Saturday, September 26, 2015

JWHA Report - Roper and Fields

Yesterday at the John Whitmer Historical Association meeting in Independence, Missouri, Matt Roper, Paul Fields, and I presented in a combined session.

My topic was "Ghosts of the Times and Seasons: Authorship of an Article describing Central America as Zarahemla's Location."

Roper/Fields spoke on "Joseph Smith, Benjamin Winchester and Central American Archaeology: Assessing the Authorship of the 1842 Book of Mormon Articles in the Times and Seasons."

First, it was an honor for me to have been invited. The organizers of JWHA put together a fantastic program and my only regret was having to present at the same time as several other outstanding presentations were underway with speakers such as Alex Baugh and Kyle Walker.

Second, I'm fine with a conclusion that Joseph Smith wrote the letters, if that's what actually happened. All I care about is the truth. What I don't want is perpetuation of a false historical narrative that Joseph Smith wrote or approved of everything in the Times and Seasons. In my view, that's what has been presented heretofore by the "consensus" scholars in the citation cartel, so I was looking forward to the presentations.

I had 30 minutes to present. I went through the 7 historical assumptions in the Roper/Fields article (located here) and showed how every one of them was wrong or misleading. I will go through each of them in upcoming days (I have so much material for this blog I'll never post it all.)

When I finished, Matt stood and discussed the Bernhisel letter for a few minutes. Here's a link to it.

I thought Matt did a pretty good job, given what he has to work with. He put half of the letter on the screen and read it to the audience. I had shown in my presentation that no one knows who wrote the letter because we don't know whose handwriting it was in. This is according to a note in the Joseph Smith Papers, which I have independently confirmed with the Church History Department and by my own examination of the handwriting. (I'm not a handwriting expert but I have prosecuted forgery cases and I've had to explain to juries how we distinguish one person's writing from another's, so I know more than a little about the topic.)

By contrast, here's what Roper/Fields said in their paper:  "The letter to Bernhisel, written in the hand of John Taylor, belongs to a class of historical documents that are extant only in the hand of scribes but are included in the Joseph Smith corpus (see, for example, Jessee, Personal Writings of Joseph Smith, 527–28, 551–52). The letter could suggest that Joseph Smith either dictated the letter or directed the apostle to write to Bernhisel on his behalf. In either case, it would be unlikely for Taylor to knowingly attribute views to the Prophet that were not his own."

Because Taylor is one whose penmanship is not in the letter, anyone who reads Roper's paper is being misled. Roper knows this, but he has never retracted his paper or corrected it until, maybe, now.

The Second Edition of the Zarahemla book has an entire chapter on the Bernhisel letter. I won't get into the detail here except to point out that 1) the Bernhisel letter compares the Stephens books to "all of the histories that have been written pertaining to the antiquities of this country;" 2) apart from this anonymous letter, there are no accounts of Joseph Smith reading any such histories, let alone all of them; 3) There are no records in Joseph's or anyone else's journal in which Joseph mentions Stephens or even Central America; 4) Wilford Woodruff was known for having read history extensively even before he joined the Church; 5) Woodruff is the only person who actually read the Stevens books; 6) A few years later when he was crossing the ocean from England he was reading a history of Russia and compared the author to Stevens; 7) the letter itself reads like a simple thank-you note, and a generic one at that. IMO, the evidence suggests that Woodruff was the source of the letter. I have some ideas on who wrote it but haven't had time to make the comparison. I can see Joseph taking one look at 2 volumes of history adding up to 900 pages and telling Woodruff, "Send him a thank you note." It's unimaginable to me that no one would have mentioned Joseph taking the time to read these books; if he had, everyone would have wanted to read and discuss them. The only book that Joseph's journal mentions him reading is the Book of Mormon, and his scribes even mention what page he was reading. Regarding the Stephens books, there is complete silence.

Still, it is possible that Joseph spent the time to read 900 pages, plus all the other histories he compared the Stephens book to, and no one thought it was unusual enough to write about. Maybe Joseph spent all his time reading. I'll let anyone familiar with his life's history, his personality, and his heavy schedule and responsibility think about that one.

That said, Roper did a good job. He took only about ten minutes.

Then Paul Fields stood.

I anticipated that they were going to revisit their stylometry study. I had a few slides on stylometry but I skipped over them in the interest of time. I did mention that stylometry has just a few weaknesses and problems. I'll do a separate post on that. But it turns out I didn't need to mention the problems with stylometry.

To no one's surprise, Fields claimed his study (Matt, apparently, had little to do with it) verified that Joseph Smith wrote the articles. But I was surprised at what happened next.

Paul Fields impeached his own methodology.

He spent around 10 minutes discussing the Federalist Papers as an example of stylomtery. It reminded me of Jeane Dixon, who had predicted the assassination of JFK and forever after rode that single success, despite a long list of failed prophecies.

(FYI, the Federalist Papers are a collection of 85 essays about the Constitution written by John Jay, Alexander Hamilton, and James Madison. They all wrote under a pseudonym of "Publius," but shortly after publication, contemporary experts recognized the writing style of the three men, who had been publicly discussing these issues. Hamilton later took credit for most of the essays, Madison took credit for some that Hamilton had claimed, and some remained unclear. Numerous authorship studies have been done, most of which attributed the disputed essays to Madison, but some claim the disputed essays were a collaborative effort, which is what I think. I've referred to the Federalist Papers in some of my books on Constitutional Law.)

Fields, of course, has had nothing to do with the examination of the Federalist Papers. Stylometrists like to cite this case because it is famous and has produced a fair overall consensus. But this is nothing like the case with the 1842 articles. First, we know there were three candidate authors. Second, the authors themselves took credit for the papers. Three, the authors had publicly discussed the issues and were well-known experts on the topic (which is why we care about their opinions). Four, the corpus (the essays themselves) discussed a clearly defined topic. Five, the essays were of approximately the same length (900-1500 words). Six, the essays were all published. Seven, they were contemporaneous. Eight, they were all addressed to the same audience.

Fields pointed out elements 4-8 as being key to an effective stylometry analysis. He didn't mention 1-3, but in my view, those were equally as important. The Federalist Papers were an easy case, actually; even people in the 1700s without computers could identify the authors.

The fallacy in Fields' presentation, which left me wondering what he was getting at, was NONE OF THESE ELEMENTS IS PRESENT IN THE 1842 ARTICLES.

Granted, the Federalist Papers is a good example of stylometry. But the reasons it is a good example do not apply to the 1842 articles. I'll go through each element and show why I question his conclusions.

1. The list of potential authors for the 1842 Times and Seasons (T&S) articles, particularly those on Sept. 15 and Oct. 1, is unlimited. A majority of the material published in the T&S came through the mail. Anyone could have sent them in. There is good evidence that the extracts in the articles was proofread against the actual books (although even then there are some strange copy mistakes), but that is not evidence of the origin of the commentary (the 900 words in dispute).

2. No one ever publicly took credit for the T&S articles.

3. Some of the potential authors of the T&S articles had publicly discussed the topic, or would in the future. I identified five of them in my presentation: Orson Pratt, Parley P. Pratt, W.W. Phelps, Benjamin Winchester, and William Smith. Arguably, Wilford Woodruff discussed the topic, although the only record is in his journal from a year previously. So far as I know, he never discussed it publicly. Besides, he and John Taylor were extremely sick in this time frame and didn't even come into the office. The one candidate for whom there is no record--zero--of ever having discussed Central America publicly (or even privately) is Joseph Smith. And, unlike the authors of the Federalist Papers, none of the T&S candidates were experts on Central America. At most, they had read the Stephens book.

4. The corpus in Fields' study is not comprised of other writings on the same topic. (Actually, no one knows what the corpus is. This was my complaint about the original Roper/Fields study. I haven't had time to read their paper; I'm only responding to their presentation here. Maybe in the paper they list every document they used in the comparison, along with their software and parameters. I hope so.) At any rate, if they confined their corpus to writings about Stephens, or even writings about Mesoamerica and the Book of Mormon, they have a corpus consisting of Orson Pratt, several anonymous articles, and a few writings by the men I mentioned above. But they'd have nothing from Joseph Smith. Actually, they'd have nothing on the specific question of Zarahemla in Guatemala because on that point, the T&S articles are unique.

Roper/Fields even included the Bernhisel letter in their corpus! And we know it is not holographic for Joseph Smith. We don't know who wrote it.

5. Since I don't know what corpus they used, it's possible Roper/Fields came up with a statistically meaningful collection of 300-word essays and comments. That's what we're looking at in the T&S: 3 short pieces totaling about 900 words. I'm eager to see this corpus. But in the first article, Roper/Fields cut up longer works to come up with comparable chunks. That begs the question, though; is a 300-word sample valid? Is it a valid assumption that all three articles in the T&S were written by the same person? Someone in the audience astutely asked about this. Fields said they were all the same (although in their published data, Roper/Fields only show them combined into one). I answered that I had looked at them separately with stylometry software and got different results for the three pieces. But I question the validity of those results anyway, given how short the pieces are and the evidence of editing input.

6. During the Q&A, the question of what prior works of Joseph Smith the authors used. All Matt would say was they used holographic writings of Joseph Smith. Holographic? So they are comparing hand-written material by Joseph Smith to a published article. That directly contradicted Fields' own criteria. Maybe the analysis of the Federalist papers used Madison's letters to his wife, but I doubt it and Fields himself said that would be produce poor data. Or maybe Roper/Fields used holographic samples from Phelps, Winchester and the others, but I'm aware of only a few letters written by Winchester that I doubt add up to 900 words each. We do have quite a few holographic letters that Phelps wrote to his wife; maybe that's what Fields used? Presumably, this is in the paper, but they didn't elaborate during the presentation.

7. Fields made a point that comparison samples should be contemporaneous, so I'm eager to see what writing samples they used from Joseph Smith during 1842, plus or minus a year, as well as what contemporaneous samples they used from the other candidates. Actually, I'm sure every historian at the conference would have attended had Roper/Fields produced such a corpus.

8. The 900 words were addressed to the readers of the Times and Seasons. I'm not aware of any holographic writings by Joseph Smith that were published as such in the Times and Seasons, but maybe there were some. Many of the other writings by the 5 candidates were addressed to unbelievers (as part of a larger argument).

Bottom line, all the criteria Fields identified that made the Federalist Papers example so compelling are absent here.

Fields claimed their results are replicable and have all been published in peer reviewed journals. So far as I know (relying on this site), the only two things Roper/Fields have published in this area are

1) their original study of the Times and Seasons, linked above, published by the Maxwell Institute, which I don't think was peer reviewed or even copy edited (although I noted Roper had fixed the typo from the conclusion in the original published article). Certainly this article does not contain enough data to allow anyone to replicate the work.

2) a study of Earl Wunderli's book that Wunderli himself effectively rebutted, IMO, here. The Roper/Fields article was published by BYU Studies, but the link is broken so I can't cite it now.

They also have published three articles in the Interpreter, which as I've shown is about as far from peer-reviewed as a publication can get. I'll go through Roper's third article as soon as I get a chance.

Maybe Fields has published more about stylometry. I'd very much like to see his other articles; they didn't come up in a google search, and they're not otherwise cited that I can find. Fields portrayed himself as an expert during the presentation, so I'm quite curious to see what his representation is based upon. When he finished his presentation, I was itching to conduct cross-examination.

Well, I'm out of time on this. Maybe I'll pick up more later.

At the end of the session, I suggested to Matt that we get together and discuss these things. I think we agree about more than we disagree about, and I'd like to narrow ad define the disagreements so we can evaluate them rationally. I barely had time to scratch the surface in my presentation. He claimed he was too busy here at the conference, and that we could meet back in his office in Utah, but only when Paul Fields is available.

We'll see if/when that materializes.


Roper/Fields want us to believe that Joseph Smith wrote these three short comments, accompanied by extensive excerpts from the Stephens books.

This is in the same issue of the Times and Seasons (15 September 1842) in which we find this:

"The following letter was read to the Saints in Nauvoo, last Sunday week, and a copy forwarded to us for publication:-and cordially we give it a hearty welcome, and a happy spread among those who love the truth for the truth's sake."

The letter referred to was written by Joseph Smith on 1 Sept 1842 and became D&C 127. Roper/Fields would have us believe  that Joseph sent this letter to himself to publish. During his presentation, Matt referred to Joseph as the nominal editor, so maybe he's conceded that Joseph wasn't acting as editor at this point. Hopefully everyone can agree that Joseph did not send D&C 127 to himself for publication.

However, Roper/Fields still want us to believe that the editor who gave Joseph's letter a "hearty welcome" also received these three articles from Joseph Smith that he didn't even identify as coming from Joseph. Is that plausible?

The next issue, 1 Oct. 1842, includes a letter from Joseph Smith dated Sept. 6 1842 that is titled "LETTER FROM JOSEPH SMITH." We're supposed to believe the editor used such a headline to tout this letter, but left Joseph's other supposed article, the one on Zarahemla, anonymous.

It defies credulity to propose that the same editor would emphasize Joseph's authorship of two letter but suppress his authorship of 3 supposed articles.

Then there is the problem that none of Joseph's papers mention him reading, discussing, or writing about the Stephens books. His journal includes the creation of the Book of the Law of the Lord and other material, but says nothing about Stephens. We're supposed to believe that while Joseph Smith was evading extradition, setting forth temple doctrine, overseeing the construction of the Nauvoo temple, welcoming new members, developing Nauvoo real estate and so forth, he was lugging around the 900-page Stephens books, making selections to extract anonymously in the Times and Seasons.

No doubt there will be some people who believe this. Apparently Roper and Fields do.

But I don't.

1 comment:

  1. Conjured up evidence meets JAG. Hmmm, I don't think that's fair. The fact that they try to discredit the very translation that they are trying to use as evidence to prove their point kinds leaves them with out much to hold on to. Besides their archaeology is based on old inaccuracies as well. Too many unanswered questions. Not much to believe in.
    On the other hand, the Heartland theory is based on new, continuous, discoveries of accurate evidence in many different scientific fields. Hmmmm, nope not fair.
    Well, let's hope that they can save face a little by agreeing on some minor, obvious points. Maybe that will be a good start.