Rejecting a quotation from publication and erasing it from the record are two different decisions. This page is the record.
What this page is
This article's evidence came from papers and books on disk, so most of its quotations verified straightforwardly. Four items did not, and they are held here rather than deleted. Three of them are in circulation and a reader may meet them elsewhere.
Item 3 was resolved during verification on 5 September 2026 and is kept here with its resolution, because the record of a wrong verdict is worth as much as the record of a right one.
Status words.
- REFUTED — the attribution or wording is demonstrably wrong. A better copy will not help.
- UNVERIFIABLE — no route to the named source has been found yet. That is a statement about this project's reach, not about the claim.
- PRINTED WITH A FLAG — the wording is genuine and correctly attributed, but the fact it asserts could not be checked, so the article prints it as somebody's claim and says the number is unverified.
Nothing on this page was established with WebFetch. Every check below was made against downloaded bytes: curl for web pages and PDFs, pdftotext -layout for extraction, NCBI E-utilities for abstracts, and archive.org's search-inside endpoint for print-disabled books.
1. Li & Durbin's PSMC caveat, in the wording that circulates
Wording as it circulates: the PSMC reconstruction spread an instantaneous crash "over several preceding tens of thousands of years."
Claimed source: Heng Li & Richard Durbin, "Inference of human population history from individual whole-genome sequences," Nature 475:493–496, 2011. The wording is quoted this way by Richard Buggs in his 29 October 2017 post, and from there by others.
Status: UNVERIFIABLE as Li & Durbin's wording, and not printed.
What was tried and what was found. The copy in the source library is the Europe PMC author manuscript. Its text reads: "the instantaneous reduction from 12,000 to 1,200 at 100kya in the simulation was spread over the preceding 50,000 years in the PSMC reconstruction." That is a different sentence from the one Buggs quotes, and it carries a specific number where his carries a vague one. One of the two versions was edited between manuscript and print. Buggs may be quoting the typeset Nature page, which the project does not have.
What the article does instead. It quotes the author-manuscript wording, names the version it read, and prints the discrepancy in the footnote. Neither version changes the argument: both say a sudden crash is smeared across tens of millennia.
What would settle it. The typeset Nature page 495, from a library or a subscription.
2. "Only seven versions of the gene existed in the ancestral population"
Wording: "A later study by Bergström et al. examined the same HLA-DRB1 gene, but used intron 2 … this study concluded that only seven versions of the gene existed in the ancestral population from which both chimps and humans supposedly came around 4–6 million years ago, and that the population had an estimated size of 10,000 rather than the 100,000 estimated by Ayala."
Claimed source: Ann Gauger, "The Science of Adam and Eve," in Science and Human Origins (Discovery Institute Press, 2012), p. 114, citing Bergström et al. 1998.
Status: PRINTED WITH A FLAG. The quotation is Gauger's, verbatim, from the PDF in the source library, and the article prints it as her words. The number seven could not be checked against Bergström.
What was tried and what was found. Bergström et al. 1998 (Nature Genetics 18:237–242) is closed at the publisher. Unpaywall reports is_oa: false; OpenAlex reports oa_status: closed; Semantic Scholar reports the open-access PDF status as CLOSED; Europe PMC answers "Subscription required." The abstract was pulled as bytes from NCBI E-utilities and does not contain the figure. Gauger's endnote cites the 1998 paper itself, not the 1996 Science technical comment, and that comment (Erlich, Bergström, Stoneking & Gyllensten, Science 274:1552–1554) is also closed, with no abstract carried by PubMed under either of its two PMIDs.
Why it matters less than it looks. Seven is more than four. Even if the figure is exactly right, it does not fit inside a founding pair's ceiling, so the chapter's conclusion does not follow from its own number.
What would settle it. The full text of Bergström et al. 1998, from an academic library or an interlibrary loan.
3. "Each block comes in a few variants, four for many parts of the genome"
Wording: "the blocks are several thousand nucleotides long, and in spite of this there is very little variation within them. Each block comes in a few variants, four for many parts of the genome."
Claimed source: Ola Hössjer, Ann Gauger & Colin Reeves, "Genetic Modeling of Human History Part 1," BIO-Complexity 2016(3), p. 11, citing Rosenfeld, Mason & Smith 2012.
Status: RESOLVED. Printed, and the number is roughly supported.
What was tried and what was found, first pass. The sentence carries no citation. The nearest reference, [64] Rosenfeld, Mason & Smith 2012, belongs to the preceding sentence about block length and is about the limitations of the human reference genome. Gabriel et al. 2002 was the paper that would settle it, and it read as closed: Science paywalled, Unpaywall is_oa: false, OpenAlex oa_status: closed, Semantic Scholar CLOSED. On that basis the article first printed the number as unsupported.
What was found on the second route. OpenAlex also listed an author copy in the LSU Scholarly Repository, https://repository.lsu.edu/animalsciences_pubs/1898. curl gets a 403 there from a Cloudflare wall; bin/cdp-fetch.py file, which drives a headful Chrome over raw CDP, returned the 1.1 MB PDF. It is the Science Express version of 23 May 2002. On page 3 of 10 it reports an average of 5.0 common haplotypes per block in Yoruban and African-American samples, 4.2 in European, 3.5 in Asian, with each single population holding "3.1 – 4.9 haplotypes with a frequency >5%" and the union across populations holding 5.3.
So the verdict flipped, in both directions at once. Hössjer's "four for many parts of the genome" is a fair description of the middle of the measured range, and calling it unsupported was wrong. The African average of 5.0 and the cross-population union of 5.3 are both above four, which is the real difficulty, and it is a smaller and sharper one than "unsupported".
Why it matters. This is the most load-bearing number in the created-diversity model. Falsifiability is staked on human chromosomes being mosaics of four founder blocks. If the true number is five or more in a substantial fraction of blocks, the model fails at those blocks.
What is still missing. Gabriel published averages, not a distribution. A fresh count from 1000 Genomes phased haplotypes, giving for every block the number of distinct haplotypes covering 90% of chromosomes, with the distribution and its maximum, is the number that would decide the model. Nobody has published it in either direction.
The lesson, recorded on purpose. Three open-access indexes agreed this paper was closed, and all three were right about the publisher and wrong about the world. An institutional repository copy behind a bot wall is not a closed paper.
4. "Doxiadis 2012 contradicts Gauger on intron trees"
Wording as it stood in an earlier draft of this article: that Gauger's sentence "It is clear that the intron sequences group according to species, whereas exon 2 sequences show no species-dependent relationships" is "contradicted by the paper it points at", on the strength of Doxiadis et al. 2012's phrase "demonstrate the common history of certain DRB genes/lineages of humans and chimpanzees".
Status: REFUTED, and withdrawn from the article.
What was found. Read whole, the Doxiadis 2012 sentence is: "all phylogenies of different intron segments show consistently that, apart from the pseudogenes, only DRB5 genes are shared by OWM and HOM, and they demonstrate the common history of certain DRB genes/lineages of humans and chimpanzees." The subject of "they" is DRB5. The sentence says intron phylogenies mostly do separate the species, with the pseudogenes and DRB5 as the named exceptions. Neither is DRB1. Further, Gauger's endnotes 7 and 8 cite Doxiadis et al. 2008, not the 2012 paper, and the 2008 abstract says "most of the primate DRB alleles investigated represent relatively young entities, possessing species-unique sequences", which is her sentence.
Why it is recorded. This was an attack on a creationist source built by quoting the tail of a sentence whose subject was elsewhere. It is the same defect this project is meant to catch in others.