The 'disappearing anachronisms' chart

The most-shared apologetic graphic says about 90% of Book of Mormon anachronisms are now confirmed; the number depends on how items are chosen and scored.

What the Church claims

Of 226 features critics have called anachronistic since 1830, 174 are now 'confirmed' and 31 'partially confirmed', so that unconfirmed items fell from about 90% in 1844 to 9.29% in 2024; time vindicates the book.

Once the evidence is in, this claim: Weak. The headline number is an artifact of how the list was built: items are selected, split and counted unweighted, and the textual and genetic problems are left off it entirely. A minority of the confirmations are real. The claim fails because a percentage is the wrong instrument for authenticity, not because every entry on the chart is wrong.

The record

  • The chart's lineage is: John E. Clark, Wade Ardern and Matthew Roper, FAIR Conference, 2005-08-04, published by Clark in BYU Studies 44:4 (about 60 criticisms from three 19th-century books, about 60% resolved, a sample Clark called about 3% of published criticisms); Roper and Kirk Magleby, 'Time Vindicates the Prophet', FAIR Conference, 2019-08 (205 items); Roper's serial in Interpreter vol. 65, an introduction and nine chapters published from 2025-05-02 to 2025-06-27, which raised the count to 226 items.
  • Roper's stated method: review of more than 1,000 critical publications from 1830 to 2024; an item is 'confirmed' when it 'turns up' in the historical, archaeological or linguistic record, 'partially confirmed' when data are not conclusive but trending, and 'unconfirmed' when no known data directly support it.
  • Published totals (2025): 1830-1844, 102 items (7 confirmed, 2 partial, 93 unconfirmed); 1845-1965, 168 items (36/10/122); 1966-2024, 226 items (174/31/21).
  • The 2019 version used different numbers: 89 items by 1844, 150 by 1965, 205 by 2019, with FAIR reporting 141 confirmed, 26 trending, 38 unconfirmed, and FAIR's own wiki noting the categories were not explicitly defined. A 2020 Saints Unscripted article gives 90/149/204 and 'about 83 percent'. The versions are not the same dataset.
  • Single criticisms are split into several chart items. Horses are four items, three of them 'confirmed' by Ice Age fossils and native traditions, while 'horses in Book of Mormon times' is only 'partially confirmed'; elephants are three items on the same pattern; iron is four items, three 'confirmed' by hematite mirrors and meteoric iron, while iron smelting is unconfirmed.
  • Loan-shifts are scored as confirmations: cows and oxen ('Confirmed Bison Loan Shift'), goats (brocket deer), swine (peccary), lions (puma). 'Glass' is confirmed because obsidian is natural glass.
  • Many items are generic features of any agrarian society: tools to hoe, plow, thrash and spin, sickles, arts, many buildings, barns, judges, lawyers, armies, cords, ladders, rations, trumpets, fainting from loss of blood. Item 109 is 'Archaeological Evidence - confirmed', citing Nahom and names, which are counted elsewhere.
  • All 21 Old World journey items are scored confirmed; seven of them (Bountiful, much fruit, wild honey, timber, ore, bellows, a mountain) come from a single candidate site in Oman.
  • The chart contains no item for Deutero-Isaiah, KJV translation errors, the Sermon on the Mount's textual base, New Testament quotations or DNA; the chapters on records and on Third Nephi do not address them.
  • Roper's own caveats: the study evaluates anachronisms 'as framed by critics', gives a 'limited' picture, and makes little attempt to test items against any specific geography model.
  • What Roper still scores unconfirmed includes New World steel, iron smelting, wheat, New World synagogues, pre-Columbian Christianity, any identified Book of Mormon city, New World metal records, Hebrew or Egyptian inscriptions in America, battle remains, and the names Timothy and Jonas.

The moving target

  • 2019: The Church takes no position on where in the Americas the events occurred and asks members not to claim prophetic support for any theory; the chart is not a Church product.

The exchange

Matthew Roper (Scripture Central / Interpreter 2025); John E. Clark (2005); FAIR

The trend is the evidence: a forger's errors should accumulate as knowledge grows, but the book's list of problems has shrunk from about 90% to about 9% unconfirmed.

The answer: The falling percentage is produced mainly by the denominator. The item count more than doubled (102 to 226), and the additions are overwhelmingly easy wins: generic civilization traits, loan-shifts scored as confirmations, etymologies scored as attestations, and single criticisms split so that the trivially true parts (horses once existed in the Pleistocene) count as separate confirmations. Meanwhile the problems that scholarship has added since 1830 (Deutero-Isaiah, KJV error carry-over, late Greek readings in 3 Nephi, DNA) are not on the chart at all. There is no weighting: 'ladders' counts the same as 'steel'. And no item is ever scored 'disconfirmed'; the only possible movements are toward confirmation.

What they get right: Some nineteenth-century criticisms really were wrong, and the chart is right to say so: cement, Old World steel, writing on metal in the Near East, barley of a kind, pre-exilic 'Jew', fortifications and a warlike Maya. Critics who still repeat those points deserve the correction.

Roper, Introduction

'Confirmed' only means the feature is no longer out of place; it does not claim to prove the book.

The answer: That caveat is sound but does not survive the chart's circulation. FAIR, Saints Unscripted and social posts present it as '83 percent debunked' or '70% confirmed'. And a feature is still out of place if it is attested on the wrong continent, in the wrong millennium, or only by renaming a different thing.

John L. Sorenson (1984); FAIR; Roper

Loan-shifting is a documented linguistic process (Spaniards called bison 'cows'; Maya called horses 'tapir').

The answer: True, and therefore it is a harmonization, not a confirmation. FAIR itself calls it 'plausibility only'. A harmonization that would work equally well for a fictional text cannot count as evidence, and it fails where the text has the animals doing things the substitutes never did (herded, pulling chariots, distinguished as domestic and wild).

What they get right: As a reason not to treat an animal name as instantly fatal, it is reasonable.

Both cases at full strength

The skeptics' case

The chart answers a question nobody serious is asking (have any 19th-century criticisms turned out wrong?) and is presented as answering the one that matters (does the evidence now favor an ancient American origin?). On Roper's own scoring, the heavy items remain unconfirmed or partial: horses in the relevant period, steel, iron smelting, metallurgy before AD 600, wheat, chariots, any identified city, any Hebrew or Egyptian writing, any Christian trace. The 174 confirmations are reached by splitting (four horse items), by renaming (bison for cows), by generic traits (judges, hoes, ladders), by proposed etymologies, and by Old World details that were available in 1829. The problems that have grown since 1830 are left off. In the 77-item audit compiled for this site, of 39 topics the chart scores 'confirmed', this review rates 8 resolved, 6 not anachronisms to begin with, 10 partially resolved, 10 contested and 5 unresolved; of the 14 items that would matter most to historicity (weight 3), all 14 are unresolved.

The apologetic case

A fabricated book about ancient America, written in 1829 when the public image of Native Americans was of 'rude savages', should have aged badly. Instead, cities, temples, kings, roads, cement, fortifications, large-scale warfare, writing, a fertile Arabian coast and an Arabian toponym NHM have all turned up. Even if one throws out every loan-shift and every generic item, the direction of travel since Stephens and Catherwood (1841) has been toward the book's picture of a complex, literate, warlike civilization, and the residual problems are concentrated in perishable or rarely preserved categories.

What would change this

A replicated, directly dated (bone collagen) horse, or any smelted iron, wheat, or Old World script from a secure pre-Columbian context in the proposed lands; or a re-scored chart that weights items, records disconfirmations and includes the textual problems, and still trends upward.

Sources

An independent project. Not affiliated with The Church of Jesus Christ of Latter-day Saints or with any organization cited here. Summaries are written here; short quotations are attributed and linked. Every page names its sources so that it can be corrected, and every page can be challenged.