Behavioral Modernity

Summary

1. The record is old and spread out. Pigment at Pinnacle Point in South Africa is dated to about 164,000 years. Shell beads at Bizmoune in Morocco are older than 142,000, burials and beads at Skhul and Qafzeh in Israel are 90,000 to 135,000, and engraved ochre, shell beads and a drawing at Blombos Cave in South Africa run from about 100,000 to 73,000. A child was buried at Panga ya Saidi in Kenya about 78,000 years ago, and the engraved ostrich eggshells at Diepkloof are 55,000 to 65,000 years old. A hand stencil in Sulawesi is older than 67,800 years, Chauvet's drawings date from 37,000 to 33,500 years, and people used Lascaux about 21,000 years ago. Neanderthals used pigmented shells by 115,000 years ago and engraved a cave wall more than 57,000 years ago. All of these are conventional ages.
2. The dates come from five or six methods that agree where they meet. Luminescence, thermoluminescence, spin resonance, uranium-series, radiocarbon and chlorine-36 rest on different physics. Their real disagreements, at Bizmoune, Chauvet and the Spanish caves, are factors of 1.5 to 2.6.
3. The "human revolution" failed. Richard Klein argued that modern behaviour began about 50,000 years ago through a genetic change, an idea he admitted "cannot be tested or falsified".¹ Sally McBrearty and Alison Brooks showed the pieces appearing in Africa tens of thousands of years earlier.² Many archaeologists now call "behavioral modernity" "theoretically obsolete".³
4. Young-earth writers read these people as fully human, and the record agrees. Ross, Brummel and Wood list artistic creativity and deliberate burial among the marks of "advanced, human cognition and therefore the image of God".⁴ Their expectation that human behaviour came with human bodies held up better than Klein's model.
5. They place all of it after Babel, inside a 700-year Ice Age, and mostly set the dates aside by the biblical bound. Answers in Genesis maps luminescence-dated carvings to "Centuries soon after the flood".⁵ Ross, Brummel and Wood concede that "none of the dating methods have a widely accepted method of recalibrating to a creationist timescale."⁴ No young-earth paper discusses the Blombos luminescence dates, the Sulawesi uranium-series dates or the Hulu Cave calibration.
6. The compression costs a great deal. On the writers' own figures the finds from Blombos to Lascaux must fit into 400 to 600 years, about 130 to 195 conventional years for each real year, and Pinnacle Point stretches it further. Every clock was started by people or after them, so any speed-up of decay would have to come after Babel, later than the scientists who proposed accelerated decay place its end: roughly 80 to 250 times today's rate if confined to the Ice Age, or 12 to 36 times on average over the whole time since Babel. Radiocarbon needs 3 to 12 times today's carbon-14 production for centuries. And at Hulu Cave the two kinds of clock were measured on the same calcite.
7. The common ground is real. Symbol-making as a mark of the image of God fits the record of who made symbols. It does not fit the young-earth record of when.

Verdict.  On the humanity of the symbol-makers the evidence supports the young-earth position.  On their dates it runs strongly against it, and no young-earth answer to the dating methods involved has been published.

Part 1: What "behavioral modernity" means, and why the term is fading

The term names a list of behaviours, not a body.  Archaeologists cannot see a mind.  They see what people left behind, and since the 1980s they have used "behavioral modernity" for the set of remains that mark the living human way of life.  Sally McBrearty and Alison Brooks gave the list in 2000:  "blade and microlithic technology, bone tools, increased geographic range, specialized hunting, the use of aquatic resources, long distance trade, systematic processing and use of pigment, and art and decoration."²  This article is about the symbolic part of that list:  engraving, pigment, beads, burial and painting.  Language itself, and the biology behind it, belong to Human Uniqueness:  Language, Mind, and FOXP2 and Did Language Begin Once?.

The late revolution, in Richard Klein's words.  Klein, a palaeoanthropologist at Stanford, gave the late-revolution model its fullest form.  His starting point was a lag.  "Human fossils from Klasies River Mouth and other African sites and from sites in Israel immediately adjacent to Africa show that people who were anatomically like us had appeared in Africa by 100,000 years ago.  Despite their modern appearance, however, these people left artifacts and animal remains which show that they were not fully modern in behavior.  It is only after 50,000 years ago that behavioral evolution caught up".¹  His explanation was biological:  "We suggest that this capacity stemmed from a genetic change that promoted the fully modern brain in Africa around 50,000 years ago."¹  He said plainly what that explanation could not do:  "it cannot be tested or falsified by experiment or by examination of relevant human fossils."¹  He defended it as "the simplest, most parsimonious explanation for the available archeological evidence."¹  Older finds did not move him.  Of a possible figurine from Berekhat Ram in Israel he wrote that, "like other occasional, supposed art objects from before 50,000 years ago, it does nothing to alter the impression of a creative explosion afterwards."¹  On burial:  "Much more elaborate graves with unequivocal ideological or religious implications show up only after 50,000 years ago".¹

The reply:  "The revolution that wasn't."  McBrearty and Brooks answered with a 111-page review of the African Middle Stone Age.  They put the revolution model's claim in one sentence:  that "modern human behaviors arose suddenly, and nearly simultaneously, throughout the Old World ca. 40-50 ka."  They called it the product of "a profound Eurocentric bias and a failure to appreciate the depth and breadth of the African archaeological record," because "many of the components of the "human revolution" claimed to appear at 40-50 ka are found in the African Middle Stone Age tens of thousands of years earlier."²  The pieces "do not occur suddenly together as predicted by the "human revolution" model, but at sites that are widely separated in space and time.  This suggests a gradual assembling of the package of modern human behaviors in Africa".²  They tied the start to the Middle Stone Age itself, "at 250-300 ka."²

A third view brought in the Neanderthals.  Francesco d'Errico and Chris Stringer set out the three positions in 2011.  The first is Klein's, a "genetic mutation that took place approximately 50 ka in Africa among already evolved anatomically modern humans."  The second is McBrearty and Brooks's, a gradual emergence "starting at least 200 ka".  The third "states that innovations indicative of modern cognition are not restricted to our species and appear and disappear in Africa and Eurasia between 200 and 40 ka before becoming fully consolidated."⁶  The third view matters for this article, because "not restricted to our species" means Neanderthals.

Why many archaeologists now drop the term.  Christopher Henshilwood and Curtis Marean argued in 2003 that the trait list itself was the problem:  many traits were "empirically derived from and context-specific to the richer European record", many were "ambiguous because other processes can be invoked, often with greater parsimony, to explain their character", and many "lack theoretical justification."⁷  John Shea went further in 2011.  He called behavioral modernity "deeply flawed at all epistemological levels," tested its core assumption on East African sites, and concluded "that behavioral modernity and allied concepts have no further value to human origins research."⁸  In 2023 Eleanor Scerri and Manuel Will reviewed the African record again, twenty-three years after McBrearty and Brooks.  They found that research had "continuously failed to find a discrete threshold for a complete 'modernity package' and that the concept is theoretically obsolete."  They also found against smooth gradualism:  "long periods of stasis and a lack of cumulative developments argue further against a strictly gradualistic nature in the record."  The record "bears no direct relation to a simplistic shift in the human brain but rather reflects similar cognitive capacities that are variably manifested."³

What the dispute is, and what it is not.  The live argument is about tempo and cause:  whether symbolic behaviour arrived as a package or piece by piece, and whether biology or population drove it.  Klein's 50,000 years is about 11 times the 4,350 years since the Flood on Andrew Snelling's young-earth date;⁵ McBrearty and Brooks's 250,000 to 300,000 years is 57 to 69 times.⁹  Nobody on either side of this debate dates the first engravings, beads or burials anywhere near the last 10,000 years.  "Even the scientists disagree" is true here, but they disagree about when inside a span of more than 200,000 years, not about whether the span exists.

The paradox the debate leaves behind.  Klein's lag between anatomy and behaviour was a puzzle for evolutionary theory, and Colin Renfrew named its later form the "sapient paradox":  "the human brain at birth, may not have changed greatly over the past 60000 years.  Yet many of the major behavioural changes that we associate with most human societies are very much more recent".¹⁰  The gradualist record shrank the lag but did not remove it, as Scerri and Will's "long periods of stasis" shows.  Part 6 returns to what each side makes of that.

Part 2: The record, oldest first

Each find below is given with its date, the method behind the date, and the excavators' own reason for calling it symbolic, followed by their own caveat.  Part 3 asks how far the dates can be trusted.  The papers write thousands of years as "ka".

South and North Africa

Pinnacle Point 13B, South Africa:  pigment at about 164 ka.  Curtis Marean's team excavated a sea cave on the south coast.  The lowest layer held shellfish, bladelets and 57 pieces of pigment.¹¹  It was dated by optically stimulated luminescence, a method that measures how long sand grains have been buried away from sunlight, to a weighted mean of 164,000 ± 12,000 years, with uranium-series dates on the flowstone that caps the deposit.¹¹  The pigment's purpose is a judgement, and the paper says so:  "Coincident with this diet and habitat expansion is an early use and modification of pigment, probably for symbolic behaviour".¹¹  Ochre also tans hides and makes glue, so ground ochre alone does not prove a symbol.

Bizmoune Cave, Morocco:  shell beads older than about 142 ka.  Thirty-three perforated sea-snail shells came from layer 4c, "making them the oldest shell beads yet recovered."¹²  The age is a minimum set by uranium-series on a speleothem, a cave carbonate that formed within the layer above most of the beads.  The paper prints its own conflict:  "OSL dates from layer 4c are uniformly younger than U/Th dates", and it leaves the cause "a topic for future research."¹²  The luminescence ages for the layer run from about 54 ka to 109 ka.¹²  So the beads are somewhere between about 60 ka and more than 142 ka, depending on which method is right.

Taforalt, Morocco:  beads at about 82 ka.  Thirteen perforated Nassarius shells, ten with red ochre residue, came from one layer in the Grotte des Pigeons.  Four methods dated the sequence:  radiocarbon on charcoal in the upper part, luminescence on 15 sediment samples and five burnt flints, and uranium-series on a flowstone below the bead layer.  "The age model constrains the main horizon containing the pierced Nassarius shells to between 73,400 and 91,500 years ago at 2σ".¹³  The authors hedge on intent:  the shells "were either perforated deliberately or, less probably, carefully selected on the beach for their large perforation, which is rare in nature".¹³

Blombos Cave, South Africa:  engraved ochre, beads and a drawing, about 100 to 72 ka.  Christopher Henshilwood's excavations produced three kinds of find.

1. Engraved ochre. In 2002 his team reported two pieces of red ochre with cross-hatched designs from the M1 phase, dated by thermoluminescence on five burnt stone tools to 77,000 ± 6,000 years, with the overlying dune sand dated by luminescence to about 70,000.¹⁴ Their verdict was firm on intent and open on meaning: "The Blombos Cave engravings are intentional images",¹⁴ which "may have been constructed with symbolic intent, the meaning of which is now unknown."¹⁴ In 2009 they added thirteen more incised pieces from layers dated about 75,000 to 100,000 years.¹⁵ They wrote their own caution into the paper: "We are aware that the above features do not guarantee the identification of intentionally made engravings",¹⁵ and called the evidence "admittedly fragmentary and the sample size limited".¹⁵
2. Shell beads. In 2004 the team reported 41 perforated tick shells, 39 of them from phase M1, dated by luminescence to 75,600 ± 3,400 years. Their case that people made the holes rests on a count: 88 percent are perforated in one spot near the lip, a type that "is absent in living populations and accounts for only 8.6% of naturally pierced shells in modern thanatocoenoses," the death assemblages on beaches.¹⁶ They concluded that the beads add "an unambiguous marker of symbolically mediated behavior".¹⁶
3. A drawing. In 2018 they reported "a cross-hatched pattern drawn with an ochre crayon on a ground silcrete flake recovered from approximately 73,000-year-old Middle Stone Age levels".¹⁷ Chemical analysis found iron-oxide pigment on a stone that contains none of its own.

The Blombos ages were refined after 2002.  The 2009 paper gives 72,700 ± 3,100 years by luminescence for the upper part of M1 and 74,000 to 78,000 by thermoluminescence.¹⁵  The differences are a few thousand years.

Diepkloof Rock Shelter, South Africa:  engraved ostrich eggshell, 55 to 65 ka.  Pierre-Jean Texier's team reported "270 fragments of intentionally marked ostrich eggshell".  The fragments come from at least 25 water containers, engraved with a small set of repeated motifs.¹⁸  Thermoluminescence, luminescence and a radiocarbon minimum together place them "securely between 55 and 65 ka."¹⁸  Their reason for calling it symbolic is the repetition and change of a few motifs over time, and they mark the limit:  "Whether specific markings always reflect representational intent is a matter of debate".¹⁸

Panga ya Saidi, Kenya:  a child's burial at about 78 ka.  María Martinón-Torres's team described "the partial skeleton of a roughly 2.5- to 3.0-year-old child dating to 78.3 ± 4.1 thousand years ago".¹⁹  Soil and micromorphology analyses "indicate that the pit was deliberately excavated", and the child lay "tightly flexed", in a position the authors call "consistent with the upper part of the body being wrapped".¹⁹¹⁹  The age rests on feldspar luminescence, including one sample from the pit fill, combined with radiocarbon and earlier quartz dates in a statistical model.  The authors note that the earlier quartz ages for these layers were too young.¹⁹  They do not claim the burial was symbolic, only deliberate.

The Levant: Skhul and Qafzeh

Two caves in Israel hold the oldest graves of anatomically modern people found so far, and the earliest shell beads outside Africa.

1. Skhul. Marian Vanhaeren's team matched the sediment stuck inside one perforated Nassarius shell to a layer "containing 10 human fossils and dating to 100,000 to 135,000 years ago". Their reason for calling the shells beads: "The remoteness of these sites from the seashore and a comparison of the shells to natural shell assemblages indicate deliberate selection and transport by humans for symbolic use."²⁰ The date comes from Rainer Grün's uranium-series and electron spin resonance work on the burials' bones and teeth: "If the Skhul burials took place within a relatively short time span, then the best age estimate lies between 100 and 135 ka", in agreement with thermoluminescence on burnt flint of 119,000 ± 18,000 years.²¹ Grün's team concluded that "the oldest known symbolic burials are those of early modern humans at Skhul and Qafzeh."²¹
2. Qafzeh. The lower layers, with hearths, graves, sea shells carried 35 km inland and lumps of red ochre, are dated by thermoluminescence on burnt flint and by electron spin resonance to about 90,000 to 100,000 years.²² The shells were found near the graves, but the excavators state that "there is no indication that they were directly associated with any of the human burials".²³ One grave is different. The adolescent Qafzeh 11, who had survived a severe head wound, was buried with red-deer antlers in contact with the skeleton. Hélène Coqueugniot's team called it "a unique case of differential treatment with convincing evidence for ritual behavior."²²

The Qafzeh papers do not all agree on the details.  The shell paper calls the holes natural in its abstract and describes percussion holes in its body, and the Qafzeh 11 paper gives the child's age at death as 12 to 13 years in one place and a brain volume like a 4- to 6-year-old's in another.²⁴  None of this moves the dates.

Before Homo sapiens?

Trinil, Java:  an engraved shell from Homo erectus, 430,000 to 540,000 years.  In 2015 Josephine Joordens's team re-examined freshwater shells in the collection of Eugène Dubois, who found the first Homo erectus fossils at Trinil in 1891.  One shell carries a geometric engraving.  Sediment inside the shells was dated by argon-argon and luminescence to between 0.43 ± 0.05 and 0.54 ± 0.10 million years.²⁵  The authors are careful:  "Although it is at present not possible to assess the function or meaning of the engraved shell, this discovery suggests that engraving abstract patterns was in the realm of Asian Homo erectus cognition and neuromotor control."²⁵  The weak link is that the shell came from a museum drawer, not a dated excavation, so its age rests on the sediment still inside it.

The claims for Homo naledi in South Africa, burial and engraving in a small-brained species, belong to Homo naledi:  Burial, Engravings, and Fire, which finds the burial evidence rated inadequate by its reviewers and the engravings undated.

Indonesia: the oldest dated paintings

The oldest dated cave paintings are not in Europe but on the island of Sulawesi.  They are dated by uranium-series on small knobs of calcite, called coralloids, that grew on top of the paint.  Calcite takes in uranium from drip water but almost no thorium, and the uranium decays to thorium-230 at a known rate, so the thorium-to-uranium ratio says when the calcite formed.  A crust on top of the paint is younger than the paint, so every date in this section is a minimum.  The series began in 2014, when Maxime Aubert's team dated crusts over twelve hand stencils and two animal paintings in the Maros karst:  "The earliest dated image from Maros, with a minimum age of 39.9 kyr, is now the oldest known hand stencil in the world."²⁶

1. Leang Tedongnge, a pig, at least 45,500 years. Adam Brumm's team dated a coralloid over a painting of a Sulawesi warty pig: "The oldest, with a minimum age of 45.5 ka, is from Leang Tedongnge."²⁷ They did not claim to know who painted it: "we are unable to definitively conclude that the dated figurative rock art depiction from Leang Tedongnge is the handiwork of cognitively "modern" members of our species", though they think it most likely.²⁷
2. Leang Karampuang, a scene, at least 51,200 years. In 2024 Adhi Agus Oktaviana's team used laser ablation to map the uranium and thorium across each crust and take the layer nearest the paint. A panel of human-like figures with a pig was "Painted at least 51,200 years ago" and is, in their words, "the earliest known surviving example of representational art, and visual storytelling, in the world".²⁸ Only one of the four crusts gave that figure; the others gave 18,700, 31,900 and 44,000 years as minimums, and the one over the pig itself gave 31,900.²⁸ The paper also reports "clear evidence of diagenesis", chemical alteration, at the surfaces of some samples and avoided those zones.²⁸
3. Liang Metanduno, Muna Island, a hand stencil, at least 67,800 years. In January 2026 the same group reported a calcite date of "71.6 ± 3.8 thousand years ago (ka), providing a minimum-age constraint of 67.8 ka for the underlying motif."²⁹ That is older than the Spanish stencil attributed to Neanderthals, discussed below. The authors say they cannot tell who made it: "There is no obvious method available to us to determine which human taxon produced the Liang Metanduno stencil", since Sulawesi had archaic humans before modern ones arrived.²⁹ And they warn against reading an order from minimums: "Differences between minima do not by themselves demonstrate that one painting predates another."²⁹

Europe: Chauvet

Chauvet, France:  charcoal drawings of lions, rhinoceroses and horses, about 37,000 to 33,500 years.  Chauvet's black drawings were made with charcoal, so the drawings themselves can be radiocarbon dated.  In 2016 Anita Quiles's team built a statistical model from 259 radiocarbon dates on drawings, wall marks and floor charcoal.  It "shows that there were two distinct periods of human activity in the cave, one from 37 to 33,500 y ago, and the other from 31 to 28,000 y ago."³⁰  To test for contamination they dated a second chemical fraction of 22 samples, which "concur with the dates of the associated purified charcoal fractions".³⁰

A second clock agrees.  The old entrance was sealed by rockfalls from the cliff above.  Benjamin Sadier's team measured chlorine-36, which cosmic rays build up in rock surfaces once they are exposed, on the scars left by the falls.  The exposure ages "demonstrate that the cliff overhanging the Chauvet cave has collapsed several times since 29 ka until the sealing of the cavity entrance prohibited access to the cave at least 21 ka ago", which they call "Remarkably agreeing with the radiocarbon dates of the human and animal occupancy".³¹  This method uses no carbon and no charcoal.  It cannot date a drawing, but it says no one has walked into the cave for at least 21,000 years on the conventional scale.

The same abstract records the dissent:  the Aurignacian and Gravettian ages for the art "are indeed still debated on the basis of stylistic criteria", citing Paul Pettitt.³¹  Pettitt's argument is that the drawings look too advanced for their radiocarbon age, which would make them younger, not older, by thousands of years.

Hohle Fels, Germany:  a carved figurine, at least 35,000 years.  Nicholas Conard reported a female figure carved from mammoth ivory from the lowest Aurignacian layer of Hohle Fels Cave:  "This figurine was produced at least 35,000 calendar years ago, making it one of the oldest known examples of figurative art."³²

Lascaux, France:  the occupation at about 21,000 years, the paintings undated.  Lascaux is the most famous painted cave, and its paintings have never been dated directly.  What is dated is the people's debris on the cave floor.  Sylvain Ducasse and Mathieu Langlais reviewed every radiocarbon result from the cave in 2019 and added new dates on reindeer bone.  Four gave reliable results, "centered on a 14C age of 17,600 uncal. BP (21.5-21 cal ka BP)".³³  The older results, "considering their standard deviations once calibrated", spread over "the 23.5-17 cal ka BP interval".³³  They state the gap themselves:  "the links between the new radiometric framework presented above and the parietal art have still to be discussed".³³

Neanderthals

The Neanderthal record is set out in Neanderthals, which covers their burials at La Chapelle-aux-Saints and Shanidar, the Bruniquel stalagmite rings, and the dispute over the Spanish cave paintings.  Four points matter here.

1. Shells and pigment at 115,000 years. At Cueva de los Aviones in Spain, Dirk Hoffmann's team dated the flowstone that seals a deposit with perforated, pigment-stained shells: "the symbolic finds made therein are 115,000 to 120,000 years old and predate the earliest known comparable evidence associated with modern humans by 20,000 to 40,000 years."³⁴
2. Eagle talons at about 130,000 years. At Krapina in Croatia, eight white-tailed eagle talons carry cut marks and polish. Davorka Radovčić's team argued that "These features suggest they were part of a jewelry assemblage".³⁵
3. Finger flutings sealed at least 57,000 years ago. At La Roche-Cotard in France, lines traced by fingers on a cave wall were sealed in when sediment closed the cave. Fifty luminescence ages date the sealing, so the engravings were "made more than 57±3 thousand years ago", before modern humans are known in the region.³⁶
4. The painted Spanish caves, in dispute. Hoffmann's team dated calcite over red paint in three Spanish caves to a minimum of 64,800 years, which "implies Neandertal authorship."³⁷ Randall White and 43 co-authors replied that "there exist more than 130 direct 14C dates for paintings in decorated caves and shelters. None of them are older than 38 ka".³⁸ The Neanderthal article weighs the dispute. Every number on either side of it is more than 38,000 years on the conventional scale.

The record in one table

Part 3: How the dates were obtained, and how far they can be trusted

What starts each clock

The record above rests on six methods, each with its own physics.  Each has its own article on this site, and the point here is what event starts each clock, because Part 5 turns on it.

1. Luminescence and thermoluminescence measure the radiation dose a grain of quartz or feldspar has absorbed from uranium, thorium and potassium in the ground around it. Sunlight empties the dose from sand; fire empties it from flint. So the clock starts when sand is buried or a flint is burnt in a hearth. See Luminescence Dating.
2. Electron spin resonance measures the same kind of dose in tooth enamel, starting when the tooth grew.
3. Uranium-series measures thorium-230 growing from uranium in calcite. It starts when the calcite forms. See Uranium-Series Dating.
4. Radiocarbon measures the carbon-14 left in charcoal or bone. It starts when the tree or animal died. Beyond the tree-ring record it must be calibrated against other clocks. See Radiocarbon Dating.
5. Chlorine-36 exposure measures what cosmic rays have made in a rock surface since it was laid bare. See Cosmogenic Nuclide Dating.
6. Argon-argon measures argon-40 from potassium decay in volcanic minerals.

Where the methods meet, they agree

1. Blombos, phase M1. Luminescence on the sediment gives 72,700 ± 3,100 years.³⁹ Thermoluminescence on burnt stone tools gives "a mean age of 74 ± 5 ka", which Chantal Tribolo's team called "in good agreement with both ESR dates on teeth and OSL dates on sediment".⁴⁰ Three methods, on sand, burnt flint and teeth, agree within their errors.
2. Skhul. Uranium-series and spin resonance on the burial layer give 100,000 to 135,000 years, which "agrees very well" with thermoluminescence on burnt flint at 119,000 ± 18,000.²¹
3. Chauvet. Radiocarbon on charcoal and chlorine-36 on fallen rock are unrelated methods, and they agree that the cave was used before 29,000 and sealed by 21,000 years ago on the conventional scale.³¹
4. Radiocarbon and uranium-series together. Radiocarbon dates older than the tree-ring record are calibrated mainly with stalagmites from Hulu Cave in China. Hai Cheng's team made "Paired measurements of 14C/12C and 230Th ages from two Hulu Cave stalagmites", "covering the full range of the 14C dating method (~54,000 years)."⁴¹ The international calibration curve, IntCal20, uses these "14C and U-Th measurements on the Hulu Cave MSD and MSL speleothems" back to 53,900 years.⁴² So the European radiocarbon dates and the Indonesian uranium-series dates sit on one timescale, tied together on the same calcite.

Where they disagree, and by how much

1. Bizmoune beads: uranium-series 142,000 against luminescence 54,000 to 109,000 in the same layer, a factor of up to 2.6, unexplained.¹²
2. The Spanish paintings: 64,800 years by uranium-series against a radiocarbon ceiling of 38,000 for cave paintings, a factor of 1.7.³⁷³⁸
3. Chauvet's style: Paul Pettitt and Paul Bahn argue that the radiocarbon dates put the black drawings "more than 20 000 years earlier than traditional stylistic models would suggest", and they caution "against reliance solely on the date of the charcoal."⁴³ Taken at its largest, that is a factor of about 2.
4. Cave-art dating in general: Pettitt and Alistair Pike, both of whom later worked on the Spanish uranium-series dates, wrote in 2007 that "Issues of sample contamination and of the heuristic relationship between materials dated and the production of the art are particularly problematic", so "one should at present be very cautious about straightforward interpretations of apparent "dates" of cave art."⁴⁴
5. Revisions: Panga ya Saidi's first quartz luminescence dates were found to be too young, not too old.¹⁹ Blombos's engraved ochre moved from 77,000 to about 73,000.

The largest disagreement in this article is the factor of 2.6 at Bizmoune.  Most are tens of percent.  A young-earth reading needs every one of these dates to be wrong in the same direction by a factor of tens to hundreds, as Part 5 shows.  The disputes are real, and none of them is that size.

Minimum ages cut one way.  Every Indonesian date, the Aviones date and the Spanish dates are minimums, from carbonate that grew after the paint or the deposit.  If they are wrong because the calcite lost uranium, they are too old;  if they are wrong because the carbonate formed long after the painting, the painting is older still.  The first is the open-system objection the Neanderthal article weighs, and it moves ages by tens of percent, not by factors of ten.

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

These were people

Young-earth writers read the whole symbolic record as the work of human beings, and they do so without strain, because their model says humans were fully human from the start.

1. Ross, Brummel and Wood, 2023. The fullest recent young-earth treatment of the human record, given at the movement's main technical conference, sets out "six distinct categories of evidence that attest to advanced, human cognition and therefore the image of God". Two of the six are "The crafting of unnecessary objects that attest to artistic creativity or preferences" and "Intentional burial of the dead".⁴ On art: "the elaborate cave art of such locations as Lascaux is associated exclusively with ancient Homo sapiens", and the authors "affirm that all H. sapiens are true humans."⁴ On Neanderthals: they "definitely exhibit the sorts of complex behaviors we would associate with human cognition and the image of God."⁴
2. Creation Ministries International. On the Taforalt beads, 2007: "Since the beginning, mankind has possessed a fully developed spiritual, technological, and cultural capacity."⁴⁵ On Neanderthal ornaments, 2006: "Neandertals and other varieties of humans show human cognition and symbolic thinking, which date back to before these varieties split at Babel."⁴⁶ On Chauvet, 2009, answering Gilles Tosello of the University of Toulouse: "Tosello is right about man's artistic expression having been 'there from the beginning'."⁴⁷
3. The Institute for Creation Research. On the Blombos ochre workshop, 2011: "modern human cognition must have been purposeful and present at the beginning."⁴⁸
4. Peter Line, 2023, on the Homo naledi claims: "if the evidence indicating H. naledi buried their dead, used fire, made symbolic engravings, and used stone tools, is confirmed, then H. naledi individuals were fully human, whether they had small or large brains."⁴⁹

This is a prediction, and it has held up better than Klein's.  A model in which humans were made fully capable has no reason to expect a long lag between modern skulls and modern behaviour, and the gradualist record shrank that lag.  Part 6 weighs how much this is worth.

These people lived after the Flood, and mostly after Babel

1. Andrew Snelling, Answers in Genesis, 2018, puts the Flood "about 4,350 years ago" and Babel "just over 4,000 years ago", after which "all human migrations and cultures have occurred".⁵ His chart assigns each dating method a window of real time. The row for "Early Human Carvings & Pottery", dated by "luminescence, electron spin resonance, and cosmogenic radioisotopes", reads "Centuries soon after the flood (~2250 BC–present)"; the row for radiocarbon reads "Later centuries after the flood (~2150 BC–present)".⁵
2. Michael Oard, the movement's main writer on the Ice Age, puts Babel "about 100 to 300 years after the Flood. By then the Ice Age was well underway."⁵⁰ His Ice Age is one event of "roughly 700 years": about 500 years to the glacial maximum and 200 to melt.⁵¹ In 2024 he mapped the conventional dates onto it: "The uniformitarian time period of 10–50 kyr ago could apply to the middle and end of the post-Flood Ice Age."⁵¹
3. Kurt Wise, 2005, reasoned from the spread of Homo erectus across three continents that "both Lower and Upper Pleistocene sediments are post-Babel in age."⁵² That places the whole Pleistocene record in this article, Pinnacle Point to Lascaux, after Babel.
4. Creation magazine says the same of particular finds. The Chauvet and Lascaux painters "were part of a group which arrived in Europe following the Babel dispersion. This occurred in the first few centuries after the Flood, some 4,500 years ago."⁵³ The Taforalt and Blombos beads "were made by humans descended from those who dispersed from the Tower of Babel after Noah's Flood, and are no older than about 4,500 years ago."⁴⁵ Engraved plaques from Jersey were made by "descendants of those travelling from Babel (c. 2000 BC)" who "settled in the English Channel, before the end of the Ice Age."⁵⁴
5. Ross, Brummel and Wood place the remains "post-Flood on the basis of geological indicators", with dispersal "most likely" after Babel.⁴

One older paper breaks the pattern.  Ken Guenter, in the Creation Research Society Quarterly in 1981, put Babel "at the close of the Neolithic" and suggested that "the C14 date of 50,000 B.P. might indicate a true age of c.4,000 B.C. or 8,000 B.C. for the Flood."⁵⁵  On his scheme the whole Palaeolithic falls between the Flood and Babel.  Even his earlier date leaves the Upper Palaeolithic inside 10,000 calendar years, and a Flood in 8000 BC is older than any biblical chronology allows:  Chris Hardy and Robert Carter put the earliest possible Flood at 3386 BC.⁵⁶

How they handle the dates

Most treatments set the dates aside by the biblical bound rather than by a measurement.

1. "The cave paintings can be no older than the global Flood, which occurred only around 4,500 years ago, and in fact are likely to be much younger than that."⁴⁷
2. "However, these dates far exceed the Bible's history, so must be disregarded."⁵⁴
3. "Ironically, the early critics were right about one thing—the unreliability of the dating assumptions."⁵³ The article does not say which assumptions.
4. Marvin Lubenow's survey of human fossils: "as a young-earth creationist, I do not accept those dates. In fact, I will demonstrate that they are nothing but science fiction."⁵⁷

Four make a point about method.

1. ICR on the Blombos workshop: "Curiously, however, they did not carbon-date any artifacts", and "The answer probably has to do with the fact that carbon dating is unreliable for artifacts older than 60,000 or so carbon-years."⁴⁸ That is the reason the excavators used luminescence instead, so it is not a fault in their work.
2. Creation magazine on the Spanish dates: "all U/Th ages can legitimately be questioned."⁵⁸
3. Line on luminescence and uranium-series dates in Laos and China: "I have no confidence in the ages obtained using any of these dating techniques as, apart from all the discrepancies, they also make invalid assumptions."⁴⁹
4. Snelling names two specific doubts about luminescence and spin resonance: scientists "cannot be sure if they really started at zero or if impurities affected how much energy or radioisotopes the samples stored."⁵ For radiocarbon he gives "a stronger magnetic field and rapid buildup of radiocarbon after Noah's flood".⁵ Wise puts "C-14 dates of Flood carbon" at 40,000 to 60,000 radiocarbon years.⁵²

Ross, Brummel and Wood state the position honestly:  "none of the dating methods have a widely accepted method of recalibrating to a creationist timescale."⁴

Where no young-earth answer exists

Searches of the full text of creation.com, the Creation Research Society Quarterly, and the site maps of the Institute for Creation Research and Answers in Genesis found no young-earth article that discusses the luminescence dates at Blombos, the uranium-series dates on the Sulawesi paintings, the chlorine-36 dates at Chauvet, or the Hulu Cave pairing of radiocarbon with uranium-series.⁵⁹  The only method-specific reply to the dating of any painting is the one sentence on uranium-series quoted above.  No young-earth writer has put a calendar date on the Sulawesi paintings.

For contrast: the old-earth reading of the same "big bang"

Reasons to Believe, an old-earth ministry, used the late revolution the other way.  Fazale Rana and Hugh Ross argued in 2005 that "When all the data is considered, human behavior consistent with God's image appears suddenly.  Its timing coincides with the appearance of humanity in the fossil record, just as the RTB model and the biblical creation account require."⁶⁰  Creation Ministries International pointed out in 2007 that the Taforalt and Blombos bead dates "clearly contradict the past teaching of RTB".⁴⁵  The gradualist record did more damage to that old-earth model than to the young-earth one, because it tied the image of God to a date that the finds then moved.

Part 5: What the compression costs

The window

Put the young-earth writers' own figures together.  Snelling puts the Flood about 4,350 years ago.  Oard puts Babel 100 to 300 years later, with the Ice Age "well underway", and gives the Ice Age about 700 years.  Wise puts all Pleistocene deposits after Babel, and Snelling and Creation magazine put the human cultures in them there too.  So everything in Part 2 that is older than the end of the last Ice Age, from Pinnacle Point to Lascaux, has to fit between Babel and the end of the Ice Age:  400 to 600 years.

Answers in Genesis has published a shorter window.  Andrew Snelling and Mike Matthews put the start of the Ice Age, "the Pleistocene", at the Babel judgment, "around a century or so after the Flood (perhaps 2250 BC)", and the extinction of the large Ice Age mammals "around 2100–2000 BC".⁶¹  On their figures the window is about 250 years, and every factor below roughly doubles.  This article uses Oard's longer window because it is the generous case.

The conventional ages of the finds in that window run from about 164,000 to about 21,000 years.  Blombos phase M3 to the Lascaux occupation alone is about 78,000 conventional years.  Fitting that into 600 real years means about 130 conventional years for every real year;  into 400 real years, about 195.  Oard's own mapping of 10,000 to 50,000 conventional years onto "the middle and end" of a 700-year Ice Age implies about 115 to 1.⁶²  No young-earth writer states these ratios.  They follow from the writers' own figures.

What started each clock

Every clock in this article was started by people, or by something that happened after them.

1. The Blombos, Diepkloof and Panga ya Saidi luminescence clocks started when sand was buried in the layers people lived on, when flint was burnt in their hearths, and when a pit was filled over a child.
2. The Sulawesi and Aviones uranium-series clocks started when calcite began to grow over paint or over the deposit.
3. The Chauvet radiocarbon clock started when the trees whose charcoal made the drawings died, and the chlorine-36 clock when the cliff fell and sealed a cave people had already used.

On the young-earth account all of this happened after Babel, so all of each clock's reading built up after Babel.  This is the argument The Oldest Homo sapiens makes for the fossils, and it applies here with more force, because many of these clocks were set by the very acts this article is about.

The young-earth mechanism for shrinking radiometric ages is a past burst of faster radioactive decay.  The scientists who proposed it say it happened in the Flood and "would have decelerated at the end of the Flood", so that decay rates "have been essentially constant during most of the post-Flood period".⁶³  They add that "there is intriguing evidence suggesting this may not be strictly true".  Snelling widens it to "during and soon after the flood",⁵ which could reach into the first centuries after Babel, and his chart puts carvings in those centuries.  Neither says how fast decay ran after Babel.  On the RATE scientists' own placement, the clocks in this article ran at today's rates;  on Snelling's, they need the speed-up priced below.

The price for the decay clocks

If the speed-up ran after Babel anyway, it can be priced.  The table uses the lower end of each published age, the earliest Babel on Oard's range, and so gives the smallest speed-up the data allow.  No young-earth writer has proposed this.⁶⁴

Three conditions come with any such speed-up.

1. The people lived through it. The dose the Blombos flints and sands recorded came from uranium, thorium and potassium in the cave floor the people sat on, and potassium-40 is in every human body. A speed-up of 100 means a hundred times today's natural radiation from those sources, for centuries.
2. Different clocks had to speed up together. At Blombos, luminescence on sand, thermoluminescence on burnt flint and spin resonance on teeth agree.⁴⁰ At Skhul, uranium-series and spin resonance agree with thermoluminescence.²¹ Luminescence and spin resonance count dose from three elements; uranium-series counts one decay chain. A speed-up that hit them by different amounts would have pulled the dates apart, and the RATE scientists reported that the speed-up "seemed to be proportional to the atomic weights of the parent radioisotopes",⁶³ which would leave light potassium-40, a large part of the luminescence dose, behind uranium and thorium. The site's uranium-series article shows from measured uranium ratios how tightly such uneven speed-ups are limited.
3. The order has to survive. Across three continents and five methods, the finds fall in one sequence, and where two methods meet they agree on it. A young-earth map from conventional to real ages must keep that order for every method at once.

The price for radiocarbon

Radiocarbon is different, and here the young-earth side has a real mechanism.  Its answer is not faster decay but less carbon-14 in the air after the Flood, building up over the following centuries.⁵  That can be priced too.  If the Chauvet charcoal, about 30,000 radiocarbon years old,⁶⁵ grew 3,900 years ago, the air then held about 4 percent of today's carbon-14.  If the Lascaux bones, at 17,600 radiocarbon years,³³ date to about 3,750 years ago, the air held about 18 percent.  Snelling has the dates reliable again by 1000 BC, so the air had to reach about 100 percent within the next 800 years.⁶⁶

1. With today's carbon reservoir in the air, oceans and living things, and today's production of carbon-14 in the upper atmosphere, starting from 4 percent the air would reach only about 15 percent in a thousand years.
2. To climb from the Chauvet level to the Lascaux level in 100 to 350 years takes 3 to 12 times today's production; to climb from the Lascaux level to today's in 800 years takes about 9 times.
3. Snelling's own reason pulls the other way. A stronger magnetic field shields the atmosphere from cosmic rays and lowers production. Hulu Cave shows the effect in reverse: during the weak field of the Laschamp excursion, carbon-14 in the air rose to "1.70 times higher" than modern.⁴¹

A smaller carbon reservoir after the Flood would ease these numbers, and no young-earth model this article found says how large the post-Flood reservoir was.  The site's Radiocarbon Dating article sets out the tree-ring record that a buildup of this kind must also explain.

The two clocks tied together

The strongest problem is where the two kinds of clock meet.  At Hulu Cave, radiocarbon and uranium-series were measured on the same layers of calcite, back to about 54,000 uranium-series years.  Through the last glacial period the radiocarbon ages run younger than the uranium-series ages, and the largest gap Cheng's team reports is 5,200 years.⁴¹⁴¹  On the young-earth account, the radiocarbon offset comes from the air and the uranium-series offset from faster decay.  Those are unrelated causes, so the paired record would have to be a coincidence held year by year for the whole span.

Wise's mapping sharpens it.  He puts "Flood carbon" at 40,000 to 60,000 radiocarbon years.⁵²  At Hulu, calcite with that much radiocarbon lies in the upper part of the record, at uranium-series ages from the 40,000s up to its end at about 54,000.  The calcite over the Karampuang scene reads 53,500 by the same method, and over the Muna stencil 71,600.²⁸²⁹  Read through the Hulu pairing and Wise's mapping, the Karampuang crust began to grow while Hulu calcite was taking in Wise's "Flood carbon", and the Muna crust before that:  at the time of the Flood or before it, over paintings that the same writers place after Babel.  The only escape is that the uranium-series clock ran at different rates in China and Indonesia.  No one has proposed that, and it would undo the agreement between methods at Blombos and Skhul.

What this does and does not show

None of this proves the conventional ages.  It shows what a young-earth chronology has to supply, and that nobody has supplied it:  a speed-up of decay after Babel, tens to hundreds of times today's rate, the same for every element, in the lifetimes of people who were painting and burying their dead;  a rise of atmospheric carbon-14 several times faster than today's production allows, against the direction of the field the model invokes;  and a reason two unrelated distortions line up layer by layer at Hulu.

Part 6: The common ground, and what each side gains and loses

The shared claim

Genesis 1:27 says God created man in his own image.  Young-earth writers take symbol-making as one visible mark of that image.  Ross, Brummel and Wood list artistic creativity and deliberate burial among their "six distinct categories of evidence that attest to advanced, human cognition and therefore the image of God".⁴  Creation Ministries International reads Neanderthal ornaments as "consistent with all humans, including Neandertals, being descended from Adam and Eve, whom God created in His image".⁴⁶  The archaeologists in Part 1 use different words for a similar line:  symbols, ornament and burial are the remains they treat as marks of the minds of living people.  Ross, Brummel and Wood add the caution that "some of these behaviors are observed in non-human animals, albeit in more rudimentary form".⁴

On what the record shows, the two sides now largely agree.

1. Symbolic objects and deliberate burials appear with anatomically modern humans in Africa and the Levant by more than 100,000 years on the conventional scale, at Skhul, Qafzeh, Blombos, Taforalt and Bizmoune.
2. Neanderthals made ornaments, used pigment, engraved cave walls and buried at least some of their dead.
3. One engraved shell may come from Homo erectus, and its meaning is unknown.
4. Nothing comparable is known from any australopith. The Homo naledi claims are the test case and are not established.

What the young-earth side gains

1. Its prediction held. A model in which people were made fully capable expects no long gap between human anatomy and human behaviour. Klein's model needed that gap and a late genetic switch to close it. Scerri and Will now describe "similar cognitive capacities that are variably manifested".³ That is close to what the young-earth writers have said all along.
2. Neanderthals behaved as people. The young-earth claim that Neanderthals were human is supported by their burials, ornaments and wall engravings, as Neanderthals concludes.
3. The explanation the field now prefers is demographic. Scerri and Will credit the patchy record to "demographic processes such as population structure, size, and connectivity".³ Small scattered groups that lose and regain skills fit a dispersal from Babel as easily as a long prehistory. This is a fit, not a test: the same demography fits the conventional timescale.

What the young-earth side loses

1. The marker carries its dates with it. If symbol-making marks the image of God, the record says image-bearers were painting, engraving, stringing beads and burying children 50,000 to 160,000 years ago by every method that reaches that far. The young-earth reading takes the marker and must reject the dates attached to it.
2. The order is a sequence, not a scatter. The finds come in one dated order across five methods and three continents, and Part 5 prices what a young-earth map of that order requires.

What the evolutionary side gains

1. A mosaic fits evolution better than a switch. Piecemeal appearance across Africa, over 100,000 years or more, is what descent with modification would lead one to expect.
2. The capacity looks old. If Neanderthals and modern humans both made symbols, the capacity was probably present in their common ancestor, which d'Errico and Stringer's third scenario allows.⁶

What the evolutionary side loses

1. Its clean explanation. Klein's genetic switch was the one proposal that tied behaviour to biology, and he admitted it "cannot be tested or falsified".¹ The field has moved to an account in which the capacity was present and its expression came and went.
2. No step-by-step evolution of the capacity is visible. The earliest symbolic traces come from people whose anatomy is already that of modern humans or Neanderthals. The record shows the capacity used, not the capacity growing.
3. The long silences remain. Scerri and Will's "long periods of stasis" and Renfrew's sapient paradox are puzzles on a timescale of hundreds of thousands of years. They are not puzzles on a short one.

A lesson from the old-earth case

Reasons to Believe tied the image of God to a date, the sudden "big bang" of culture, and the dates moved.⁶⁰⁴⁵  The young-earth writers tie it to the people instead.  That is the safer theology.  It does not answer the chronology.

Objections and responses

1. "The dates rest on unprovable assumptions." Every method has assumptions, and Snelling names the real ones for luminescence: whether the clock started at zero, and whether impurities changed the dose.⁵ Burnt flint is zeroed by fire, and the assumptions of different methods are different, so where luminescence, thermoluminescence, spin resonance, uranium-series and radiocarbon meet, as at Blombos, Skhul and Chauvet, their agreement tests them. The disagreements that do occur are tens of percent, and at most a factor of 2.6.
2. "Even the experts disagree about Chauvet, the Spanish caves and Bizmoune." They do, by factors of 1.5 to 2.6. The young-earth chronology needs factors of about 80 to 250 for the same finds, in the generous case of Part 5. Quoting the disputes as support for that would misuse them.
3. "Minimum ages are not real ages." True, and every one of them makes the art older, not younger. The young-earth reading needs the crust to be too old by tens of thousands of years. Calcite that loses uranium does read too old, which is why the Neanderthal dates are disputed; the measured effect is tens of percent, as Neanderthals sets out.
4. "Accelerated decay during the Flood accounts for the old ages." It cannot reach these clocks. Each was started by people, or after them, and on the young-earth account people lived after Babel. The RATE scientists place the speed-up in the Flood and say it "decelerated at the end of the Flood".⁶³
5. "Radiocarbon dates before 1000 BC are inflated by the post-Flood build-up of carbon-14." This is a real young-earth mechanism, and the article prices it: 3 to 12 times today's production of carbon-14 for centuries, against the direction of the stronger field Snelling invokes. It does not touch the luminescence and uranium-series dates, which carry most of the record before 40,000 years.
6. "Scratched ochre and holed shells need not be symbols." Partly right, and the excavators say so: Henshilwood's team wrote that their criteria "do not guarantee the identification of intentionally made engravings", and Texier's that representational intent "is a matter of debate".¹⁵¹⁸ The strongest cases are the bead perforations, which occur on 88 percent of the Blombos shells and 8.6 percent of naturally holed ones,¹⁶ the repeated motifs on 25 Diepkloof containers, and the dug grave at Panga ya Saidi. Whatever their meaning, they were made by people, and they are dated.
7. From the skeptic's side: "Young-earth writers call anything human, so their claim cannot be tested." Ross, Brummel and Wood give criteria in advance, and admit that some of the behaviours appear in rudimentary form in animals.⁴ Their model would be in trouble if the same behaviours turned up securely with australopiths, which is why the Homo naledi claims matter. And their prediction of no lag between human bodies and human behaviour was at risk under Klein's model and survived.
8. "Neanderthal art proves Neanderthals were human, so the young-earth side wins." It supports their humanity. It does not shorten their dates, which come from the same uranium-series and luminescence methods as everything else here.
9. "Behavioural modernity is an abandoned idea, so there is nothing to explain." The term is fading; the finds and their dates are not. The question this article tests is the dates.
10. "These people could have lived before the Flood." Ross, Brummel and Wood answer that hearths, tool-making sites and stone arrangements are "ephemeral remains" that "would not survive the cataclysm of a global Flood".⁴ The burials at Skhul, Qafzeh and Panga ya Saidi and the painted caves are of that kind.

Issues and unanswered questions

1. Sources not read in full. The main text of the 2018 Blombos drawing paper, Zenobia Jacobs's 2013 Blombos chronology, Chantal Tribolo's 2013 Diepkloof chronology, the 1988 Qafzeh thermoluminescence paper, Pettitt and Pike 2007 and Pettitt 2008 were not obtained. The article cites their abstracts or later summaries, and Tribolo 2013, which revisits the Diepkloof dates, is not used.
2. A circulating quotation not used. A passage attributed to Pettitt and Pike 2007, on black dots at Candamo Cave dated 32,000 to 35,000 radiocarbon years by one laboratory and 15,000 to 16,000 by another, could not be checked and is left out. If it is accurate, it is the clearest case of a factor-of-two disagreement on one cave painting.
3. Inconsistent numbers inside sources. Oktaviana's 2024 paper gives the Sipong 4 hunting scene a minimum of 48,000 years in one place and implies 49,000 in another. Brumm's 45,500 for the Tedongnge pig is the central value, not the two-sigma minimum the later papers use, which would be about 45,200. The Qafzeh papers disagree on the age of Qafzeh 11 and on how the shells were holed. None of these moves a date by more than a few percent.
4. The size of the post-Flood carbon reservoir. The radiocarbon arithmetic assumes today's. A young-earth model that specified a smaller reservoir would need less production. None was found.
5. The dose to the people. A post-Babel speed-up of 80 to 250 times would have given the people at these sites that many times today's natural dose from the ground and from their own potassium. This article does not compute what that would have done.
6. The young-earth literature not reached. Peter Line's "Developments in paleoanthropology no. 6" in the November 2025 Journal of Creation appears in searches for Sulawesi and behavioural modernity and is not yet open to non-subscribers. It may contain the first young-earth discussion of the Sulawesi dates.
7. The Trinil link. The engraved shell's age rests on sediment inside the shells from a nineteenth-century collection. A direct date on the shell itself would settle whether Homo erectus engraved.

Verdict

On what these people were, the evidence supports the young-earth position.  Engraving, beads, pigment and burial appear with the earliest anatomically modern humans in Africa and the Levant and with Neanderthals, not as a late switch 50,000 years ago.  The late "human revolution" that once seemed to show a human mind arriving long after the human body has been given up by most of the field, and young-earth writers who said humans were capable from the start were closer to the record than Richard Klein was.

On when they lived, the evidence runs against it.  The finds form one dated sequence, Pinnacle Point and Blombos through Sulawesi and Chauvet to Lascaux, measured by five or six methods that rest on different physics and agree where they meet.  Their disputes are tens of percent.  The young-earth writers place all of it after Babel and before the end of a 700-year Ice Age, which requires compressing about 78,000 conventional years into 400 to 600 real ones.  The decay clocks would need a speed-up of roughly 80 to 250 times if it was confined to the Ice Age, or 12 to 36 times on average over the whole time since Babel, later than the scientists who proposed accelerated decay place its end, in the lifetimes of the people who made the art.  The radiocarbon clock would need 3 to 12 times today's carbon-14 production for centuries, and the Hulu Cave pairing of the two clocks would have to be a coincidence.  No young-earth writer has proposed any of this, and most of the published replies set the dates aside by the biblical bound rather than by a measurement.  On the present evidence, the dated record of symbolic behaviour is one of the clearer problems for a young-earth chronology of human history.

Sources