The literature

This page exists so that nobody has to take the rest of the article on trust.  Every source it rests on is listed here with what it actually measured, who wrote it, how it was obtained, and which way it cuts.

A catalogue of the literature on luminescence dating — optically stimulated luminescence (OSL), thermoluminescence (TL), infrared-stimulated luminescence on feldspar (IRSL and pIRIR), and electron spin resonance (ESR) where it shares the same trapped-charge physics. It covers the method's own foundational papers, the two famous Australian test cases (Jinmium and Lake Mungo), the known-age validation studies, the failure modes the field itself publishes, and the young-earth engagement with all of it.

It is deliberately not a reading list of things that agree with this site. Sources are tagged by the direction they cut, and a reader who wants only the strongest material against the young-earth argument can filter for undercuts and read those. On this topic most of the strong material is undercuts, and that is stated rather than hidden.

How to read an entry

Measurement or model. The most important distinction. An equivalent dose measured on 800 quartz grains is an observation. A "lifetime average water content" is an assumption, and the age depends on it linearly. A dose rate computed from gamma spectrometry plus published conversion factors is a measurement passed through a small model. Every entry says which it is.

Both terms are measured on the sample. This is what makes luminescence unlike most geochronology. The numerator (the accumulated dose, D_e, in grays) is measured by stimulating the actual grains and counting photons. The denominator (the environmental dose rate, in grays per thousand years) is measured from the actual sediment's uranium, thorium and potassium. Neither is inherited from a global calibration. That matters for any argument about whether the clock could have run at a different speed.

Access route. Stated for every entry, because "I could not find it" is not "it does not exist," and because the next agent should not repeat a failed route. Where a source was OCRed from a page scan, that is said, along with the resolution and tool.

Direction. Whether the source supports, undercuts, or complicates the young-earth argument. Assigned by what the source shows, not by what its author believes. Several mainstream papers are tagged supports because they document real, large, published errors in the method.

Camp. Mainstream, creationist, old-earth Christian, counter-apologetics, or unaffiliated. Disclosure, not verdict.

Field definitions

Method mechanics and constants

Guérin, Mercier & Adamiec 2011 — Dose-rate conversion factors: update

Ancient TL 29(1):5-8. DOI 10.26034/la.atl.2011.443. CC BY 4.0. topic: dose-rate · camp: mainstream · type: measurement · access: OA · direction: background

What it measured. Nothing in the field. It recomputes, from the Brookhaven National Nuclear Data Center's Evaluated Nuclear Structure Data Files as downloaded 5 November 2009, the energy released per disintegration in the ²³²Th, ²³⁸U and ²³⁵U chains and by ⁴⁰K and ⁸⁷Rb, and converts those into dose rates per unit parent concentration. These are the constants every luminescence age in the world is divided by.

The numbers this project uses (Tables 1-3, infinite matrix, secular equilibrium, Gy ka^-1):

The paper also states the ⁴⁰K half-life used, 1.248 Ga, and notes it is "2.3% lower than previously published" following Grau Malonda & Grau Carles 2002 and Kossert & Günther 2004. Differences from the previous standard (Adamiec & Aitken 1998) are 0.5-3.5 % depending on chain and radiation type — the paper's own summary is that the update "may reach a few percent on the final obtained age calculation."

Access route that worked. https://www.ancienttl.org/article/download/6707/5817. Ancient TL is fully open access under CC BY and is the single best free route into this literature. Its own search endpoint is https://www.ancienttl.org/search/search?query=...; note that the search result page lists article IDs in document order, not the order a naive sort -u produces, which caused one wrong PDF to be fetched here before the mapping was checked.

Direction. Background. Neither side disputes these constants. They matter because they are the denominators in every number in the article, and because they let the arithmetic be redone from scratch rather than quoted.

Murray, Arnold, Buylaert, Guérin, Qin, Singhvi, Smedley & Thomsen 2021 — Optically stimulated luminescence dating using quartz

Nature Reviews Methods Primers 1:72. DOI 10.1038/s43586-021-00068-5. topic: review, mechanism, saturation · camp: mainstream · type: review · access: repository-AM · direction: mixed

What it is. The current authoritative statement of the method, written by eight of the people who built it, including Andrew Murray (co-author of the SAR protocol) and Kristina Thomsen (pIRIR). It is the right primary for any general claim about how OSL works, what it can do, and where it fails.

What it concedes, in its own words. From the "Outlook" section: laboratory dose response curves "often increase to large doses (hundreds of Gy), but natural doses either do not increase above ~200 Gy, or if they do, they often appear to be an underestimate." And: "Limits on the upper dose range of the OSL signal are also poorly understood." And, on the physics: "The controls on luminescence saturation with dose are poorly understood." The realistic floor on accuracy is stated as "the probable minimum of ~5%."

On water content, Box 1: "a 1% increase/decrease in lifetime average water content typically leads to a 1% increase/decrease in derived age." The worked example there takes a measured 9 % water against a saturation value of 21 %, adopts 15 ± 3 % for the lifetime average, and calls the resulting age uncertainty "~3%."

On overall accuracy, section "Overall accuracy": the performance of quartz OSL "to date samples from a variety of deposition environments, each with independent age control, has been evaluated by others" — the three compilations are Murray & Olley 2002 (Geochronometria 21:1-16), Madsen et al. 2005 (Marine Geology 214:251-268) and Rittenour 2008 (Boreas 37:613-635) — "These data sets are combined in Fig. 11b, for 249 ages there is now a considerable body of data suggesting that OSL dating is accurate over a time span of more than four orders of magnitude."

Access route that worked. The accepted manuscript is deposited on HAL at insu-03418831. HAL sits behind an Anubis proof-of-work wall that returns a "Making sure you're not a bot!" page to plain curl; bin/anubis-fetch.py https://insu.hal.science/insu-03418831/document out.pdf solved it in 15 ms and returned the 124-page PDF. Caution on locators: the deposited version is the accepted manuscript with its own line numbering, not the published pagination. Cite it by section heading, not by page, unless the published version is checked.

Direction. Mixed. It is the strongest available demonstration that the method works over four orders of magnitude, and it is also the most quotable mainstream admission that the high-dose end of the method is not understood.

Riedesel 2025 — Not fade away: the persistence of fading in feldspar luminescence

Ancient TL 43(1). CC BY 4.0. topic: feldspar-fading · camp: mainstream · type: review · access: OA · direction: mixed

What it establishes. Feldspar luminescence leaks. The abstract states that fading "has been shown to be ubiquitous in feldspars, affecting both thermoluminescence as well as optically (infrared) stimulated luminescence signals," that the accepted explanation is quantum-mechanical tunnelling from the trap's ground state to a nearby recombination centre, and that "fading causes an underestimation of the luminescence age." The introduction notes that measurements "all over the globe have shown the ubiquitous nature of this process ... questioning the validity of the term 'anomalous' fading."

The correction. The fading rate is the g-value, "the percentage of luminescence lost due to fading per decade of time," fitted from repeated regenerative doses measured after different storage delays (Auclair et al. 2003, eq. 1 in this review). The standard correction is Huntley & Lamothe 2001; a second model, Kars et al. 2008, uses a tunnelling parameter rho-prime and can build a natural dose response curve.

Access route. https://www.ancienttl.org/article/download/7620/8205. Text layer present, no OCR needed.

Direction. Mixed. A young-earth reader should note that feldspar ages require a laboratory-measured correction whose model has known pathologies (the review shows the Huntley-Lamothe form going negative at long times for large g). A skeptic should note that the field discovered, quantified, published and corrected this itself, and that quartz OSL — the workhorse — does not fade.

The two Australian test cases

Fullagar, Price & Head 1996 — Early human occupation of northern Australia: archaeology and thermoluminescence dating of Jinmium rock-shelter

Antiquity 70:751-773. topic: case-jinmium · camp: mainstream · type: measurement · access: NOT OBTAINED (full text) · direction: supports

The claim. TL ages of 50-75 kyr for quartz sand associated with buried pecked cupules, and 116-176 kyr for the underlying artefact-bearing deposits. That would have roughly doubled the accepted antiquity of human occupation of Australia, and it was reported worldwide as such.

Access. The full text was not obtained. Routes tried: Cambridge Core article page (paywalled, "Article purchase — Temporarily unavailable"); OpenAIRE (record only, DOI 10.1017/s0003598x00084040); Semantic Scholar (openAccessPdf: CLOSED); scholar.archive.org (no full-text copy); Flinders and Griffith repositories (metadata only). The pagination and the age ranges are confirmed from two independent citing sources: Spooner 1998's reference list (Antiquity 70: 751-73) and the Roberts et al. 1998 Nature abstract, which states the ranges verbatim.

Direction. Supports — this is the young-earth argument's best single exhibit, and it was produced by mainstream archaeologists in a mainstream journal.

Spooner 1998 — Human occupation at Jinmium, northern Australia: 116,000 years ago or much less?

Antiquity 72(275):173-178. DOI 10.1017/S0003598X00086361. topic: case-jinmium · camp: mainstream · type: measurement · access: abstract-only · direction: mixed

What it did. Nigel Spooner re-assessed one of the later dates in the Jinmium sequence and, in Antiquity's own editorial summary of the piece, "His interpretation calls into doubt some of the earlier claims." The editorial note also concedes the general point plainly: "some TL dating is notoriously difficult to interpret."

Access route. Cambridge Core renders the abstract and the full reference list to a browser but not to curl; bin/cdp-fetch.py page <url> followed by cdp-fetch.py eval 'document.body.innerText' returned both. The reference list is itself valuable — it is where Prescott & Purvinskis 1991 and Murray, Olley & Caitcheon 1995 were found.

Direction. Mixed. It is a mainstream correction of a mainstream error, published nine months after the error.

Roberts, Bird, Olley, Galbraith, Lawson, Laslett, Yoshida, Jones, Fullagar, Jacobsen & Hua 1998 — Optical and radiocarbon dating at Jinmium rock shelter in northern Australia

Nature 393:358-362. DOI 10.1038/30718. topic: case-jinmium, single-grain · camp: mainstream · type: measurement · access: abstract-only (on disk) · direction: mixed

The verdict, verbatim from the abstract (already on disk at sources/web/roberts-1998_optical-radiocarbon-dating-jinmium.html):

"Claims for ancient rock art and an early human presence at this site were based on thermoluminescence ages of 50–75 thousand years (kyr) for quartz sands associated with buried circular engravings (pecked cupules) and on thermoluminescence ages of 116–176 kyr for the underlying artefact-bearing deposits. Here we report substantially younger optical ages for quartz sand ... Using conventional (multiple-grain) optical dating methods, we estimate that the base of the deposit is 22 kyr. However, dating of individual grains shows that some have been buried more recently. The single-grain optical ages indicate that the Jinmium deposit is younger than 10 kyr. This result is in agreement with the late-Holocene ages obtained for the upper two-thirds of the deposit from radiocarbon measurements. We suggest that some grains have older optical ages because they received insufficient exposure to sunlight before burial. The presence of such grains in a sample will cause age overestimates using multiple-grain methods, whether using thermoluminescence or optical dating."

Note the size of the correction: 116-176 kyr down to under 10 kyr, a factor of at least twelve, and the intermediate multiple-grain optical estimate of 22 kyr was itself still more than twice too old. Note also the last sentence, which is a general warning about the method and not a special plea about one site.

Access. Abstract verbatim from the Nature page; full text paywalled. Not obtained. Routes tried: Nature (subscription), Semantic Scholar (CLOSED), scholar.archive.org (no copy), OpenAIRE (DOI only), UOW Research Online (Cloudflare 202 to curl).

Direction. Mixed, and it is the most important single entry here. It records the error and the correction and the mechanism, all inside two years and all in the mainstream literature.

Roberts, Galbraith, Olley, Yoshida & Laslett 1999 — Optical dating of single and multiple grains of quartz from Jinmium rock shelter, northern Australia: Part II, results and implications

Archaeometry 41(2):365-395. DOI 10.1111/j.1475-4754.1999.tb00988.x. (Part I is Galbraith, Roberts, Laslett, Yoshida & Olley, same volume, DOI 10.1111/j.1475-4754.1999.tb00987.x.) topic: case-jinmium, single-grain · camp: mainstream · type: measurement · access: abstract-only · direction: mixed

Abstract, verbatim:

"Quartz sediments from the floor deposit at Jinmium rock shelter have been investigated using the multiple-grain and single-grain optical dating methods described by Galbraith et al. (1999, this volume). Here we present the results of this dating programme and argue that the artefact-bearing sediments were deposited within the last 10000 years. This time interval is consistent with the radiocarbon chronology but is much younger than previous claims for initial human occupation during the Pleistocene. Analysis of individual grains revealed also that the characteristic saturation doses of some grains are unusually high, which may permit dating of deposits older than a few hundred thousand years. Such grain-to-grain differences raise doubts, however, about the validity of using multiple-grain samples to investigate the phenomenology of quartz luminescence."

Access route. Wiley's article page is JavaScript-rendered and returns a bare <title> to curl; bin/cdp-fetch.py page followed by eval 'document.body.innerText' returned the abstract, authors, volume, issue and page range. Pagination correction: several secondary sources give this as Archaeometry 41:11-20. It is 41(2):365-395. The 11-20 range belongs to a different paper.

Direction. Mixed.

Roberts, Yoshida, Galbraith, Laslett, Jones & Smith 1998 — Single-aliquot and single-grain optical dating confirm thermoluminescence age estimates at Malakunanja II rock shelter in northern Australia

Ancient TL 16(1):19-24. DOI 10.26034/la.atl.1998.287. CC BY 4.0. topic: case-jinmium, known-age, cross-method · camp: mainstream · type: measurement · access: OA (OCR-scan) · direction: undercuts

Why this matters more than it is usually given credit for. It is the control experiment for Jinmium, run by the same team, on the same continent, in the same year. If the Jinmium failure were a failure of TL as such, Malakunanja II should have failed too. It did not.

Dose rates used, from the original 1990 study: 0.77 ± 0.09 Gy ka^-1 for KTL162 and 0.91 ± 0.08 Gy ka^-1 for KTL164.

Its own conclusion, verbatim: "These single-aliquot and single-grain optical dating results support our earlier TL-based findings that people were present in Australia at 50-60 ka. They also show that correct burial ages can be obtained using TL methods, provided the sediments are bleached sufficiently before burial to reset the dating signal. The Jinmium sediments failed to meet this requirement (Roberts, 1997; Spooner, 1998), whereas adequate bleaching at Malakunanja II is indicated by the modern TL age for the near-surface sample (Figure 1), the similarity of palaeodoses obtained from TL signals (at 325°C and 375°C) with markedly different susceptibilities to light (Figure 4), and the concordance of TL and ¹⁴C ages at this site (Figure 1)."

Access route. Ancient TL article 6551, PDF at https://www.ancienttl.org/article/download/6551/5661. The PDF is a page scan with no text layer; the cover sheet is text but pages 2 onward are images. Extracted with pdftoppm -r 300 -png then tesseract. Page numbers in the OCR match the printed running heads (Ancient TL vol 16 No.1 1998, folios 19-24), so locators are anchored.

Direction. Undercuts. Any argument that treats Jinmium as proof that luminescence dating does not work has to explain Malakunanja II, where three methods (TL, single-grain OSL, radiocarbon) agree and the near-surface control returns zero.

Thorne, Grün, Mortimer, Spooner, Simpson, McCulloch, Taylor & Curnoe 1999 — Australia's oldest human remains: age of the Lake Mungo 3 skeleton

Journal of Human Evolution 36(6):591-612. DOI 10.1006/jhev.1999.0305. topic: case-mungo, cross-method · camp: mainstream · type: measurement · access: abstract-only (on disk) · direction: supports

The claim, verbatim from the abstract (on disk at sources/papers/thorne-1999_lake-mungo-3-skeleton-age-abstract.txt): "the best age estimate for the human skeleton is obtained through the combination of U-series and ESR analyses yielding 62,000+/-6000 years. This age is in close agreement with OSL age estimates on the sediment into which the skeleton was buried of 61,000+/-2000 years."

Direction. Supports. Four apparently independent numbers agreed to within 2 % and were later replaced by a figure 55 % lower. Already worked up in agents/memory/radiometric-failure-modes.md; the point there — that ESR, Th/U and Pa/U all read the same uranium in the same open-system bone and share one uranium-uptake assumption, so they are not independent — carries over unchanged and should not be re-derived.

Bowler, Johnston, Olley, Prescott, Roberts, Shawcross & Spooner 2003 — New ages for human occupation and climatic change at Lake Mungo, Australia

Nature 421:837-840. DOI 10.1038/nature01383. topic: case-mungo · camp: mainstream · type: measurement · access: abstract-only (on disk) · direction: mixed

Verbatim from the abstract: "Mungo III ... has been variously estimated at 30 thousand years (kyr) old, 42–45 kyr old and 62 ± 6 kyr old, while radiocarbon estimates placed the Mungo I cremation near 20–26 kyr ago. Here we report a new series of 25 optical ages showing that both burials occurred at 40 ± 2 kyr ago and that humans were present at Lake Mungo by 50–46 kyr ago."

The point to make from it. The disagreement was not OSL against OSL. It was OSL on the burial sediment against ESR plus U-series on the bone. Thorne et al.'s own 61 ± 2 kyr "OSL age estimates on the sediment" came from a different stratigraphic attribution, and it was resolved by dating the sediment that actually filled the grave. Note that Spooner is an author on both papers, and Olley and Roberts — the Jinmium single-grain team — are authors here.

Direction. Mixed. The method corrected a published age by 35 %, which is a real error; and the correction came from the same method, applied to a better-chosen target, which is how a working technique behaves.

Known-age and modern-analogue validation

Prescott & Purvinskis 1991 — Zero thermoluminescence for zero age

Ancient TL 9(2):19-20. DOI 10.26034/la.atl.1991.176. CC BY 4.0. topic: known-age, bleaching · camp: mainstream · type: measurement · access: OA (OCR-scan) · direction: mixed

What it measured. Two Australian sediments whose true age is near zero, dated by TL. Sample CA20S/1 from Carlo Blow is a dune currently encroaching on living vegetation, so "the time since the sample was last exposed to light can scarcely exceed a decade." Sample EB1S/02 is a currently mobile dune surface at the Roonka archaeological site.

The result, and it cuts both ways. Under the conventional total-bleach protocol of Singhvi et al. 1982, CA20S/1 "gave an age of several thousand years" and EB1S/02 "an age of some 400 years." Both are wrong, and both are wrong in the same direction as Jinmium. Under the selective-bleach protocol using the rapidly-bleaching 325 °C peak, both returned modern age. The paper's own conclusion: "the proposed selective bleach technique gives zero age for modern samples for which total bleach methods give non-zero ages and satisfies at least this criterion of Wintle and Huntley." The stated detection limit is "a true ED of about 0.1 Gy would be detectable."

Access route. Ancient TL article 6440, PDF https://www.ancienttl.org/article/download/6440/5550. Scanned pages, no text layer; pdftoppm -r 300 -png then tesseract. Folios 19-20 confirmed from the running heads.

Direction. Mixed, and unusually useful. It is a mainstream paper stating in plain words that an accepted TL protocol returned thousands of years for a deposit less than ten years old — and it is also the paper that fixed it and set a numerical zero-age criterion.

Alexanderson 2007 — Residual OSL signals from modern Greenlandic river sediments

Geochronometria 26:1-9. DOI 10.2478/v10003-007-0001-6. topic: known-age, bleaching · camp: mainstream · type: measurement · access: OA · direction: mixed

What it measured. Equivalent doses on sediment of known zero or near-zero age from four rivers on Jameson Land, East Greenland — the hardest possible bleaching environment, glaciofluvial water under a high-latitude sun.

The numbers (abstract, p. 1, and body, p. 4): river-bed sand ~0.1 Gy; silt (10-63 µm) ~1.1 Gy; a spring-flood sidebar ~1 Gy, "which corresponds to ca 500 years (dose rate 1.90±0.07 Gy/ka; measured with gamma spectrometry)"; and a subaerial reference deposit "completely zeroed (ED=0.00±0.07 Gy)."  Locator correction, 2026-08-30: these three quotations are on p. 4, not p. 5.  The even-page running heads in this journal print the folio to the left of "H. Alexanderson", and the header carrying "4" sits directly above the passage; the "5" that suggested otherwise is a y-axis tick label in Figure 4.

Its own honest framing: "Incomplete bleaching of this amount (1 Gy) is generally not a significant source of error for sediments of Pleistocene age," and "coarse grain-size fractions are better bleached and more suitable than fine grains for OSL dating of glaciofluvial/fluvial sediments."

Direction. Mixed. A modern sample returning 0.00 ± 0.07 Gy is a clean pass. A modern sample returning 1 Gy — half a millennium of apparent age — is a clean statement of the size of the incomplete-bleaching problem in exactly the depositional setting where it is worst.

Thomsen, Murray, Buylaert, Jain, Helt-Hansen & Aubry 2016 — Testing single-grain quartz OSL methods using sediment samples with independent age control from the Bordes-Fitte rockshelter

Quaternary Geochronology 31:77-96. DOI 10.1016/j.quageo.2015.11.002. topic: known-age, single-grain, cross-method · camp: mainstream · type: measurement · access: repository-AM · direction: supports

What it measured. Four sediment samples from Roches d'Abilly, France, with radiocarbon AMS control in the 22-48 ka range, dated by single-grain quartz OSL, multi-grain quartz OSL, and feldspar pIRIR₂₉₀.

The concession, verbatim from the abstract: "It is shown that had these samples been analysed in the absence of other age control, standard modelling decisions based on the shape of single-grain dose distributions would have led to significant misinterpretation of results and a corresponding >40% underestimation in age. If we instead ignore this standard decision process and apply weighted average and mixing models then the most likely results deviate from the expected ages by >10%. Finally, we show that by careful consideration of the luminescence characteristics of individual grains, we are able to obtain good agreement with the independent age control by applying alternative rejection criteria but this is at the cost of reducing the accepted grain population by more than an order of magnitude, with the corresponding inevitable decrease in precision."

And from the conclusions: "we conclude that, for these quartz samples, with doses in the range ~60 to 110 Gy, multi-grain analysis gives the most accurate ages; there is no advantage in the use of single grains."

Access route. DTU Orbit accepted manuscript, reachable only through the Internet Archive: https://scholar.archive.org/work/oa3itvsjzbhozg4rau5lyc534y/access/wayback/http://orbit.dtu.dk/files/127328942/QUAGEO_D_13_00083R4_no_reply.pdf. The live orbit.dtu.dk URL 403s. scholar.archive.org itself throws a "Session Verification" JavaScript wall at curl; bin/cdp-fetch.py page gets past it. The AM carries its own "Page N of 36" numbering, not the published 77-96.

Direction. Supports, strongly, as a young-earth exhibit — this is the field's own leading laboratory saying that the standard analysis procedure, applied blind, would have been out by more than 40 % on samples with good independent control. It is also, read fairly, a demonstration that multi-grain OSL agreed with radiocarbon across 22-48 ka.

Not obtained, with routes tried

Aitken 1985 — Thermoluminescence Dating (Academic Press) and Aitken 1998 — An Introduction to Optical Dating (Oxford)

access: NOT OBTAINED · direction: background

The two standard textbooks. Neither is on archive.org in any form. Routes tried: archive.org/advancedsearch.php on title and on creator:Aitken (0 hits); archive.org full-text search for the exact phrase "introduction to optical dating" (0 hits); OpenLibrary, which holds catalogue records for both (1985 and 1998) with ia: None, meaning no scanned copy exists to borrow or to search inside. The printdisabled search-inside route in CLAUDE.md therefore does not apply — there is no item. Their content is reachable second-hand: Guérin et al. 2011 restates Aitken's infinite-matrix and radon-escape framework; Riedesel 2025 quotes Aitken 1985 on trap lifetime and on the g-value definition; Roberts et al. 1998 cites "Aitken, 1985: Appendix B" for adding random and systematic uncertainties in quadrature. If a later agent needs Aitken directly, try a library route (hoopla, Evergreen Indiana, or an academic ILL) rather than repeating the archive.org search.

Huntley, Godfrey-Smith & Thewalt 1985 — Optical dating of sediments

Nature 313:105-107. access: NOT OBTAINED. Semantic Scholar openAccessPdf: CLOSED; not on scholar.archive.org; Nature back catalogue paywalled and not opened by the registered-reader account. Its existence, pagination and role are confirmed from Spooner 1998's reference list.  Considered and not used in the article.  The article describes optical dating without needing to date its invention, and citing a paper nobody here has read would add a name and no evidence.

Murray & Wintle 2000 — Luminescence dating of quartz using an improved single-aliquot regenerative-dose protocol

Radiation Measurements 32(1):57-73. DOI 10.1016/S1350-4487(99)00253-X. access: NOT OBTAINED. The most-cited paper in the field. Semantic Scholar CLOSED; Elsevier paywalled; Aberystwyth and Wageningen repositories hold metadata records only. The protocol is described in full in Murray et al. 2021 and is summarised in Alexanderson 2007 Table 1, so the substance is available even though the paper is not.  Now used at one remove, and the reason changed on 30 August 2026.  Two of the five young-earth objections — Walker's "the sensitivity of the crystal to radiation has not changed" and DeYoung's "calibration of the technique is uncertain" — are objections to precisely the thing SAR was built to fix.  The article therefore quotes Murray et al. 2021 lines 531-536 for the mechanism (a fixed test dose after each measurement, dividing out the sensitivity change) and lines 571-576 for the size of the correction in their worked example: 3.8 Gy corrected against 1.8 Gy uncorrected, and 25 Gy against 14 Gy for a sedimentary quartz aliquot.  The primary is still not obtained and the article does not cite it directly.

Murray & Olley 2002 — Precision and accuracy in the optically stimulated luminescence dating of sedimentary quartz: a status review

Geochronometria 21:1-16. access: NOT OBTAINED. This is the compilation that underpins the "OSL is accurate to about 10 %" claim, and it is the one worth another attempt. Geochronometria's own online archive (geochronometria.com) appears to start around volume 26 (2007); volume 21 is not there. geochronometria.pl PDF paths 404. CORE search 403s. BASE returned no PDF. Untried routes for a later agent: ResearchGate (has a copy, requires login); the Silesian University of Technology repository; a request to the authors.

Fullagar, Price & Head 1996, Roberts et al. 1998 Nature, Bowler et al. 2003 Nature, Thorne et al. 1999

Full texts not obtained; verbatim abstracts held for all four. See the individual entries above for routes.

Olley et al. 2006 — single-grain optical dating of grave-infill at Lake Mungo

Quaternary Science Reviews 25:2469-2474. access: NOT OBTAINED, not attempted beyond a note. This is the paper that most directly closes the Lake Mungo question by dating the sediment that actually filled the graves. A later agent should get this before the article is written.  It was not obtained, and the article was written without it.  Considered and not used, and the consequence is stated openly: the Lake Mungo case is cross-referenced to radiometric concordance rather than re-argued here, and the article claims nothing about Mungo beyond the Thorne 1999 and Bowler 2003 abstracts, both of which are on disk.

Creationist engagement

Rewritten 30 August 2026 after a five-venue sweep of the creationist literature.  The earlier version of this page said the fullest young-earth treatment was a magazine article and that creation.com and ICR could not be searched.  Two of the three claims were wrong, and both were wrong because a search box returned nothing.  Five sources are now known and all five are catalogued below, oldest first.  The searches that establish the remaining negatives are tabulated at the end, with their controls, so a reader can repeat every one.

DeYoung 1994 — Age of the Arizona Meteor Crater

Creation Research Society Quarterly 31(3):153-158 (December 1994).  PDF at https://crsq.creationresearch.org/791/1/Age%20of%20the%20Arizona%20Meteor%20Crater.pdf, on disk as sources/papers/deyoung-1994_age-arizona-meteor-crater.pdf. topic: creationist · camp: creationist · type: review · access: OA · direction: supports

What it is.  A survey of dendrochronology, rock erosion, meteorite radioactivity and thermoluminescence applied to Barringer Crater, by a physicist at Grace College.  It is the only worked young-earth case study of a thermoluminescence date found anywhere.

What it says.  He describes the reset correctly — "At the time of collision, the rocks were shock-heated, which 'reset the clock'" — reports Sutton's measured age of 49,900 ± 2,900 years, and rejects it for two reasons: "First, calibration of the technique is uncertain. ... Second, the underlying physical mechanism of thermoluminescence is poorly understood."  He then explains away the convergence of independent age estimates on ~50 ka as "tracking," "the tendency of published data to cluster around a particular value, similar to peer pressure."

Access route.  curl -sL with a browser user agent, pdftotext -layout.  Folios anchored twice on the running heads: PDF page 2 reads "154 CREATION RESEARCH SOCIETY QUARTERLY" and PDF page 6 reads "158", so printed page = PDF page + 152.  All the thermoluminescence text is on PDF page 4, printed p. 156.  Note the CRSQ EPrints trap: the repository's citation for this item reads "31 (3): 3", where the last number is the article's position in the issue, not its first page.

Direction.  Supports, and used in the article.  The calibration objection was answered by the SAR protocol six years later; the mechanism objection is partly conceded by Murray et al. 2021; the "tracking" move is answered as unfalsifiable and self-defeating.

Walker 2003 — The dating game, and its "Dating methods" sidebar

Creation 26(1):36-39 (December 2003).  https://creation.com/en/articles/the-dating-game.  Already on disk at sources/web/walker-2003b_the-dating-game.txt and sources/web/walker-2003c_the-dating-game-creation.html (indexed under radiometric-concordance). topic: creationist · camp: creationist · type: popular · access: OA · direction: supports

Previously recorded here as "considered and not used."  That was a mistake and it is now used.  The article body is a retelling of the Lake Mungo redating and adds no measurement, which is why it was set aside.  But attached to it is a boxed sidebar, "Dating methods," and that sidebar is the canonical Creation Ministries International statement on TL, OSL and ESR.  It names four assumptions: the initial stored dose, that nothing but environmental radiation changed the signal, that the environmental dose rate was constant, and "that the sensitivity of the crystal to radiation has not changed."  It also groups ESR with TL and OSL explicitly, which is the article's own framing before it is the article's.

Why the sidebar was missed the first time.  The site search could not be reached at all, and the article was found through the radiometric-concordance research on Lake Mungo, where only the body text mattered.  A source can be on disk and still be missed, if it was catalogued for a different article's purpose.

Access route.  curl -sL with a browser user agent, tags stripped locally.  The page prints "First appeared in Creation 26(1), Pages 36 - 39, December 2003"; its "Published 21 Aug, 2013 | Updated 30 Dec, 2023" line is a web-posting date and must not be cited as the date of publication.

Direction.  Supports.  The sensitivity-change assumption is the best of the four and is answered by the SAR test-dose normalisation, published in 2000.

Cupps 2016 — Examining Thermoluminescence Dating

Acts & Facts 45(5):12 (May 2016), Institute for Creation Research.  https://www.icr.org/article/examining-thermoluminescence-dating/; issue PDF https://www.icr.org/i/pdf/af/af1605.pdf, on disk as sources/papers/cupps-2016_examining-thermoluminescence-dating.pdf. topic: creationist · camp: creationist · type: popular · access: OA · direction: supports

A dedicated ICR article on one of this page's two methods, and the first version of this article missed it.  Cupps holds a PhD in nuclear physics from Indiana University-Bloomington.  One printed page.  He frames luminescence as the method that fills the gap between radiocarbon's 50,000-year ceiling and the million-year floor of the other methods, states the age as total measured dose over an assumed dose rate, and puts the whole objection on the denominator.

The strongest thing about it.  He does not run the parent-and-daughter three-assumption checklist.  He goes straight to the dose rate, which is the only assumption a dosimeter has.  His framing is cleaner than Snelling's and the article says so.

The weakest thing about it.  His support for dose-rate variability is the American Nuclear Society's consumer "Radiation Dose Calculator" (his note 7), and the claim it supports — that dose rates "vary with location, season, solar activity, and even time of day" — is about a person's external exposure, not about a buried grain.  His list of the "crystalline materials" in pottery (note 4: CaF₂, LiF, CaSO₄, Li₂B₄O₇, Ca₃(BO₃)₂, KBr, feldspars) is the standard list of manufactured TLD badge phosphors, not of the minerals actually dated.

Access route.  curl -sL with a browser user agent, pdftotext -layout on the 24-page issue PDF.  Folios anchored twice: PDF page 11 carries the foot "MAY 2016 | ACTS&FACTS 11" and PDF page 12 carries "12 ACTS&FACTS | MAY 2016", so printed page = PDF page.

Direction.  Supports, and used.

Snelling 2018 — How Reliable Are Dates for Human Fossils?

Answers magazine, 1 September 2018. answersingenesis.org/geology/radiometric-dating/how-reliable-are-dates-human-fossils/ topic: creationist · camp: creationist · type: popular · access: OA · direction: supports

The most-read young-earth statement on the subject.  (This entry previously called it "the fullest young-earth treatment of luminescence found anywhere," which is no longer true: DeWerd 2020 is nine pages of technical argument and this is a magazine article.)  Snelling describes the mechanism accurately, with a rechargeable-battery analogy, and then applies Answers in Genesis's standard three-assumption framework: that the original number of unstable atoms can be known, that the rate of change was constant, and that the daughter atoms were all produced by decay.

His specific objections to luminescence, verbatim: "Do the minerals show when the fossil bone was buried, or did the minerals first get their stored energy somewhere else before being buried with the bone? Did the mineral grains, typically quartz or potassium-feldspar, have impurities that could have affected the rate of storage of electrons? Were these minerals totally reset to zero before being buried with the bone?"

His objection to ESR, verbatim: "Has the radiation dose been small enough so that the minerals could cope with it? (If the radiation dose was too large, then the minerals' 'batteries' would have overfilled, the extra 'charge' would be lost, and thus the dates obtained would be highly erroneous.)"

His conclusion, verbatim: "If the dating methods say 2 million years, it may actually be a hundred years or just ten years."

What is right about it and what is not. The three questions about bleaching, impurities and resetting are the real failure modes, and Jinmium is a documented case of exactly the first and third. The saturation objection to ESR is also a real effect. But the article's framing misfires on luminescence specifically. Assumption 1, "the original number of unstable atoms can be known," does not apply: luminescence does not measure a parent-daughter ratio and does not need an initial inventory. Assumption 3, contamination by daughter atoms, does not apply either. The only one of the three that bites is assumption 2, and the article does not do the arithmetic on what changing the rate would require. That arithmetic is in research/dose-rate.ts and research/_brief.md.

Direction.  Supports, as the most-read statement of the young-earth position.  Its weakest move — importing the parent-daughter assumption list wholesale onto a method that has no parent-daughter ratio — should be stated plainly rather than repeated, and it should be stated as a criticism of this framework rather than of the movement, because four of the five young-earth sources on this method never use that list.

DeWerd 2020 — Dating Uncertainties with Thermoluminescence

Creation Research Society Quarterly 56(4):212-220 (Spring 2020).  PDF at https://crsq.creationresearch.org/1249/1/Dating%20Uncertainties%20with%20Thermoluminescence.pdf, on disk as sources/papers/dewerd-2020_dating-uncertainties-thermoluminescence.pdf. topic: creationist · camp: creationist · type: model · access: OA · direction: supports

This is the fullest and best-qualified young-earth treatment of the method that exists, and the first version of this article did not know it existed.  Larry A. DeWerd is Professor of Medical Physics at the University of Wisconsin-Madison and runs a radiation calibration laboratory; thermoluminescent dosimetry is his working field.  He opens by naming Cupps 2016 as the prior review.

The argument, and it is a good idea.  Published luminescence ages quote a Type A uncertainty — the scatter of repeated measurements.  A calibration laboratory would build a full NIST-style budget (Taylor and Kuyatt 1994) including Type B judgement terms.  DeWerd builds one, Table II, p. 219: eleven line items, combined 42.72% at k = 1, expanded 85.44% at k = 2.  His worked laboratory example, Table I, is a brachytherapy dose-rate constant measured with LiF:Mg,Ti.

Where it fails, and all three are checkable.  (1) His own 85.44% on 20,000 years gives 2,912 to 37,088 years; he prints "2,900 to 10,000" and does not print the upper bound, and the summary repeats the 10,000 figure, so it is not a typographical slip.  (2) An uncertainty band is symmetric and cannot bias a whole literature downward; if his terms were real and random, half of all published TL ages would be too young.  (3) Most of his eleven terms — NIST traceability, PMT spectral match, calibration dose, reproducibility, energy response — are absolute-dosimetry terms that cancel in a substitution measurement, because the natural and calibrating signals go through the same grains and the same optics.  Keep only the environmental terms and the honest expanded uncertainty is about ±42%, a factor of 1.4.  His claim that "all of the dates determined using TL are generally less than 10,000 years" is contradicted by Aitken 1985, which he cites, and by the TL ages at Malakunanja II and Jinmium.

What should be kept.  The demand for a Type B budget is legitimate and the field has not fully met it.  And his summary contains a sentence his own side rarely quotes: "It would appear that the most accurate dating method for artifacts would be via the TL process."

Access route.  curl -sL with a browser user agent, pdftotext -layout.  Folios anchored twice: PDF page 1 carries the printed citation line "Creation Research Society Quarterly 2020 56:212–220", and PDF pages 3 and 9 carry the running heads "214" and "220", so printed page = PDF page + 211.

Direction.  Supports, and used at length.

Silvestru 2003 — The Significance of Caves in the Post-Flood History of Mankind

Proceedings of the Fifth International Conference on Creationism, pp. 551-560 (2003).  PDF at https://cedarville.tind.io/record/21024, on disk as sources/papers/silvestru-2003_significance-of-caves-post-flood.pdf. topic: creationist · camp: creationist · type: measurement · access: OA · direction: background

Why it is here.  It is the only one of the 457 ICC paper PDFs whose full text mentions luminescence dating at all, and it is therefore the entirety of the young-earth technical-conference engagement with the method.  What it says, on p. 556, is this: "There exists a solid and well-documented literature regarding the issue of reliability of 'absolute dating' and alternative methods (thermoluminescence, fission tracks, dendrochronology etc.).  Their main point is that they should not be taken as infallible."  That is a parenthesis in a list.  The paper's actual subject is cave art, speleothems and whether cave artists would have painted around formations that were not yet there.

Access route.  Folios anchored on the printed page numbers, which run 551-560 across the ten-page PDF; the quoted passage sits between the folio markers for 555 and 556.  Read with pdftotext -layout.

Direction.  Background.  Cited in the article only to establish, precisely and auditably, how thin the engagement is.

Lubenow 2004 — Bones of Contention, revised edition

Baker Books, Grand Rapids, 2004.  ISBN 0801065232.  Read from the print-disabled archive.org item bonesofcontentio0000lube_s0n6 with bin/iapdsearch.py.  Nothing stored in the library; the row in index.tsv carries file = -. topic: creationist · camp: creationist · type: book · access: book · direction: background

Previously recorded as "not attempted."  Now obtained and searched, and the result is a negative worth printing.  The book mentions thermoluminescence four times: once of substance, reporting a mainstream TL date of 90,000 years for Qafzeh (p. 171), and three times in its bibliography.  "Electron spin resonance" appears once, in an endnote citing Rainer Grün on Petralona (p. 384).  "ESR" as a standalone token appears nowhere in the book.

The finding.  Lubenow devotes a chapter to a "coverage gap" — "the period from about 40,000 years ago to about 200,000 years ago on the evolutionist's time scale" — which he says radiocarbon and potassium-argon cannot reach (p. 289).  That is exactly the window luminescence and ESR were built to fill, and he does not name them in that chapter.  Cupps 2016 names the same gap from the other side and treats it as the method's motive.

Access route, with the leaf trap closed.  bonesofcontentio0000lube_s0n6_scandata.xml returns HTTP 200 with no login and carries an explicit <pageNumber> for 396 of the 406 leaves.  Printed page = leaf − 2, checked at thirteen leaves from leaf 18 to leaf 386.  The _djvu.txt derivative 403s, as expected for a print-disabled item.  A second copy of the 1992 first edition exists at bonesofcontentio0000lube (302 leaves) and was not needed.

Direction.  Background.  Used in the article to show that the most widely read young-earth book on hominin dating names the gap and never engages the methods that fill it.

Woodmorappe 1999 — The Mythology of Modern Dating Methods. NOT OBTAINED.

Master Books, Green Forest, AR, 1999. topic: creationist · camp: creationist · type: book · access: NOT OBTAINED · direction: unknown

Routes tried, 30 August 2026.  Open Library holds a catalogue record for the work with no ia identifier attached, meaning no Internet Archive scan exists.  openlibrary.org/search/inside.json?q="Mythology of Modern Dating Methods" returns 31 full-text hits and every one of them is another book citing Woodmorappe, chiefly John Reed's Rocks Aren't Clocks (rocksarentclocks0000reed); none is the book itself.  archive.org/advancedsearch.php finds no item under the title.  The book appears never to have been scanned.

Why it still matters.  It is the most systematic young-earth attack on dating in book form and it has a chapter on non-radiometric methods.  Until it is read, the survey above has one real hole, and the article says so.  A physical copy or an interlibrary scan is the only remaining route.

Phillips, Zreda, Smith, Elmore, Kubik, Dorn & Roddy 1991 — Age and geomorphic history of Meteor Crater. NOT OBTAINED.

Geochimica et Cosmochimica Acta 55(9):2695-2698.  DOI 10.1016/0016-7037(91)90387-K. topic: cross-method · camp: mainstream · type: measurement · access: paywalled · direction: unknown

Identified while answering DeYoung 1994 and deliberately not used.  This is an independent cosmogenic ³⁶Cl and ¹⁴C dating of the same crater, published three years before DeYoung wrote and not cited by him.  Its result would be the cleanest possible test of his "tracking" claim, because cosmogenic ³⁶Cl shares no assumption with thermoluminescence.  Its number was not obtained: Elsevier's page is closed, Semantic Scholar reports the abstract elided by the publisher, and no open copy was found.  The article therefore answers DeYoung on the logic of the "tracking" move rather than on this measurement, and this entry records the paper so a later agent does not have to rediscover it.

Considered and not used

• "Age of Aboriginal Art Disputed," creation.com, undated. On disk as sources/web/cmi-2006f_age-of-aboriginal-art-disputed.html. Now cited in the article as a fairness point: CMI reported the Jinmium correction accurately, by quoting a mainstream luminescence specialist rather than dismissing the method. Its date could not be pinned from bytes and the article prints none. The page shows "Published 28 Feb, 2006", which is the Meilisearch published_on web-posting date; its own text refers to a press report "of February 18" and a "Hot Topic ... a few months back", and Jinmium was published in December 1996, so the underlying piece is probably February 1997. That is an inference, not a verified date. The Spooner figures it quotes — a factor of ten, a maximum of 20,000 years — are press comment, not his 1998 Antiquity paper.
• Oard 2023, "Greenery beneath Greenland Ice Sheet?", Journal of Creation 36(2). Says "the luminescence dating technique is questionable at best," citing DeWerd 2020, and its subject is the Camp Century core. It adds no argument beyond DeWerd, whom the article engages directly. It is how DeWerd was found, which is worth recording: a footnote in an unrelated article, not a search box.
• "What date would you like?", Creation 45(2):10 (April 2023). A news paragraph noting that ESR dates on the Sterkfontein fossils "ranged from ∼1 to 4 My." Belongs to cosmogenic nuclide dating, not here.
• Oard 1998, "How reliable are quaternary dating methods?", CEN Technical Journal 12(3):258-259. The methods it disputes are tephrochronology, palaeomagnetism and fission track, none of them luminescence. Its portable quotation — "Dating methods other than radiocarbon are not yet reliable nor widely accepted and used" — is from the mid-1990s, before the SAR protocol and before single-grain OSL, and quoting it flat would be the selective quotation this site criticises elsewhere.
• Twenty-four further creation.com articles returned by the body-text sweep below — Denisova Cave, Homo floresiensis, the White Sands and Tularosa footprints, the Namib sand sea, the Caspian Sea, Hofmeyr, the Jishi Gorge flood, and the Developments in paleoanthropology series — all cite an OSL, TL or ESR date in passing while reporting a mainstream disagreement. Every one was listed and none makes a methodological argument this article does not answer. Recorded as a block so nobody opens them again.

The negative findings, and the exact searches

Rebuilt 30 August 2026 on enumeration rather than on search boxes.  The table below is the whole basis for every "the young-earth literature has not engaged this" claim in the article.  Each row gives the route, the control that proves the route works, and the result.

What each result means, stated plainly.

• The ICC negative is now strong. It is established at body-text level over all 457 paper PDFs, not at title level over 474 listing entries as this page previously said. The six matches are four cathodoluminescence references (an imaging technique for quartz and diamond growth zoning), one bacterial bioluminescence discussion, and Silvestru 2003, whose two matches are his parenthesis and a Valladas et al. 1987 reference. In thirty-seven years the movement's technical conference has never had a paper on luminescence dating.
• The Creation Matters and CGS negatives are strong, both with passing controls.
• The Journal of Creation Letters negative is weak on its own and should be reported as such: it is a title index, and zircon returns zero at title level too. What carries the creation.com finding is the body-level Meilisearch sweep.
• No venue's search box can establish a negative, and ICR's is the proof. Queried on 30 August 2026, https://www.icr.org/home/submitsearch/?f_keyword_all=thermoluminescence answered "Your search returned no results" — and so did the control zircon. ICR hosts an article with thermoluminescence in its title. ICR is therefore explicitly excluded from establishing any negative in this article, and its one positive result came from the sitemap.

Three traps recorded so nobody hits them again.

1. creation.com's Meilisearch endpoint. Results live under hits[]._formatted, not hits[]. types=articles, plural, silently returns zero — it must be types=article. And published_on is the web-posting date, not the print date: Walker's December 2003 article reports as 2023-12-30. Cite the "First appeared in…" line on the article page.
2. CRSQ EPrints citations. Every record page prints something like 56 (4): 1, and the last number is the article's position in the issue, not its first page. Pin folios from the PDF running heads; post-2013 PDFs also carry a full citation line on page 1.
3. Print-disabled archive.org books. _djvu.txt 403s but _scandata.xml returns 200 with no login and gives an explicit <pageNumber> per leaf, which is how the Lubenow page numbers above were pinned. BookReaderJSIA.php returned "MetaData file not found!" for that item, so _scandata.xml is the more reliable route.

Superseded and recorded for the record.  The first version of this page reported: creationicc.org?s= returns "Nothing Found" for luminescence, thermoluminescence and the control zircon, so that box indexes proceedings-level listing pages rather than paper text; answersresearchjournal.org and creationresearch.org returned no article hits for thermoluminescence; icr.org's /search/?q= URLs 404; and creation.com "resisted every route tried" behind a Nuxt front end and Cloudflare.  All four observations were accurate about the search interfaces.  All four were useless as evidence about the literature, and two of them led this page to a wrong conclusion.  That is the reason the table above reports routes and controls rather than queries and hit counts.

Computation done for this article

research/dose-rate.ts

Deno TypeScript. Computes dry infinite-matrix dose rates from U, Th, K and Rb concentrations using the Guérin et al. 2011 factors; applies the HF-etch (alpha removed), Mejdahl 1979 beta attenuation for 90-125 µm grains, Zimmerman 1971 water attenuation and the Prescott & Hutton 1994 cosmic term; converts a measured D_e to an age; and shows the sensitivity of the age to water content and to a hypothetical past dose-rate multiplier. Run with deno run html/unpublished-ai/luminescence-dating/research/dose-rate.ts. Full output and its interpretation are in research/_brief.md.

Added with the 2026-10-20 correction

Humphreys 2000 (humphreys-2000), "Accelerated Nuclear Decay:  A Viable Hypothesis?", the RATE volume of 2000, pp. 333-380.  Read from the source library PDF.  RATE's own statement of when decay was accelerated:  "three episodes:  one in the early part of the Creation week, another between the Fall and the Flood, and the third during the year of the Genesis Flood" (p. 333, and the Figure 2 caption, p. 341).  Nothing after the Flood.  Cited in the article for that position.

Oard 2021 (oard-2021f), "Much Greater Cosmic Rays During the Ice Age and Before," Creation Research Society Quarterly 58(1):30-48.  Read from the source library PDF.  The one young-earth author found who puts accelerated decay after the Flood:  "during the Flood and Ice Age" (p. 30) and "accelerated radiometric decay continued for a while after the Flood" (p. 39).  He gives no acceleration factor, and concedes it "presents a problem of too much radiation during the Ice Age" (p. 44).  Cited in the article for that position and tested against the required factors.