“Stylistic Drift” during Book of Mormon Dictation

I’ve posted before on Worlds Without End about some examples of “stylistic drift” that occurred during Book of Mormon dictation. As I then explained, Joseph Smith didn’t restart his dictation from the beginning of the Book of Mormon after Martin Harris lost the first 116 pages. Rather, Smith picked up the dictation where he’d left off: at the beginning of the Book of Mosiah. Only after translating from Mosiah to the end of the Book of Mormon did he come back to translate 1 Nephi through Words of Mormon. When we arrange the text according to this Mosiah-first dictation sequence, we observe that at the beginning of Book of Mormon dictation Smith tended to use the terms “therefore,” “whosoever,” and “insomuch,” whereas by the end of the dictation process he instead preferred the synonymous terms “wherefore,” “whoso,” and “inasmuch.” This implies that Smith exerted at least some influence on the vocabulary of the English Book of Mormon rather than just passively reciting an existing English text projected on the surface of his seer stone.

In order to strengthen this hypothesis of monotonic stylistic drift, a few years ago I devised a way of statistically testing it in the aggregate. The results were consistent with the hypothesis, and I posted them on my personal blog. Recently I was contacted by BYU statistician G. Bruce Schaalje, who had discovered my work after independently performing a similar analysis with a similar result. Bruce looked at my data and suggested a few simple, important refinements, so it seems an update is in order. I’m grateful to him for his help.

To mitigate reader boredom, I’ll begin with an abbreviated explanation of my results. Those interested in the full details can find them after the chart below.

Here’s the short version. First I divided the Book of Mormon into twenty-one sections of ten chapters each, which I arranged according to their Mosiah-first dictation sequence. These twenty-one sections are represented by the x-values in the chart below. Then I measured each section’s relative stylistic similarity (based on the frequencies of about sixty common words) to the front and back ends of the Book of Mormon (when arranged according to dictation sequence). These are represented by our chart’s y-values. Higher y-values indicate greater similarity to the front end of the Book, while lower y-values indicate greater similarity to the back end. If the hypothesis of monotonic stylistic drift is correct, then we should end up with a fairly straight trend line with a negative slope. As our x-values increase, our y-values should decrease at a fairly constant rate. As you can see, this is basically the pattern we observe. The first data point’s a bit of an outlier, so we do get some deviation there from the predicted slope. But overall, the pattern is consistent with monotonic stylistic drift. This is significant because monotonic stylistic drift is what we would expect if the English text of the Book of Mormon were composed by a single author or translator in a Mosiah-first sequence of dictation. If the English text had multiple authors/translators or was composed in a different sequence, we wouldn’t really expect to see this pattern in the data.[1]

Here’s a fuller description of my method for those interested in the gory details:

1. First, I divided the Book of Mormon into twenty-one sequential ten-chapter sections, starting with the first chapter of Mosiah. I excluded Book of Mormon chapters that parallel chapters in the King James Bible. I also excluded from my final analysis a twenty-second section consisting of only five leftover chapters (Jac. 7, Enos 1, Jar. 1, Omni 1, and Words 1). This section was an outlier, probably because the text sample was too small.

2. Second, I generated a list of all the articles, conjunctions, pronouns, and prepositions that are found in all twenty-one of my ten-chapter sections. In the literature on stylometry these words are sometimes referred to as “non-contextual” words. Bruce Schaalje prefers to call them “low content” words, because they aren’t really non-contextual. The theory is that by studying texts’ usage of such low-content words, we may be able to detect their authors’ distinctive patterns of speech or thought with relatively little interference from other variables such as subject matter or genre.

3. Next, I computed what we might call the “stylistic distances” between each ten-chapter section and each individual chapter of the Book of Mormon. Basically, these “distances” were a measure of how differently the words from Step 2 are used in each pair of texts. More specifically, I used a measure known as “Delta.” Delta is an averaged and normalized measure of differences in word-frequencies between two texts.[2]

4. I now had a spreadsheet with 21 columns (representing the 21 ten-chapter sections) and 215 rows (representing the 215 individual Book of Mormon chapters). I then found the regression slope of each column.[3] Crudely put, these regression slopes represent each ten-chapter section’s relative similarity to the first- and last-dictated chapters of the Book of Mormon. A higher slope indicates a greater stylistic similarity to the first-dictated half of the Book of Mormon, while a lower slope indicates a greater stylistic similarity to the last-dictated half of the Book of Mormon.

5. I plotted these regression slopes as y-values on the graph above, then generated a polynomial trendline. If the Book of Mormon’s style changed monotonically during the course of dictation, the resulting trendline should be a nearly straight line with a negative slope. As you can see, this prediction is mostly borne out in my results.

NOTES

[1] To download my data, click here. I performed two control tests, one on the Book of Mormon arranged according to 1 Nephi-priority, and another on the King James Bible. Neither test produced a pattern consistent with monotonic stylistic evolution, though the KJV chart is actually pretty interesting in its own right, exhibiting text clusters that likely reflect the influence of both genre and different translation committees. Follow the hyperlinks to see the charts.

[2] For an introduction to Delta, see John Burrows, “‘Delta’: A Measure of Stylistic Difference and a Guide to Likely Authorship,” Literary and Linguistic Computing 17 (3): 267–87.

[3] In the regression analysis for each column, I excluded the rows corresponding to chapters that are part of that column’s corresponding ten-chapter section. This was to avoid a sort of “false positive” effect. Delta distances between a ten-chapter set and the chapters comprising the set would tend to be smaller than delta distances to other chapters, potentially resulting in a linear, downward sloping plot even in the absence of style evolution by the author. I’m grateful to Bruce Schaalje for bringing this problem and its solution to my attention.


Comments

“Stylistic Drift” during Book of Mormon Dictation — 19 Comments

  1. Fascinating. One question: Why are you plotting such higher order polynomials when a straight line (to my eye) seems to be the best fit. The KJV especially: the dots just seem to have a general downward trend. (With perhaps one outlier on the upper right) Not go up, go down, flatten, go down then go back up as this higher order polynomial suggests.

    But other then that curiosity, awesome post.

  2. Joseph,

    To my eye, the KJV data seem to group into three clusters, with each cluster having an upward rather than downward trend. You could certainly fit a straight line with a negative slope to the data, but that, I think, is an artifact of the way the different genres are clustered at different ends of the book. In other words, the “New Testament Epistles” cluster is more negative because the back end of the Bible is comprised largely of New Testament Epistles. The “Old Testament Historical Books” cluster is more positive because the front end of the Bible is comprised largely of Old Testament Historical Books. The Wisdom literature clusters in the middle, because the middle of the Bible is comprised mostly of Wisdom literature. When you look at the individual clusters, though, they don’t trend in a negative direction. I chose a higher order polynomial to show this quasiperiodicity in the data. You can see it even better with a moving average.

    Blake,

    There’s variation in the data because measuring authorial style with word frequencies is far from an exact science. Word frequencies are sensitive to a lot of factors, authorial style being only one of them. However, you’re mistaken in your assertion that there is no trend.

  3. Christopher: Well, we have a plethora of very sophisticated word-print studies done prior to yours that point to intra-author unity among authors identified as such in the text. Yours is not nearly as careful and sophisticated as the prior computer word-print studies.

    Your numbers seem to be just as consistent with changes in authorial frequency of words during differing times. As you know, I am quite skeptical of the assumptions underlying the entire word-frequency methodology and your results only strengthen my skepticism.

  4. Blake, I agree that the various attempts at authorship attribution using statistical methods are highly dubious for many reasons. And you’re right that my results are merely consistent with monotonic translator drift; they do not absolutely disprove (though they do render somewhat less probable) the possibility of multiple authors or translators of the English text.

    My results, of course, have even less bearing on the question of whether a multiply-authored ancient source text underlies the English translation. The method simply isn’t designed to answer that question. It’s specifically designed, in fact, to filter out such “noise” from other variables—to see if monotonic drift is detectable beneath other signals such as genre, subject-matter, and multiple narrators or original-language authors. It isn’t designed to detect the presence or absence of anything but monotonic stylistic drift itself. So while it does perhaps have negative implications for the Spalding-Rigdon hypothesis, it doesn’t have any particular implications for the hypothesis that the Book of Mormon is ancient.

  5. In my hurry I seem to have overlooked your suggestion that my data might be explained by “changes in authorial frequency of words during differing times.” That explanation doesn’t work because the monotonic drift is only evident when the Book of Mormon is arranged in its dictation sequence. We don’t see the trend when we arrange it in its published textual sequence, which roughly follows the chronology of its ancient composition. It’s an interesting suggestion, though, and I guess this is one respect in which my study might have some small bearing on the ancient authorship question: the method doesn’t seem to detect a monotonic chronological drift in Nephite use of low-content words. But then, I don’t know that we’d really expect such drift to be detectable in the English translation.

  6. Interesting bit of research. It would be neat to see someone do another word frequency multiple underlying authorship test, but control for drift and see how that affected the results. I don’t imagine it would be especially easy to construct, but it would be terribly interesting.

  7. Blake and Chris, I am interested in word print studies and related subjects. Different systems have been used at various times some with pretty good results. I think what Chris and others are doing is great and necessary in order to determine the limitations of various methods.

    Chris, have you applied your system to other works, such as the Doctrine and Coevenants and maybe other non LDS works?

    Glenn

  8. Glenn,

    I did do a study of the Urantia book a few years back. See here. Needs revising and updating, though.

    Basically my conclusion—by which I still stand—was that detecting monotonic stylistic drift is about the only thing Delta is good for when it comes to highly genre-distinctive texts. Statistical authorship attribution of a target text using Delta just doesn’t seem to work unless you have texts of the same genre to compare it to. I suspect this is also true of other statistical methods of authorship attribution, not just Delta.

  9. Chris, I know that statistical studies were very effective in showing that it was indeed Sholokhov who wrote “Silent Flows the Don,” despite Solzhenitsyn’s claims to the contrary. The study was bolstered by new research which demonstrated that he was well acquainted witht the prototypes and events pictured in the novel. Word studies only go so far, and are most effective when suplementing other types of evidence.

  10. Allen, wordprint studies can certainly be useful under the right conditions. The Sholokhov case satisfied those conditions better than either the Urantia Book or the Book of Mormon, because both of these latter texts adopt unusual vocabulary, syntax, and poetic meter that are not found in other texts definitely known to have been composed by the target texts’ candidate authors. The Book of Mormon also makes heavy use of biblical quotation, which complicates matters further. You might be able to find individual syntactical peculiarities which are shared between the Book of Mormon and a candidate author’s other texts, but I simply have no confidence in attempts to draw meaningful conclusions from a comparison of syntax or vocabulary at an aggregate level.

    The one previous application of “wordprint” technology to the Book of Mormon that I do think is potentially fruitful was Dr. John Hilton’s comparison of two texts within the Book of Mormon—Alma and Nephi, IIRC. Hilton found that the syntactical difference between them exceeds the syntactical difference we would normally find between two texts by the same author. I have some reservations about this finding, but it’s at least based on a comparison of apples to apples, and so cannot be dismissed lightly.

  11. “You might be able to find individual syntactical peculiarities which are shared between the Book of Mormon and a candidate author’s other texts, but I simply have no confidence in attempts to draw meaningful conclusions from a comparison of syntax or vocabulary at an aggregate level.”

    For sure. Various genres does complicate things immensely.

  12. If the conclusion was an apologetic of the authenticity of the Book of Mormon codex as the small plates of Nephi being less abridged by Mormon than all the Mosiah on material from the large plates, and thus should have a different stylistic voice, I think the responses might be different.

  13. Pingback: Sunday in Outer Blogness: Gay Family Easter Edition! » Main Street Plaza

  14. This implies that Smith exerted at least some influence on the vocabulary of the English Book of Mormon rather than just passively reciting an existing English text projected on the surface of his seer stone.

    Joseph served as an interpreter of multiple speakers, so you are not going to have him exerting some influence on the vocabulary, but total influence. It was wholly put into his own words. So stylistic drift is to be expected. The BOM translation was an on-the-spot interpretation, as if the text was spoken in one language and then immediately interpreted into another language. There was no dictionary consultation to find more precise words that the interpreter might not be familiar with using. It was simple dictation, as he felt out (not saw) the right words.

    So, it is to be expected that there will be differences due to the multiple, original authors’ writing styles and also stylistic drift due to the singular interpreter.

    Also, it is to be expected that the Small Plates of Nephi were placed by Mormon at the end or bottom of the Plates of Mormon, not inserted in the middle of them. So it makes sense that this is why the drift is shown.

    When the Plates of Mormon are re-translated so that we get Mormon’s original Title Page (which he wrote at the beginning of the record), the Book of Lehi and the beginning of the Book of Mosiah, I expect we’ll also get a proper view of the plates’ placement, and that it will show that the Small Plates of Nephi were placed at the end of the entire record.

  15. Joseph served as an interpreter of multiple speakers . . . he felt out (not saw) the right words.

    That’s a matter of some dispute. Royal Skousen has argued for a “tight translation,” in which Joseph was merely a passive mouthpiece. In this model, Joseph exerted no influence over the vocabulary; he merely read off the words that appeared on the surface of his seer stone. Skousen also believes the translation was extremely literal, which accounts for the Hebraisms in the text. Other LDS scholars, of course, have taken a position closer to your own. The OP is best thought of as a contribution to that debate—a contribution that supports your position.

    Also, it is to be expected that the Small Plates of Nephi were placed by Mormon at the end or bottom of the Plates of Mormon, not inserted in the middle of them. So it makes sense that this is why the drift is shown.

    I’m not sure I follow. Are you saying the stylistic drift I’m observing might be from Mormon rather than Joseph Smith? The Small Plates weren’t abridged by Mormon, so we shouldn’t see his style continuing in the Small Plates.

  16. Royal Skousen has argued for a “tight translation,” in which Joseph was merely a passive mouthpiece. In this model, Joseph exerted no influence over the vocabulary; he merely read off the words that appeared on the surface of his seer stone.

    No, of course not. The translation was done by faith, not by knowledge. There were no words appearing before his physical eyes. The only authoritative text we’ve been given (through Joseph Smith) about the process of translation is found in D&C 9, which only speaks of a “feeling.” Everything outside of that text is speculation, David Whitmer’s statements notwithstanding. So, we know a word was “thought of” by the translator, which is what people do when they study stuff out in their minds, they think of things. As this was a translation, the things being thought of were words. Then the translator asked the Lord if the word he had thought of was right. If right, he got a feeling. Oliver failed because he didn’t trust his feelings, but instead wanted the word to be revealed to him (instead of him being the one who thought up the word). Oliver failed because he tried to translate by knowledge, and not by faith, as Joseph did. Joseph trusted (had faith in) his feelings, using the “eye of faith” (the mind’s eye), whereas Oliver did not but wanted a visual or audible witness, or some other manifestation stronger than a feeling, that the word was given of God.

    So, in this particular, I disagree with Skousen. Joseph most certainly did not read off words that magically appeared before his eyes.

    Skousen also believes the translation was extremely literal, which accounts for the Hebraisms in the text.

    In this instance I agree with Skousen. This was a machine translation, akin to taking a text and running it through the Alta Vista Babel Fish. This was to be an additional witness that the text was of ancient date. Unfortunately, young Joseph obviously saw how the literal translation didn’t flow so well in English and so made changes to make it read better.

    I’m not sure I follow. Are you saying the stylistic drift I’m observing might be from Mormon rather than Joseph Smith?

    No, I’m saying it was from Joseph, but that the drift was shown from the Plates of Mormon to the Small Plates because of where Mormon placed the Small Plates. In other words, Joseph didn’t skip over the Small Plates and then continue with Mosiah and then when he was done with the Plates of Mormon return back to where the Small Plates were inserted (between the Book of Lehi and the Book of Mosiah) and translate them. No, he simply kept translating the record and after he finished translating the Plates of Mormon, he saw the Small Plates (’cause they were at the bottom of the stack of plates) and translated them, too. Then he placed the translation of the Small Plates at the beginning, took Moroni’s second Title Page, found at the end of the Book of Moroni, and put it at the very beginning (where Mormon’s first Title Page was originally, and edited the Book of Mosiah so that it began where the Words of Mormon ended.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>