Restoring the Original Text of the Book of Mormon

Royal Skousen
August 2010

Restoring the Original Text of the Book of Mormon

[Editor’s Note: This presentation was made by Dr. Skousen at the FAIR Conference, on August 5, 2010. Slides were kindly provided by Dr. Skousen.]


What I’m going to be talking about today is restoring the original text of the Book of Mormon. I’ve been working on the critical text project of the Book of Mormon since 1988 and there are a number of works that have been published.


There are two main goals in this project. One is the attempt to restore the original English-language text, to the extent it can be done. This is the text, I believe, that Joseph Smith received through the instrument (either the Nephite interpreters, later called the Urim and Thummim, or the seer stone) and was able to read off. The second part of the project is to determine the history of the text, the kind of changes that it has undergone—both accidental and editorial (many of the latter have been of a grammatical nature).


Most of the work in recovering the original text involves two manuscripts. The most important one is the original manuscript, the one that Joseph Smith dictated to his scribes. Of this manuscript, 28 percent is extant. The other manuscript is called the printer’s manuscript and is a copy of the original manuscript. This second manuscript is the one that was prepared to take to the printer in 1829-30 to set the type. In the project we have included 20 printed editions of the Book of Mormon, 15 LDS editions, one private edition from 1858, and four RLDS editions.


I point out here that the photographs you will be seeing, also when viewed online, are protected by copyright, as described in this slide.


As I indicated, 28 percent of the original manuscript is extant. (In calculating this percentage, I exclude the 116 pages that were lost by Martin Harris in 1828.) In 1841 Joseph Smith placed the original manuscript in the cornerstone of the Nauvoo House, a hotel being built in Nauvoo. And the manuscript lay there in the cornerstone for the next 41 years until in 1882 Lewis Bidamon, the second husband of Emma Smith’s, after her death, retrieved the manuscript. Most of it was severely damaged by water that had seeped in as well as by mold that ate away a lot of the manuscript. Bidamon gave most of the larger manuscript portions to LDS people, and so 25 of that 28 percent has ended up in the archives of the LDS Church. There is half a leaf at the University of Utah. And the equivalent of a leaf in fragments is held privately. Most important for this project has been the discovery of two percent of the text that Wilford Wood bought from Charles Bidamon, the son of Lewis Bidamon, in 1937.


Let me show you some of those fragments. Here we have the lump of fragments as they were observed on 30 September 1991. At the time we weren’t sure if this really was the original manuscript, or what it might be. It turned out to be, for the most part, from the original manuscript.


Here Robert Espinosa, head of conservation at BYU’s Harold B. Lee Library, is beginning the very difficult task of teasing apart these fragments.


This is one of the fragments from 2 Nephi 7-8, all rolled up.


First, it was unraveled, and you can see on the edges where the mold had eaten away parts of the leaf. You can also see the large water stain in the center, from water that had originally gotten into the cornerstone.


After the fragment was leveled and photographed, you can see basically what it is. The text is in the hand of Oliver Cowdery; the ink was originally black and has turned brown over time.


We found that black and white ultraviolet photography brought out the text best of all. This fragment of the original manuscript is very interesting. When Oliver copied this particular portion of the text into the printer’s manuscript, he made six changes, of which five were accidental. In this case, he was copying from an Isaiah quotation, which is difficult enough. This number of errors was unusual for him; he was probably getting tired as he was making the copy for this part of the text. But he also made one conscious change here; this was of a grammatical nature.


One of the biggest discoveries of the critical text project was the finding that for one sixth of the Book of Mormon text the printer’s manuscript was not the manuscript taken to the printer; instead it was the original manuscript. And we can see this quite well from these pencil marks here in this color photograph of Helaman 15:9-14. These pencil marks were placed there by the typesetter. About one third of the time he marked up his manuscript in advance of doing the typesetting. We find evidence that he used the original manuscript from Helaman 13:17 through the end of Mormon.


Here is the black and white ultraviolet photograph for the same part of Helaman 15; as one would expect, it doesn’t show the pencil marks very well.


Click for Slide 14) The reason this finding is important for the text is that it means that from Helaman 13:17 to the end of Mormon there are two firsthand copies of the original manuscript. We have very little of the original manuscript remaining for this part of the text. It’s just a small percentage. But for this part, we have two firsthand copies, which basically means that when those two copies—the printer’s manuscript and the 1830 edition—agree, then that’s probably what the original manuscript read. And when they disagree, then one of the readings is probably the correct one. But that can be a difficult problem, trying to determine which reading is the correct one.


Here’s what the printer’s manuscript looks like. This is the first leaf, the bottom portion of which has been worn away, about one and a half lines. This page is in the hand of Oliver Cowdery.


Here is a blown-up section from that first page, and you will note that for these lines there are a number of corrections. On the third line, two words, to be, are crossed out; and written above the crossout is a grammatical correction that Joseph Smith made, the change to the word is. Joseph made this change (and others like it) when he edited the printer’s manuscript for the 1837 edition, the second edition of the Book of Mormon. On the next line—you can barely see it—there is a capital letter P. That was added by the typesetter. It’s above the line, written in pencil, and it means: put a new paragraph here.

There are a couple of other corrections here that were made by Oliver Cowdery when he originally wrote down the text for the printer’s manuscript. Sometimes he missed some words or wrote something wrong, which he then corrected, sometimes by inserting words above the line. In the last line shown here, Oliver originally wrote the word that, then he crossed it out and wrote the above the crossout. So you get these kinds of corrections by the original scribe in the printer’s manuscript. And you also get corrections like these in the original manuscript.


Next is what’s called a facsimile transcript. In doing the critical text project of the Book of Mormon, we produced these transcripts, which faithfully reproduce what’s actually on the manuscripts. You will note, by the way, that the word prophets was misspelled as prophits in the next-to-last line. And destroyed was not spelled correctly in the last line. We leave it to you to figure out what the reading should be, and most the time there isn’t any problem.


In 2001 the Foundation for Ancient Research and Mormon Studies (FARMS) published the facsimile transcripts of the original and printer’s manuscripts. Here are the two volumes, these three large blue books which are up here, and they reproduce all the known portions of the original manuscript as well as basically the complete printer’s manuscript. From these books you can read what’s on the actual manuscripts.


Since 2001 I have continued work on three other volumes. Volume 4 was completed last year. This volume is called Analysis of Textual Variants of the Book of Mormon. There are six books in that volume, namely, these maroon books which are up here. They represent my work in going through the text, basically from verse to verse, looking at all of the variants (and potential variants) in the text as well as looking at the evidence, in trying to determine what the original reading might have been.

Volume 3 is The History of the Text of the Book of Mormon, which I am currently working on. And volume 5 will be a computerized collation that will be made available with volume 3. In a moment, I’ll show you what that collation looks like.


What I discovered, as I was getting near the end of completing the six books of volume 4, was some potential problems with the project as originally conceived. One of the problems was that I was having difficulty in getting people, even some academic people doing research on the Book of Mormon, to cite the findings of the critical text project. Here are these lovely six books, very heavy—I know they are heavy, I carried them in here. As a complete set, the books are also rather expensive, and people who were writing articles on the Book of Mormon were quite oblivious to what the actual reading should be—or at least what the critical text project had to say about the reading.

Another problem that I noticed was that with the completion of volume 4 anyone could go out and produce a Book of Mormon, taking the findings of volume 4 and referring to the resulting book as the original text. I decided that I would prefer to do this myself, and so I arranged with Yale University Press to publish in August 2009 The Book of Mormon: The Earliest Text. This book basically represents in one volume the original text to the extent it can be determined. When you take off the dust jacket of this book, you will see that the hard cover matches the maroon cover of volume 4. This was done intentionally, to show the connection to volume 4 of the critical text: the Yale edition derives from the decisions made in volume 4.


There are two important innovations in the Yale edition. The first is that it is set in sense-lines—not in small paragraphs where each verse is its own little paragraph, nor in narrow double columns. My idea, as originally conceived, was to break the lines in the text so that they would represent in some general way how Joseph Smith dictated the text—namely, in phrases and clauses, but none of them so long that they could not be easily read.

Here we will look at some of the pages of the Book of Mormon as they appear in the Yale edition. One of the things I have discovered since its publication is that readers particularly like this format. Many readers have seen things in the text that they have never seen before, simply because it is laid out in sense-lines. Nor do readers get fatigued as they do when reading a two-column text that frequently breaks in the middle of words (by hyphenation) or in the middle of phrases and clauses, a process that puts a lot of stress on the reader in trying to negotiate the text. With the sense-lines, it’s also much easier to keep on reading. Some readers have discovered they can now read several chapters at a time instead of just the one chapter at a time that they were used to reading.

The second innovative aspect of the Yale text is that it is the first edition that has attempted to publish the original English-language text, to the extent that it can be determined. This particular text, it turns out, does not have what we call a copytext. It is not a revision of some particular edition of the Book of Mormon. Instead, everything has been done from scratch, so to speak; in fact, the text is directly derived from the computerized collation, as I will show in a moment.


Here is the title page of the Yale edition.


And here you see the title page of the Book of Mormon itself. The breaks in the text occur at places where you could reasonably pause, especially if you were reading the text off to someone else, as dictation. You will note, by the way, that the statement about the book being translated by Joseph Smith is not on the title page. That’s because it wasn’t on the original title page of the Book of Mormon. Instead, I place that attribution on the preceding page, on what we call the half title. This is the appropriate place to acknowledge Joseph Smith as the translator.


Here is the beginning of First Nephi. You’ll notice that the text just says “the Book of Nephi”—that’s because the manuscripts tell us that that was the original title for the book. There are, it turns out, four books of Nephi in the Book of Mormon. Later Oliver Cowdery or other editors added first, second, third, and fourth. And so this is the way the text originally read, and we reproduce it that way.


This is the next page, where you can clearly see the sense-lines. There are also extra lines of space that represent the paragraphs that I have broken the text into. You will note that overall this is a clear text, a plain text. There is not much editorial intrusion beyond the sense-lines and the paragraph breaks. I did put the LDS chapter and verse numbers out in the left margin because readers need to reference where they are in the book.


And here I show another page, one where there’s a transition from one book to another, from Mosiah to Alma.


I now want to go over how this text was constructed. As I said, it doesn’t have a copytext. Here is the computerized collation for that portion of 1 Nephi that goes from chapter 3 to chapter 4. It’s an instructive example for several reasons. You will notice in this collation that at the beginning of each line, before each textual variant, there is something in curly brackets and that stands for the kind of variant that will follow. Then there is something in square brackets, and this gives the actual variant. The very first variant in this sample from the collation is a number. In fact, it’s a verse number. The numbers 0 and 1 stand for the two manuscripts; and they don’t have any verses, so the variant shows that there is no number in the original and printer’s manuscripts. And then you have A through T, which are the printed editions of the Book of Mormon (from the 1830 edition to the 1981 edition), and the majority of them have verse numbers. And you can go down through the collation in this way. The underlined text represents what I believe is the original text (but with the spelling we expect today in standard English). The text in bold stands for the reading of the original manuscript. In other portions of the collation you won’t see any bolding, and that’s because the original manuscript is not extant for those portions.


From the computerized collation I had a computer program extract the underlined text. This means that I did not key in the text for the Yale edition. So this is what you get for that portion of the collation covering the transition between 1 Nephi 3 and 4. This is much like what the 1830 typesetter was confronted with—just a long string of words with no sentence breaks except at the beginning of the original chapter breaks.


And so from that long string of words, I constructed the sense-lines and put extra lines in for where I thought there should be paragraph breaks—in those places where new paragraphs would help the reader.


The next stage was that I added the LDS chapter and verse numbers in the left margin. One of the things you’ll notice right here is that chapter 4 comes in the middle of a paragraph. And if you read this portion of the text, you’ll see that there shouldn’t be a chapter break there, not even a paragraph break.


And finally, after the chapter and verse numbers had been added, I put in the punctuation (as well as the capitalization), but from scratch, without looking at the current punctuation or the original 1830 punctuation. I found this to be the most difficult task in preparing the Yale edition. I’m frankly amazed of what John Gilbert, the 1830 typesetter, was able to do here, especially given the difficult syntactic structure of the Book of Mormon. About two thirds of the time, Gilbert determined the punctuation as he was setting the type. He did a really fine job, although he probably overpunctuated the text (but this is what was expected in those days). What I wanted to do here was to do the punctuation from scratch. As far as I know, the Yale edition is only the second time that the punctuation has been done completely from scratch—or at least in a published edition.


The Yale edition also has an appendix that shows the significant changes in the text. There are 719 of these at the end of the book. I should point out that this appendix is not simply a comparison between the current standard text and the Yale edition. Instead, it’s a representation of important changes that have occurred in the history of the text. In many instances, the Yale edition agrees with the current standard text. So the appendix is not intended to be a listing of differences, but of significant variants in the history of the text.


One of the things we get at the beginning of the appendix is what’s called a stemma, which shows the textual relationship between the editions of the Book of Mormon. You will notice here that there are two textual traditions that split off from the 1837 Kirtland edition, the second edition. One is the RLDS textual tradition—that’s on the left. And on the right is the LDS textual tradition. And these relationships in the stemma show the copytext for each printed edition since the 1830 edition. If you look at the 1920 LDS edition there, its copytext was the 1911 LDS edition. This means that the 1920 typesetter worked off a copy of the 1911 edition, one that was presumably marked up with changes. Thus the stemma here shows the copytext relationships between the various editions.


Now we turn to the actual list of the 719 significant textual changes. This is the beginning of the list, which you can’t read very well here. For each set of alternative readings, I indicate with a pointer the one I have accepted as the original reading. For each case, I always list the reading of the original manuscript, if it exists, plus the reading of the printer’s manuscript and the reading in the 1830 edition. I then list any other edition that deviates from its copytext—that is, any place where the editors or typesetters for that edition decided on some other reading. And so you can actually reconstruct the whole history in each case, providing you refer to the stemma that shows the copytext relationships for the editions. In the Yale edition, I wanted the list of changes to appear in the appendix, not intruding upon the text itself.


Here’s the second page of the list of significant changes.


The Yale edition, as already noted, derives from volume 4 of the critical text. In volume 4, there are 5,280 cases of variation that I considered. And it turns out that 2,241 of these differences show up in the Yale edition. This last count excludes most cases of grammatical variation in the text. Nor is this number particularly important because most of these changes aren’t earthshaking.

However, there are a couple of number counts that are important. One is that there are 606 changes in the Yale edition which have never appeared in any standard printed edition of the Book of Mormon, in neither the LDS or the RLDS textual traditions. And if you look at those readings that account for the number 606, you will see that the vast majority come from the manuscripts. In fact, over half of the new readings (304 of them) come from the original manuscript. And 187 come from the printer’s manuscript (these are cases where the original manuscript is not extant). There are also two new readings in the title page that come from other early copies of the title page. None of these 493 readings have ever been implemented in any of the standard printed editions. You may also note that there are 113 conjectures. I will come back to conjectures in a moment.


What I want to point out here is the significance of the original manuscript in restoring the original text. For the six books of volume 4 of the critical text, the six maroon books that we have over here, I went through the 491 new readings that come from the two manuscripts and I divided them up according to which book in volume 4 they are discussed. For three of the books (the first, fourth, and fifth), large portions of the original manuscript are extant (each of these books is marked with an asterisk). And for each of those books, look at the number of changes that show up. For the first and the fourth one, we have about 75 percent of the original manuscript, and we get almost 100 changes for both of these books. For the fifth book, we have about 25 percent of the original manuscript, and we get a proportional amount of changes. For the three other books of volume 4, we have hardly anything of the original manuscript, and it shows. What this really means is that by human means we aren’t going to be able to recover as much of the original text for these parts of the text. Since we don’t have the complete original manuscript, it makes a real difference. We need to keep that in mind when we refer to the original English-language text of the Book of Mormon.


A second count that is quite important when considering the Yale edition is that there are 241 readings that make a difference in meaning. Now by difference in meaning I mean that if we translate the text into another language there will be a change in the words—that is, there will be some word difference, no matter what the language. For each of these 241 readings, the change makes a difference in meaning, not just in phraseology.

A good example of this is found in 1 Nephi 12:18: “and a great and a terrible gulf divideth them / yea even the sword of the justice of the Eternal God”. This is the way the original manuscript reads, but Oliver Cowdery miscopied this as “yea even the word of the justice of the Eternal God”. In other words, he replaced sword with word. And that’s the reading that’s been retained in the text ever since.


Yet when we look at the rest of the Book of Mormon, we discover that there are seven references to “the sword of God’s justice”, but there are no examples of “the word of God’s justice”. In particular, note the example in Ether 8:23 (“the sword of the justice of the Eternal God”). It’s identical to the original reading in 1 Nephi 12:18. When we consider the translation of this change in words, I know of no language where sword and word are the same word. Every translation is going to end up making the change here. This is what I mean then by a change in meaning. Of course, this change doesn’t make a huge difference in meaning. You can deal with God’s justice being his word. But that isn’t what the text originally read. It read sword.


Some people have asked whether any textual change ever alters the doctrine—and the answer is, no. Whenever a change ever involves doctrine, we find that the change restores the correct doctrine. An example of this is found in Alma 39:13, where Alma is talking to his son Corianton and tells him to go back to the Zoramites and “acknowledge your faults and repair that wrong which ye have done”. When Oliver Cowdery finished writing this page in the original manuscript, he accidentally dropped some ink on the page. And the letter p in repair has a drop of ink right on top of the ascender for the p, which ends up making the p look like it’s crossed. In fact, the p ends up looking like a t. And Oliver’s r‘s and n‘s often look alike. So when Oliver came to copy this part of the text into the printer’s manuscript, he copied it as “acknowledge your faults and retain that wrong which ye have done”. That reading doesn’t quite work, and so the 1920 committee decided to just remove the word retain because it didn’t make any sense. Thus they ended up having Alma say to Corianton that he should go back and “acknowledge your faults and that wrong which ye have done”. In other words, “go back and say you’re sorry”. But the necessity of repairing his wrong had now been removed from this passage.


When we look at other parts of the Book of Mormon text, we indeed find that when people confess their sins, they do everything they can to repair the wrongs or the injuries they have done. Here’s one in Mosiah 27:35: “zealously striving to repair all the injuries which they had done to the church / confessing all their sins”.


And here’s one in Helaman 5:17: “they came forth and did confess their sins … and immediately returned to the Nephites to endeavor to repair unto them the wrongs which they had done”. So by putting back the word repair in Alma 39:13, the correct doctrine of repentance is restored. The doctrine hasn’t been changed.


You will also find 15 new readings for Book of Mormon names in the Yale edition. I think the most interesting one is that the actual name for the surviving son of king Zedekiah was Muloch, not Mulek, the implication being that Zedekiah named this son after the pagan god Moloch that they sacrificed children to, which suggests a rather ominous aspect to king Zedekiah’s character.


Earlier I pointed out that the Yale edition has 113 conjectural emendations that are new, and some people have been critical of this. But I think it’s worth noting that in every printed edition of the Book of Mormon there are readings that are the result of conjectural emendation. A conjecture is introduced into the text whenever a typesetter, a scribe, or an editor doesn’t like the particular reading of his copytext and doesn’t like any of the other readings that might have appeared in earlier editions or in the manuscripts, and so he decides on a new reading. That’s a conjecture. (Here I exclude emendations involving grammatical editing.)

What we find is that conjectures are quite common in the Book of Mormon, and in many instances they are necessary. Sometimes the original manuscript has such a bad reading that no one is going to accept it. Consider, for instance, 1 Nephi 7:5: “the Lord did soften the heart of Ishmael and also his hole hole“. That’s the way the original manuscript reads, hole hole. It’s even the corrected reading in the original manuscript, which means that that is what the scribe decided on. When Oliver Cowdery copied this into the printer’s manuscript, he just couldn’t accept what the original manuscript read. And so he decided that hole hole was household, writing “the Lord did soften the heart of Ishmael and also his household“. My belief is that the original text here actually read “the Lord did soften the heart of Ishmael and also his whole household“. That would explain why the original manuscript ended up having two instances of hole, one standing for whole, the other for the hold of household.


In support of this reading, in looking at the rest of the text of the Book of Mormon, when it refers to a patriarch and his household, the text always refers to his entire household. In corresponding contexts, we have either “all his household” or “his whole household” (the reading in Alma 22:23). But the point here is that you can’t accept the reading of the original manuscript, hole hole. There must be a conjecture here, either household or whole household (or some other possibility). For that phrase in 1 Nephi 7:5, every text of the Book of Mormon is going to have some kind of conjectured reading.


Now when we look at the current standard text, we find that there are 654 conjectured readings. On the other hand, there are 354 in the Yale edition. It turns out that the Yale edition accepts a lot of difficult readings that have otherwise been removed over time from the standard text. In volume 4 of the critical text, I considered overall 1,346 cases of conjectural emendation. About one fourth of them were accepted. It’s also worth noting that when we compare the Yale edition with the current standard text there are 187 conjectures that both texts agree on. So there is considerable agreement between the two texts in addition to the differences.


Now let’s look at some of the individuals who had the most influence in introducing conjectural emendations into the text. Oliver Cowdery, the main scribe for both manuscripts, made 131 conjectures, of which the Yale edition accepts about 30 percent of them. For instance, in 1 Nephi 7:1 the original manuscript reads: “that his sons should take daughters to wife that might raise up seed”. When Oliver copied this into the printer’s manuscript, he added the pronoun they, thus “that his sons should take daughters to wife that they might raise up seed”. I think that this emendation is correct, but you’ll have to read about this in volume 4 if you want to know the arguments for including the pronoun. You won’t find that discussion in the Yale edition, although the change is listed in the appendix. The arguments are all in volume 4.


Here is one of Oliver Cowdery’s conjectural emendations that I think he got wrong. In 1 Nephi 13:24 the original manuscript reads “it contained the fullness of the gospel of the Land“, which seems impossible. When Oliver copied this passage into the printer’s manuscript, he changed “the gospel of the Land” to “the gospel of the Lord“. He obviously couldn’t accept the word land here, and he undoubtedly thought Land looks like Lord. In actuality, the reading of the original text was probably “the gospel of the Lamb“. The original scribe apparently misheard lamb as land, but without the d at the end being pronounced, which he then wrote as Land in the original manuscript. At every other place in the Book of Mormon, in four places in 1 Nephi 13, it consistently reads “the gospel of the Lamb“, never “the gospel of the Lord“. Of course, “the gospel of the Lord” is possible, but that isn’t the way the Book of Mormon expresses it.


John Gilbert, the compositor for the 1830 edition (the one who set the type) made a total of 167 conjectures, of which a large percentage, 47 percent, are accepted in the Yale edition. The reason so many are accepted is that in many cases Gilbert was confronted with a manuscript reading that was unacceptable yet it was easy enough to figure out the correct reading. That’s why the percentage of acceptance is so high for him. Here’s an example from 1 Nephi 17:48 where I think he was right. His manuscript copy read “and whoso shall lay their hands upon me shall wither even as a dried weed“. Gilbert interpreted the word weed as an error for reed, and thus he set the text as “even as a dried reed“.


On the other hand, in Alma 5:35 Gilbert replaced the verb put with hewn, giving “and ye shall not be hewn down and cast into the fire” rather than “and ye shall not be put down and cast into the fire”, the reading of the printer’s manuscript. Normally the Book of Mormon text refers to people being hewn down and cast into the fire. Even so, the occurrence of put down was more likely an error for the visually similar cut down, so that the original text (and original manuscript) probably read “and ye shall not be cut down and cast into the fire”. The word cut was likely written with a capital C in the original manuscript, with the result that the scribe who copied the text into the printer’s manuscript misread the capital C as a capital P, thus introducing put as the verb.


Joseph Smith made a large number of conjectures in his editing for the second edition of the Book of Mormon (published in 1837). For the third edition (published in 1840), he made a few more conjectures. In most of these cases, Joseph was simply trying to remove difficult readings from the text. Many of these original, difficult readings are, nonetheless, acceptable. I suspect he often thought, “that reading is difficult for people to understand, so let’s change it to this”.


Here in Ether 3:9 is a change in the 1837 edition where Joseph correctly inserted the word not, changing “for were it so / ye could not have seen my finger” to “for were it not so, ye could not have seen my finger”. On the other hand, in Mosiah 21:28 Joseph Smith replaced king Benjamin with king Mosiah in order to deal with a perceived problem in chronology. I think, in this case, Joseph’s emendation was unnecessary. You can read the arguments in volume 4.


We can also consider the number of conjectures in the more significant editions since 1840 and identify how many of them are accepted in the Yale edition:

1849 Orson Pratt 8 2 25%
1852 Franklin and Samuel Richards 17 5 29%
1879 Orson Pratt 9 4 44%
1905-11 German Ellsworth 8 2 25%
1920 James E. Talmage 130 22 17%
1981 scriptures committee 10 4 40%

Note in particular the high number of conjectural emendations in the 1920 edition, largely the result of James E. Talmage’s decision to emend difficult readings in the text. The large majority of these emendations were unnecessary, although they made the text easier to read. Only about one conjectured reading out of six in the 1920 edition is accepted in the Yale edition.


A number of LDS scriptural scholars have independently made suggested emendations prior to the critical text project. In many respects, these are good suggestions, and a rather high percentage, 48 percent, have been accepted:

Paul Cheesman 1 1
Stan Larson 7 3
Hugh Nibley 1 1
W. Cleon Skousen 1 1
Sidney B. Sperry 3 3
John Tvedtnes 10 2


In my work as editor of the critical text project, I have proposed 401 conjectures, 103 of which I have accepted—or 26 percent (about one reading out of four).

Of particular help in this project have been people who have independently sent me suggestions for change or identified readings that seemed strange in some way. In all, 42 individuals have corresponded with me and have recommended 173 changes, of which 36—or 21 percent (about one out of five)—have been accepted. Thus the Yale edition reflects a tremendous amount of input from careful readers of the text. Here I list eight people who basically went through the whole text looking for readings that seemed problematic: David Calabro, Joanne Case, Lyle Fletcher, Ross Geddes, Heather Hardy, Paul Huntzinger, Brent Kerby, and Greg Wright. You will find their suggestions discussed in volume 4, and in many instances their suggested changes made it into the Yale text.


Finally, I want to talk about the evidence that the text of the Book of Mormon is a precise English-language text, and that when it was given through the instrument to Joseph Smith, he could see that text and he read it off to his scribe. B. H. Roberts thought that reading off the text was too easy, but of course B. H. Roberts himself never received a text from the Lord in that way. There’s a lot of evidence the translated text of the Book of Mormon was controlled down to the very word, in fact, to the very letter (at least for the Book of Mormon names).

The first type of evidence involves the occurrence of Hebrew-like constructions that are unacceptable in English and have consequently been edited out of the text. One example is the extra use of the conjunction and that follows a subordinate clause and comes right before the main clause. Recently I’ve discovered that these extra and’s do not occur if the subordinate clause is simple. They only occur when there is some complexity in the subordinate clause, either an extra clause or a phrase that interrupts the flow of the text. When that happens, the Hebrew-like and usually appears. Here are some examples involving various subordinate conjunctions:

1 Nephi 8:13
and as I cast my eyes around about
that perhaps I might discover my family also
and I beheld a river of water

Helaman 13:28
and because he speaketh flattering words unto you
and he saith that all is well
and then ye will not find no fault with him


3 Nephi 23:8
and when Nephi had brought forth the records
and laid them before him
and he cast his eyes upon them and saith…

Mormon 3:4
and it came to pass that after this tenth year had passed away
making in the whole three hundred and sixty years
from the coming of Christ
and the king of the Lamanites sent an epistle unto me


Moroni 10:4
and if ye shall ask with a sincere heart with real intent
having faith in Christ
and he will manifest the truth of it unto you
by the power of the Holy Ghost

The last example is one of the most famous passages in the Book of Mormon.

All of these extra and’s have been removed from the Book of Mormon text—and that’s because they are such bad English. But what does the occurrence of these extra and‘s say about the translation? It says that Joseph Smith had to have seen the and‘s. If he had just been getting ideas in his mind, there would have been no reason for him to have put these and‘s in—they’re non-English, and they haven’t occurred in any known dialect of English or in the history of the language. And their occurrence isn’t just an accident since there are so many examples of the extra and. Moreover, there is this wonderful passage from Helaman 12:13-21 where seven of them occur, virtually one after another, in the original text. These extra Hebrew-like and‘s were really there. Joseph could see them, and so he read them off.


Another type of evidence, one even more surprising (and controversial to some) is that the vocabulary in the Book of Mormon, the meanings of the words themselves, come from the 1500s and 1600s. To be sure, there are examples of archaic Book of Mormon word usage that can also be found in the 1611 King James Bible, such as the phrase “to cast an arrow”, which means ‛to shoot an arrow’. This is found in Alma 49:4: “the Lamanites could not cast their stones and their arrows at them”. But we also find that expression in the King James Bible, in Proverbs 26:18: “as a mad man who casteth firebrands arrows and death”.


Another example that also occurs in the King James Bible is the verb require with the meaning ‛to request’. In Enos 1:18 the Book of Mormon text reads “thy fathers have also required of me this thing”—in other words, Enos’s fathers requested this thing of the Lord. Similarly, in the King James Bible, in Ezra 8:22, Ezra refrains from requesting troops from the Persian king: “for I was ashamed to require of the king a band of soldiers and horsemen to help us against the enemy in the way”.


A third example is the use of the verb wrap with the meaning ‛to roll’, in 3 Nephi 26:3: “and the earth should be wrapped together as a scroll”. Compare this with the usage in 2 Kings 2:8: “and Elijah took his mantle and wrapped it together and smote the waters”.

All of these meanings can be found in the Oxford English Dictionary (referred to as the OED). For each of the three verbs already mentioned, their archaic meanings were typical of 1600s language and can therefore be found in the King James Bible. And one could argue, then, that they are in the Book of Mormon simply because Joseph Smith knew his Bible that well.


The problem with this proposal is that there is archaic 1500s and 1600s usage in the Book of Mormon text that is not found in the King James Bible. Consider the original occurrence of the conjunctive but if in Mosiah 3:19: “for the natural man is an enemy to God … and will be forever and ever but if he yieldeth to the enticings of the Holy Spirit”. Here but if means ‛unless’ and that meaning occurred in Early Modern English (for this meaning the OED gives citations dating from 1200 to 1596). We have this 1580 example from Sir Philip Sidney: “He did not like that maids should once stir out of their fathers’ houses but if it were to milk a cow.” The editors for the 1920 edition decided to emend the Book of Mormon reading in Mosiah 3:19, replacing but if with unless, which is semantically correct and makes the text understandable for modern readers.


Another Book of Mormon example uses the verb commend in a sentence with the meaning ‛to recommend': “and now I would commend you to seek this Jesus” (Ether 12:41), which in today’s English would read “and now I would recommend you to seek this Jesus”. The OED gives a date in the 1600s for this usage, which has now died out.


Another very interesting example of archaic usage in the original text is the phrase “to counsel someone” with the meaning ‛to counsel with someone’. There are two examples of this usage in the original text, for which the 1920 committee added the preposition with (which is correct as far as the meaning goes). But when we go back to Early Modern English, we get uses of the phrase “to counsel someone” with the meaning ‛to counsel with someone’, as in this 1547 example from John Hooper: “Moses … counseled the Lord and thereupon advised his subjects what was to be done.”


Another interesting example in the Book of Mormon is the use of the verb depart in the original text with the meaning ‛to part, divide, or separate’. This meaning for depart was regularly used in English Bibles up to the 1611 King James Bible. But by then that meaning for depart had become archaic, so the King James translators systematically eliminated that use of depart from their translation, so there are no examples in the King James Bible of what had regularly occurred in earlier English translations. Yet the Book of Mormon has this particular use of depart in Helaman 8:11 in the printer’s manuscript: “to smite upon the waters of the Red Sea and they departed hither and thither”. The 1830 typesetter just couldn’t believe that departed was correct, so he replaced the word with parted (thus he set “to smite upon the waters of the Red Sea and they parted hither and thither”). We have examples from the 1557 New Testament of the Geneva Bible like “they departed my raiment among them” (John 19:24), translated in the 1611 King James Bible as “they parted my raiment among them”.


In the following example from Helaman 9:17, language usage from the 1500s and 1600s leads us to consider assigning the meaning of ‛to expose’ to the verb detect: “and now behold we will detect this man and he shall confess his fault”. Such usage can be found, for instance, in this example from Richard Hooker in 1594: “The gentlewoman goeth forward and detecteth herself of a crime.”


The adjective extinct now refers to the death of a species, but in Early Modern English it could refer to the death of a person, as we find in a 1675 English translation of Machiavelli’s The Prince: “the Pope being dead and Valentine extinct“. And we find such usage in the original (and current) text of the Book of Mormon: “and inflict the wounds of death in your bodies that ye may become extinct” (Alma 44:7).


Here is an interesting example from Helaman 7:16 where the text uses the verb hurl but it more likely refers to dragging rather than throwing: “yea how could ye have given away to the enticing of him who art seeking to hurl away your souls down to everlasting misery and endless woe”. And the OED provides a 1663 citation from Robert Blair where hurl is assigned the meaning ‛to drag or pull with violence': “The new creature was assaulted, hurled, and holed as a captive.” And this is what we expect in Helaman 7:16, that Satan will drag us down to hell.


The expression “to pitch battle” no longer exists as such in modern English; today we have it only in the set phrase “a pitched battle”. In fact, we think of a pitched battle as a full-fledged one, but originally what it referred to was a fully set battle. Interestingly, the Book of Mormon uses the original, now archaic, syntactic expression in Helaman 1:15: “and they came down again that they might pitch battle against the Nephites”. In Early Modern English there is Christopher Marlowe’s 1590 example in the passive, “Our battle, then, in martial manner pitched.” But such general use of the verb phrase “to pitch battle” no longer exists in English.


Finally, I give an example of an unexpected extension of the noun rebellion in Mosiah 10:6: “and he began to stir his people up in rebellion against my people”. In today’s English, we think of the word rebellion as hierarchical, that rebellion occurs in opposition to higher authority. But this example from the Book of Mormon refers to the Lamanite king as stirring up his people, the Lamanites, against the people of Limhi, a Nephite people that are in virtual slavery to the Lamanites. The meaning of the phrase “in rebellion” in Mosiah 10:6 seems to simply mean ‛in opposition'; there the phrase “in rebellion” lacks any kind of hierarchical implication. Thus far I have found only one example (and an early one at that) with this more general meaning for the noun rebellion (used by Gilbert Haye in 1456, as cited in the OED): “if man should have this rebellion and contrariety, any against another, when they are of diverse complexions?”


The third type of evidence that shows the preciseness of the original text of the Book of Mormon deals with various set expressions and word forms that were consistently used in the original text but no longer are in the standard text. In over a hundred different expressions and word forms, the text has developed various exceptions to its original consistency, exceptions that we might call “wrinkles in the text”. The original translated text is so consistent in this respect that it doesn’t look like it’s the result of a translator freely choosing how he should translate a given expression or word form each time he comes across it. Here are some examples; I first give the usage in the current standard text, then in the original text:

whatsoever, never whatever 72 to 2 in the current text; 75 to 0 in the original text
conditions, never condition 12 to 2 in the current text; 14 to 0 in the original text
this time, never these times 60 to 1 in the current text; 61 to 0 in the original text


“observe to keep the commandments”, never “observe the commandments”

10 to 1 in the current text; 11 to 0 in the original text
“thus ended a period of time”, never “thus endeth a period of time” 43 to 4 in the current text; 47 to 0 in the original text
“to do iniquity“, never “to do iniquities 21 to 1 in the current text; 22 to 0 in the original text


“if it so be that …”, never “if it be so that …”

36 to 2 in the current text; 38 to 0 in the original text
“to have hope“, never “to have hoped 17 to 1 in the current text; 18 to 0 in the original text
“the Nephites and the Lamanites”, never “the Nephites and Lamanites” 14 to 1 in the current text; 15 to 0 in the original text

Thus the original text appears to be a fully controlled text. The Yale edition restores more than one hundred of these kinds of systematic phrases and word choices.


Another type of evidence for the systematic nature of the original text can be found in identical citations that come from completely different parts of the text. One well-known pair of citations involves a reference to Lehi’s vision of the heavenly scene. It is first quoted in 1 Nephi 1:8 as “and he thought he saw God sitting upon his throne surrounded with numberless concourses of angels in the attitude of singing and praising their God“, then the same precise language, word for word, is used considerably later, in Alma 36:22: “yea and methought I saw—even as our father Lehi saw—God sitting upon his throne surrounded with numberless concourses of angels in the attitude of singing and praising their God“.


Another example of identical citation involves a liturgical expression found in Mosiah 3:8, which originally read as follows: “and he shall be called Jesus Christ the Son of God / the Father of heaven and of earth / the Creator of all things from the beginning“. Later, in Helaman 14:12, we get the same language, word for word: “and also that ye might know of the coming of Jesus Christ the Son of God / the Father of heaven and of earth / the Creator of all things from the beginning“. Interestingly, in Mosiah 3:8, the 1830 typesetter accidentally deleted the preposition of before the noun earth, giving “the Father of heaven and earth” rather than the correct “the Father of heaven and of earth”. So in the current text these two liturgical citations are no longer identical.


There is also evidence for letter for letter control over the spelling of Book of Mormon names. The witnesses of the translation process indicated that whenever the scribe had difficulty in spelling a name correctly, Joseph Smith would spell it out for him. And we can find clear evidence of this spelling out of names in the original manuscript. For instance, in Alma 33:15, when Oliver Cowdery had to spell the name Zenoch for the first time in the original manuscript, he initially spelled it as Zenock. Then he immediately crossed out the misspelling, Zenock, and wrote inline the correct spelling, Zenoch. Later on, in Helaman 1:15, Oliver originally spelled the first occurrence of the name Coriantumr as Coriantummer, a phonetic spelling. Again, he crossed out the misspelling and then wrote inline the correct Coriantumr. In this instance, Joseph would have been required to spell out the name letter for letter in order to get the impossible sequence mr at the end of the name.


In conclusion, there are three goals that have guided me in producing The Book of Mormon: The Earliest Text. First of all, I wanted to present the text in an inviting format and provide a clear text with minimal editorial intrusion, one that would be fully accessible and easy to read—and for both LDS and non-LDS readers. Second, I wanted to publish the most accurate text possible, one with readings based on the two manuscripts and the earliest editions. And finally, I wanted to provide access to all the significant textual changes that the text has undergone over the years, from the manuscripts and early editions up to the current LDS and RLDS editions. But since I did not want these variants to intrude upon the text itself, I placed them in an appendix.

As I have studied the original text and the evidence of how it was transmitted, it’s become very clear to me that this text was revealed to Joseph Smith word for word and that he could actually see the spelled-out English words. We do not know the mechanism that allowed Joseph to receive the text in this precise way, but the evidence argues that the text was a specific English-language translation that was revealed through him. It is indeed a marvelous work and a wonder.

Thank you.

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>