Material worked up for Luminescence Dating and not printed in the article as fact. Nothing here is deleted. Rejecting a claim from publication and erasing it from the record are different decisions.
Everything on this page falls into one of two classes: a number whose source was never read, or a locator the evidence on disk will not support. There are no fabricated quotations here and no quotation whose wording failed verification. Every quotation the article prints was matched against bytes on disk or bytes downloaded and read in the writing session.
"OSL dating is accurate to about 10%"
Where it comes from. Murray, Arnold, Buylaert, Guérin, Qin, Singhvi, Smedley and Thomsen, "Optically stimulated luminescence dating using quartz," Nature Reviews Methods Primers 1:72 (2021), section "Overall accuracy," which combines three earlier compilations into its Figure 11b and reports 249 ages accurate "over a time span of more than four orders of magnitude." The 10% figure itself traces to the first of those compilations, Murray & Olley, "Precision and accuracy in the optically stimulated luminescence dating of sedimentary quartz: a status review," Geochronometria 21:1-16 (2002).
What was tried. The Geochronometria online archive at geochronometria.com, whose holdings appear to begin around volume 26 (2007); geochronometria.pl, whose PDF paths 404; CORE, which 403s; BASE, which returned no PDF. ResearchGate holds a copy behind a login. The two other compilations, Madsen et al. 2005 (Marine Geology 214:251-268) and Rittenour 2008 (Boreas 37:613-635), were not attempted at all.
How the article handles it. It quotes Murray et al. 2021's summary sentence verbatim, states in a footnote that none of the three underlying compilations was obtained, and says the accuracy claim rests on that summary rather than on the papers themselves. This is the largest single gap in the article's evidence.
The percentages inside Murray et al. 2021 Figure 11b
What is missing. The caption for Fig. 11b states that it plots "n = number of samples, followed by the % of samples within ± 2σ of the line of unit slope." Those percentages are drawn inside the figure, not written in the text, and the version read here is the HAL accepted manuscript rather than the published article.
How the article handles it. No percentage from Fig. 11b appears anywhere in the article. A later agent with the published version should add the real numbers, because they would let the article state OSL's hit rate against independent control as a measurement rather than as a summary phrase.
The water, beta-attenuation and cosmic-dose constants used in research/dose-rate.ts
The three constants. Water attenuation factors of 1.50 alpha, 1.25 beta and 1.14 gamma (Zimmerman 1971, Archaeometry 13:29-52, as standardised by Aitken 1985). Beta attenuation of about 0.94 for etched 90-125 µm quartz (Mejdahl 1979, Archaeometry 21:61-72). A cosmic-ray term of about 0.15 Gy ka-1 at 2 m burial (Prescott & Hutton 1994, Radiation Measurements 23:497-500).
What was tried. None of the three originals was attempted. Both Aitken textbooks were chased on archive.org by title, by creator:Aitken, by full-text search and through OpenLibrary; no scanned item exists for either, so the search-inside route for print-disabled books does not apply and the article does not cite Aitken as a source it has read.
How the article handles it. A footnote names all three as second-hand and says so. The Guérin et al. 2011 conversion factors, which carry the weight in the dose-rate arithmetic and set every number in the article's tables, were read from the source PDF.
The 2.4 mGy per year world-average background dose
Where it comes from. Attributed to the UNSCEAR 2008 report; used in research/dose-rate.ts only to express the required sediment dose rates on a familiar scale.
What was tried. Nothing. The figure was carried over from the research dossier, which flagged it as general knowledge.
How the article handles it. A footnote says the report was not consulted and that nothing in the argument turns on the exact value. The conclusion drawn from it, that a 4.8x-background environment is survivable, would not change for any plausible value.
The page number for the Malakunanja II Conclusions quotation
The problem. The companion literature review places the "correct burial ages can be obtained using TL methods" quotation on p. 24 of Ancient TL 16(1). The Ancient TL PDF is a page scan with no text layer, and the tesseract output interleaves the article's two columns: running heads appear at output lines 1, 162, 221, 385, 441 and 582 while folios appear as 19, 20, 20, 21, 22, 23, 24. Depending on which sequence is trusted, the quotation lands on p. 22, p. 23 or p. 24.
How the article handles it. It cites the passage by section name, "Conclusions," and claims no page. A right quote at a wrong page is the commonest citation defect in this corpus, and a guess would have been one. A later agent with the printed journal, or with a cleaner scan, should pin it.
Whether any ICC paper discusses luminescence in its body text
What was established. All nine International Conference on Creationism proceedings listing pages were harvested, giving 474 papers from 1986 to 2023. No paper title contains luminescence, thermoluminescence, zircon or electron spin. The CedarCommons repository search, which reaches abstracts, returns exactly one record for thermoluminescence: Silvestru 2003, whose treatment is a single parenthesis.
What was not established. Nobody has grepped the 474 PDFs themselves. A paper could discuss luminescence in its body without saying so in its title or abstract.
How the article handles it. It states the harvest and its scope precisely, and does not claim that no ICC paper mentions the method.
creation.com
What was tried, 30 August 2026. https://creation.com/search?q=thermoluminescence and /search-legacy?q= both 302 to a /en/ locale path. https://creation.com/en/search?q=thermoluminescence returns HTTP 200 and about 164 KB of Nuxt application shell behind Cloudflare, with the query echoed into the search <input value> and no result list anywhere in the HTML. A headful Chrome render through bin/cdp-fetch.py did not produce a result list either.
How the article handles it. It says plainly that creation.com is unaudited. It does not treat the failure to reach the search as evidence that creation.com has published nothing on the topic.
What would resolve it. The Nuxt bundle at /_nuxt/*.js calls a search API from the browser. Reading that bundle for the endpoint and calling it directly is the untried route.
Two book-length creationist treatments
John Woodmorappe, The Mythology of Modern Dating Methods (ICR, 1999), which has a chapter on non-radiometric methods, and Marvin Lubenow, Bones of Contention (Baker, 1992/2004), which discusses ESR and TL at hominin sites. Neither was attempted. Until one of them is read, the article's statement that the technical young-earth engagement with luminescence is close to nil should be read as covering the conference and journal literature, which was searched, and not the popular book literature, which was not.
Olley et al. 2006 on the Lake Mungo grave infill
Quaternary Science Reviews 25:2469-2474. This is the paper that most directly closes the Lake Mungo question, by dating the sediment that actually filled the graves rather than sediment nearby. Not obtained, and not attempted beyond a note. The article therefore claims nothing about Lake Mungo beyond what the Thorne et al. 1999 and Bowler et al. 2003 abstracts say, both of which are on disk, and cross-references radiometric concordance for the rest.