The Oldest Homo sapiens

"I'm not really qualified to comment on thermoluminescent dating, but I know when inferences are built on inferences which are built on other inferences.  That's a time to be reasonably skeptical." — Todd Charles Wood, young-earth biologist, on the Jebel Irhoud date, 2017¹

"Thus, the real age of these bones is only probably around 4,000 years, and the bones are those of Noah’s descendants." — David Menton, Answers in Genesis, 2017²

Summary

The oldest fossils assigned to our own species come from Morocco, South Africa, Ethiopia, Israel and Greece.  On the anatomy, the young-earth side and the mainstream largely agree:  these were people.  The whole dispute is about when they lived.  This article sets out the fossils, how each was dated, how well the dates hold, what the genetic evidence adds, and what young-earth writers say about them in their own words.  It then prices what a young-earth timeline needs.

The fossils are human on any reading.

1. Jebel Irhoud, Morocco, is the oldest. Its faces and teeth fall inside the range of living people, but its braincases are long and low, "not globular."³ Fourteen flints burnt in the people's fires were dated by thermoluminescence. The layer with most of the fossils gives 315,000 ± 34,000 years.⁴ A tooth from the site, dated by a different method, gives 286,000 ± 32,000.⁴
2. Florisbad, South Africa, has one tooth dated to 259,000 ± 35,000 years. Lee Berger and John Hawks argue that the tooth may not belong to the skull it is shown with, and that the date is too loose to prove much.⁵
3. Omo I, Ethiopia, lies below a layer of volcanic ash that chemistry ties to an eruption dated to 233,000 ± 22,000 years, so Omo I is at least that old.⁶ Its maximum age is still open.⁶
4. Apidima 1, Greece, is a partial skull dated to more than 210,000 years. Its own describers call it incomplete and distorted, and both its dating and its identification as Homo sapiens have been questioned, including by one of its own authors on the identification and by a uranium-series specialist on the date.⁷⁸⁸⁹
5. Misliya, Israel, a jaw with teeth, is 177,000 to 194,000 years old by three methods on three materials: burnt flint, a mineral crust on the jaw, and the tooth enamel.¹⁰
6. Herto, Ethiopia, the skulls of two adults and a child with stone tools and butchered hippopotamus bones, is bracketed between 160,000 and 154,000 years by argon dating of volcanic material in and above the fossil layer.¹¹

The dates hold up better than young-earth writers say, and not perfectly.

7. Where two methods were applied to the same horizon, they agree within their errors. When dates were revised, they moved by factors of 1.2 to 1.8, and at Irhoud and Omo they moved older.¹²
8. The one documented case of an "oldest modern human" that may be off by as much as a young-earth reading needs is a set of teeth that may not have belonged to their layer. Teeth from Fuyan Cave, China, published at 80,000 to 120,000 years, gave radiocarbon ages of 2,500 to 9,400 years in 2021.¹³ The original team disputes this.¹⁴ At Irhoud the clock dates the people's own fires, and at Omo and Misliya the bones themselves were tested and found not to be later burials.¹⁵¹⁰
9. The genetic evidence agrees with the fossils in order of magnitude. The deepest split among living human populations is put at 260,000 to 350,000 years by one study,¹⁶ and at 120,000 to 135,000 years by another that models the history differently.¹⁷ Both scale with an assumed mutation rate.

What young-earth writers say, and what it costs.

10. Young-earth writers agree these fossils are human, and place them after Noah's Flood and after Babel. Answers in Genesis gives Irhoud "probably around 4,000 years."² Creation Ministries International calls Herto "probably post-Babel and thus post-Flood."¹⁸ Marcus Ross, Peter Brummel and Todd Wood concede that "none of the dating methods have a widely accepted method of recalibrating to a creationist timescale."¹⁹
11. The required compression is 28 to 80-fold. Every fossil date in Part 1 is 35 to 80 times the 4,350 years Andrew Snelling of Answers in Genesis gives since the Flood, and 28 to 64 times the longest Flood date any young-earth chronology allows.²⁰
12. The young-earth physics does not reach these clocks. Young-earth scientists place accelerated radioactive decay in the Flood year, and wrote in 2016 that it "decelerated at the end of the Flood";²¹ Snelling stretched it in 2018 to "during and soon after the flood."²² The Irhoud and Misliya clocks were started by fires lit by people, and the Omo clock by an eruption that buried ground where a person had already died. On the young-earth account all of that happened after Babel. To fit, decay would have had to run about 72 times faster than today for the whole time since the Flood, or about 300 times faster for a thousand years of it, while people lived on the ground being dated.²³ At 72 times, their own radiation dose would be no higher than some people in Ramsar, Iran, live with today; at 300 times it would be about 2.6 times the highest external dose measured there, and at 3,000 times about 26 times.²⁴ No young-earth writer has proposed this, and none names a faulty input to any of these measurements. The one young-earth number that could move these clocks into the years of fast decay, Robert Brown's radiocarbon curve, does it by putting these people within a decade of the Flood, before Babel, where the same writers' reading of Genesis 11 has no one in Africa.²⁵

Verdict:  these are human fossils, and young-earth writers and the mainstream agree on that.  The dates are where the young-earth position has no answer yet.  They are consistent across methods, they were started by human activity that a young-earth chronology places after Babel, and the main mechanism the young-earth literature offers for compressing them is placed by its own authors in or just after the Flood year.  Nobody has worked out how it could run after Babel, through the lives of the people whose fires it would have to date.

Part 1: The fossils, oldest first

"Oldest Homo sapiens" depends on what counts as sapiens.  None of the fossils below looks exactly like a living person in every part.  Each mixes features found in people today with features found in older human fossils, and the mainstream papers say so.  The young-earth writers who discuss them count all of them as human, and so does this article.  The table gives the short version;  the sections after it give the evidence.

Which fossil is "oldest" moves by about 100,000 years with the definition.  Chris Stringer of the Natural History Museum, London, gives the narrow one:  "the generally accepted first appearance of ‘anatomically modern humans’ (that is to say fossils that predominantly share the skeletal morphology of extant humans)" is "the Omo Kibish 1 skeleton and the somewhat younger Herto material."  On that definition Irhoud is not sapiens:  Stringer wrote that "there is enough preserved of JI1 to indicate that it does not represent anatomically modern H. sapiens."²⁶  Irhoud becomes the oldest Homo sapiens only on the broader definition, membership of the lineage that leads to us, which Stringer also uses.  He grants that splitting the lineage into named species does not work well:  "unless there is truly punctuational change at each species origination, there would undoubtedly be blurred and probably unworkable boundaries between helmei and sapiens, and between steinheimensis and neanderthalensis."²⁶  This fits the young-earth reading that these were all one variable human population, and the Middle Pleistocene Muddle article takes the naming question further.  It does not touch the dates, which are the subject of the rest of this article.

The ± figures are each paper's own.  For Irhoud the 315 ± 34 is one standard error,⁴ and for Omo and Misliya the ranges are two standard errors.¹⁰⁶

Jebel Irhoud, Morocco

What was found.  Mining in the Jebel Irhoud hills, 55 km southeast of Safi, cut into a cave fill in 1960, and an almost complete skull turned up in 1961.  The excavations that followed produced a second braincase and a child's jaw.³  A team led by Jean-Jacques Hublin of the Max Planck Institute for Evolutionary Anthropology dug again from 2004 to 2011 and found a second partial skull, a nearly complete adult jaw, a maxilla, bones and teeth.  Layer 7 holds "the remains of at least five individuals (three adults, one adolescent and one immature, ca. 7.5 years old)."³  The people left Middle Stone Age tools made by the Levallois method, and burnt bones are 19 to 25 percent of the animal bones in layers 5 to 7.⁴

What is modern and what is not.  Hublin's team found "a mosaic of features including a facial, mandibular and dental morphology that aligns the Jebel Irhoud material with early (EMH) or recent anatomically modern humans (RMH) and a more primitive neurocranial and endocranial morphology."³  The face is short and tucked under the braincase, and "all the possible reconstructions of the new facial remains of Irhoud 10 fall well within RMH variation, as does Irhoud 1."³  The braincase is the archaic part:  "In contrast to their modern facial morphology, the Irhoud crania retain a rather primitive overall shape of the braincase and endocast, i.e. unlike those of RMH, they are elongated and not globular."³  The brain was already full-sized.  A 2018 study of brain casts measured Irhoud 1 at 1,375 ml and Irhoud 2 at 1,467 ml, and concluded that the shape, not the size, changed later:  "Brain shape, however, evolved gradually within the H. sapiens lineage, reaching present-day human variation between about 100,000 and 35,000 years ago."²⁷  An earlier study found that the Irhoud child's teeth grew on a modern human schedule.²⁸

The date.  Of the 320 stone artefacts larger than 2.5 cm, 37 percent show signs of heating.  Of 45 heated pieces tested, 14 passed checks that the fire had been hot enough to reset them.⁴  Daniel Richter's team dated those 14 by thermoluminescence.  Individual flints gave 240 ± 35 to 378 ± 30 thousand years.  The layer averages are 302 ± 32 thousand for layer 6 and 315 ± 34 thousand for layer 7, which holds nearly all the fossils.⁴  At two standard errors, layer 7 is 247,000 to 383,000 years old.⁴  The team also re-dated the same Irhoud 3 tooth that had been dated to 160,000 ± 16,000 years in 2007, and got 286,000 ± 32,000 (or 281,000 +37,000/−29,000 under a second uptake model).  They give the reason the first figure was low:  "The published ESR age of the Irhoud 3 fossil was based on an external dose-rate measured on a small sediment sample of imprecise stratigraphic context."⁴

Its limits, in the authors' words.  The thermoluminescence and tooth dates are not fully independent, because both rely on the same measurement of the gamma rays coming from the surrounding sediment:  "Because of the shared γ-dosimetry, we here consider these methods as quasi-independent and argue that the ESR data is, therefore, supporting the TL data.  Nevertheless, other independent age control would be desirable."⁴  There was no volcanic ash or cave crust to date by a third method.  The link between the dated flints and the bones is by layer, not by contact:  the team presumes the fires and the burials "were linked in time, but we cannot establish this beyond the resolution of an individual stratigraphic deposit."⁴

Florisbad, South Africa

The front of a skull and face was found at this spring site in 1932.  In 1996 Rainer Grün and colleagues dated a tooth from the site by electron spin resonance to 259,000 ± 35,000 years.⁵  Chris Stringer of the Natural History Museum, London, notes that the fossil has been read three ways:  as like the older Broken Hill skull, as "an early member of the H. sapiens clade," or as a separate species.  He calls it "too incomplete for definitive assignment," but thinks that "like the Irhoud material this fossil probably represents an archaic part of the H. sapiens clade."²⁶  Lee Berger and John Hawks challenged that in 2023.  They argue that the tooth is not shown to belong to the skull:  "We find that the contemporary records raise doubt about the contemporaneity of the tooth and calvaria."  The spring reworked its sediments:  Grün's own team put the scatter of their dates down to "reworking of material by spring action."  And the yearly dose for the tooth was taken from "the average concentration of radioactive elements of white sand layers," which lie far above the peat level the fossils are reported from.⁵  Their conclusion is that "The current data do not demonstrate the presence of H. sapiens in southern Africa at ~260 ka."⁵  They also note that the error bar is wide enough to allow "both much more recent and much more ancient ages."⁵  Their "much more recent" means tens of thousands of years, not a few thousand:  their own comparison is with a date "more than 30 ka younger than" 226,000 years.⁵  Florisbad is the weakest-dated fossil in this article, and it does not carry any argument here.

Omo Kibish, Ethiopia

A team led by Richard Leakey found the Omo I skull and skeleton and the Omo II braincase in 1967.²⁹  Omo I looks more modern than Omo II, though the two come from similar levels.³⁰  The fossils were first dated to about 130,000 years by uranium-series on shells, a figure Michael Aubert's team later called "controversial".¹⁵

In 2005 Ian McDougall, Francis Brown and John Fleagle dated feldspar crystals from pumice in a volcanic layer below the fossils by argon-argon.  That gave "an older limit of 198 +/- 14 kyr (weighted mean age 196 +/- 2 kyr) on the hominids," and a layer above gave a younger limit of 104 ± 7 thousand years.  Their preferred age was "195 +/- 5 kyr."³⁰

In 2022 Céline Vidal and colleagues at the University of Cambridge went at it from above.  A volcanic ash called the KHS Tuff "overlies Member I, where Omo I was retrieved around 1.4 m lower down section, and is demonstrably younger than the fossils."⁶  They matched the ash's glass chemistry to the deposits of a large eruption of Shala volcano, sampled 350 km northeast of the Omo site, and dated feldspar crystals from those deposits:  "By dating the proximal deposits of this eruption, we obtain a new minimum age for the Omo fossils of 233 ± 22 kyr."⁶  This also answers a standard young-earth worry about argon dates, that a crystal can carry argon from the magma and look too old.  An isochron plot, which separates trapped argon from argon made by decay, gave 219 ± 27 thousand years from the same 71 crystals, "indistinguishable from the accepted age," with a trapped-argon ratio of 314 ± 24, close to that of air.⁶  What they could not do is set a maximum:  "The challenge remains to obtain a robust maximum age for Omo I."⁶

Whether the bones belong to the layers dated is checked directly.  Aubert's team dated the Omo I skull itself by uranium-series, and reported that "the human remains do not belong to a later intrusive burial."¹⁵

Apidima, Greece

Two partial skulls were found in a cave in southern Greece in the late 1970s.  In 2019 Katerina Harvati's team reconstructed them by computer and dated them by uranium-series.  Their abstract says the skulls "have remained enigmatic owing to their incomplete nature, taphonomic distortion and lack of archaeological context and chronology."  They report that "Apidima 2 dates to more than 170 thousand years ago and has a Neanderthal-like morphological pattern.  By contrast, Apidima 1 dates to more than 210 thousand years ago and presents a mixture of modern human and primitive features."⁷  Both ages are minimums.  The two skulls lay a few inches apart in the same rock, and earlier workers had taken them to be one age, at least 160,000 years.⁸  Creation Ministries International makes the most of this:  the two were "found embedded in the same breccia, separated by a few inches, yet were ‘dated’ 40,000 years apart!"⁹

Both the date and the identification are disputed by mainstream specialists.  Warren Sharp of the Berkeley Geochronology Center, a uranium-series specialist, called the team's data for the 210,000-year result "imprecise and highly scattered."  Rainer Grün of Griffith University, one of the paper's authors, answered that "this, in my view, is the best interpretation of the results."⁸  Creation Ministries International quotes Lizzie Wade's report of Sharp's figures in Science:  "the Apidima 1 samples actually returned dates ranging from more than 300,000 years old to less than 40,000 years old."⁹  On the anatomy, Apidima 1 is the back of a braincase and nothing else.  Juan Luis Arsuaga, who described the Sima de los Huesos skulls from Spain, said "I cannot see anything suggesting that the individual belongs to the sapiens lineage," and "To make such a claim, you need a face."  Chris Stringer, one of the paper's own authors, agreed that the missing parts matter:  "We don’t have the frontal bone, browridge, face, teeth, or chin region, any of which could have been less ‘modern’ in form."⁸  This article treats Apidima 1 as a claim, not a finding.  Sharp's complaint, as reported, is that the results scatter too widely to trust any one of them;  he is not reported proposing an age of his own for Apidima 1.

Misliya, Israel

An upper jaw with its teeth, Misliya-1, was found in a cave on the western slope of Mount Carmel.  Israel Hershkovitz's team published it in 2018 as Homo sapiens, on the evidence of its teeth and the shape of the jaw.¹⁰  Of the premolars, for example, they write that "Like in H. sapiens, Misliya-1 premolars display tall crowns and relatively expanded occlusal polygons."¹⁰  Three dating methods were used on three different materials.  Nine burnt flints from the same layer, near the fossil, gave a thermoluminescence mean of "179±48 ky (n=9; 2σ error)."  The tooth enamel gave 174 ± 20 thousand years by combined uranium-series and electron spin resonance.  A calcite crust grown over the jaw itself gave a minimum of 185 ± 8 thousand by uranium-thorium.  The team combined the crust minimum and the enamel maximum into a range "between 177 ky (=U-TH 185 ky-8 ky) and 194 ky (=US-ESR 174 ky+20 ky)."¹⁰  A uranium-series date on the tooth itself gave 70,200 years.  Uranium can soak into a tooth long after death, so that figure is a minimum, and the team argue it rules out a recent burial:  "The U-series date of 70 ky on the Misliya-1 itself excludes the possibility of recent intrusion."¹⁰  Not every specialist accepts the older range.  Warren Sharp, the uranium-series specialist who doubts the Apidima date, was reported in 2019 as "contending that the fossil can’t be older than 70,000 years."⁸  The report does not give his reasons, and no published version of the argument was found.  If he is right and the tooth's own 70,000-year figure is close to its age, Misliya-1 is still 16 times the time since the Flood.

Herto, Ethiopia

Tim White's team found the Herto fossils in the Middle Awash in 1997.  In Stringer's summary the most important are "a nearly complete adult skull, an immature calvaria and parts of another cranial vault, probably adult," all very large, "the adult skull having a capacity of approximately 1450 cm3."  The adult skull is longer than any of more than 5,000 modern crania, but its high, rounded shape fits our species.²⁶  White's team called the people "morphologically and chronologically intermediate between archaic African fossils and later anatomically modern Late Pleistocene humans."³¹  They gave them a subspecies name, idaltu, which Stringer thinks "does not seem justified."²⁶  The skulls carry "bone modifications indicative of deliberate mortuary practices," the animal bones show "repeated, systematic butchery of hippopotamus carcasses," and the tools mix Acheulean and Middle Stone Age types.  "The fossils and artefacts are dated between 160,000 and 154,000 years ago by precise age determinations using the 40Ar/39Ar method."¹¹  The two ends of that range come from different places.  Pumice and obsidian pebbles inside the fossil sandstone give a maximum of about 160,000 years.  The ash that caps the sandstone was matched by chemistry to a layer at Konso, 800 km south, that lies below a tuff dated to 155,000 ± 14,000 years.⁶

Later finds, and one that fell apart

The fossils of Skhul and Qafzeh in Israel are dated by "luminescence, ESR and U-series" to about 90,000 to 120,000 years.²⁶  In China, 47 teeth from Fuyan Cave were published in 2015 as "dated to more than 80 kyr and with an inferred maximum age of 120 kyr."³²  That claim is now in serious doubt, and Part 2 explains why.

Part 2: How the dates were obtained, and how far they can be trusted

The methods behind these dates each have an article of their own on this site, and that is where the young-earth objections to each method are taken up:  luminescence dating, uranium-series dating, and argon dating in Radiometric Concordance and Radiometric Anomalies.  This part says what each method measured at these sites, where two methods meet, and how the figures have moved.

The clocks, in plain words

Thermoluminescence on burnt flint (Irhoud, Misliya).  Flint, like quartz, traps a little energy from the natural radiation around it, which comes from uranium, thorium and potassium in the soil and from cosmic rays.  Strong heating empties the traps.  Heating a sample again in the laboratory releases the stored energy as light, and the light measures the dose received since the last fire.  In Richter's words, "Thermoluminescence dating of heated flint artefacts is based on the accumulation of metastable charges (palaeodose) in the crystal lattice by ionizing radiation since the last heating of the rock."  The age is the dose divided by the yearly dose, "with the assumption that the dose-rate was constant over the entire burial time."⁴  At Irhoud the yearly gamma dose was measured in place with 47 small dosimeters buried in the layers, and averaged 807 ± 107 millionths of a gray a year in layer 7.⁴  The layer-7 flints hold 340 to 560 gray each.⁴  This is not a decay clock with a parent and a daughter.  It is a radiation meter, and the luminescence article explains why two of the three usual young-earth objections to radiometric dating do not apply to it.

Electron spin resonance on tooth enamel (Irhoud, Florisbad, Misliya) works on the same principle in tooth enamel.  It is harder, because a tooth soaks up uranium from groundwater after burial, and the age depends on when that happened.  Richter's team give two models for the Irhoud tooth, 281 and 286 thousand years, and say their range covers "most possible scenarios of how the uranium diffused into the dental tissue."⁴

Uranium-thorium on cave crusts (Misliya).  Calcite crystallizing from water takes in uranium but almost no thorium.  Uranium-234 then decays to thorium-230 at a known rate, and the ratio gives the time since the crust formed.  A crust grown over a fossil gives a minimum age for the fossil.  The same decay can be measured in bone and teeth (Omo I, Misliya-1, Apidima), but bone keeps taking up uranium from groundwater after burial, so a date on bone is read as a minimum.³³

Argon-argon on volcanic crystals, and ash fingerprinting (Omo, Herto).  Potassium-40 decays to argon-40.  Molten rock loses its argon, so the clock in a feldspar crystal starts when the eruption throws it out.  Each eruption's glass also has its own chemical fingerprint, so an ash layer at one site can be matched to its volcano and to ash layers elsewhere.  The known weakness is argon carried up from the magma, which makes a crystal look too old.  The isochron check described in Part 1 is the standard test for it, and at Shala it found no significant excess.⁶

Where two methods meet

The Misliya team wrote that "Three methods (combined US-ESR, U-Th, TL) yielded mean dates which coincide (they are in a range from 174 ky to 185 ky)."¹⁰  Those three methods rest on different things:  a dose of radiation in flint, a dose in enamel corrected for uranium uptake, and the decay of uranium-234 into thorium-230 in a crust.

Omo is the one real conflict.  A tuff below the fossils gave 196,000 years in 2005, and an ash above them was matched in 2022 to an eruption 233,000 years old.  Both cannot be right.  An earlier proposal had matched the Omo ash to the ash at Herto and put both near 172,000 years;  the Herto team rejected it from their own stratigraphy.⁶  Vidal's team side with their own figure, "Given the uncertainty in the age and stratigraphic placement of the Nakaa'kire Tuff, as well as its heterogeneous lithology and geochemistry."⁶  The disagreement is 37,000 years, about 19 percent.  It is a real error in one of the two papers, and on either figure Omo I is close to 200,000 years old.

Which way the dates moved

At Irhoud, Richter's team describe the whole site's new figures as "almost double the previous age estimates for the lower part of the deposits."⁴  The same tooth was dated twice, and the second figure is higher because the radiation dose in its surroundings was measured properly the second time.⁴  These revisions are a few tens of percent to a factor of two.  They are what "even the scientists disagree" amounts to at these sites, and they run in the wrong direction for a young-earth reading.

How a date can be wrong by tenfold: Fuyan Cave

One claim in this literature did collapse by as much as a young-earth reading needs, and it is worth telling in full.  In 2015 Wu Liu, María Martinón-Torres and colleagues described 47 human teeth from Fuyan Cave, southern China.  A small stalagmite that grew on top of the fossil layer gave about 80,000 years, so the teeth were older than that.  The upper limit of 120,000 years was not measured;  the abstract calls it "an inferred maximum age."³²

In 2021 Xue-feng Sun and colleagues dated two teeth from the site directly by radiocarbon, and tested them for ancient DNA.  The human teeth gave 2,510, 2,540 and 9,380 calibrated years before present.¹³  The flowstones they dated over the same layer still came out old, 95,000 to 168,000 years,¹³ so they argued that the teeth had worked their way down into much older cave fill:  "Still, it is clear that they were all buried in cave sediments during the Holocene and not 120 to 80 ka, as suggested previously for Fuyan Cave."¹³

The 2015 team answered that one of the two "human" teeth is not human:  "Critically, we confirm that FY-HT-2 is not human but belongs to an herbivore."¹⁴  Thomas Higham and Katerina Douka, radiocarbon specialists not involved in either study, wrote that the radiocarbon samples fail the usual tests for well-preserved bone protein:  "We hold that the radiocarbon data are essentially unreliable, and the results should be considered minimum ages."³⁴  Sun's team replied, holding to their reading that "U-Th ages cannot represent the burial age of the paleoanthropological materials at Fuyan Cave."³⁵  The dispute is open.

The lesson cuts both ways.  For the young-earth side:  a published "oldest modern human" can be wrong by a factor of ten or more, and specialists may take years to notice.  For the skeptic:  the error at Fuyan, if it is one, is not a failed clock.  The flowstones were dated correctly.  The question is whether the teeth belong to the layer the flowstones cap.  That is why it matters how each fossil in this article is tied to its date.  At Irhoud the main clock dates the people's own fires, and one human tooth was dated directly.  At Omo and Misliya the bone and tooth themselves were dated by uranium-series, and both teams report that the fossils are not later intrusions.¹⁵¹⁰  At Herto the fossil sandstone holds dated volcanic pebbles and is capped by a matched ash.  Florisbad has no such check, which is one reason Berger and Hawks doubt it.

Part 3: The genetic side, briefly

Living people carry their own record of how long ago their ancestors' populations separated.  The site covers the details in Mitochondrial Eve, Y-Chromosome Adam and Noah, Adam, Eve, and Human Genetic Diversity and Out of Africa or Out of Babel?.  Here the only question is how the genetic dates compare with the fossils.

The deepest split among living people.  In 2017 Carina Schlebusch's team sequenced the genome of a boy buried about 2,000 years ago at Ballito Bay, South Africa, whose ancestry had not yet been mixed with that of later herders and farmers.  Comparing it with other genomes, they estimated "the first modern human population divergence time to between 350,000 and 260,000 years ago."¹⁶  A 2023 study by Aaron Ragsdale and colleagues fitted a different kind of model, in which several weakly separated populations kept exchanging migrants, and found that "The earliest population divergence among contemporary populations occurred 120,000 to 135,000 years ago and was preceded by links between two or more weakly differentiated ancestral Homo populations connected by gene flow over hundreds of thousands of years."  They argue that "model misspecification explains the variation in previous estimates of divergence times."¹⁷  The two estimates differ by a factor of about 2.5, because they answer different questions under different models.  Both stand 28 to 80 times above a Flood date.²⁰

The single-line clocks.  The site's Y-chromosome article finds that a mutation rate measured only in fathers and sons, with no fossil or radiometric input, puts the root of the human Y tree at about 163,000 years.  The mitochondrial article puts Mitochondrial Eve near 150,000 years.  These single lines of descent are expected to be younger than the deepest population split, and they are.

How the genetic and fossil dates relate.  They are not fully independent.  Every genetic date is a count of differences divided by a mutation rate, and Schlebusch's team say so:  "although our assumptions on mutation rates and generation times influence the exact chronological estimates, our results notably increase the time depth for the deepest split for modern humans on a relative scale."¹⁶  Some published rates are calibrated against fossil ages, and some, like the pedigree rates, are not.  What the comparison shows is consistency, not proof:  the Irhoud fossils at about 315,000 years sit at or before the deepest genetic split, and Vidal's team note that their Omo date is "congruent with most models for the evolution of modern humans, which estimate the origin of H. sapiens and its divergence from archaic humans at around 350–200 ka."⁶

The young-earth genetic case.  Nathaniel Jeanson of Answers in Genesis puts the origin of the major ethnic groups at Babel, "around ~4000 years ago," and writes that his results "underscore the biblical model of human origins in which the major ethnic groups originate simultaneously at the Tower of Babel event."³⁶  With Ashley Holland he argued that Y-chromosome mutation rates from high-coverage sequencing "confirm a 4,500-year history for human paternal ancestry."³⁷  Their own figure caption admits the gap at exactly the point that matters here:  the predictions "captured all but the A00 branch length values,"³⁷ and A00 is the deepest African branch of the Y tree.  Robert Carter of Creation Ministries International rejects the clock outright:  "I do not assume the 'molecular clock' works and have good reason to question it."³⁸  The site's Y-chromosome article tests Jeanson's rate and finds that it comes from a filtering choice, not from sequencing coverage.  On the genetic side, as on the fossil side, the young-earth reading needs every published rate to be wrong by a factor of 28 to 80.

Part 4: What young-earth writers say, in their own words

This part states the young-earth position before anything is priced against it.  It quotes the writers who discuss these fossils directly, oldest first within each point.

They were people

Young-earth writers who discuss these fossils agree that they were human, and mainstream anatomy gives them no reason to think otherwise.  Marvin Lubenow, author of the young-earth survey of human fossils Bones of Contention, rejects the "African Eve" model but calls the Herto finds "three remarkable early Homo sapiens fossil skulls."³⁹  Carl Wieland and Jonathan Sarfati of Creation Ministries International called Herto "just one more example of fossil human bones."¹⁸  Of Omo II's older-looking features, Wieland wrote that "they are a part of the range of variation within the human gene pool after Babel."⁴⁰  Todd Wood, a young-earth biologist who has published statistical tests of which fossils group with humans, wrote of Irhoud:  "From a creationist perspective, these fossils are interesting but not all that different from what we've already known as human."¹  David Menton of Answers in Genesis called them "Homo sapiens (fully human people)."²  Marcus Ross, Peter Brummel and Wood wrote in 2023 that "anatomical similarity to living humans and abundant evidence of advanced human culture give unanimous testimony of the humanity of even the “oldest” H. sapiens remains on the conventional timelines," naming Herto and Irhoud, the latter "found with evidence of controlled fire, stone tools, and cooking."¹⁹

They lived after the Flood, and after Babel

The same writers place these people after Noah's Flood and after the dispersal from Babel.  Lubenow:  "Humans did not first appear in Africa.  After the Genesis flood, the first humans appeared in Turkey or Armenia."³⁹  Wieland and Sarfati on Herto:  "probably post-Babel and thus post-Flood."¹⁸  Brian Thomas of the Institute for Creation Research gives the reason:  human remains belong in "Deposits that date from the post-Flood Ice Age until the present," because people spread "after the dispersal event at the Tower of Babel, an event that occurred a few hundred years after the Flood."⁴¹  Ross, Brummel and Wood:  "The historical setting of these remains is post-Flood on the basis of geological indicators, evidence of in situ habitation, and the provenance of source rocks employed as tools.  Furthermore, given the universal perspective presented in Genesis 11, the dispersal of humans most likely occurs post-Babel."¹⁹

Three writers put numbers on it.  Snelling, Answers in Genesis's research director, dates Babel "just over 4,000 years ago" and the Flood "about 4,350 years ago."²²  Jeanson places Babel "around ~4000 years ago."³⁶  Menton applies this to Irhoud directly:  "a supposed TL age of 315,000 ± 34,000 years makes these bones those of Homo sapiens (fully human people) who lived during the Ice Age, which biblically occurred after, and as a result of, the global Flood cataclysm.  Thus, the real age of these bones is only probably around 4,000 years, and the bones are those of Noah’s descendants."²

One young-earth writer's view sits uneasily with this.  The physicist Jake Hebert wrote in Creation Ministries International's Journal of Creation in 2024 that "There is very strong evidence for a 'high' Flood/post-Flood boundary in the rocks; generally speaking, high or higher than the Mid-Pleistocene," and limited his own study to the Late Pleistocene and Holocene for that reason.⁴²  Every fossil in this article is Middle Pleistocene by its measured age.  If the boundary sits that high, some of these deposits could be Flood sediment.  Ross, Brummel and Wood argue the opposite from the sites themselves:  hearths, stone tool manufacturing sites and animal bone deposits are "ephemeral remains" that "would not survive the cataclysm of a global Flood."¹⁹  No young-earth source read for this article reconciles the two.

What they say about the dates

Here the young-earth writers differ, and most say little.

Rejected outright, with general reasons.  Lubenow:  "as a young-earth creationist, I do not accept those dates.  In fact, I will demonstrate that they are nothing but science fiction."³⁹  He also argues that, even granting the methods, "the dates used to establish human evolution are, for all practical purposes, worthless,"³⁹ and that the methods of his day had a "coverage gap" from about 40,000 to 200,000 years.³⁹  Peter Line of Creation Ministries International, on the Omo argon dates:  "Both the potassium-argon (K-Ar) and argon-argon (40Ar/39Ar) methods have pitfalls, and so the above 40Ar/39Ar ages are not accepted here.  Aside from this, that these types of fossils keep getting redated indicates uncertainty about the dates by the evolutionists themselves, as why redate them otherwise."²⁹  Wieland keeps the order but not the numbers:  "The fact that we don't accept the absolute value of the radiometric age … does not alter the fact that they are located such as to be the same age."⁴⁰

Specific objections to the Irhoud date.  Menton makes four.  The error bar is wide:  "the stated error margin of this claimed TL age is more than ± 10%."  The dates have changed:  "As the old Chinese proverb says, “A man with two clocks never knows what time it is.”"  The method was not used on bone:  "But note it was not the Homo sapiens’ bones that were dated but rather the spear points and tools assumed to belong to them."  And the method is mostly used on pottery, where "the dating error can be considerable."²  Creation Ministries International makes a similar point about Apidima, quoting the uranium-series specialist Warren Sharp:  "You have this huge spread of apparent ages, and you don’t know if any of them are any good."⁹

The mechanism.  Snelling's chart for human fossils sorts the dating methods into three rows.  For argon and uranium dates the stated problem is decay rate:  "Multiple lines of evidence show that radioactive elements decayed rapidly during and soon after the flood, so these “early” dates must be calibrated to the biblical timeframe—a matter of months, years, or decades, not millions of years."  For luminescence and electron spin resonance, the problem is different:  scientists "cannot be sure if they really started at zero or if impurities affected how much energy or radioisotopes the samples stored."²²  Two years earlier, Snelling and his co-authors at the Institute for Creation Research had placed the end of accelerated decay at the end of the Flood:  "such accelerated decay would have decelerated at the end of the Flood.  Hence, creation scientists think, as do uniformitarian scientists, that decay rates have been essentially constant during most of the post-Flood period."²¹

Candour.  Wood:  "there's a huge amount of uncertainty for creationists when re-interpreting a combination of thermoluminescent dating, molecular clock inferences, and other forms of fossil dating.  I'm not really qualified to comment on thermoluminescent dating, but I know when inferences are built on inferences which are built on other inferences.  That's a time to be reasonably skeptical."¹  Ross, Brummel and Wood:  "Other radiometric methods are still under investigation, and none of the dating methods have a widely accepted method of recalibrating to a creationist timescale."¹⁹  Rupe and Sanford take the opposite tack and use the new date against evolution.  For them, the 315,000-year Irhoud date "greatly confounds any claims of an evolutionary progression within the genus Homo," and paleoanthropologists "can no longer trust the techniques they have used to date hominin bones and artifacts."⁴³  They print the conventional dates and give no young-earth date for any of these fossils.

What none of them does

No young-earth source found for this article names a faulty input to any measurement in Parts 1 and 2:  a wrong dose rate, a contaminated sample, an unbleached flint, a misidentified ash.  None engages the Misliya agreement between three methods, or the Shala ash chemistry.  None gives a young-earth date for any of these fossils by a stated method, except Menton's "probably around 4,000 years," which is read off the biblical chronology, not off the evidence at Irhoud.  Searches behind this finding are listed in the literature review.

Part 5: What a young-earth reading has to explain

The size of the gap

The young-earth position, in its own words above, is that these people lived after Babel, about 4,000 years ago.  The table divides each measured age by the time since the Flood.  The Flood date used is Snelling's 4,350 years, and in the last column the longest any young-earth chronology allows, 5,411 years before 2026, from Chris Hardy and Robert Carter's survey of the biblical texts.⁴⁴  Using the Flood date rather than Babel makes every factor smaller than the young-earth position actually needs, since these people lived after Babel.  The factors are this article's arithmetic, not any young-earth writer's.²⁰

Set this against the disagreements inside the mainstream literature.  The largest revision at these sites is a factor of 1.8, at Irhoud, and it made the fossil older.  The Omo conflict is about 19 percent.  The error bar Menton calls too wide is about 11 percent at one standard error.  The young-earth reading needs every one of these figures to be too high by a factor of 28 to 80.  "The scientists disagree" is true;  they disagree by tens of percent, and the gap to a Flood date is thousands of percent.

What started each clock

Every clock in this article was started by an event that, on the young-earth account, happened after Babel.  That is the heart of the problem, and it does not depend on any assumption about decay rates.

• Irhoud and Misliya: people's fires. A thermoluminescence clock in flint starts at zero when the flint is heated. The Irhoud flints were heated in the fires of the people whose bones lie with them, the "controlled fire" that Ross, Brummel and Wood count as a mark of humanity.¹⁹ On the young-earth account those people lived after Babel, so everything the flints have absorbed, 340 to 560 gray each,⁴ has been absorbed since Babel.
• Irhoud 3 and Misliya-1: tooth enamel. Enamel grows inside a living child's jaw. Its radiation clock starts when the tooth forms, so there is no inherited signal to argue about. Snelling's objection that luminescence and spin-resonance clocks may not have "started at zero"²² does not apply to a tooth. The Irhoud 3 tooth reads 286,000 ± 32,000 years and the Misliya tooth 174,000 ± 20,000.
• Omo I: an eruption over buried bones. The Shala ash fell on ground where Omo I already lay buried.⁶ If Omo I lived after Babel, the eruption was later still, and the 233,000 years of argon in its crystals built up after Babel.
• Misliya: a crust on the jaw. Calcite grew over the jaw after it was buried, and holds 185,000 ± 8,000 years of uranium decay.¹⁰
• Herto: an ash over the dead. The ash that caps the Herto sandstone settled after the people died, and its match at Konso lies below a tuff dated to 155,000 ± 14,000 years, which erupted later still.⁶

What the young-earth physics offers for that window

The main mechanism the young-earth literature proposes for shrinking radiometric ages is a past episode of faster radioactive decay, worked out by the RATE project.  Its own authors put that episode in the Flood year and say it "decelerated at the end of the Flood," leaving decay rates "essentially constant during most of the post-Flood period."²¹  Snelling's 2018 article widened it to "during and soon after the flood."²²  ICR's Thomas puts Babel "a few hundred years after the Flood."⁴¹  On the young-earth writers' own placement of both events, then, the clocks in this article ran entirely or almost entirely after accelerated decay had stopped.  This is the same finding the site reached for corals and cave formations in Uranium-Series Dating, for the cave fills in The South African Australopiths, and for post-Flood dunes in Luminescence Dating.  The difference here is that the clocks were set by human hands.

If the speed-up ran after Babel anyway

No young-earth writer has proposed this, but it can be priced.  At Irhoud, the flints must absorb 315,000 years' worth of today's radiation in the time since their fire.  If the extra decay was spread over the whole 4,350 years since the Flood, the decay of uranium, thorium and potassium in the cave fill had to run about 72 times faster than today.  If it was packed into a shorter window, with today's rate the rest of the time, it had to run faster still:²³

Three conditions come with it.

1. The people lived through it. The radiation the flints recorded came from the uranium, thorium and potassium in the cave fill, the same ground the people sat and slept on, and potassium-40 is in every human body. Whatever the speed-up factor was, everyone at the site received that many times today's natural radiation from those sources, for the whole window. Today that dose is about 1.1 millisieverts a year: the gamma rays measured in layer 7, 0.81 thousandths of a gray a year,⁴ plus 0.29 millisieverts from the potassium, uranium and thorium inside the body.⁴⁵ Sped up 72 times over the whole 4,350 years, that is about 79 millisieverts a year. The most exposed people in Ramsar, Iran, live today with up to 131 millisieverts a year from external sources, plus up to 72 from radon.⁴⁶ So the longest window gives a dose that some people live with today, and this condition does not rule it out. The shorter windows are another matter: about 340 millisieverts a year if the speed-up lasted 1,000 years, and about 3,400 a year if it lasted 100, which is 26 times the highest external dose any Ramsar resident receives.²⁴ The long window has its own problem, which is the RATE authors' statement that decay rates have been "essentially constant during most of the post-Flood period."²¹
2. Different elements had to speed up together. The Misliya dates rest on a dose from uranium, thorium and potassium in flint, on uranium-234 decaying to thorium-230 in a crust, and on a dose in enamel, and they agree. The Irhoud flint and tooth ages agree. A speed-up that hit different elements by different amounts would have pulled these apart. The RATE authors themselves reported a pattern in which the speed-up "seemed to be proportional to the atomic weights of the parent radioisotopes,"²¹ which would have spread the clocks apart, not kept them together. The site's uranium-series article shows from measured uranium ratios how tightly such a non-uniform speed-up is limited.
3. The order has to survive. Across the sites, the ages fall in one sequence: Irhoud, then Florisbad, Omo, Apidima, Misliya, Herto, then Skhul and Qafzeh. On average, the fossils with the most rounded, modern braincases come later in that sequence, though with exceptions, set out under the objections below.²⁷ A young-earth map from measured ages to real ages must keep that order for luminescence, spin resonance, uranium-series and argon alike.

None of this proves the conventional ages.  It states what the young-earth alternative has to supply, and that nobody has yet supplied it.

If the people lived in the first years after the Flood

The other way to bring these clocks inside a speed-up is to move the people earlier, into the years just after the Flood, where Snelling puts rapid decay "during and soon after the flood."²²  The one young-earth curve with numbers points that way.  Robert Brown's 1992 radiocarbon curve, worked through in Neanderthals and Denisovans, puts the end of the Neanderthal tool industry, about 34,000 to 36,000 radiocarbon years, 7 to 10 years after the Flood.  It puts Babel, 101 years after the Flood on the Hebrew text, at a radiocarbon age of about 18,000.²⁵  Every site in this article is far older in the conventional sequence than the last Neanderthals.  If a young-earth timeline keeps that order, as Wieland does for the two Omo skulls,⁴⁰ then on Brown's curve these sites fall within a decade of the Flood.  That is inside the window Snelling gave in 2018, though not the one he and his co-authors gave in 2016, which ends with the Flood.²¹

This route has three costs.  It puts people in Morocco, Ethiopia, South Africa and Israel, burning fires and butchering hippopotamus, within ten years of the Flood and about a century before Babel.  Ross, Brummel and Wood reject that from Genesis 11:  "given the universal perspective presented in Genesis 11, the dispersal of humans most likely occurs post-Babel."¹⁹  It uses Brown's curve past the limit Brown himself set:  "Use of Equation 8 for C-14 ages greater than 34,000 is highly questionable."⁴⁷  And conditions 2 and 3 above still apply.  No young-earth writer has proposed this placement for these fossils.  It is set out here because it is the only published young-earth number that reaches them, and because it shows the choice a young-earth reading faces.  After Babel, the clocks ran after fast decay had stopped, on its own authors' account.  Before Babel, Genesis 11 as these writers read it has no people in Africa to light the fires.

Objections and responses

"The error margin is more than ten percent, so the date is not precise."  This is Menton's point, and the figure is right:  315,000 ± 34,000 is about ±11 percent at one standard error, and the two-standard-error range is 247,000 to 383,000 years.⁴  But a young-earth age of about 4,000 years is not at the edge of that range.  It is 1.3 percent of the central value, which would be an error of more than 98 percent.  Imprecision of 11 percent does not open a door that wide.

"A man with two clocks never knows what time it is."  The two Irhoud clocks agree within their errors, 315,000 and 286,000 years.  At Misliya three methods agree within about six percent.  The earlier Irhoud figure of 160,000 years was replaced because its dose rate came from "a small sediment sample of imprecise stratigraphic context,"⁴ and the replacement made the site older.  Disagreement between clocks is how errors are found, and at these sites every correction moved the date away from a young earth.

"They dated the tools, not the bones."  Menton is right that the Irhoud thermoluminescence date is on flints, and Rupe and Sanford are wrong to describe "a jawbone that dates 315,000 years old."⁴³  The link between flints and bones is by layer, and Richter's team say they cannot tighten it "beyond the resolution of an individual stratigraphic deposit."⁴  Two things limit the force of the objection.  The flints were burnt in human fires, so they date people being there, which on the young-earth account is post-Babel human activity in any case.  And a human tooth from the site was dated directly, to 286,000 ± 32,000 years.⁴

"These fossils keep getting redated, which shows the scientists are unsure."  Line's point.  The revisions are real, and Part 2 lists them.  They are factors of 1.2 to 1.8, and all but one went older.  The exception is Fuyan, where the dispute is whether the teeth belong to the layer, and that is still open.  Uncertainty of that size does not reach a factor of 50.

"Radioactive decay was faster in the past."  This is the main mechanism the young-earth literature offers.  Its authors put it in the Flood year or just after.  The clocks here were set after Babel, by fires, by eruptions over buried bones, and by teeth growing in children's jaws.  Part 5 gives the speed-up a post-Babel version would need, from 72-fold to more than 3,000-fold depending on the window, and the conditions that come with it.  No young-earth physicist has proposed it.

"Luminescence clocks may not have started at zero."  Snelling's objection.²²  An incompletely reset flint does read too old, and the mainstream checks for it:  Richter's team tested 45 heated flints and kept only the 14 that passed.⁴  A leftover signal would have to supply about 98.6 percent of the measured dose for the true age to be 4,350 years, in flints chosen because they passed the test for full resetting.  Tooth enamel has no inherited signal at all, and the Irhoud and Misliya teeth agree with the flints.

"Impurities may have changed how much energy the samples stored."  Also Snelling's.  Each sample's response to radiation is measured directly in the laboratory, by giving it known doses and measuring the light;  Richter's team used "a multiple aliquot additive regeneration slide approach" for this.⁴  A sample whose sensitivity was changed by impurities shows it in that test.

"Argon dates on young volcanic rock come out far too old."  Lubenow cites lavas from historic eruptions of Mount Ngauruhoe in New Zealand:  "The obvious problem is that argon gas was already in these rocks when they solidified."³⁹  Line rejects the Omo argon ages because the methods "have pitfalls."²⁹  The problem is real, and the site's Radiometric Anomalies article sets out how large it is and how it is detected.  At Omo the isochron test on the Shala crystals found trapped argon close to the ratio in air, and gave the same age as the standard calculation.⁶  At Herto the maximum comes from pebbles inside the fossil sandstone and the minimum from a separate tuff above its matched ash, and the two fall in the right order.

"Fuyan shows that 'oldest modern human' claims can be wrong."  Yes, if Sun's team are right, and this article says so in Part 2.  The Fuyan problem was the link between the teeth and the dated layer, not a failed clock.  Omo I, Misliya-1 and the Irhoud 3 tooth were dated directly.

"The Middle Pleistocene might be Flood sediment, so these are pre-Flood people."  Hebert's high Flood boundary points that way.⁴²  But these are living sites:  hearths, burnt flint, butchered hippopotamus, tool-making floors.  Ross, Brummel and Wood argue that such "ephemeral remains would not survive the cataclysm of a global Flood."¹⁹  A young-earth reading has to choose, and the evidence at Irhoud and Herto favours a post-Flood setting, which brings back the timing problem.

A skeptic's objection:  "If young-earth creationists admit these are humans, they have admitted humans are 300,000 years old."  Only if the dates are granted.  The young-earth claim is that the bodies are human and the ages are wrong.  This article's finding is that the first half is well supported and the second has no worked answer.

A skeptic's objection:  "The fossils show a gradual change toward the modern skull, which is evolution."  The Irhoud faces are already modern, and on average the braincases become rounder over time.²⁷  The trend is real, but it is a tendency, not a ladder.  Simon Neubauer's team measured brain casts of 20 fossil Homo sapiens in three age clusters, about 300,000 to 200,000, 130,000 to 100,000 and 35,000 to 10,000 years, with none in between, and found that age "explains 21.9% of shape variance."²⁷  The rest is variation among people living at the same time.  And the order has crossings.  A person buried about 13,000 years ago at Iwo Eleru, Nigeria, had a long, low braincase "outside the range of modern human variability," most like Irhoud, Skhul and Qafzeh, while Qafzeh 9, from the group of about 100,000 years, falls with the Upper Palaeolithic Europeans.³³  Omo I and Omo II, from similar levels, differ the same way.³⁰  Young-earth writers read all of this as variation inside one human kind, and the crossings fit that reading better than a strict sequence would.  The skeptic's reply is fair too.  Iwo Eleru's describers explain it by "late survival of “archaic” features and possibly deep population substructure in Africa,"³³ and a trend with older forms surviving in places is what a branching population history predicts.  What young-earth writers have not explained is why the average shape changes in time order on every dating method.

A young-earth objection:  "The genetic dates assume the molecular clock."  They do, and Schlebusch's team say the clock's rate sets the exact figures.¹⁶  The site's genetics articles test the young-earth alternatives;  Jeanson and Holland's own Y-chromosome clock does not fit the deepest African branch.³⁷

Issues and unanswered questions

1. Several primary papers were read only in abstract. White 2003 and Clark 2003 (Herto), McDougall 2005 (Omo), Aubert 2012 (the direct date on Omo I), Smith 2007 (Irhoud 3) and Harvati 2019 (Apidima) were not available in full. Aubert's actual uranium-series numbers, Clark's sample counts, and Harvati's own caveats should be added when the full texts are read.
2. Grün 1996, the Florisbad date, was not read at all. Its figure is reported here through Berger and Hawks, who are its critics. The published version of their 2023 note was not compared with the preprint read.
3. The Omo conflict is unresolved. A tuff below Omo I gave 196,000 years and an ash above it was matched to an eruption 233,000 years old. Vidal's team prefer their own figure; no reply from McDougall's group and no re-dating had been published by October 2026.
4. The critiques of Apidima 1 are known only from the press. Sharp's, Arsuaga's and Stringer's remarks are quoted from National Geographic and from Creation Ministries International's quotation of Lizzie Wade in Science. No peer-reviewed reassessment of Apidima 1 was found, and Marie-Antoinette de Lumley's 2019 monograph, which classes both skulls differently, was not obtained. Sharp's reasons for putting Misliya-1 under 70,000 years are not known.
5. The Fuyan dispute is open, and no independent re-dating of a securely provenanced tooth has been published.
6. The dose-rate fractions at Misliya were not obtained, so the test of whether a speed-up that favoured heavy elements could keep the three Misliya methods in agreement was not computed. It could be, from the supplement's dosimetry tables.
7. Jeanson's Traced (2022) was not reached. Its timeline for the deepest African lineages is known here only through Carter's review.
8. A small young-earth misstatement. Ross, Brummel and Wood write that the Herto remains "are dated to >200,000 years ago on the conventional timeline."¹⁹ Clark's team dates them to 160,000 to 154,000 years.¹¹ Nothing in their argument depends on it.
9. The young-earth camp has not settled where the Middle Pleistocene sits relative to the Flood. Hebert's high boundary and the post-Flood placement of these sites by Answers in Genesis, Creation Ministries International, and Ross, Brummel and Wood are not reconciled in any source read here. Brown's radiocarbon curve, if the conventional order is kept, puts these sites within a decade of the Flood: close to Hebert's boundary, and a century before the post-Babel placement.
10. The radiation dose in condition 1 leaves out radon, which depends on the cave's air and averages 1.15 millisieverts a year worldwide. Adding it would roughly double the doses given there.²⁴

Verdict

The fossils.  Jebel Irhoud, Florisbad, Omo Kibish, Misliya, Herto and probably Apidima 1 are human fossils on any reading.  The mainstream papers call them early Homo sapiens with a mixture of modern and older features, and every young-earth writer who discusses them calls them people descended from Noah.  On that question the evidence and the young-earth position agree.

The dates.  They come from thermoluminescence, electron spin resonance, uranium-series and argon dating, on flint, enamel, crust and volcanic crystals.  Where two methods meet on one horizon, they agree within their errors, except at Omo, where they differ by about 19 percent.  The revisions have run older by factors of 1.2 to 1.8.  The one claim that may have collapsed, at Fuyan, failed on the link between fossil and layer, which the sites here test directly.  The genetic split times fall in the same range.

The young-earth answer.  Young-earth writers place these people after Babel, about 4,000 years ago.  That needs every date here to be 28 to 80 times too old.  They give general reasons to doubt the methods, and Menton's point that the Irhoud date is on flints is fair.  None names a faulty input to any of these measurements.  The clocks were started by human fires, by eruptions over buried bones and by teeth growing in children, all after Babel on the young-earth account.  The main mechanism on offer, accelerated decay, is placed by its own authors in or just after the Flood year.  The one published young-earth number that could move these clocks into those years, Brown's radiocarbon curve, does it by putting these people within a decade of the Flood and before Babel, which the same writers rule out from Genesis 11.  Until someone proposes and prices a post-Babel speed-up, the dates of the oldest Homo sapiens stand unanswered.

Sources