test, it’s impossible for them to state, with any degree of certainty at all, that Rigdon “was the main architect?”

    Any number of people may have matched better than Rigdon.

    In medical studies, this is known as a sample size so small as to be worthless. Any medical researcher would be embarrassed to publish a paper with a control group of 2 and a study group of 5.

    Here’s an analogy of this weakness in the study:

    Let’s say that I claim to be a doctor, and I write a paper on Obstetrics. An observor is not convinced that I am truly a doctor – and disputes that I actually wrote the paper.

    So he collects 5 people from my neighborhood and subjects them to a rigorous test of their knowledge of obstetrics and medical terminology. He also includes another 2 people who are known airline pilots as his “control group.”

    Then, after completing his rigorous and exhaustive testing, he finds one person among the 5 who has a somewhat better knowledge of obstetrics and a somewhat better command of medical terminology than the other 4 – and promptly ordains him as the true doctor and author of the paper. And does so without even testing me!

    2. I also find it very interesting as to the 5 specific members of the study group the authors chose. Not exactly what we in medicine would term a random sample.

    Their interesting selection has the flavor of, “Well, we’re pretty sure the real author of the Book of Mormon was one of these 5 people, so they’re the only ones we’re going to test.”

    The mere selection of the individuals to comprise the study group bespeaks massive bias upfront.

    3. And finally, even if their conclusion is 100% correct, we are still left to explain the impressive number of ancient Hebrew linguistic “bullseyes” in the Book of Mormon.

    These include such ancient Hebrew literary conventions as merismus, exquisite chiasmus, simile curses, authentic ancient Middle Eastern proper names for people and places, synonymous parallelism, antithetic parallelism, repeated alternate, the prophetic perfect tense, poetic climax, compound prepositions, plural amplification, the construct state, repetition of the definite article, use of the cognate accusative, the use of many “ands” in serial, and repetition of the possessive pronoun.

    I find it highly doubtful that even Sidney Rigdon was sufficiently conversant in ancient Hebrew to craftily include all of these remarkable linguistic “footprints” within the Book of Mormon.

    So even if they manage to shift attention to an allegedly different “author,” they are still left to explain the many and varied evidences of the Book of Mormon’s fundamental antiquity.

    But perhaps Sidney Rigdon was just a spectacularly good guesser of future information unknown at his time.

  39. Steven Danderson Post author

    Hi Zax!

    You said, “Picking probabilities as done in the Jockers study is not rolling the dice to see who the ‘winner’ is, but rather should be thought of as a method of constructing meaningful dice for such a determination.”

    Normally, that is true. However, Professor Criddle’s biases inform that construction, which distort that determination. By not giving us a p level (or at least a significance level, which would give us a ball park figure for that p level), we are left with only subjective research, at best. This, I think may explain why we Latter-day Saints who know some statistics are unimpressed by the conclusions of Professor Criddle and his fellows.

    Moreover, Professor Criddle’s biases lead directly to bad history, which is the cause of the inverted null and alternate hypotheses. And yes, Zax, Professor Criddle DOES have the burden of proof, since HE is the one asserting a thesis that is NOT generally accepted.

    A parallel, I think, is Fawn Brodie’s scandalous accusation that our nation’s third President, Thomas Jefferson, fathered children with his slave, Sally Hemmings. While that thesis may be true, it was up to Ms Brodie to prove that thesis, since she made it. Since, of course, Several of Jefferson’s near-relatives had access to Ms Hemmings (and others had access to other slaves who, in turn, had access to Ms Hemmings), proving that thesis may be less possible than one might think.

    You are right, Zax, that in wordprint analysis, relative probabilities may be more relevant than absolute ones, especially if one is exhaustive in the pool of alleged authors, and confident beyond a reasonable doubt that at least one of them actually did write the Book of Mormon. To paraphrase Sherlock Holmes telling Dr. Watson, “When you eliminate the impossible, the remainder, no matter how improbable, must be the truth.”

    The problem is, though, that Professor Criddle and company were NOT exhaustive in testing alleged authors. Thus, beyond telling us that Messrs Lonfellow, Verne, and others didn’t write the Book of Mormon, we have nothing that tells us who did write it.

    To be fair, while Solomon Spaulding and/or Sidney Rigdon may have written the Book of Mormon, until we can be exhaustive in testing all possible alleged authors, my reading of the data (which show that the probability of either man being the author is less than 50%, let alone the roughly 90% which would overcome reasonable doubt!) compels me to be extremely skeptical, if not dismissive of that thesis.

    As Dr. Lindsay points out, the more one looks, the more one is distressed by the shoddy scholarship shown by Professor Criddle and his fellows. And I am NOT saying this because I am a Mormon who doesn’t like contrary evidence. I would think that the disdain that I have for Rodney Meldrum’s unauthorised revelation masquerading as research [See http://www.fairblog.org/2008/10/05/usingand-misusing-scholarship-and-revelation/ would indicate that I dislike faux-scholarship, no matter what the source.

  40. Steven Danderson Post author

    Hi Jeff!

    I really appreciate your kind words on your blog. You are right about the sad fact that, the more one looks, the more problems one should have with the Criddle “study” (Professor Joskers and Ms Witten were merely led by the nose by Professor Criddle–which is a problem there!).

    This looks too much like the global warming hysteria hyped by environmental extremmists. Yes, earth’s temperatures are higher than thirty years ago, and yes, cities are heat sinks, and yes, pollutants have a positive corellation with temperatures, but, by overstaing the actual threat, and by rigging the evidence, people could be led either to hold science itself in contempt, or to do stupid things advocated by the extremists–neither of which would make us any better off! Indeed, such pseudo-research makes us WORSE off–MUCH worse!

  41. Steven Danderson Post author

    Hi Steve!

    First, let me say that I REALLY like your analogy of the obstetrics paper! ;)

    Second, you’re right that Grandin or Harris or Anthon could have written the Book of Mormon.

    Third, I must, however, defend the Criddle study (I never though that I would do that! ;) ) from one of your points. While Criddle, et al., gather samples from too few people, the sample size from each author they did test appears to be adequate. After all, in his book, “The Signature of God,” Dr. Robert Hamsom was able to differentiate the Saviour from the Apostle Matthew with one 500-word block from each. Moreover, one 300-word block from the Saviour and Luke enabled me to distinguish the two. (Note that I am NOT saying that such a small sample size would be definitive, nor am I advocating using such a small sample!!)

    Moreover, the sample need not be random for the results to be valid. All that is required is for the sample to be representative. It is just that, usually, large random samples are the best way to get valid results.

    In the case of Book of Mormon authorship, it is unnecessary to get samples from the entire population of the USA, or even the whole of Manchester and/or Palmyra, NY. When Professor Don Foster of Vassar College correctly identified Joe Klein as the author of “Primary Colors,” he obviously didn’t test every one of the literally thousands of peole at the 1992 Democratic National Convention, let alone the millions who lived in New York City. Since the author knew things only “insiders” could know, Foster only needed to test a dozen or so people–or fewer.

    It is the same with the Book of Mormon. It is not needful to test everybody; only the ones who had opportunity.

    Do I make sense?

  42. Steven Danderson Post author

    Those of us who have read Whitsitt:

    http://sidneyrigdon.com/wht/1891WhtB.htm

    have known for a long time that Spalding and Rigdon wrote The Book of Mormon:

    http://www.fairblog.org/2008/12/14/lies-expletive-deleted-lies-and-statistics/comment-page-1/

    The problem is, Byron, Sidney Rigdon didn’t meet Joseph Smith until December 1830–two years AFTER the Book of Mormon’s publication.

    While it is certainly possible for Ridgon, Smith, and Parley P. Pratt to be lying about the post-publication meeting (and true that they lived not too far from each other!), the burden of proof is on those who make the claim–not on Joseph Smith and the Latter-day Saints.

Comments are closed.