Companion to Flood Legends Around the World. Every source that was worked up is listed here with what it contributes and which way it cuts. Sources considered and not used are marked as such, with the reason.
The corpora
Isaak, Mark. "Flood Stories from Around the World." TalkOrigins Archive, 1996–2002. https://www.talkorigins.org/faqs/flood-myths.html 262 summarised stories in nine regions, each with a source citation. This is the article's working corpus. Chosen because its collecting rule is stated, permissive, and unrelated to resemblance to Genesis: "collect every flood story I find." Isaak is a critic of creationism, which is a bias worth naming, but the direction of his bias would be to exclude Genesis-like stories, and he does not — Genesis itself is in the corpus and scores 11 of 11. Two limitations he states himself and this article repeats: the entries are English summaries, so a compressed-out motif is invisible; and some entries merge several tellings from one culture, which inflates that entry. Extracted to research/isaak-corpus.json. Cuts both ways — it is the instrument, not a witness.
Liguori, Nick. Echoes of Ararat: A Collection of Over 300 Flood Legends from North and South America. Master Books, 2021. Serialised free by Answers in Genesis as "Flood Legends from the Americas," Parts 1–3 (13, 20, 27 April 2021). Used for the modern creationist case, and it is the best version of it. Liguori is a civil engineer working from primary ethnography — Catlin, Schoolcraft, Curtis, Dorsey, Kroeber, Bancroft, Clark, Bureau of American Ethnology bulletins, mission journals, nineteenth-century travel accounts — not from a chain of apologetics quotation. He has found real material that most compilations miss, and Frazer would have been glad of it. Two problems: (1) the interpretive frame is applied hard — a released rabbit becomes "clearly a memory of Noah's dove," "huge, serpentine sea and land animals" become "a reference to dinosaurs," and a chief who splits the earth becomes "catastrophic plate tectonics"; (2) the claim that these traditions predate missionary contact is not defended, and for the Americas it is not defensible. The book itself was not obtained — in print, not on archive.org, not on hoopla, no digital lending copy located. The three AiG articles are the accessible substitute and are substantial. Cuts for the general claim, against it on inspection of the strongest single datum (see Shoemaker below).
Nelson, Byron C. The Deluge Story in Stone. Augsburg, Minneapolis, 1931. https://archive.org/details/delugestoryinsto0000byro_a4b6 The origin of the percentages, traced here for the first time in this project. Appendix I, "The Significance of Flood Traditions," pp. 165–170, plus Fig. 38 on p. 169. Lending-only scan; read via archive.org's search-inside API (bin/iasearch.py), which returns paragraphs with raw leaf indices. Leaf-to-page offset of 20 established from the book's own index, which gives "Flood Traditions, The Significance of, 165" (Appendix I opens at leaf 185) and "Appendix I. chart, 169" (the chart caption is at leaf 189) — two independent anchors agreeing. Fig. 38 itself is an image and its cells were not read; what was read is Nelson's prose about it, which is what matters. Cuts against the percentages and for the qualitative argument: his four objections to the missionary hypothesis are better than their reputation, and objection 3 is simply correct.
Frazer, James George. Folk-Lore in the Old Testament, vol. 1, ch. IV "The Great Flood," pp. 104–361. Macmillan, 1918. https://archive.org/details/folkloreinoldtes01fraz Public domain, full OCR text on disk. The foundation compilation: every flood tradition then known, with citations to the ethnographers who collected each. Used for four things: (1) the flood traditions themselves; (2) the Africa denial at p. 333 — now known to be too strong; (3) the contamination list at p. 337 n. 2; (4) documented contamination cases and the Im Thurn and Loftus material. Frazer's own verdict, p. 335, is that apart from the Hebrew story and modern missionary-influenced instances, there are no decisive grounds for deriving the world's flood traditions from the Babylonian. He is thus a witness against the simple diffusion story in either direction — against Genesis-as-source and against Babylon-as-source. Page locators verified by reconstructing the running heads from the OCR; see research/ for the method.
The quantitative and phylogenetic literature
d'Huy, Julien. "At the Origin of Flood Mythologies: Synthesis of Three Papers." RMN Newsletter 15–16 (2020–21), pp. 70–76. Open access. The most important secular source for this article. Phylogenetic analyses (Mesquite, PAUP*, parsimony and maximum likelihood) of the flood section of Berezkin's motif catalogue. Key results used: the reconstructed centre of the flood tradition is Southeast Asia, not the Near East; the tree structure matches a double peopling of the Americas; retention index 0.52 indicates partial package transmission; motif C5A "Bird-scouts" is an independent global motif present in Australia. His statement that "most of the myths of the flood around the world do not appear to be related to the Christian or Muslim traditions" is quoted in the article because it corrects a common skeptical overreach. Cuts against a Genesis-centred diffusion, for the claim that these stories are genuinely old and genuinely related.
Thuillard, Marc; d'Huy, Julien; Berezkin, Yuri E.; Le Quellec, Jean-Loïc. "A Large-Scale Study of World Myths." Trames 22(4) (2018), pp. 407–424. Open access, on disk. Background on the Berezkin database's scale (2,264 motifs, 934 peoples, ~10,000 sources read) and the methods. Consulted, not cited in the article — it supplies context rather than a load-bearing claim.
Witzel, Michael E. J. "Pan-Gaean Flood myths: Gondwana myths — and beyond." In New Perspectives on Myth, PIP-TraCS 5 (2010), pp. 225–242. Open access via Harvard DASH. Used for one thing, but an important one: p. 234 n. 15, reporting that Baumann's 1936 survey criticised "the then (as today) prevailing opinion that the flood myth is hardly found in Africa" and found the myth "basically spread, in pockets, all over sub-Saharan Africa." This is what corrects Frazer. Witzel's own Laurasian/Gondwana framework is a large and contested theory that this article does not adopt or need. Cuts against Frazer's Africa denial, against the article's own case only mildly, since the African material remains structurally unlike Genesis.
Berezkin, Yuri E. and Duvakin, E. N. World Mythology and Folklore: Thematic Classification and Areal Distribution of Motifs — Analytical Catalogue. http://ruthenia.ru/folklore/berezkin The underlying database for d'Huy's work. Not queried directly. Motif definitions (C2, C4, C5A, C6A, C6C, C6D, C8, C19, C32, C34) are taken from d'Huy's quotations of it. A direct query would be the obvious next step for anyone extending this article, because the catalogue would allow the regional distribution to be recomputed on a corpus twenty times larger than Isaak's.
Le Quellec, Jean-Loïc. Avant nous le Déluge! L'humanité et ses mythes. Éditions du Détour, 2021. Not obtained — French, in print, no digital copy located through Evergreen Indiana, hoopla, Libby or archive.org. Cited in the article only at second hand through d'Huy, and flagged as such. It is reportedly the fullest recent treatment of the African question and would strengthen or weaken §"The Africa problem" more than anything else on this list. Open request.
Oral tradition and real events
Nunn, Patrick D. and Reid, Nicholas J. "Aboriginal Memories of Inundation of the Australian Coast Dating from More than 7000 Years Ago." Australian Geographer 47(1) (2016), pp. 11–47. Full text obtained. 21 Aboriginal stories describing named coastal features as formerly dry, each matched against the Australian postglacial sea-level curve, with minimum and maximum ages in their Table 2. The single strongest piece of evidence in this whole field for the durability of oral memory, and it comes from geography and linguistics, not apologetics. Cuts for the principle behind the flood-legend argument; cuts hard against the young-earth chronology, because the minimum ages run 7,250–11,730 cal BP. Both halves are stated in the article.
Mesopotamia and candidate real floods
Kramer, Samuel Noah. "Reflections on the Mesopotamian Flood." Expedition 9(4) (1967). Free at penn.museum. Used for the archaeology: Woolley's Ur silt at ≥3500 BC is too early for Ziusudra; Kish (Langdon & Watelen, 1928) and Shuruppak (Schmidt, 1931) have flood layers above Jemdet Nasr material c. 3200–3000 BC; Mallowan's 1964 Iraq review concludes for a real but "by no means universal" disaster around 3000 BC. Cuts against a global flood and against Woolley; cuts for a real Mesopotamian event behind the Mesopotamian story.
Mallowan, M. E. L. "Noah's Flood Reconsidered." Iraq 26 (1964), pp. 62–83. Not obtained directly — Cambridge Core paywall; JSTOR holdings not checked against the free account. Cited through Kramer 1967, who summarises its conclusion. Worth getting if this article is developed further, since it is the standard reference for the flood-layer stratigraphy.
Wu Qinglong et al. "Outburst flood at 1920 BCE supports historicity of China's Great Flood and the Xia dynasty." Science 353 (2016), pp. 579–582, with three technical comments in Science 355 (2017), p. 1382 (Han; Wu Wenxiang et al.; Huang et al.). Abstract and comment metadata verified through Crossref; full texts not obtained. Cited in the article strictly as a contested proposal.
Ryan, W. B. F. et al. "An abrupt drowning of the Black Sea shelf." Marine Geology 138 (1997), pp. 119–126, against Giosan, Filip & Constatinescu, "Was the Black Sea catastrophically flooded in the early Holocene?" Quaternary Science Reviews 28 (2009), pp. 1–6. Citations verified through Crossref; neither full text obtained. Mentioned in one sentence and flagged as contested; nothing in the article rests on it.
The Chinese-characters claim
Xu Shen 許慎, Shuowen Jiezi 說文解字 (c. AD 100), 舟部, entry 船. Read at the Chinese Text Project, https://ctext.org/dictionary.pl?if=en&char=%E8%88%B9, with the page bytes downloaded and the characters read locally. 「船：舟也。从舟，鉛省聲。」 This is the decisive source and it is free. A Chinese lexicographer of about AD 100 analyses the character as boat-radical plus phonetic. Corroborated on the same page by the Guangyun fanqie and the Jiyun's reading 音沿. Cuts decisively against the "vessel for eight mouths" argument.
Batten, Don. "CMI misrepresents ancient Chinese language?" Creation Ministries International, 2006. CMI's published defence. Reproduced fairly in the article, including its three grounds, none of which is historical. CMI has not retracted the argument; its 2006 "Chinese Characters and Genesis" page still runs it. Cuts against — the strongest available statement of the creationist case here does not engage the primary source that refutes it.
Nelson, Ethel; Broadberry, Richard; Tong Chock, Ginger. God's Promise to the Chinese (1997); Nelson, Broadberry & Wang, The Beginning of Chinese Characters (2001). Not obtained. These are the works CMI points to. Getting them would let the argument be answered on its own terms rather than through CMI's summary, but the Shuowen settles the specific claim about 船 regardless of what they say.
The live apologetic sources
Apologetics Press, "Legends of the Flood." Carries the percentage block, attributed to James Perloff, Tornado in a Junkyard (1999), p. 168. Also carries the 船 claim and the assertion that "the legends were recorded long before any missionaries arrived." Used as the citable form of the live claim.
Perloff, James. Tornado in a Junkyard: The Discovery of Design. Refuge Books, 1999, p. 168. Not obtained — the intermediate link in the chain. Its content is known from Apologetics Press's verbatim quotation and its dependence on Nelson 1931 Fig. 38 is reported by secondary sources. Open item: confirm at primary that Perloff's chart credit reads "adapted from B.C. Nelson, The Deluge Story in Stone, Appendix 11, Flood Traditions, Figure 38." Nothing in the article depends on Perloff being the link rather than some other populariser; what matters is that the numbers are Nelson's, and Nelson says what his sample is.
Answers in Genesis, "Flood Legends" hub; Institute for Creation Research, "Why Does Nearly Every Culture Have a Tradition of a Global Flood?" Surveyed for the current state of the argument. Neither adds a datum the article uses. ICR's article URL redirects; the intended text was not retrieved and nothing rests on it.
Osanai, Nozomi. A Comparative Study of the Flood Accounts in the Gilgamesh Epic and Genesis. AiG, 2005. Considered and not used here. It belongs to the Genesis-versus-Gilgamesh dependence question, which this project treats in Noah's Ark: The Logistics and in the noahs-ark-logistics memory file, where the primary texts (Lambert & Millard's Atra-hasīs, George's Babylonian Gilgamesh Epic, the eBL edition of Gilgamesh XI, Finkel's Ark Tablet, Berossus) are already worked up with verified quotations. Duplicating it here would produce two divergent treatments of the same texts.
Provenance of the Iroquois anniversary
Shoemaker, Henry Wharton. "History of Tamarack Swamp." Reading Times, 6 January 1912. Liguori's source for the Iroquois 4,000-year commemoration. The column itself was not read — it is a 1912 Pennsylvania newspaper and was not located in an accessible archive. The article's argument against the datum is therefore internal (an absolute anniversary in Ussher's reckoning is not oral-traditional material) with provenance as corroboration, and it says so.
Bronner, Simon J. Popularizing Pennsylvania: Henry W. Shoemaker and the Progressive Uses of Folklore and History. Penn State Press, 1996. Not obtained; known through its publisher's description, the Penn State Libraries' finding aids, and secondary summaries. The consistent report is that Shoemaker's collecting was attacked in his own lifetime by trained folklorists, that Richard Dorson used him as the type case of "fakelore," and that Bronner finds him more concerned with picturesqueness and political usefulness than authenticity. This is stated in the article as provenance evidence and explicitly not as proof of fabrication in this particular column, because the column was not read. Getting Bronner, and getting the Reading Times page, would settle it.
What is missing
Ordered by how much they would change the article.
- Le Quellec 2021. The Africa section rests on Baumann-through-Witzel and on Isaak's seventeen entries. Le Quellec has done the count properly and in a language the project can read with effort.
- A provenance-coded corpus. The contamination correlation reported in the article is partly circular because Frazer's flags were partly resemblance-based. Coding every story in Isaak's corpus for date of first recording, recorder's profession, and years of mission presence in that region would make the test clean. This is a bounded job — 262 stories, all with citations — and nobody appears to have done it.
- Berezkin's catalogue, queried directly. Would allow the regional and motif analysis to be redone on ~900 peoples instead of 262 stories.
- Liguori's Echoes of Ararat, as data. If its 300-plus American traditions were released with their motifs coded, the recount could be repeated on the creationist corpus and the two answers compared. That comparison is the single most useful experiment available in this subject.
- Mallowan 1964 at primary, for the flood-layer stratigraphy.
- Perloff 1999 p. 168, to close the citation chain at both ends.