How Was the Book of Mormon Produced?

Summary

The Book of Mormon was dictated aloud, in about three months, mostly with Joseph Smith's face in a hat. That last detail was contested for a century and a half. It is not contested now. The Church of Jesus Christ of Latter-day Saints published it in its Gospel Topics essay "Book of Mormon Translation."¹ On 4 August 2015 the Church newsroom announced that the Ensign article "Joseph the Seer" and the introduction to Revelations and Translations, Volume 3 would carry "never-before-seen photographs of a seer stone Joseph Smith likely used in the translation of the Book of Mormon."² Critics who present the hat as a suppressed secret are describing the situation of 1980, not of today.

Four separate questions get blurred together and are kept apart here.

How was the text physically produced? Very well established. Joseph Smith dictated continuously to a scribe. He used either the two stones in silver bows found with the plates, called the "interpreters" in the Book of Mormon and later called the "Urim and Thummim," or a single seer stone he had owned since before he had the plates.¹ Witnesses who disagreed about almost everything else agree on the hat. The plates themselves were usually not open in front of him, and were often not even in the room.

What does the 1826 legal record show? Much less than either side usually claims. Something happened before Justice Albert Neely in Bainbridge, New York, in March 1826, and it concerned Joseph Smith's stone-gazing for Josiah Stowell. Beyond that the documents are thin, late, and disputed. Calling it a "conviction" outruns the evidence. So does calling it nothing.

How fast was the dictation, and is the pace by itself an argument? The pace is real and it is impressive. About 268,000 words came out between early April and the end of June 1829.³¹ That is roughly 3,700 to 4,100 words per working day, sustained for three months, with no known drafts. It is not, however, without parallel in the documented history of oral composition and dictation. The pace is a fact that any explanation must fit. It is not by itself a proof of anything.

Is the small-plates episode a providential design or a convenient recovery? Genuinely ambiguous, and this article does not pretend otherwise. After the 116-page manuscript was lost in June 1828, Joseph Smith did not re-dictate it. He said God had foreseen the loss more than two thousand years earlier and had a replacement already engraved. The replacement covers the same period from a different angle. Read one way that is a striking piece of internal design. Read another way it is the only move a dictating author could safely make. Both readings survive the evidence.

The article's real question is the last one. A 268,000-word narrative with hundreds of named characters, a tracked internal chronology, and a consistent theological argument came out of the mouth of a young farmer with little formal schooling, in about three months, with no surviving notes. "He was uneducated" is not an answer to that, and this article does not give it. The serious naturalistic accounts — William Davis's work on early American extemporaneous sermon culture, long prior mental composition, and Joseph Smith's actual literacy and reading — are presented at their strongest. So is the Latter-day Saint answer, including the scribes' insistence that he never re-read what had already been written and simply resumed where he had left off.

The claim being tested

The Latter-day Saint claim is not a single proposition. At minimum it contains these:

1. An ancient record existed on metal plates, and Joseph Smith physically possessed it.
2. The English text of the Book of Mormon is a translation of that record, produced by divine power rather than by ordinary composition.
3. The process worked through physical instruments — the interpreters, and later a seer stone — which displayed English text.
4. Joseph Smith could not have produced the text himself.

These are separable. Evidence about the process (proposition 3) is not automatically evidence about the source (proposition 2). Showing that a naturalistic account is possible (against proposition 4) is not the same as showing that it is true. And the presence of the plates in the house is not evidence that they were used, since on nearly every eyewitness account they were not being read from.

This article deals with production. Whether the resulting text reads like an ancient record is a separate investigation. What the formal witnesses saw is another. Whether an earlier manuscript by Solomon Spaulding lies behind it is a third, and the answer there is that the transmission chain is not established.

The strongest case for the Latter-day Saint claim

A believing Latter-day Saint should recognize this section. It is the case at full strength.

The scribes were in a position to know, and they never backed down. Emma Smith, Martin Harris, Oliver Cowdery, and David Whitmer all watched the dictation from a few feet away. Three of the four later left or were pushed out of the church Joseph Smith founded. David Whitmer spent nearly fifty years outside it and wrote a book attacking Brigham Young's succession. Oliver Cowdery was excommunicated in 1838. Martin Harris drifted through several movements. None of them ever said the text came from a manuscript. Emma Smith was interviewed by her son in February 1879, weeks before her death that April, and the interview was published in the Saints' Herald on 1 October 1879. Asked "Had he not a book or manuscript from which he read, or dictated to you?" she answered: "He had neither manuscript nor book to read from." The Church's own essay quotes the same exchange.¹ Asked whether he could have had one without her knowing, she said, "If he had had anything of the kind he could not have concealed it from me."⁴

The continuity detail is hard to fake and hard to explain. Emma Smith added something that a coached witness would be unlikely to invent, because it is specific and mundane. In the same 1879 interview she said that "when returning after meals, or after interruptions, he would at once begin where he had left off, without either seeing the manuscript or having any portion of it read to him. This was a usual thing for him to do."⁴ A person reciting from memorized notes would need to find his place. A person reading from a hidden manuscript would need the manuscript.

The manuscript evidence supports dictation, not copying. The surviving original manuscript contains errors of the kind a scribe makes when hearing words, not the kind a scribe makes when copying them by eye. The Gospel Topics essay makes exactly this argument, and it is the correct kind of argument: it is a claim about physical evidence that could in principle have come out the other way.¹ Royal Skousen, who has run the Book of Mormon Critical Text Project since 1988, built much of the case.⁵

There was no punctuation. The original manuscript has essentially none. The Church's own essay states that Joseph Smith did not call for periods, commas, or question marks, and that the typesetter supplied them later.¹ A person reading aloud from a prepared document would read the punctuation, or at least pause at it in ways a scribe would record. A person hearing text for the first time and speaking it in units would not.

The pace is genuinely extraordinary. Whatever explains it, it needs explaining. Three months of sustained oral output at roughly 3,600 to 4,100 words per working day, while also being baptized, moving households twice, negotiating with printers, obtaining a copyright, and receiving visitors, is not an ordinary human performance.

Joseph Smith did not take the easy way out over the lost pages. In June 1828 the first 116 pages disappeared. The easiest recovery for a fraud would have been to re-dictate them, since nobody could produce the original for comparison. He explicitly refused. The revelation now printed as Doctrine and Covenants 10 forbids it: "you shall not translate again those words which have gone forth out of your hands." He then produced a substitute covering the same period from different records, and the Book of Mormon text itself contains an internal explanation for why such a substitute existed.

The plates

Almost everything known about the physical object comes from people who described it decades later, and most of them described it through cloth.

Joseph Smith's own published description is the fullest. He wrote it for John Wentworth, editor of the Chicago Democrat, and it was printed in Times and Seasons on 1 March 1842. The records "were engraven on plates which had the appearance of gold, each plate was six inches wide and eight inches long and not quite so thick as common tin. They were filled with engravings, in Egyptian characters and bound together in a volume, as the leaves of a book with three rings running through the whole. The volume was something near six inches in thickness, a part of which was sealed."⁶

That description creates an arithmetic problem worth stating plainly. Six inches by eight inches by six inches is 288 cubic inches. Gold has a density of about 19.3 grams per cubic centimetre, which is about 0.698 pounds per cubic inch. A solid block of gold that size would weigh about 201 pounds.⁷ Nobody in the record carried anything like that. Martin Harris, interviewed by Joel Tiffany in 1859, said he had "hefted the plates many times, and should think they weighed forty or fifty pounds."⁸ Emma Smith said she moved them around on the table while doing her housework.

Two adjustments close the gap, and both are used by Latter-day Saint writers. Stacked leaves bound with rings are not solid; a plausible packing fraction is somewhere near half. And "the appearance of gold" is not gold: a gold-copper alloy of the kind used in pre-Columbian America, often called tumbaga, has a density well below pure gold's. Run those together — 288 cubic inches, half filled, at about 11 grams per cubic centimetre, or 0.397 pounds per cubic inch — and the weight comes to about 57 pounds.⁷ That is heavy but liftable, and it is in the neighbourhood of Harris's forty to fifty. The calculation is honest arithmetic, but note what it is: a reconstruction chosen to fit the testimony, not an independent measurement. No one weighed the object.

The stone box is described in Joseph Smith's canonized 1838 history. He found the plates under a large rounded stone on the west side of a hill near Manchester, New York, and describes prying it up with a lever: "The box in which they lay was formed by laying stones together in some kind of cement. In the bottom of the box were laid two stones crossways of the box, and on these stones lay the plates and the other things with them" (Joseph Smith—History 1:52). He dates their delivery to 22 September 1827 (Joseph Smith—History 1:59). Joseph Knight Sr., writing sometime between 1833 and his death in 1847, remembered Joseph describing "a plane Box" that he uncovered and opened.⁹

Who lifted it? Emma Smith moved the covered plates on the table. Martin Harris gave dimensions from handling: the plates "were seven inches wide by eight inches in length, and were of the thickness of plates of tin; and when piled one above the other, they were altogether about four inches thick; and they were put together on the back by three silver rings, so that they would open like a book."⁸ Note the mismatch with the Wentworth letter: Harris made the stack about four inches thick and seven inches wide, Joseph Smith wrote six inches thick and six inches wide. Neither man was measuring, and Harris was speaking thirty-two years after the fact.

The central point about the plates for this article is negative and is not disputed by anyone. During most of the dictation the plates were not being consulted. Emma Smith said they "often lay on the table without any attempt at concealment, wrapped in a small linen table cloth, which I had given him to fold them in," that she "once felt of the plates, as they thus lay on the table, tracing their outline and shape," and that she "moved them from place to place on the table, as it was necessary in doing my work."⁴ Martin Harris said they were "usually kept in a cherry box made for that purpose."⁸ Isaac Hale, who was not allowed to look inside, "was allowed to feel the weight of the box."¹⁰ David Whitmer said the words appeared on the stone. Joseph Knight said the words appeared in the hat. Whatever the plates were, they were not functioning as a text being read.

The interpreters and the "Urim and Thummim"

The terminology in this story shifted, and the shift matters, because the later term absorbed the earlier one and hid a distinction.

The earliest layer uses "interpreters" and "spectacles." The Book of Mormon itself calls them interpreters (Mosiah 28:14–20). Joseph Smith's 1838 history describes "two stones in silver bows—and these stones, fastened to a breastplate, constituted what is called the Urim and Thummim" (Joseph Smith—History 1:35). Oliver Cowdery, testifying under oath in 1831, described "two transparent stones, resembling glass, set in silver bows," and said that by looking through these Joseph "was able to read in English, the reformed Egyptian characters." A resident of Union Village, Ohio, reported after Cowdery's visit in the fall of 1830 that the translation came by "two transparent stones in the form of spectacles thro which the translator looked on the engraving."¹ Joseph Smith's own 1842 account calls it "a curious instrument which the ancients called 'Urim and Thummim,' which consisted of two transparent stones set in the rim of a bow fastened to a breastplate."⁶ Martin Harris, who saw the instrument, gave measurements: the two stones were "about two inches in diameter, perfectly round, and about five-eighths of an inch thick at the centre," joined by "a round bar of silver, about three-eighths of an inch in diameter, and about four inches long," and were "white, like polished marble, with a few gray streaks."⁸ The Ensign draws the obvious conclusion: "As described, this seeric device would have been bulky."¹¹

"Urim and Thummim" is a biblical term for a priestly divination device. It does not appear in the Book of Mormon. On the Church's own account it entered Latter-day Saint usage for these instruments around 1833, and once it arrived it spread to cover everything. The 2015 Ensign article states the point without hedging: "By 1833, Joseph Smith and his associates began using the biblical term 'Urim and Thummim' to refer to any stones used to receive divine revelations, including both the Nephite interpreters and the single seer stone. This imprecise terminology has complicated attempts to reconstruct the exact method by which Joseph Smith translated the Book of Mormon."¹¹

That is a significant admission, and it is the Church's, not a critic's. It means that a nineteenth-century source saying "Urim and Thummim" cannot be assumed to mean the interpreters. Joseph Knight's account is the clearest case: he wrote that Joseph "put the urim and thummim into his hat,"⁹ which is not something one does with a bulky pair of stones mounted on a breastplate, and which the Church now reads as referring to the seer stone.¹¹

The change also means that a Latter-day Saint who grew up hearing that Joseph translated with the Urim and Thummim was not being lied to. He was hearing a term that had, by 1833, stopped distinguishing the two objects. The distinction was not so much concealed as dissolved.

The seer stone in the hat

This is now official. The Gospel Topics essay states it directly: "According to these accounts, Joseph placed either the interpreters or the seer stone in a hat, pressed his face into the hat to block out extraneous light, and read aloud the English words that appeared on the instrument."¹ The same essay says that the seer stone was one Joseph Smith "discovered in the ground years before he retrieved the gold plates," and that "as a young man during the 1820s, Joseph Smith, like others in his day, used a seer stone to look for lost objects and buried treasure."¹¹

On 4 August 2015 the Church went further and published photographs. The newsroom announcement for Revelations and Translations, Volume 3 said that the volume introduction and the accompanying Ensign article "both include never-before-seen photographs of a seer stone Joseph Smith likely used in the translation of the Book of Mormon."² It described the object and traced its custody: "The stone he used in the translation was often referred to as a chocolate-colored stone with an oval shape. The stone was passed from Joseph Smith to scribe Oliver Cowdery and then from Cowdery's widow, Elizabeth Whitmer Cowdery, to Phineas Young. Young then passed it on to his brother, Brigham Young, the second president of the Church. After President Young died, one of his wives, Zina D. H. Young, donated it to the Church."²

The eyewitness record behind that concession is worth assembling with its dates and its distances, because not all of it is equally good.

Three observations follow from that table, and two of them cut against the usual critical presentation.

First, the timing is mixed, and the article's earlier draft got this wrong by treating the whole record as late. The clearest first-person hat descriptions from inside the household — Emma Smith, David Whitmer, Martin Harris — do come from the late 1850s through the 1880s, from witnesses in their seventies and eighties, often through interviewers with agendas. Emma Smith was interviewed by her son, who was president of a church that competed with Brigham Young's. David Whitmer was defending his own claim to authority. But Isaac Hale's statement is dated and signed in March 1834, five years after the dictation, and Joseph Knight's holograph cannot be later than 1847. The hat is not only a late memory.

Second, the convergence is what carries the weight. The witnesses were estranged from one another and from the institution. Isaac Hale thought the whole thing "a silly fabrication of falsehood and wickedness."¹⁰ Joseph Knight believed it completely. They were interviewed separately, in different decades, by people with incompatible goals. They converge on a specific, unflattering, physically odd detail. Convergence under those conditions is the strongest kind of testimony a historian gets from imperfect sources.

Third, they do not all describe the same thing happening on the stone. David Whitmer said one character appeared with its English interpretation beneath it, and the character would not disappear until the English had been written and read back correctly. Joseph Knight said a whole sentence appeared, and would not go away until it was spelled right. Martin Harris, on Edward Stevenson's 1881 account, described sentences appearing and said "Written" when he had finished writing. Those are different mechanics — character-by-character versus sentence-by-sentence — and they are not reconcilable by any obvious harmonization. Skousen's manuscript evidence, which puts at least twenty words in view at a time, fits Harris and Knight and does not fit Whitmer.⁵ What survives the disagreement is the hat, the stone, the darkness, and the dictation. What does not survive is a confident account of what appeared.

The Church's essay is careful here, and correctly so. It says "some accounts indicate that Joseph studied the characters on the plates. Most of the accounts speak of Joseph's use of the Urim and Thummim (either the interpreters or the seer stone), and many accounts refer to his use of a single stone."¹ That is an accurate summary of a divided record.

Treasure-seeking and the 1826 legal record

The Book of Mormon claim and the money-digging record are connected by one object. The stone Joseph Smith used to translate is, on the Church's own account, the stone he had used to look for buried treasure.¹¹ That is not a hostile inference. It is the Gospel Topics essay's sentence.

What the record supports beyond that is narrower than critics usually assert, and the best guide to it is now the Church's own documentary edition. The Joseph Smith Papers publish the case as State of New York v. JS–A, with an editorial introduction, the two surviving official documents, and two appendix items.¹³ The account below follows that edition, because it is both the most recent professional treatment and the one produced by editors who have every institutional reason to soften the episode and did not.

The employment is documented and undisputed. Josiah Stowell, a farmer of Bainbridge, Chenango County, New York, hired Joseph Smith in late 1825 to help search for a reputed Spanish silver mine on the Susquehanna. The Joseph Smith Papers editors state that Stowell sought him out "having heard, that he was in possession of certain means"—seer stones—"by which he could discern things, that could not be seen by the natural eye," and that Joseph Smith "had developed a reputation as someone who could locate valuable objects buried in the earth using a seer stone."¹³ Isaac Hale, Emma's father, hostile and writing in 1834, says the same from the other side: the diggers "boarded at my house," and Joseph Smith's "occupation was that of seeing, or pretending to see by means of a stone placed in his hat, and his hat closed over his face."¹⁰

A legal proceeding took place on 20 March 1826. Joseph Smith appeared before Albert Neely, a justice of the peace, on a charge under New York's disorderly persons statute. That statute listed "all persons pretending . . . to discover where lost goods may be found" among ten categories of offenders, and the word "pretending" did the work: the law assumed that seer-stone use was deceptive whether or not the user believed in it.¹³

What survives, and what does not. Neither Neely's docket book nor the case papers survive. Two official documents do, both fee bills, both held by the Chenango County Historical Society in Norwich, New York.¹³ Neely's bill dates the proceeding, calls Joseph Smith "the Glass Looker," describes the crime only as an unspecified misdemeanour, and totals $2.68 without itemizing. Constable Philip De Zeng's bill is itemized: serving a warrant, subpoenaing twelve witnesses, "Attendance with Prisoner two days & 1 Night," "Notifying two Justices," and travel. Those entries partly confirm that the statutory procedure was followed.¹³

The printed docket entry, and why it cannot be trusted on its own. Three publications printed what purports to be Neely's docket entry: Charles Marshall in Fraser's Magazine after copying papers in Utah in 1872; Bishop Daniel S. Tuttle in Philip Schaff's Religious Encyclopaedia in 1883; and the anti-Mormon Utah Christian Advocate in 1886. The chain runs through Emily Pearsall, Neely's niece and a Methodist missionary in Utah, who "tore the leaves" out of the record and who died in 1872. The Joseph Smith Papers editors judge that "each printing was apparently made independent of the others," that the transcript's details match the two surviving 1826 bills well enough to suggest "that the published transcript was based on an authentic source," and that "without the original source, it remains unknown how accurately any of the published versions represents Neely's original docket entry."¹³ That is the correct weight to give it: probably derived from something real, not testable.

William D. Purple's 1877 recollection. Purple, a Bainbridge physician, published an account on 28 April 1877 saying he had attended and taken notes at Neely's request. The Joseph Smith Papers editors note that the article is presented as his "historical reminiscences," not his 1826 notes, that he interleaves later commentary, and that it reads like the written form of his repeated "public and private rehearsals" of the story. It agrees with the printed docket entry in places and disagrees in others.¹³

What terminology does that support? Not "conviction." No surviving 1826 document records a verdict, Neely's fee bill contains no entry for making the record of conviction that the 1825 justice's manual required, and there is no evidence Joseph Smith was ever committed to the bridewell. The Joseph Smith Papers editors conclude that this "strongly suggests that Neely did not convict JS," while adding that "the lack of verifiable contemporary records renders tentative any conclusion about the case's outcome."¹³

The earlier draft of this article recommended calling the proceeding an examination rather than a trial. That is now retired as too weak. The Joseph Smith Papers editors, who read the fee bills against the New York statute and the 1825 justice-of-the-peace manual, write that Joseph Smith "was most likely arrested and tried as a disorderly person," and De Zeng's bill for a warrant, twelve subpoenaed witnesses, and two days' custody describes something more than a preliminary look at the evidence.¹³¹³ Trial is the right word. Conviction is not.

The sources also disagree about the outcome, and that disagreement should be shown rather than smoothed over. Two early accounts by people who were not present say the court "condemned" Joseph Smith but that he was allowed to escape. Oliver Cowdery, also not present, reported in 1835 that the court "honorably acquitted" him. The printed docket entry has Neely finding "the defendant guilty." Purple has the court discharging him for want of evidence.¹³ Four accounts, four outcomes, none of them from the room and from 1826.

What follows for the Book of Mormon? Less than critics hope and more than defenders like. It does not show that the Book of Mormon is fraudulent; a person can be a village scryer and also receive a revelation, and the Church's own essay says as much. What it does show is continuity of method — and Isaac Hale, who watched both, drew the connection himself in 1834: the way Joseph Smith read the plates "was the same as when he looked for the money-diggers, with the stone in his hat, and his hat over his face."¹⁰ The same young man, using the same kind of object, in the same posture, moved from finding lost property to producing scripture. That continuity is a fact the Latter-day Saint account has to absorb, and since 2015 it has largely absorbed it by reframing seer stones as legitimate and biblical rather than by denying the history.¹¹

The scribes

The Joseph Smith Papers source note to the original manuscript names the scribes precisely, and the distinction it draws is worth keeping. The handwriting actually present on the surviving leaves is Oliver Cowdery's, John Whitmer's, one unidentified scribe's, and Joseph Smith's own. Reported as scribes, but with no handwriting known to survive, are Emma Hale Smith, Martin Harris, Samuel Harrison Smith, Christian Whitmer, and Reuben Hale.¹⁴ That is eight named people plus one unidentified. Emma Smith's own 1879 list was shorter — "Myself, Oliver Cowdery, Martin Harris, and my brother, Reuben Hale" — which is a reminder that she was recalling the Harmony period she shared.⁴

Martin Harris (spring 1828) was a prosperous Palmyra farmer, the first scribe of the lost 116 pages, and later the financier of the printing. He was not a disinterested party: he mortgaged his farm against the printing bill. Isaac Hale, watching from next door, described Harris writing while Joseph Smith interpreted, and remembered "that Harris wrote down one hundred and sixteen pages, and lost them."¹⁰

Emma Hale Smith wrote during the Harmony period, in 1828 and again in 1829. Her 1879 testimony is the single most valuable scribal account precisely because she had every reason, by 1879, to distance herself from Brigham Young's church and no reason to protect Joseph Smith's reputation on this specific point.

Reuben Hale, Emma's brother, wrote for a short period in Harmony. Joseph Knight remembered that "his wifes Brother would sometimes write a little for him through the winter."⁹ His participation matters because the Hale family was hostile: Isaac Hale gave a signed statement in 1834 calling the whole enterprise "a silly fabrication of falsehood and wickedness."¹⁰ Reuben's involvement shows the work was not hidden from that household.

Oliver Cowdery arrived on 5 April 1829, having heard about the plates while boarding with Joseph Smith's parents in Palmyra, and began writing two days later, on 7 April (Joseph Smith—History 1:66–67). He wrote the overwhelming majority of the surviving text and later copied nearly all of it again to make the printer's manuscript. His own published description — "These were days never to be forgotten—to sit under the sound of a voice dictated by the inspiration of heaven, awakened the utmost gratitude of this bosom! Day after day I continued, uninterrupted, to write from his mouth" — is the closest thing to a contemporary scribal statement in the record. It is printed in the canonized history (Joseph Smith—History 1, note following verse 75).

John Whitmer and Christian Whitmer wrote during the June 1829 stretch at the Whitmer farm in Fayette, New York. Their involvement is what moves the story from a two-man operation in Harmony to a household of observers in Fayette, and it is why the Fayette period produces so many later witness accounts.

The scribal roster matters for one reason above all. If the text were being read from a prepared manuscript, the number of people who would have had to be either fooled or complicit rises with every scribe. Eight or nine is a large number for a conspiracy and a small number for a mass delusion. That is a real problem for the crudest fraud theories, and it is why serious naturalistic accounts do not use them.

The lost 116 pages

In June 1828 Martin Harris borrowed the manuscript of the first portion of the translation — Joseph Smith called it 116 pages, and it was known as the Book of Lehi — to show his family, most likely to satisfy his wife Lucy that the money he had committed was not being wasted. The pages never came back. Isaac Hale, next door, remembered the same thing from the outside: "It was said, that Harris wrote down one hundred and sixteen pages, and lost them."¹⁰ Joseph Knight Sr., who was close to the family, wrote the plainest surviving summary: Harris "urged him By fair promises that he would be Careful and he would Return it again. But he Being free with it some person go[t] hold of it and Cept [kept] it so that he never Could obtain it again. There fore Joseph Lost his privilige for a while."⁹

Two revelations respond to the loss and are now canonized.

Doctrine and Covenants 3, dated July 1828, is a rebuke. It is worth noting how severe it is, because a fabricated revelation issued to cover a disaster would not usually be this harsh toward its author: "thou art still chosen, and art again called to the work; Except thou do this, thou shalt be delivered up and become as other men, and have no more gift." Verse 12 states the offence: "when thou deliveredst up that which God had given thee sight and power to translate, thou deliveredst up that which was sacred into the hands of a wicked man." Verse 14 gives the consequence: "this is the reason that thou hast lost thy privileges for a season."

Doctrine and Covenants 10 supplies the solution and is the more interesting document. It first explains why re-translating would be a trap. Verses 30 to 33 argue that whoever holds the pages has altered the wording and is waiting: "if you should bring forth the same words they will say that you have lied and that you have pretended to translate, but that you have contradicted yourself."

Then it supplies the replacement, in verses 38 to 45: an account of the same period "is engraven upon the plates of Nephi," it is "more particular," and Joseph is instructed to "translate the engravings which are on the plates of Nephi, down even till you come to the reign of king Benjamin, or until you come to that which you have translated, which you have retained." The text he was to publish is what became 1 Nephi through Omni — the "small plates."

Was the small-plates solution providence or recovery?

This is the article's most genuinely balanced section, and readers on both sides should find it unsatisfying in the same way.

The strongest form of the design argument. The Book of Mormon contains, in its own narrative voice, an explanation for the existence of the duplicate record — and that explanation is embedded in a passage whose narrator says he does not know why he is doing it. In Words of Mormon, the abridger Mormon says he found a second small set of plates covering the earliest period, and included them: "And I do this for a wise purpose; for thus it whispereth me, according to the workings of the Spirit of the Lord which is in me. And now, I do not know all things; but the Lord knoweth all things which are to come; wherefore, he worketh in me to do according to his will." On the Latter-day Saint reading, a fourth-century editor was prompted to preserve a redundant record for a reason he could not see, and the reason surfaced in 1828 when Martin Harris lost the abridgement of that same period.

The structural fit is genuinely tight. The small plates run from Lehi's departure to the reign of king Benjamin. Doctrine and Covenants 10:41 instructs Joseph to translate the plates of Nephi "down even till you come to the reign of king Benjamin." The seam closes exactly. And Words of Mormon then hands off to Mosiah in a way that lets the two sources meet without a gap. That is not a sloppy patch.

The strongest form of the recovery argument. Everything above is also exactly what an author in trouble would have to produce.

Consider the constraints. Joseph Smith could not re-dictate the lost text, because he could not reproduce 116 pages verbatim and the holders of the manuscript could publish it against him. He could not simply skip the period, because the book would then begin in the middle. He needed a version of the same events that was allowed to differ in wording and content from a document he could not see. "A different record of the same period, by a different author, with a different emphasis" is the unique solution to that problem. The fact that it is elegant does not distinguish design from necessity, because necessity here permitted only one shape of answer.

The Words of Mormon passage is the crux, and it can be read either way. On the design reading it is an ancient editor's unexplained prompting. On the recovery reading it is exactly the seam one would expect: a short bridging text, written last, whose job is to explain why the two blocks of material fit together, and which pre-empts the obvious question ("why is this here twice?") by attributing the redundancy to a purpose the narrator declines to specify. Bridging passages that say "I do this for a wise purpose" and "I do not know all things" are what an author writes when joining two sections.

There is one further consideration, and it points slightly toward the recovery reading without settling it. The dictation-order evidence, described below, indicates that Words of Mormon was among the very last material produced. If so, the passage explaining the duplication was written after both blocks it joins already existed. That is what a seam looks like. It is also, of course, exactly where a translator would encounter it, since the small plates were translated last and Words of Mormon sits between them and Mosiah.

The verdict here is that neither reading is forced. A skeptic who calls the small-plates episode an obvious patch is ignoring how well it fits. A believer who calls it proof of foreknowledge is ignoring that the constraints allowed no other move.

The order of dictation

The surviving text was almost certainly not dictated in the order it is printed.

The reconstruction that most historians of the text now accept runs: Mosiah → Moroni → the Title Page → 1 Nephi → Words of Mormon. In other words, the dictation resumed in June 1828's aftermath at the point where the lost pages had ended, ran to the end of the book, and only then went back to produce the replacement opening.

The evidence for it is of several kinds.

The revelation itself. Doctrine and Covenants 10:41 tells Joseph to translate the plates of Nephi "until you come to that which you have translated, which you have retained." The natural reading is that a body of translated text already existed and the small plates were to be produced to meet it.

The Title Page. Joseph Smith described the title page as a literal translation of the last leaf of the plates. If it was translated in sequence, it would have come at the end of the Mosiah-to-Moroni run, before the small plates. The copyright was secured on 11 June 1829 and the application used the title page text, which puts the title page before that date and gives a hard point to hang the sequence on.¹⁵

A believing scholar builds it into his own calendar. John W. Welch's 2018 reconstruction allots fifty-three of the seventy-four available days to Mosiah through Moroni and twenty-one to 1 Nephi through Words of Mormon — that is, it puts the small plates at the end.¹⁶ The order is not a critics' hypothesis.

The scribal and manuscript evidence. Brent Lee Metcalfe's 1993 essay "The Priority of Mosiah: A Prelude to Book of Mormon Exegesis," in New Approaches to the Book of Mormon, is the standard critical statement of the case from internal evidence. It was not opened for this article and is named here only so a reader can go to it;  no claim in this section rests on it.

What does the order imply? Two things, in opposite directions.

For the Latter-day Saint account, it implies that a great deal of the text was produced without the opening being available to consult. If the small plates came last, then all the foreshadowing in 1 Nephi and 2 Nephi that pays off in Alma and Helaman was written after its payoff. That is easier for an author than for a translator only if you assume the author was working forward; an author writing the opening last can retrofit foreshadowing freely, which is a point for the naturalistic side.

For the naturalistic account, it implies something less convenient. A composing author who wrote the opening last had the whole later narrative available in his memory when he did so. Retrofitting is easy. But it also means the earlier 200,000 words were composed without a written outline of the beginning, which is the harder half of the feat.

The order is a real finding and it constrains both stories. It is not a knockdown for either.

The pace of dictation

Here the arithmetic should be shown rather than asserted, because the figures that circulate are often quoted without their assumptions.

The word count. A count of the Book of Mormon text of 1 Nephi through Moroni, using the public-domain electronic text and counting only alphabetic word tokens (so that chapter and verse numbers are excluded), gives 268,359 words. Including the title page and the witnesses' testimonies brings it to 269,177. The commonly quoted figure of roughly 269,000 words is therefore sound. Of that total, 1 Nephi through Omni — the small plates, dictated last — accounts for 67,384 words, or 25.1 percent. Mosiah through Moroni accounts for 200,114 words, or 74.6 percent. Words of Mormon is 861 words.³

The window. Oliver Cowdery arrived on 5 April 1829, and Joseph Smith's own history fixes the start: "Two days after the arrival of Mr. Cowdery (being the 7th of April) I commenced to translate the Book of Mormon, and he began to write for me" (Joseph Smith—History 1:66–67). The Joseph Smith Papers date the whole document "circa 12 April 1828–circa 1 July 1829," which puts the end of the work at about 1 July 1829.¹⁴

The calendar. From 7 April 1829 to 30 June 1829 inclusive is 85 calendar days. Twelve of those are Sundays. Removing Sundays leaves 73 days. Further known interruptions have to come out of that: the baptisms of 15 May 1829 (Joseph Smith—History 1:71–72), the move from Harmony to the Whitmer farm in Fayette, the trip to Palmyra that secured the copyright on 11 June 1829,¹⁵ approaches to printers, and the experience of the Three and Eight Witnesses in late June. Each of those costs days, but no source counts them. Deducting about eight days brings the plausible number of actual working days toward 65. That figure is an estimate, not a count, and the article uses it as one.³

The division.

The defensible range is therefore roughly 3,600 to 4,100 words per working day, with the extremes at about 3,200 and about 4,700 depending on how many days one deletes. Every figure in that table rests on the same total and differs only in the denominator, which is why the denominator assumptions have to be stated. Welch's 2018 survey of the published estimates found figures of 60, 63, 65, 74, 75, 85, and 90 days in print, all from writers looking at the same three months.¹⁶ The spread is a measure of how much the answer depends on assumptions rather than documents.

What does that feel like? At a comfortable dictation speed of about 25 to 30 words per minute — a pace set by how fast a scribe can write with a quill, not by how fast a person can talk — 4,000 words takes about two and a half hours of actual dictation. Spread over a working day with interruptions, that is not physically extreme. The extraordinary part is not the hourly rate. It is the sustaining of coherent narrative output at that rate, day after day, for three months, without notes.

Comparisons. Documented rates of rapid literary production exist and should be put on the table honestly.

Dictated composition is faster than handwritten composition, because the bottleneck moves from the hand to the mouth. Beyond that general point, this article declines to put numbers on comparison cases it has not checked. The oral-formulaic tradition described by Milman Parry and Albert Lord — illiterate South Slavic singers composing epic narrative in performance from an inherited stock of formulas rather than reciting memorized text — is the parallel usually reached for, and Lord's The Singer of Tales is its standard account. Neither Lord's book nor any other study of oral composition was opened for this article, so no figure for what those singers produced per day appears here.

What can be said from the Book of Mormon's own text is that its heavy formulaic texture — "and it came to pass," "and now I say unto you" — is the kind of feature that lowers the cost of composing aloud, whatever the source of the text.

And the limits of the comparison are clear without opening a book on it. A South Slavic singer worked inside a tradition of songs he had heard performed hundreds of times. There is no known tradition of Nephite epic that Joseph Smith could have absorbed. The formulaic texture of the Book of Mormon is King James English, which he certainly had absorbed, but the plot is not a traditional story. The precise question — whether any documented case of sustained oral composition matches the Book of Mormon on length, speed, internal chronological consistency, and absence of an inherited plot, all at once — is open, and this article does not answer it.

The honest verdict on pace. It is fast. It is not unique. Rapid sustained oral composition is a documented human capacity. What has no clear parallel is the combination: this speed, at this length, in this register, producing this degree of internal consistency, with no drafts and no known apprenticeship in the form. The pace argument is strong evidence that something unusual happened. It is weak evidence about what.

The manuscripts

Two manuscripts were made, and their fates are almost opposite.

The original manuscript is the one written down as Joseph Smith spoke. It ran to about 600 pages in all, of which roughly 488 were dictated after the 1828 loss.¹⁴ On 2 October 1841 the Saints laid the cornerstone of the Nauvoo House, a stone chest "about 10 x 14 inches, and eight inches deep," and Joseph Smith's journal records that he placed in it coins, a Bible, printed copies of the Book of Mormon, the Times and Seasons and the Doctrine and Covenants, and the original manuscript.¹⁴ It stayed there just over forty years. In 1882 Lewis Bidamon, Emma Smith's second husband, was demolishing the eastern wing of the building, came across the cornerstone, and found the box "filled with a number of written and printed documents, the most of them mouldy and more or less decayed," the manuscript among them.¹⁴

What survived was then dispersed. The Joseph Smith Papers editors date the acquisitions and name the principal recipients: Sarah Granger Kimball took ten leaves in 1883; Joseph W. Summerhays a leaf in 1884; Franklin D. Richards portions of sixty-nine leaves in 1885; Andrew Jenson parts of eight leaves in 1888; and Wilford Wood numerous fragments of twenty-nine leaves in 1937.¹⁴ Portions of 232 pages survive out of 488; the other 256 are gone entirely, and seven of the fifteen books — Jarom, Omni, Words of Mormon, Mosiah, 4 Nephi, Mormon and Moroni — have no surviving original leaf at all.¹⁴¹⁴ That is about 28 percent of the text, the figure the Church's own essay gives.¹ The Church History Library holds roughly 90 percent of what remains.¹⁴

The printer's manuscript is Oliver Cowdery's copy, made so that the original could be kept safe while pages went to the print shop. It was produced between roughly August 1829 and January 1830. It is nearly complete: the Church's own announcement says it "is virtually complete, as only three lines of the original text are missing."² It was consulted again for the 1837 Kirtland edition.

Its custody chain is unusually clean. Oliver Cowdery gave it to David Whitmer shortly before Cowdery's death in 1850. Whitmer kept it until his own death, and it passed through his family to the Reorganized Church of Jesus Christ of Latter Day Saints, renamed Community of Christ in 2001. On 20 September 2017 Community of Christ sold it to the Church of Jesus Christ of Latter-day Saints, which announced the acquisition and thanked "the generous donors who provided the means" but gave no price.¹⁷ Elder Steven E. Snow, then Church Historian and Recorder, explained its value: "The printer's manuscript is the earliest surviving copy of about 72 percent of the Book of Mormon text, as only about 28 percent of the earlier dictation copy survived decades of storage in a cornerstone in Nauvoo, Illinois."¹⁷ The figure of $35 million was widely reported at the time; no source giving it was opened for this article, and it is therefore not stated here as a fact.

What the manuscripts show. This is where Royal Skousen's Book of Mormon Critical Text Project, running since 1988, produced results that neither side had before. His 1998 summary in the Journal of Book of Mormon Studies states four findings that bear directly on production.⁵

The errors are auditory, not visual. A scribe copying by eye makes one family of mistakes — skipped lines where two lines end with the same word, transposed letters, confusion of similarly shaped letters. A scribe writing from dictation makes a different family — homophone substitutions, word-boundary errors, mishearings. The original manuscript shows the second family, and Skousen makes the comparison concrete: at Alma 30:52 the printer's manuscript turns the original's "always" into "also," a mistake of the eye, and "this kind of error does not appear in the original manuscript, because the scribes were not copying from another written source but were hearing the words dictated by Joseph Smith."⁵ This is the strongest single piece of physical evidence that the text was spoken aloud and taken down, and it is why the Church's essay cites it.¹

Names were spelled out. When an unfamiliar proper name first appears in the original manuscript, there are places where a phonetic spelling is struck out and the correct one written on the same line — at Alma 33:15 Oliver Cowdery wrote "Zenock," crossed it out, and immediately wrote "Zenoch."⁵ That is what a speaker does when giving a scribe a word the scribe has never heard. It is not what a person reading from a manuscript does, and it is not what a scribe copying from a manuscript does.

The dictation came in units of at least twenty words. Skousen's evidence is scribal anticipation: at Alma 56:41 Oliver Cowdery jumped ahead, wrote "we saw the Lamanites," struck it out, and wrote the intervening clause. Skousen concludes that "Joseph had at least twenty words in view as he was dictating."⁵ That is a working detail about the mechanics of production, and it is consistent with the witnesses' descriptions of sentence-by-sentence delivery, though not with David Whitmer's character-by-character version.

There was no punctuation. Confirmed by the manuscript itself and conceded in the Church's essay.¹ The punctuation in every printed Book of Mormon was supplied by the typesetter.

Publication

Joseph Smith secured the copyright on 11 June 1829. The surviving recipient's copy is a printed form filled in by hand, issued by the Northern District of New York, probably at Utica.¹⁵

The printing was done by Egbert Bratt Grandin in Palmyra, New York. Grandin was twenty-three and newly established, and he took some convincing. The obstacle was money: the job was priced at $3,000, an enormous sum for the time. Joseph Smith's own history records the deal in a sentence: "we went to Palmyra, Wayne County, N. Y: Secured the Copyright; and agreed with Mr Egbert Grandon to print five thousand Copies, for the sum of three thousand dollars."¹⁸ Martin Harris secured it by pledging a 151-acre portion of his farm on 25 August 1829, and on 16 January 1830 he and Joseph Smith signed an agreement giving Harris "equal privilege" to sell copies until he had been reimbursed.¹⁸

The press run was therefore 5,000 copies, which was a very large first edition by the standards of 1830 American publishing.

The typesetter was John H. Gilbert, and his role is larger than a typesetter's normally is. He supplied the punctuation, which the Church's own essay concedes: "Because Joseph Smith did not call for punctuation, such as periods, commas, or question marks, as he dictated, such marks are not in the original manuscript. The typesetter later inserted punctuation marks when he prepared the text for the printer."¹ Gilbert wrote his own account of the work in a memorandum dated 8 September 1892, which is the source the Joseph Smith Papers cite for the printing arrangements.¹⁸ The first printed Book of Mormon is therefore a collaboration in a narrow but real sense: the words are Joseph Smith's dictation as copied by Oliver Cowdery, and the sentence boundaries are John Gilbert's editorial judgment. Anyone arguing from the grammar or sentence structure of the 1830 edition needs to know which parts came from whom.

The book went on sale in Palmyra on 26 March 1830, when Grandin advertised it for sale retail and wholesale in his bookstore in that day's Wayne Sentinel.¹⁸ The Church of Christ was organized less than two weeks later, on 6 April 1830.

Later editions and textual change

The Book of Mormon has been edited. Latter-day Saints sometimes deny this and should not; the Church's own publications describe the process. Critics sometimes present the number of changes as though it implied wholesale rewriting, and it does not.

The major stages:

1830, Palmyra. The first edition, set by Gilbert from the printer's manuscript, with chapter divisions long and few, no verse numbers, and Gilbert's punctuation.

1837, Kirtland. The second edition. Joseph Smith and Oliver Cowdery worked over the text against the printer's manuscript. Most of the changes are grammatical: "which" corrected to "who" for persons, verb forms regularized, and similar tidying of nonstandard English. A small number touch sense. The best known is at 1 Nephi 11:18 and nearby verses, where the 1830 reading "the mother of God" became "the mother of the Son of God." The present text reads "Behold, the virgin whom thou seest is the mother of the Son of God, after the manner of the flesh."

1840, Nauvoo. The third edition, again with Joseph Smith's involvement, restoring some readings from the original manuscript.  It also changed one word at 2 Nephi 30:6.  The 1830 edition reads "save they shall be a white and a delightsome people."¹⁹  The 1840 edition reads "a pure and a delightsome people."²⁰  Royal Skousen judges that the change "seems to be a conscious one and was probably made by Joseph Smith as a part of his editing for the 1840 edition."²¹

1879 and 1920. Orson Pratt's 1879 edition introduced the modern short chapters and the verse numbering still used. The 1920 edition added further apparatus. These are structural rather than textual.

1981. A substantially re-edited edition with new chapter headings, a new introduction, and a small number of textual changes. The most discussed is at 2 Nephi 30:6, where "white and delightsome" became "pure and delightsome"; the present text reads "save they shall be a pure and a delightsome people."  The wording was not new in 1981.  It is the 1840 reading.  The Latter-day Saint editions had missed it for 141 years because they descended from the 1837 edition through the British printings,²¹ and the Church's 1879 edition still reads "a white and delightsome people."²²  In Skousen's words, "the LDS text followed the earlier reading until the 1981 edition, when pure (the 1840 reading) was adopted."²¹  So the charge that the Church invented "pure" in 1981 is wrong.  The defence that "white" only ever meant "pure" does not hold either:  Skousen's critical text restores "white" as "the earliest extant reading."²¹  Article 13 gives the history of the verse in full.

2013. Mostly a revision of the study apparatus rather than the scripture text. The Church's own published summary says that "more than 99 percent of the corrections are in the study aids (chapter summaries, footnotes, Topical Guide, Bible Dictionary, and triple combination index)," and describes the changes to the text itself as "some minor typographical errors in spelling and punctuation … corrected to ensure fidelity to the earliest manuscripts and editions."²³

How many changes, and how substantive? The figure that circulates most is about 3,913, from Jerald and Sandra Tanner's compilation 3,913 Changes in the Book of Mormon, which set the 1830 edition against a modern one and counted differences. That book was not opened for this article, and the number is reported here as a published claim rather than as a verified count. What can be said with confidence is how such a count has to work: it counts every difference of any kind, and in a book whose punctuation was supplied by a typesetter in 1830 and whose grammar was regularized in 1837, the overwhelming majority of differences will be punctuation, capitalization, spelling, chapter division, and grammar.

The fair statement is this. The Book of Mormon's text has been edited continuously since 1830, mostly in ways that any nineteenth-century book went through, and in a small number of places in ways that touch doctrine or sense. Neither "it has never been changed" nor "it has been changed nearly four thousand times, so it cannot be scripture" is an accurate description. The Church now publishes both manuscripts, which lets anyone check.

Could Joseph Smith have composed it?

This is the article's real question, and it deserves to be stated at full strength before any answer is attempted.

The difficulty, stated fairly. A man of about twenty-three, with roughly three years of formal schooling spread across his childhood, dictated aloud a narrative of about 268,000 words in roughly three months. He did it in front of hostile in-laws and skeptical neighbours as well as believers. He produced hundreds of proper names, most of them not from the Bible, and used them consistently. He tracked an internal chronology across a thousand years of fictional history and across two interleaved record traditions. He sustained a pseudo-archaic English register. He produced no drafts that anyone has found, and every scribe who was asked said he worked without notes and without having earlier text read back to him. Emma Smith's detail about resuming after meals is the sharpest form of the problem, and she put the contrast bluntly: Joseph Smith "could neither write nor dictate a coherent and well-worded letter, let alone dictating a book like the Book of Mormon."⁴

"He was uneducated, therefore he could not have done it" is not a good argument and this article will not make it. Formal schooling is a poor predictor of verbal facility, and the history of literature is full of people with little of it. But the argument above is not about schooling. It is about the specific combination of speed, length, coherence, and absence of drafts.

The serious naturalistic accounts.

Extemporaneous preaching culture. William L. Davis's Visions in a Seer Stone: Joseph Smith and the Making of the Book of Mormon (University of North Carolina Press, 2020) is the most substantial recent attempt, and it is the one this article treats as the serious opponent. The book was not opened for this article. What follows is a statement of the argument as it is generally reported, offered so that the article engages the strongest naturalistic case rather than a weak one, and it is marked as second-hand for exactly that reason. No sentence below is presented as Davis's words, and a reader checking this section should go to the book.

The argument turns on a technique widely taught and widely practised in early American religious oratory, called "laying down heads." A preacher would work out in advance a skeleton of numbered topics — the "heads" of a discourse — with subordinate points beneath them, memorize the skeleton rather than the words, and then compose the actual language extemporaneously in delivery. The technique was taught in homiletic manuals and was common in the revival preaching Joseph Smith grew up around, and it produces exactly the capability in question: fluent, structured, sustained oral composition from a compact mental outline.

On this account the Book of Mormon shows the marks of the method in its own text — sermons that announce their divisions in advance, narrative sections that pre-announce what they will cover, and the recurrent "I say unto you" transitions that are the verbal furniture of a preacher working from heads. Joseph Smith did not memorize a text or read from one; he carried an outline, refined over years, and clothed it in language at dictation speed. Davis has also argued, in an earlier article on Joseph Smith's formal education, that the folk image of an illiterate ploughboy understates the schooling, the print, and the reading culture actually available in the Smith household.

Long prior composition. The angel's first visit is dated to September 1823. The plates were obtained in September 1827. The dictation ran to mid-1829. That is close to six years during which the story was, on any account, being thought about. A naturalistic account does not need spontaneous invention in 1829; it needs six years of mental development plus three months of transcription.

Formulaic texture. The Book of Mormon's style makes rapid production easier, not harder. Its heavy use of a small set of connective formulas, its parallelism, and its King James syntax all reduce the cognitive load of composing aloud. This is the same mechanism that makes oral epic possible.

The strongest Latter-day Saint answer to all of this.

The scribes are the answer, and specifically the negative claims they make.

They say he did not have a manuscript. Emma Smith: "He had neither manuscript nor book to read from."⁴ They say he did not review what had been written. Emma Smith again: he "would at once begin where he had left off, without either seeing the manuscript or having any portion of it read to him."⁴ They say the dictation was continuous. Oliver Cowdery: "Day after day I continued, uninterrupted, to write from his mouth" (Joseph Smith—History 1, note following verse 75).

Those claims bear directly on Davis's model, and they bear on it in a specific place. Laying down heads explains fluent oral composition. It does not obviously explain resumption without review across a 268,000-word narrative with tracked chronology and hundreds of names. A preacher working from heads delivers a sermon of perhaps forty minutes and then stops. He does not resume a partially completed sermon three days later at the exact word.

The Latter-day Saint answer also notes something about the absence of drafts. Six years of mental development is a hypothesis, not a document. No outline, no notebook, no marked Bible, and no draft chapter has ever been found, and the Smith household was searched by hostile parties more than once. The naturalistic account rests at this point on an unrecovered private process.

Where this leaves the question.

Three things are true at once, and the article states all three.

First, Davis's mechanism is real, documented, and available to Joseph Smith. It removes the crudest form of the impossibility argument. Nobody should now say that sustained fluent oral composition was beyond a young man of that time and place, because the culture taught it.

Second, the mechanism has not been shown to scale to this task. There is no documented case of a person using laying-down-heads technique to produce a quarter-million-word connected narrative with a tracked internal chronology in three months. Demonstrating the technique is not the same as demonstrating the feat.

Third, and cutting the other way, the difficulty of the feat is an argument for something unusual, not specifically for divine translation. "No naturalistic account is fully worked out" and "therefore the supernatural account is established" are separated by a large gap. The Book of Mormon's origin has to be settled by what the text itself contains — its anachronisms, its dependence on the King James Bible, its nineteenth-century theological concerns, its relation to its claimed ancient setting — and those are the subject of the authorship article, not this one.

The strongest Latter-day Saint responses

The Gospel Topics essay. The Church's essay "Book of Mormon Translation" is a genuinely good document and should be treated as such. It concedes the seer stone, concedes the treasure-seeking, concedes the terminological drift, concedes the absence of punctuation, and concedes that the accounts of the process differ. It then argues that physical instruments of revelation are biblically normal, citing the rod of Aaron, the brass serpent, anointing oils, the Ark, and the clay Jesus made with saliva.¹ That argument is not a dodge. If one grants the possibility of revelation at all, the objection "but he used a rock" has limited force.

Where the essay is weaker is in what it does not say. It does not tell the reader that the earlier standard Latter-day Saint account, taught for over a century in Church curriculum and art, showed Joseph Smith reading the plates directly with the interpreters. The essay corrects that picture without acknowledging that it was the picture. Under the authority framework of article 1, that matters: correlated manuals and Church-commissioned art are not scripture, but they are what members were taught, and a correction is a historical event that deserves acknowledgement.

Royal Skousen's critical text work. This is the most valuable scholarship on the subject from any quarter, and it happens to be by a believing Latter-day Saint at Brigham Young University. Skousen's manuscript analysis produced the auditory-error finding, the spelled-out-names finding, and the twenty-word dictation unit, and he summarizes the whole as "tight control" — the claim that Joseph Smith "apparently received the translation word for word and letter for letter."⁵ He also argues, more controversially, that the original text contains Early Modern English grammatical structures that were obsolete by 1829 and that Joseph Smith is unlikely to have produced. That argument belongs to the authorship article and is contested there.

The witnesses argument. The strongest faithful position is not about the plates but about the scribes. Seven people wrote at the dictation. Several later became enemies of Joseph Smith or of his successors. None of them ever described a manuscript. That is a real evidential fact and it is why the sophisticated naturalistic accounts posit prior mental composition rather than a hidden document.

John W. Welch on the timing. Welch's 2018 article in BYU Studies Quarterly is the standard Latter-day Saint reconstruction of the translation calendar, and it is careful in exactly the way this subject needs. He gives the raw span as eighty-five days from 7 April to 30 June, excludes eleven full days on which Joseph Smith was demonstrably travelling or otherwise occupied, and arrives at seventy-four days as the maximum available — fifty-three for Mosiah through Moroni and twenty-one for 1 Nephi through Words of Mormon.¹⁶¹⁶ That split is worth noticing, because it means a believing scholar publishing in a Church university's journal builds the Mosiah-first dictation order into his own chronology. Welch also catalogues the published estimates and shows how far apart they are.¹⁶ The point where his treatment should be resisted is rhetorical rather than factual: Welch presents the pace as evidence of divine assistance, and the pace by itself does not carry that inference.

Common objections and responses

"The Church hid the seer stone and the hat"

This was substantially true until the 1980s and is not true now. The Gospel Topics essay states the method plainly.¹ The Church announced the seer-stone photographs on 4 August 2015 and printed them in the October 2015 Ensign.² The Joseph Smith Papers publish the manuscripts.¹⁴ A critic who presents the hat as a current cover-up is arguing against a position the Church abandoned. The legitimate remaining question is not whether the Church now tells the truth about the method, but why it took until 2015, and what that delay implies about how correlated Church curriculum handles inconvenient history. The 2015 Ensign article gives part of the answer itself: in their proselyting, "Joseph and other early members chose not to focus on the influence of folk culture, as many prospective converts were experiencing a transformation in how they understood religion in the Age of Reason."¹¹ A choice made in 1830 hardened into a curriculum. That is a question for article 1.

"Joseph Smith was convicted of fraud in 1826"

The evidence does not support the word "convicted," and this article does not use it. Joseph Smith was tried before Justice Albert Neely on 20 March 1826 under New York's disorderly persons statute, for stone-gazing in Josiah Stowell's employment. What survives from 1826 is two fee bills. Everything about the verdict reaches us through a docket page torn out by the justice's niece and printed three times between 1873 and 1886, and through a physician's recollection published fifty-one years later. The Joseph Smith Papers editors, working from the fee bills and the statute, conclude that the absence of a record-of-conviction fee and of any commitment to the bridewell "strongly suggests that Neely did not convict JS," while cautioning that any conclusion about the outcome is tentative.¹³ The strongest defensible statement is that Joseph Smith was arrested, held two days, and tried for glass-looking; that he was not jailed afterward; and that he continued his work. "Fraud" is also the wrong word: the charge was being a disorderly person, and the statute reached anyone who "pretended" to find lost goods whether or not he believed in his own gift.¹³

"The plates would have weighed two hundred pounds, so the story is impossible"

The arithmetic behind that objection is correct for solid gold and is shown above. It is not correct for a stack of thin leaves bound with rings, and it is not correct for the gold-copper alloys actually used in pre-Columbian America. A reconstruction using half packing and an alloy density of about 11 grams per cubic centimetre gives roughly 57 pounds,⁷ which is in the neighbourhood of the forty to fifty pounds Martin Harris estimated from hefting them.⁸ The objection fails. What should be noted instead is that the reconstruction was built to fit the testimony rather than derived independently.

"Nobody could dictate 269,000 words in three months"

People can and do. The pace is about 3,700 to 4,100 words per working day, which at quill speed is a few hours of actual dictation. Sustained rapid oral composition is documented in oral-epic traditions and in professional dictation. This objection is the weakest one in the critical arsenal and should be retired. The genuine difficulty is not the rate but the combination of rate with narrative coherence and the absence of drafts.

"The Book of Mormon has been changed 3,913 times, so it is not scripture"

The inference is wrong even if the count is right. The overwhelming majority of any such tally has to be punctuation, spelling, capitalization, chapter and verse divisions, and grammatical regularization, because the 1830 text's punctuation was supplied by a typesetter who was not working from dictation and the 1837 revision was largely a grammar pass. Substantive changes exist and should be discussed by name — the 1837 "mother of the Son of God" clarifications, the 1981 "pure and a delightsome" reading — but there are few of them. A book whose spelling was standardized is not thereby discredited. The Bible has been through the same process on a far larger scale. This article does not endorse the 3,913 figure, because the compilation it comes from was not opened.

"The witnesses all told the same story, so it must be true"

They did not all tell the same story. They agree that Joseph Smith put a stone in a hat, pressed his face into it, and dictated. They disagree about whether one character or one sentence appeared at a time, and about whether text stayed until copied correctly. The convergence on the hat is strong precisely because it is a convergence of otherwise-conflicting testimony. Overstating the agreement weakens the argument that the agreement supports.

"Emma Smith was a believer, so her testimony is worthless"

Emma Smith is the single best witness in the record and her interests cut in complicated directions. In 1879 she was affirming the Book of Mormon while denying, in the same interview, things Brigham Young's church taught about her husband. She was not defending the Salt Lake institution. She had watched the dictation from a few feet away, for months, in her own house, in 1828 and again in 1829. Her testimony can be doubted on the ground that it was given fifty years later, by an elderly woman, to her son, for publication in her son's church's newspaper. It cannot be dismissed as institutional loyalty.

What evidence would change the conclusion?

Evidence that would strengthen the Latter-day Saint claim. A contemporary document from 1828 or 1829 — a letter, a journal, a scribal note — describing the dictation as it happened would enormously improve a record that currently depends on recollections from the 1870s and 1880s. A demonstration that the Early Modern English features Skousen identifies are systematic, ungrammatical in 1829 usage, and internally consistent would be hard to explain by composition. A controlled demonstration that an equivalently educated person cannot produce comparable output under comparable conditions would give the pace argument evidential force it currently lacks.

Evidence that would weaken it. Any pre-1829 draft, outline, or source document in Joseph Smith's or an associate's hand would end the discussion. A demonstration that the "laying down heads" technique scales to quarter-million-word connected narrative — ideally a documented historical case, not an argument from possibility — would close the main gap in Davis's account. Additional archival recovery on the 1826 proceeding, particularly an intact docket page or a court record independent of the Pearsall transmission, would settle a question currently held open by bad provenance.

Evidence that would change nothing. More late recollections of the hat will not change anything; the hat is conceded. More counts of textual changes will not change anything; the counts are known and mostly trivial. The production question will not be settled by production evidence alone. It will be settled, if at all, by what the text turns out to contain.

Sources