Cradle of Humankind, Gauteng, South Africa

Rising Star Cave

About 335,000 to 236,000 years (a bracket on the Dinaledi fossils, not one clock on every bone)

Four views of a fossil skull with a small vault, large eye openings and worn teeth, on a black background. — The Lesedi Chamber cranium, four views, with a scale bar. The vault is small for a human of this date. The published age bracket of about 335,000 to 236,000 years belongs to the Dinaledi fossils, a second chamber in the same cave, not to a number printed on this skull. John Hawks, Marina Elliott, Peter Schmid and colleagues, CC BY 4.0. Shown unedited.

A human with a brain the size of a much older ancestor lay in a chamber that living cavers enter through a squeeze about 20 centimetres wide. The fossils date between about 335,000 and 236,000 years ago. Whether anyone carried the dead in, and whether anyone cut the marks on the wall, is the fight.

Why this is extraordinary

Rising Star is a dolomite cave in the Cradle of Humankind, west of Johannesburg. In 2013 a deep chamber, Dinaledi, produced more than a thousand bones of a hominin that did not match the neighbours. The body is a mixture. Hands and feet are close to later humans, fingers keep some curve, and the brain is not the brain of a human of the last few hundred thousand years. The Dinaledi crania are published at about 465 to 560 cubic centimetres, in the range of the australopithecines and far below the roughly 1,400 of people now. The first description counted at least 15 individuals, adults and children, in that one chamber. A second chamber, Lesedi, later produced more of the same species, including the skull in the photograph. The catch is that the Lesedi bones have not been given the same published age.

The way in is part of the argument. The 2015 geological report puts the Dinaledi Chamber about 30 metres below the surface and about 80 metres, in a straight line, from the nearest entrance. The last passage is a fissure: about 12 metres down, with squeezes as tight as about 20 centimetres. Teams mapped the routes and did not find another opening. They also reported no sign that a large entrance once stood over the chamber and then sealed. That negative is important, and it is not the same as a proof that no entrance ever closed. Rock overhead can hide a passage that no longer exists.

The date came two years after the name. Uranium-series ages on flowstone, electron spin resonance on teeth, and luminescence on the sediments were combined into a bracket: the Dinaledi fossils were deposited between about 335,000 and 236,000 years ago. Uranium-series is a clock in the uranium that decays inside a tooth or a mineral. Electron spin resonance is a second clock in the crystal, counting damage from natural radiation. Luminescence on the sediments is a light clock in the buried grains, reset by sunlight. The ends of that bracket are different samples, and they are not one birthday for every bone. A 2024 review, a decade after the discovery, finds little quarrel with the decision to name one species, and a real quarrel with the age, the burial story and the behaviour.

In 2023 the excavation team claimed deliberate burials in the Dinaledi Chamber and the Hill Antechamber, and cross-hatched marks on a dolomite face nearby, and tied both to this small-brained species. Reviewers answered that the graves had not been shown in the usual way: the features were not fully dug, the bone positions were hard to read, and the marks had no date of their own. A later revision of the burial paper added a closer study of how the bones lay. At least one reviewer then accepted repeated, patterned deposition, and warned against reading a symbol into it. The engraving paper's reviewers, on a revision they described as slight, still say the marks are not dated and not shown to be the work of this species. A mark without an age can be much later. The chamber is difficult for everyone, including the excavators, and that difficulty cuts both ways: it makes an accidental scatter harder to picture, and it makes a perfect excavation record harder to produce.

The same set of claims includes fire, so that people could move bodies through a dark cave. A 2023 review describes that fire claim and argues the other way. It says no hearth analysis of the kind used on burnt deposits has been presented, naming a microscope study of the ash, a chemical test of burned sediment, and a magnetic test of heated soil. It says a radiocarbon date on one apparent hearth came back very young and is not yet in a published table. It also notes that charcoal in a cave can be a wildfire. The fire claim and that review sit beside the graves. Neither is asked to close the other.

The name is a cave system in the Cradle of Humankind, not a claim about the sun. No burial in the deep chambers is published with a direction toward a solstice or a star, and the cross-hatched marks reported nearby have no sky reading in the sources used here. A chamber reached through a long fissure is a hard place to sight a horizon. The argument is the small brain, the hands, and whether the graves were dug on purpose. The dates are on the bones and the sediments. The sky is not what they are made of.

A simple side view of a narrow descent into a chamber below the ground surface. — The published squeeze, drawn as a sketch. About 12 metres down, about 20 centimetres at the tightest, the chamber about 30 metres below ground. Not a survey. Diagram: Our Deep History.

What this changed

The shift, in short. The weighing further down is the longer argument.

Older textbook view: A brain under about 600 cubic centimetres meant an early grade of human, and an early date. By 300,000 years ago, southern Africa was supposed to be the world of large-brained people. Mortuary care and cave engravings were kept for our species, or for late Neanderthals.
New evidence: A new species, with human-like hands and feet and a small vault, comes from a chamber reached today only by a very tight route, and at least 15 individuals are in the first Dinaledi sample. A second chamber holds more. The 2023 papers add claimed graves and cross-hatched marks, and the reviews do not all accept those additions.
How it is dated: A bracket, not a single method. Uranium-series on flowstone, electron spin resonance on teeth, and luminescence on sediments place the Dinaledi fossils between about 335,000 and 236,000 years. Different samples hold the two ends. The Lesedi skull is the same species by its bones, and it does not yet have that published bracket. The marks on the rock have no radiometric age at all.
What moved: A small-brained Homo survived in South Africa into the time of early Homo sapiens. The simple ladder, big brain then complex life, lost a rung. Burial and engraving would take another rung, and they are not yet secure enough to move it.
What is still disputed: How the bodies arrived: carried in, or through an entrance that later closed, or by a process the dig has not separated. Whether the 2023 burial claim survives outside the reviewers who now accept it. Whether the cross-hatching is human, and of this age. Whether every bone sits inside the 335,000-to-236,000 bracket. The species name itself is the part with the least pushback.

Why it matters

If the bodies were put there on purpose, complex treatment of the dead does not require a modern brain. If they were not, a small-brained species still shared southern Africa with the early members of our own line, and the chamber still needs an explanation.

The skull is from the Lesedi Chamber, and the bone table, the holotype pieces, the hand and the jaw are the Dinaledi fossils. The plan shows both chambers. The working station is inside the cave and is not the fossil room. The diagram is the squeeze, from the 2015 description.

Photo tour

8 photographs, unedited. On a phone, swipe the row. Choose one to look closely.

Many fossil bones laid on a black table, with a partial skeleton arranged in the centre and extra bones sorted around it. — Dinaledi bones laid out for the first description. The central arrangement is a composite. The pieces around it are more individuals, not one person's spare parts.Lee Roger Berger research team, CC BY 4.0

A skullcap, jaws and teeth in several labeled views, with a black-and-white scale bar. — The holotype pieces from Dinaledi, a partial vault, jaws and teeth, with a scale bar. These are parts. They are not one face glued together.Lee Roger Berger research team, CC BY-SA 4.0

Two views of a fossil hand beside a ruler, and an inset of finger bones in cave soil. — An adult right hand, palm and back, with a ruler, and an inset of the bones in the ground. The hand is more human-like than the brain size. The inset is the find, not a reconstruction.Tracy L. Kivell, Andrew S. Deane, Matthew W. Tocheri, Caley M. Orr, Peter Schmid, John Hawks, Lee R. Berger and Steven E. Churchill, CC BY 4.0

A fossil jaw with five worn teeth, cracked and still coated in brown sediment. — A jaw soon after recovery, with cave sediment on the bone and cracks across it. The preservation is why so many pieces could be studied. It is not, by itself, a grave.Patrick Randolph-Quinney, CC BY 2.5 ZA

A line plan of a cave system with entrances marked and two chambers labeled UW101 and UW102. — Plan of the cave. Dinaledi is marked UW101, Lesedi UW102, and three entrances are marked. The scale bar is 25 metres. This is a horizontal plan. It does not show the vertical squeeze.John Hawks and colleagues, CC BY-SA 4.0

People at a table of computers inside a cave, under hanging flowstone, with cables across the ceiling. — A working station inside the cave during the expedition, monitors under flowstone. This is not the fossil chamber. The bones are farther in, beyond passages this frame does not show.Simon Fraser University Communications, CC BY 2.0

Four replica skulls in a row, labeled Homo habilis, Homo erectus, Homo floresiensis and Homo naledi, with notes on brain size. — Comparison: replica crania, labeled on the figure as Homo habilis, an early Homo erectus, Homo floresiensis and Homo naledi. The printed notes mark naledi as a small brain volume. These are replicas, not the Dinaledi originals.Chris Stringer, Natural History Museum, CC BY 4.0

What is still unexplained

These sit beside the consensus account, with the same visual weight. None of them is marked solved.

Observed A small brain, late Dinaledi cranial capacities are published at about 465 to 560 cubic centimetres. The date bracket is the later Middle Pleistocene, and a brain of that size was not supposed to be current then.

Open A door that is hard to use About 12 metres of fissure, squeezes near 20 centimetres, about 30 metres down. That is the route now, and a closed older entrance is not ruled out by the failure to find one.

Open Burial and marks, under review Deliberate burial was sharply criticized, then revised, and has persuaded some reviewers as patterned deposition without a proven symbol. The cross-hatched marks still have no date, so they are not yet this species' work.

Open Fire in the passages The burial and engraving papers, as a 2023 review describes them, include a fire claim: a light was needed to carry bodies through the dark cave, and the light was fire. That review argues that no hearth study of the kind used on burnt deposits has been set out, that a radiocarbon test on one apparent hearth came back very young and is not yet in a full table, and that undated charcoal in a cave can be a wildfire. The claim and the review are both on the page. A visitor weighing what a small brain is being asked to do can set the fire beside the graves, and neither reading closes the other.

Not in these sources A deep chamber, not a horizon Cross-hatching on the dolomite can look like grids from other African caves, and those grids are not published as star maps in their own rooms. Here the marks have no date of their own, which is already part of the argument over the graves. A mark without a date cannot be pinned to a named year in the sky. The cave's name should not be asked to do the work of an alignment.

Location and dating

Location: Rising Star cave system, Cradle of Humankind, Gauteng, South Africa, about 2 kilometres west of Sterkfontein. The coordinate is the published entrance area, 26°01′13″ S, 27°42′43″ E. The Dinaledi Chamber is deep inside.
Period label: Current consensus — open to revision Dinaledi fossils bracketed about 335,000 to 236,000 years ago. Lesedi undated by that same publication. Marks undated. Current consensus on a small-brained species in a difficult chamber — open on burial, engraving, and the exact route the dead used.
How this dating was done: Uranium-series on flowstone, electron spin resonance on teeth, and luminescence on sediments, combined as a bracket for the Dinaledi deposit. The stratigraphy of the chamber, the order of layers, which says what came before what, is the claim that the fossils sit inside that package rather than having intruded later. The engravings were not dated by any of these methods.
What could change it: A direct age on the flowstone that truly seals each skeleton, and a date on the rock crust over the marks, would test the bracket and the pictures separately. A mapped passage, now blocked, that once opened into the chamber would reopen the route.

Published ranges, drawn as bars. A bar is a claim about a span, not a point.

Dinaledi bracketContested

335000 BCE236000 BCE

Dinaledi bracket. About 335,000 to 236,000 years, from flowstone, teeth and sediments. The ends are different samples. A 2024 review still treats the age as argued.

What is actually there

Dinaledi crania about 465 to 560 cubic centimetres.
At least 15 individuals in the first Dinaledi sample, and more of the species in the Lesedi Chamber.
Hands and feet closer to later humans than the brain size suggests. Some finger curvature remains.
Access about 30 metres down, with a 12-metre fissure and squeezes near 20 centimetres.
Age bracket about 335,000 to 236,000 years for Dinaledi only.
Burial claims revised after severe review. Engraving claims still undated.
A fire claim sits with the graves. A 2023 review argues the hearth tests are missing, and that one radiocarbon result was very young and is not yet in a full table.
No grave here is published toward a solstice.

Open questions

Question	Status
Were the bodies carried in?	Open. The living route is hard. A vanished entrance is not disproved.
Is there a grave?	Open. Some reviewers of the revision see patterned deposition. A 2024 review still calls burial controversial. The marks have no date of their own. A fire claim and a 2023 review of that claim sit beside the graves.
Does a small brain forbid this behaviour?	Open as a principle. It is not a measurement. The behaviour has to be shown first.
Do the cross-hatched marks record a sky?	Not in the sources used here. They have no date of their own, and no star name.

The current narrative

Where publication stands today. It is the best-supported reading, and it is open to revision. It is not a verdict.

The current narrative — open to revision

A small-brained species, late, in a hard chamber

Homo naledi is a real species with a small brain and a more human body, present in the Dinaledi Chamber between about 335,000 and 236,000 years ago. The bodies are in a place that is difficult to reach today. Deliberate burial and wall engraving are published claims, not yet the settled reading. Patterned deposition has convinced some reviewers after revision. It has not convinced the field.

What supports this

The skeleton repeats across many individuals, which is why the species name drew little pushback.
The age bracket comes from more than one method, and it overlaps early Homo sapiens elsewhere in Africa.
The modern route really is a tight fissure, and no second entrance was found in the survey.
A revised burial paper persuaded at least one reviewer that the bone positions are repeated and patterned, without requiring a symbol.
The excavation papers, as a 2023 review describes them, include fire as the light for carrying bodies through the dark cave.

What pushes back

The age ends are different samples. A 2024 review still lists the age among the controversial claims.
Reviewers of the first burial papers did not see a demonstrated grave. The engraving paper still lacks a date.
Failing to find an entrance is not proof that none ever existed.
A 2023 review argues that the fire claim has no hearth analysis of the kind used on burnt deposits, that one radiocarbon result was very young and is not yet in a full table, and that cave charcoal can be a wildfire.

What would settle it

The species, the small brain, the age bracket, and the difficult chamber are the published description. Burial, marks, and fire each have a claim and a review, and this room does not choose among those pairs.

Emerging narratives

Other published readings. Each may be right. None is adopted here unless the evidence carries it.

An emerging narrative

Mortuary care without a modern brain

The dead were brought in, more than once, and the cross-hatching is part of the same habit. A brain of about half a litre was enough for a practice textbooks reserved for large-brained humans.

What supports this

Many individuals, across ages, in a chamber with no easy modern door, is a poor fit for one accident.
The revision's supporters see patterned deposition, the minimum a burial claim needs.
If the marks are the same age as the bones, they sit beside the bodies rather than in an empty cave.
The same papers, as the 2023 review describes them, argue that fire lit the way through the dark passages.

What pushes back

The marks are undated. Reviewers have said they could be recent.
Patterned bones are not yet a dug grave in the view of the wider 2024 assessment.
A closed shaft, or water, can gather bodies without a funeral. That reading is not proven either. It is not ruled out.
The 2023 review’s hearth objections, the very young radiocarbon result it says is unpublished, and wildfire charcoal are the case beside the fire claim.

What would settle it

Burial, marks, and fire are published claims with published reviews beside them. This room does not close any of the three.

Should the current view be reevaluated?

Open to revision

The ladder that required a large brain before a late date does need reopening, because the fossils and the bracket are strong enough. The further claim, that this species buried its dead and engraved the wall, is not strong enough to move the consensus. Some reviewers have come toward patterned deposition. The engraving has not.

What drives this

Brain size about 465 to 560 cubic centimetres at a late Middle Pleistocene date
A surveyed route with squeezes near 20 centimetres
Severe review of the 2023 burial and engraving papers, and a partial softening on burial only
Marks that still have no age
A fire claim, and a 2023 review that argues the hearth tests and a full radiocarbon table are not yet in print

This is the museum’s own weighing of the evidence. It is not a visitor vote, and it is separate from the slider.

Weigh it yourself

Anonymous. Nothing here is stored except on this browser. Your lean does not change the assessment beside it.

A small-brained species, late, in a hard chamber Unsure Mortuary care without a modern brain

If the contested reading is right

What would follow if a reading that is not the consensus turned out to be true, and what could settle it. This explains the stakes. It does not adopt the reading.

The bodies were buried, and the marks are theirs

Treatment of the dead, and perhaps a graphic mark, would no longer be evidence of a large brain. Early Homo sapiens in Africa would have shared the land with another human who also did these things.

Why it is still worth a test

A date on the crust over the marks, and a fully recorded section through one claimed grave that a second team can read.

The bodies arrived without a funeral

The species and the date would stand. The behaviour claimed for the small brain would not. Another route, or another process, would still have to gather so many individuals in that chamber.

Why it is still worth a test

A blocked passage into the chamber, or a sediment study that shows the bodies came in with debris rather than as placed corpses.

Sources

Homo naledi, a new species of the genus Homo from the Dinaledi Chamber, South Africa eLife, 2015
Geological and taphonomic context for the new hominin species Homo naledi eLife, 2015
The age of Homo naledi and associated sediments in the Rising Star Cave eLife, 2017
Evidence for deliberate burial of the dead by Homo naledi, with the reviews eLife
Rock engravings and their relation to Homo naledi, with the reviews eLife
What we know and do not know after the first decade of Homo naledi Nature Ecology and Evolution, 2024
No scientific evidence that Homo naledi buried their dead and produced rock art Journal of Human Evolution, 2023. Argues that the fire claim lacks hearth tests of the kind used on burnt deposits, and that one radiocarbon result was very young and unpublished.
A summary of that 2023 review The review’s public summary. It describes the fire claim as a light for carrying bodies, and the charcoal as possibly a wildfire.

Image credits

The Lesedi Chamber cranium, four views, with a scale bar. The vault is small for a human of this date. The published age bracket of about 335,000 to 236,000 years belongs to the Dinaledi fossils, a second chamber in the same cave, not to a number printed on this skull.. John Hawks, Marina Elliott, Peter Schmid and colleagues, CC BY 4.0. Shown unedited.
Dinaledi bones laid out for the first description. The central arrangement is a composite. The pieces around it are more individuals, not one person's spare parts.. Lee Roger Berger research team, CC BY 4.0. Shown unedited.
The holotype pieces from Dinaledi, a partial vault, jaws and teeth, with a scale bar. These are parts. They are not one face glued together.. Lee Roger Berger research team, CC BY-SA 4.0. Shown unedited.
An adult right hand, palm and back, with a ruler, and an inset of the bones in the ground. The hand is more human-like than the brain size. The inset is the find, not a reconstruction.. Tracy L. Kivell, Andrew S. Deane, Matthew W. Tocheri, Caley M. Orr, Peter Schmid, John Hawks, Lee R. Berger and Steven E. Churchill, CC BY 4.0. Shown unedited.
A jaw soon after recovery, with cave sediment on the bone and cracks across it. The preservation is why so many pieces could be studied. It is not, by itself, a grave.. Patrick Randolph-Quinney, CC BY 2.5 ZA. Shown unedited.
Plan of the cave. Dinaledi is marked UW101, Lesedi UW102, and three entrances are marked. The scale bar is 25 metres. This is a horizontal plan. It does not show the vertical squeeze.. John Hawks and colleagues, CC BY-SA 4.0. Shown unedited.
A working station inside the cave during the expedition, monitors under flowstone. This is not the fossil chamber. The bones are farther in, beyond passages this frame does not show.. Simon Fraser University Communications, CC BY 2.0. Shown unedited.
Comparison: replica crania, labeled on the figure as Homo habilis, an early Homo erectus, Homo floresiensis and Homo naledi. The printed notes mark naledi as a small brain volume. These are replicas, not the Dinaledi originals.. Chris Stringer, Natural History Museum, CC BY 4.0. Shown unedited.
The published squeeze, drawn as a sketch. About 12 metres down, about 20 centimetres at the tightest, the chamber about 30 metres below ground. Not a survey.. Diagram: Our Deep History.

← South Africa World map