The positive 'evidences' apologists lead with

Chiasmus, Hebrew-looking names, cement and similar 'hits' are real talking points, but the apologists' own reference works concede how soft several of them are.

What the Church claims

The Book of Mormon contains ancient Near Eastern features (literary forms, names, details) that Joseph Smith could not have known in 1829 and that therefore confirm its antiquity.

Once the evidence is in, this claim: Contested. Several items on the list are real and are not easily waved off, Nahom and the short chiasms among them. But the method that produces the list cannot separate a genuine ancient signal from the hits any long text yields against a large corpus, and the misses are never counted against it. The claim is unsettled rather than established.

The record

  • Edwards and Edwards (BYU Studies, 2004) calculate P = 0.00018 that Alma 36's eight-element chiasm arose by chance, and P = 0.52 and 0.98 for proposed chiasms in the Doctrine and Covenants and Book of Abraham.
  • Earl Wunderli (Dialogue, 2005) shows Welch's Alma 36 scheme uses two of six occurrences of 'deliver', omits other repeated words, and pairs sections of 213 and 48 words.
  • BYU's Onomasticon concedes that the Bar Kokhba 'Alma ben Yehudah' has been re-read as Aramaic 'allima' ('strong') and that its etymology 'is not dependent' on that document.
  • The Onomasticon notes that the Akkadian reading behind 'sheum' is 'now rejected' by Assyriologists and that the word would have to reach the Nephites through the Jaredites.
  • The Mulek/Malkiyahu proposal rests on reading 'son of Hammelech' (Jer. 38:6) as 'son of the king'; the Onomasticon concedes the phrase may be a title and that the name was printed 'Muloch'/'Mulok' until 1879.
  • Lime plaster and structural concrete are well attested in Mesoamerica and expand from about 100 BC, roughly matching Helaman 3:7.

The moving target

The Church has never published a list of Book of Mormon evidences, and its trend has been away from doing so. In the 1950s and 1960s it funded the New World Archaeological Foundation in the expectation of confirmation; since the 1990s its statements have said instead that the book is a religious record whose truth is known by the Spirit and not settled by scholarship. The evidences below are collected by independent organizations, and the Church's position is that the reader does not need them.

The exchange

John W. Welch (1967/1969); Boyd and Farrell Edwards

Chiasmus: Alma 36 is a long, deliberate inverted parallelism centered on Christ. The form was unknown to Joseph Smith, and statistics show it is not accidental.

The answer: The statistic tests the wrong thing. It asks whether a given eight-element pattern would arise from random ordering once the elements are chosen, not whether an analyst free to choose which repeated words count could find such a pattern in repetitive prose. Wunderli documents exactly that freedom in Welch's Alma 36. Chiasmus is also a natural product of oral, sermonic, KJV-saturated speech; simple chiasms occur throughout English literature. The Edwards paper's finding that D&C chiasms are chance-level actually cuts against the claim that Joseph's revelatory texts are chiastic by design.

What they get right: Short, tight chiasms (for example Mosiah 5:10-12) are genuinely present and the book's structural complexity is greater than critics in 1830 assumed.

Hugh Nibley; Paul Hoskisson, Stephen Ricks, John Tvedtnes, Jeffrey Chadwick; BYU Onomasticon; Scripture Central

Names: Alma is attested as a Jewish male name; Mulek matches Malkiyahu son of King Zedekiah; Jershon means 'place of inheritance' and is punned on in Alma 27:22; sheum is an authentic ancient grain word.

The answer: This is a base-rate problem. The book introduces a large stock of new proper names; Semitic languages build words on short consonantal roots with flexible vowels; and the search covers Hebrew, Aramaic, Egyptian, Akkadian, Sumerian, Arabic and Eblaite across three millennia. Under those rules matches are guaranteed, and misses are never counted. The reference work built by the apologists themselves concedes that the flagship Alma document has been re-read, that the sheum reading is rejected by Assyriologists, and that the Mulek link is unproven. Jershon is the best of the group because the wordplay is in the text, but 'inherit' language is also thick in the KJV Joseph knew.

What they get right: Critics who once mocked 'Alma' as a Latin woman's name were wrong that it could not be a Semitic male name. Jershon is a neat fit.

Scripture Central; Matthew Roper

Details once mocked have been vindicated, such as cement in Helaman 3.

The answer: Mesoamerican lime cement is real, and that criticism was poor. But the text describes whole cities of cement houses built because timber was scarce, in a land northward, by people who also had Old World crops, animals and steel; a vindicated detail does not vindicate the package. The scoring is also one-sided: see the separate anachronisms entry for the items that have not moved.

What they get right: Some nineteenth- and early twentieth-century criticisms have not aged well, and skeptics should retire them.

Both cases at full strength

The skeptics' case

None of the positive evidences is the kind of evidence historians use to establish that an ancient people existed: a site, an inscription, a datable artifact, an external reference. They are literary and linguistic parallels, and parallels are cheap when the search space is huge and only hits are reported. The strongest literary claim, Alma 36, depends on selective pairing; the strongest name claims are hedged or withdrawn in BYU's own Onomasticon. What remains is suggestive at most, and it has to be weighed against the systematic absences documented under archaeology and DNA.

The apologetic case

Individually weak parallels can be cumulatively significant. A dictated text that contains working chiasms, plausible Semitic wordplay, a correct South Arabian toponym and a correct guess about Mesoamerican cement is doing better than an 1829 frontier fabrication should, and critics have no detailed account of how Joseph produced each of these.

What would change this

A pre-registered test: specify in advance the languages, periods and matching rules for names, then count hits and misses against a control text such as Tolkien or the Book of Jasher forgeries.

Closely related

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.