Monday, October 12, 2015

More stylometry fun

Despite my skepticism about stylometry as applied to the 900 words of Mesoamerican articles in the Times and Seasons, and despite Brother Roper's ongoing refusal to make his data, his software, and his assumptions public (not to mention his 9-month long refusal to share them with me), I thought it would be interesting to experiment a little more.

I tested Joseph Smith's holographic writings against a sample of Winchester's writings and D&C 127-128, which were published in the same issues of the Times and Seasons as the 3 Mesoamerican articles (the 900 words), and the  3 American Antiquities articles from May, June and July 1842.

[Actually, there are four extracts from Josiah Priest's book American Antiquities in this time frame. They appear in the May 1, June 1, June 15,and July 15 editions--all after William Smith started the Wasp and, I think, was also working on the Times and Seasons. The only person whom we know owned a copy of Priest's book in this time frame was Benjamin Winchester. Brother Roper has never produced any evidence that Joseph Smith owned, read, or even saw American Antiquities. Brother Roper and others have suggested that Winchester merely copied from Parley P. Pratt's Voice of Warning, but I have shown that Winchester used a different edition of American Antiquities than Pratt did.]

The results of my experiment indicate that Joseph's holographic writings are closest to D&C 127-128, and that Winchester's writings are closest to the 900 words and the American Antiquities articles.

For anyone who looks at the historical context, this is no surprise. D&C 127 and 128 were letters by Joseph that were delivered to the editor of the Times and Seasons for publication. Because Woodruff and Taylor were both out with serious illness, Brother Roper insists Joseph delivered these letters to himself. (Think about that one a moment.)

By contrast, I think the historical evidence shows William Smith and W.W. Phelps were actually editing the Times and Seasons in September 1842 and that Winchester mailed the American Antiquities articles and the 900 words from Philadelphia (along with the announcement about his Concordance). I also think Phelps edited the 900 words.

I proposed all of this based on the historical evidence and the writing style. Now the stylometry corroborates this position.

As always, I invite those interested to conduct their own experiments and see for themselves.

Here are the results. On each of these dials, author #1 is on the left, author #2 is on the right, and the needle points to which one the unknown author most resembles. Of course, this analysis tool is simplified for presentation, but it does give an indication. I just plugged in the respective writings and here's what came through.

1. Joseph Smith (Author 1) vs. Benjamin Winchester (Author 2) and D&C 127-128 as the unknown author:



This indicates Joseph Smith was the likely author of D&C 127 and 128.

2. Joseph Smith (Author 1) vs. Benjamin Winchester (Author 2) and the 900 words as the unknown author:




This indicates Benjamin Winchester was the likely author of the 900 words.


3. D&C 127-128 (Author 1) vs. Benjamin Winchester (Author 2) and the 900 words as the unknown author:





This indicates Benjamin Winchester was the likely author of the 900 words.

1. Joseph Smith (Author 1) vs. Benjamin Winchester (Author 2) and American Antiquities articles as the unknown author:




This indicates Benjamin Winchester was the likely author of the American Antiquities articles.






No comments:

Post a Comment