#History#Secrets Deep in the frozen wilderness of Interior Alaska, archaeologists uncovered a site that shattered our understanding of human history. Dating back 11,500 years, the discoveries at the Upward Sun River site—including the ancient remains of infants and tools unlike anything seen before—forced scientists to completely rewrite the timeline of how humans first populated the Americas.
Was the Bering Land Bridge theory wrong all along? Who were the "Ancient Beringians," and how did a forgotten population survive the brutal end of the Ice Age?
In this video, we dive deep into the genomic evidence, ancient burial rituals, and groundbreaking archaeological finds that challenge everything textbook history taught us about the First Americans.
The discovery of ancient child burials at Alaska's Upward Sun River site has revolutionized the understanding of the first Americans. DNA analysis of the \"Sunrise Girl\" revealed the Ancient Beringians, a distinct genetic lineage that split from other Native American ancestors thousands of years earlier than previously thought, effectively debunking the \"Clovis First\" migration theory.
Above a braided river in the interior of Alaska,
a low bluff of windblown silt looks out over a
valley where winter falls far below freezing
and the summer sun barely sets. If you stood here
eleven and a half thousand years ago, you would not hear silence.
You would smell wood smoke
and salmon, because people returned to this one bluff
for thirteen thousand years. Six camps
Now there is only wind, drifting silt across an open trench.
In that trench, a hearth.
In the hearth, the burned bones of a three-year-old child.
Directly beneath it, a pit stained red with ochre.
Inside the pit, two infants, one of them a fetus.
And then came the detail that made no sense.
Laid beside them were hunting weapons.
Antler rods a foot and a half long,
every piece coated in the same red pigment.
The excavators called the hearth a cremation.
A federal archaeologist told the Associated Press
the child might have been a meal, not a funeral.
Ritual. Cannibalism. Girls who hunted. A people who vanished.
The theories rushed into the silence.
Then, in 2018, a team at the University of Copenhagen took
a sliver of bone from behind the ear of the six-week-old girl
and sequenced her entire genome.
Once, this story belonged to bone and burned soil.
Now, it would belong to DNA.
What came back did not fit any map of the first Americans.
So who were these children
and what happened to the people they came from?
For decades, the story of the first Americans fit on a postcard.
Big game hunters, stone spear points,
a walk out of Asia, across the land bridge,
and down a narrow corridor between two giant ice sheets.
That was the accepted version. Clovis First, roughly 13,000 years ago.
One wave of people, one route, one clean beginning.
It sat in textbooks. It hung on museum labels.
It sat so comfortably in the public mind that
questioning it felt almost rude.
And honestly, it was a beautiful story. Simple stories always are.
But the story had a problem, and the problem kept getting older.
A site called Monte Verde
in southern Chile was dated to about 14,500 years ago.
That is before Clovis, thousands of miles
If people were already living near the bottom of South America,
How did they get there so fast?
And by which road?
So, researchers stopped staring at the stone tools.
They started looking at the mud.
In 2016, a team publishing
in Nature pulled sediment cores from two lake beds,
Charlie Lake in British Columbia and Spring Lake in Alberta,
right in the bottleneck of the supposed route.
Mud keeps a diary.
Pollen, plants, animals, all of it settles to the bottom,
layer by layer, year after year.
The team read those layers, and the verdict was blunt.
Before about 12,600 years ago, the Corridor was unlikely
to have supported the life that travelers would need.
The First Americans, they concluded, probably did not come that way.
Read that again. The Corridor was not open.
The front door of the Clovis story had been bolted shut
at the exact moment the story said people were walking through it.
But that did not end the argument.
Scientists haggled over the corridor's timing,
patched the model, nudged the dates.
Clovis first, but maybe a few stragglers before.
The postcard survived by being amended.
A story that useful rarely falls quietly.
Still, the question underneath only got louder.
So how did anyone get south?
Who were they?
And what did their ancestors look like, genetically,
before they spread across two continents?
Stone tools can tell you what people made.
They cannot tell you who those people were,
who they were related to, or who they left behind.
Bones can, but not by looking at them,
by reading what is locked inside.
Nobody could do that for the earliest northern people.
Nobody had ever found the right bones.
Then, in the interior of Alaska,
at a place called Zasana, somebody did.
In English, the site is called Upward Sun River.
The bones belonged to a baby girl who lived maybe six weeks.
The community that cares for the site gave her a name,
Zacite Anen Tidegai.
Sunrise Girl Child.
And when scientists finally read her DNA,
she did not slot into the map.
She redrew it.
Now, you've seen the title, Destroys Everything We Knew.
Let me be straight with you, that is too strong.
Her genome did not blow up the entire story
of how people reached the Americas.
In one important way, it actually backed up a piece of it.
But it did break something else.
Something people had quietly assumed for years.
That the people of ancient Alaska were just
an early draft of everybody who came after.
They weren't.
The genome says they were something else entirely.
Ben Potter, the archaeologist
who led the excavation at the University of Alaska Fairbanks,
put it plainly.
We didn't know this population existed.
If the earliest people in that valley belonged
to a branch nobody had mapped,
then every neat line drawn from ancient Alaska
to the living present had to be redrawn.
Think about what that sentence means.
A scientist who had spent years digging in that exact valley,
one of the people best placed on earth to understand it,
saying the people in the ground
were a surprise. But here's the thing,
the DNA is not even the strangest part of this story.
The strangest part is how these children were found,
how they were buried, what was placed beside
them, and an accusation made about
the site years before any genome was sequenced that I honestly
did not expect to find in the record.
Because before the babies, there was a three-year-old,
and before the DNA, there was an argument.
It began with a road. In 2006, archaeologists were
surveying possible routes for a road through the Tanana River Valley.
They found a site on a bluff
Stone tools. Burned bone.
Layers of occupation stacked like pages in a book.
The route was changed. The road went around it.
One line moved on a map,
and that single decision is the only reason anything
that follows still exists. Locally,
the place is Zasana.
Over the following years, Potter's team mapped six separate occupation
layers there. The oldest reached back about 13,200 years.
The youngest ran into recent times.
People kept coming back to this exact bluff again
and again for thousands of years.
Hold on to that detail, it matters later.
Then, in 2010, the trowels reached a hearth.
Inside it were burned human bones.
A child, about three years old.
Someone had placed this child in a fire.
And the remains stayed in that hearth for more than 11,000 years.
Archaeologists named the child Zasa Chege Tsenin,
Upward Sun River Mouth Child.
In 2011, the find was described in the journal Science.
The team called it a cremation, a deliberate funeral,
which is a heavy enough claim on its own.
Then came the accusation.
A Bureau of Land Management archaeologist named
Michael Kunems spoke to the Associated Press.
He said he saw no more evidence
of cremation than of something much darker,
that the child may have been cooked and eaten.
He pointed out that the bones missing from the hearth
were the ones carrying the most flesh.
Stop for a second.
This was not a stranger on a forum.
This was a government archaeologist talking to a national news service,
floating cannibalism about a three-year-old child
in the homeland of living people.
Potter rejected it.
The body, he said, still had soft tissue on it when it burned.
That changes how bones should look after a fire,
and he argued the evidence fit a cremation, not a meal.
That is the official explanation,
and it stands as the excavator's position.
But notice what happened next.
The one child at the center of the argument,
the burned child everyone wanted answers from,
gave up no DNA. Nothing. The bones were too damaged.
The most controversial burial at the
site stayed silent. And the argument simply stayed an argument.
So the team kept digging.
And in 2013, directly beneath that hearth,
something turned up that nobody argued about at all.
A pit. Red ochre staining the soil.
And inside it, two skeletons
so small that the first look must have been a shock.
One was a baby who lived perhaps six weeks.
The other was a fetus, somewhere between 30 and 33 weeks along.
Never born alive, or barely.
Laid in the ground together,
under the fire where an older child had been burned.
And beside them, something that did not belong with babies.
Hunting gear.
Four antler rods, called foreshafts.
Think of them as the detachable front end of a hunting dart or spear.
Each was between about 45 and 52 centimeters long.
Three were scratched with X-shaped marks.
Two stone points were still set in their hafts, made of basalt.
Every piece was coated in red ochre.
Antler, basalt, pigment.
Tools made for killing large animals,
placed beside children who would never hold them.
Ochre is not a casual material.
Someone prepared it, carried it,
and coated every single weapon in it before the dirt went back in.
Whatever happened at that pit, nobody rushed it.
And look at where it happened.
The three-year-old was burned in a hearth.
The infants lay in a pit directly beneath it.
Same spot, layered on top of itself.
Radiocarbon could not tell the two apart.
Close in time.
Possibly very close.
Why bury more children beneath
the place where another child had been burned?
The site does not say.
The community gave the second baby a name too.
Yalkanen Ti'id Gai, Dawn Twilight Girl Child.
Notice the names. In the scientific papers,
these three are USR1, USR2, and USR3, catalog labels.
The names came from the Tanana people, in their own language.
One set of names fits in a spreadsheet, the other fits in a family.
So why this burial? Why those weapons?
Here is where an interesting theory enters.
In 2020, a study led by Haas
and colleagues cited these burials as part of a larger argument.
That women in ancient hunter-gatherer
societies took part in big-game hunting,
and that girls could be buried with hunter's tools.
It is a real paper, and a serious claim.
But Potter does not read the grave goods that way.
By 2024, he was arguing that the weapons were offerings.
Gifts, he suggested.
From the fathers, fathers, hold that word.
The DNA will have something to say about it.
Then there was the season.
The hearth held fish bones, chum salmon,
among the earliest signs of salmon use
anywhere in the hemisphere.
It also held the bones of juvenile ground squirrels.
Together, those details point to one time of year,
late July, early August, the height of summer.
Summer was when food was plentiful.
It was also when this family buried its children.
Researchers read that as a sign of how brutal infant death could be,
even in the season of plenty. Imagine that scene.
Long light, river running, and a family digging a small pit.
Now here is where you should ask
the question the story has been circling.
Who got to decide what happened to these remains?
Because this was never a case
of scientists simply taking what they found.
A formal agreement covered the work.
The National Science Foundation, the state of Alaska,
the Healy Lake Traditional Council,
and the Tanana Chiefs Conference were all part of it.
And when the question came up of whether
to excavate or leave the burials in the ground,
the tribe chose excavation. The reason was the
risk of vandalism. A site this famous cannot stay quiet forever.
So the bones came out,
with permission, under an agreement.
And the community that gave that permission
wanted to learn what the bones could say.
Not every excavation begins with a conversation.
This one did.
In a field that has had its share of disputes over who gets to dig,
that agreement is part of the story of how this evidence was gathered.
Which brings us to the question that could actually be tested.
Not what the weapons meant.
Not what happened to the child in the fire.
A simpler one.
Who were the baby's mothers?
The first answer came in 2015,
and it did not go the way anyone expected.
A team led by Tackney, publishing in the journal PNAS,
looked at mitochondrial DNA.
That is the DNA passed down only through mothers.
Tiny, circular, and sturdy enough to survive when
almost everything else is gone.
Think of it as a maternal barcode. The two infants had
different barcodes. One carried a lineage called C1B.
The other carried B2.
Different mothers.
So these were not twin sisters, or a mother who lost two daughters.
They were two babies from two women,
buried in the same pit, under the same hearth, with
the same ochre, and the same weapons.
Which suggests a family joined by something other than a single mother.
That word again, fathers.
And then another oddity.
Both of those maternal lineages are rare
or absent in northern North America today.
In a place that has been continuously home to living peoples,
the mothers of these children
left no obvious trace. Researchers noticed. The question hung there.
Now, before the big genome, there was a more basic question.
Were these bones even as old
as they seemed? Because everything depends on it.
This is where carbon dating comes in,
and I will keep it simple.
Living things absorb a form of carbon from the air.
When they die, that carbon starts to decay at a known rate.
Measure what is left, and you can read how long ago something died.
The team did not rely on a single result.
They ran 41 radiocarbon dates across the site.
And the dates from the hearth
and the pit pooled together into one tight result.
About 9,973 years of radiocarbon time, give or take 27.
Convert that to calendar years
and you get roughly 11,610 to 11,270 years ago.
The hearth and the pit could not be told apart
by radiocarbon. Close in time.
Possibly very close. That part is verified.
Then came another test that sounds almost
impossible. They checked what the mothers had been eating.
Here's the trick. The chemistry of
your food ends up written into your bones and tissues.
Nursing and unborn babies carry their
mother's signature. A team led by Halfman measured it,
and the result was clear. About 30 to 32%
of those mothers' diet came from salmon. Not a rare treat. A staple.
Which is a remarkable thing to learn about women
who lived 11 and a half thousand years ago,
from the bodies of babies who barely lived.
Everything was lining up.
The age was real.
The site was real.
The season.
The food.
The fire.
All consistent.
All solid.
But dating a burial and reading a person are different jobs.
Dating tells you when.
It does not tell you who.
And the one child who might have answered,
the burned three-year-old, had already given nothing.
If the babies failed too,
the site would stay a beautiful silent puzzle.
Which is exactly when the real shock arrived.
Because the researchers wanted a full genome.
Not just the maternal barcode, but the entire genetic record.
And that is hard.
DNA breaks down in the soil.
So they went for the best place in the human skeleton to find it.
The petrous bone.
A dense knob of bone behind the ear, protecting the inner ear.
It preserves DNA better than almost anywhere else.
They took a sample from the baby girl,
Sunrise Girl child, and sent it to labs in Copenhagen.
Look closer.
This is where it gets strange.
In January 2018, a team led
by Moreno Mayar published the result in Nature.
They sequenced her genome to a depth of about 17 times coverage.
That means they read each piece of her DNA,
on average, 17 separate times.
For an ancient sample that old, that is extraordinary.
And the answer broke the pattern.
When geneticists compared her to every Native American genome they had,
she did not fall inside any of them.
She sat outside, at the base of the family tree,
related to all Native Americans,
but belonging to none of the known groups,
a cousin line that had never been sampled.
They named it Ancient Beringians.
The numbers behind that name are worth slowing down for.
The founding population of all Native Americans first split
from East Asians around 36,000 years ago.
Gene flow between the two continued until about 25,000 years ago.
Around the same stretch, between 25 and 20,000 years ago,
they mixed with a population called Ancient North Eurasians,
people whose ancestry shows up in Siberia.
Read that carefully.
It means the ancestors of Native Americans
were already a blend before they ever entered the Americas.
East Asian on one side, Siberian on the other.
Not one pure line, a mixture even then.
And then came the cut.
Somewhere between about 22,000 and 18,000 years ago,
the ancient Beringians branched
away from the ancestors of every other Native American,
during the Ice Age, long before
the corridor opened, long before Clovis.
That is a split nobody had seen before,
a branch with a name, a date, and a body in
a hearth.
So, was the title right to say previously unknown?
For the genetics, yes.
Population geneticists had expected structure like
this somewhere in the deep past.
But expecting a ghost and photographing one are different things.
Until Sunrise Girlchild, no ancient Beringian had ever been sampled.
And the rest of Native America? That branch too.
The northern and southern lines separated, the study says,
around 17 and a half to 14 and a half thousand years ago,
probably south of the ice sheets. And here is where the baby's name
finally makes sense. Scientists describe a lineage like this
as basal. Picture a family tree drawn on a wall.
Most branches fork again and again at the top.
A basal branch peels off near the bottom and keeps going on its own.
Sunrise Girlchild sat on one of those,
related to everyone, but separated from them far earlier
than anyone expected. Pause there,
because this is the moment the official story really started
to crack. Not because the Clovis story had
been wrong about everything, but because the family
tree no longer matched the map.
The Alaskan baby was not just an early version of the people who
spread south. She belonged to a lineage that had
been apart from them for thousands of years.
But the genome had a second answer.
And this one was personal.
Who were those two babies to each other?
In 2018, the University of Alaska Fairbanks reported
that they were most likely first cousins.
Potter's 2024 account widens that to half-siblings or first cousins.
Different mothers related through family.
That fits the maternal barcodes.
And it puts the weapons back on the table as a question of kinship.
One reading, Potter's,
is that the Spears were offerings from the Fathers.
The other, the one Haas and colleagues leaned on,
is that girls in these societies could be hunters.
The DNA cannot settle which.
But it settles something else.
These were not strangers who happened to die the same summer.
These were kin, buried together,
by people who knew exactly who they were.
Now back to the title. Does this destroy everything?
Not quite, and this is the reversal,
because the genome also showed that ancient Beringians
and every other Native American share one founding population, one.
Not several waves from different places,
one group of ancestors out of Asia splitting into
branches as they spread.
The old Clovis story said one wave, one route, one late beginning.
The genome said one source, yes,
but a much older beginning, and a family tree that branched early
So the Clovis timeline broke. The single source idea survived.
The simple version of the story did not die.
It just got harder to defend.
And there was more. The branch did not vanish with the babies.
In 1949, a researcher named
Helga Larsen excavated Trail Creek Cave in Alaska
and recovered a baby tooth, a single incisor.
It sat in storage for decades.
Then, in 2016, Jeff Rasek,
an archaeologist with the National Park Service, rediscovered it.
They dated it, about 8,085 radiocarbon years old, give or take 40.
That is thousands of years after the infants at Upward Sun River.
And when its DNA was read, in a study published in Science in 2018,
it carried ancient Beringian ancestry.
Rasek said the lineage had persisted to at least 9,000 years ago.
A people who kept their own identity
for at least 2,500 years after those babies were buried.
So the ancient Beringians were not a one-off.
They were a population, one that lived, hunted,
and raised families in Alaska for millennia.
Potter connected them in 2018 to a cultural tradition
known as the Denali Complex,
which spread across Alaska and the Yukon
from about 12,500 years ago to about 6,000.
And Siberia kept feeding the story.
Ancient genomes there,
including a 31,600 year old individual from a site called Yana,
and a man called Kolyma I, around 10,000 years old,
tied Northeast Asia into this picture in
ways nobody had mapped before.
But here is where the answer, once again, made the next question
worse. Because if the ancient Beringians lasted until
at least 9,000 years ago, what happened
after that? Did they die out? Were they absorbed?
Did they simply become someone else?
And that is where the real fight begins.
The first question in this fight is about location.
The theory that explains all
of this is called the Beringian standstill.
The idea is that the ancestors
of Native Americans got stuck for thousands of years,
isolated, in Beringia,
the lost landscape that once joined Siberia and Alaska.
They changed genetically there,
the theory says, and only later moved south.
It is a clean idea.
Picture Beringia, the exposed ground that joined Siberia to Alaska
when so much ocean water was
locked in ice. Not a thin bridge, a wide,
cold country with its own climate and its own animals.
Now picture a small population living there, cut off from both sides.
Genes drift, lineages split.
That is the standstill, a long pause that leaves a fingerprint.
But Potter himself has pointed out,
there is absolutely nothing in the genetics that requires
the standstill to have happened in Beringia at all.
The data shows a pause. It does not show where.
Then there is the tree itself. A preprint by Flegontov and colleagues
argues that ancient Beringians
do not form the deepest American lineage.
If that holds, the neat picture
of one early branch peeling off first gets messier.
That is a challenge, not a verdict,
but it is a reminder that the family tree is a model,
not a photograph. Then there is the question of fate.
Potter has proposed, in 2024,
that the ancient Beringians may have been absorbed by the ancestors
of Dené peoples, the Athabascan-speaking communities of the north.
Some evidence points
toward overlap. Na-Dene speakers carry 10
to 25 percent ancestry from a group called Paleo-Eskimos,
according to a 2019 study in Nature.
But that is a different line of ancestry.
And a 2026 review by Bethany Potter
and colleagues finds no later gene flow between ancient Beringians
and the northern and southern Native American branches.
So the ancient Beringians did not simply merge into the people
who spread through the Americas.
Whether they merged into anyone else, or faded,
or simply changed names, remains disputed.
That is the honest answer.
Theory, not fact.
And since we are being honest, here is a detail most videos skip.
Even the numbers move depending on where you read them.
The baby's age is given as 6 to 12 weeks in one account
and about 6 weeks in another.
Her genome is reported at 17 times coverage in Nature
and 14 times in a later review.
The split date appears as a range,
22 to 18,000 years ago, and as a single estimate, 20,900.
None of that breaks the finding,
but it is a reminder of what you were looking at.
Not a stone tablet, a model rebuilt with every new paper.
Meanwhile, the timeline for the whole continent keeps shifting
and not in the direction the old story wanted.
In 2023, a team publishing in Science argued that human footprints
in White Sands, New Mexico are between 21 and 23,000 years old.
Pollen from plants in the layers dated to around 23,000 years ago.
Sediment dating gave a minimum of about 21,500.
If true, people were walking across New Mexico far earlier than
the Clovis story allows,
and the reaction was immediate.
Michael Waters, a leading expert at Texas A&M,
told Smithsonian Magazine, I really remain skeptical.
He suggested the seed dates could
be off by seven or eight thousand years.
Then there is Chiquihuite Cave in Mexico. A 2020 paper
in Nature claimed stone artifacts there date to between 26
and a half thousand and 33,000 years
Chatters and colleagues, including Potter himself,
responded in 2021 that those stones are geofacts.
Natural limestone rubble from the cave walls, or from water. Not tools.
Even the old anchor, Monte Verde, at about 14,500 years,
lands right at the edge of the window the Nature paper gave
for the split between the northern and southern branches.
17,500 to 14,500 years ago. Probably south of the ice sheets.
Does that prove anything? No.
But it makes the tidy timeline harder to defend.
The family tree and the footprints are being drawn on the same wall
by people who do not agree on the scale.
So we have a continent where the old story is gone,
and the replacement stories are being fought over line by line.
White Sands says 21,000.
Chiquite says 30,000.
And a skeptical establishment says,
Wait, who benefits when a date moves?
Who loses?
For some careers, a new date is a triumph.
For others, it is an embarrassment.
For the old textbooks, it is a rewrite.
But for the people whose ancestors lived through this,
it is not an abstract question at all.
Here is a thought that should bother you.
That site was found because of a road survey,
a line moved on a map.
How many other river terraces in how many valleys
have never been surveyed at all?
How many ancient Beringian graves
are still sitting under the soil,
waiting for a road that never comes?
Nobody knows.
And that brings us back to a fire pit in Alaska,
and to what happened to the children.
The remains of those infants, and of the three-year-old,
are held today in a confidential, secured location.
Their final disposition has not been decided.
Not publicly, at least.
Not yet.
And here is a detail that surprises people.
A federal law protects Native American remains.
NAGPRA.
It does not technically apply here because the land is state land.
The agreement between the community
and the researchers is not a legal requirement.
It is a choice, a partnership, built on trust,
holding a priceless thing in place.
First Chief Joanne Polston
of the Healy Lake community expressed the community's interest
simply.
She said,
I would like to learn everything we can about this individual.
Everything.
That is the word.
Now think about the three-year-old again,
the burned child, the one at the center of
a public accusation.
The one child of the three who yielded no genome at all.
The most argued over burial at the site
is the only one that cannot answer.
Maybe that is the real irony.
Everything we learned about ancestry came from the two babies.
Who owns the past when the evidence changes?
The university that excavated it?
The state that holds the land?
The scientists whose names are on the paper?
The living communities who may
or may not descend from them? Or no one at all?
People like to call them a lost people.
It makes a good phrase. It might also be wrong.
Disappearing from a genome
and disappearing from the world are not the same thing.
Nothing in the evidence says a whole population simply vanished.
It says their signature changed,
and we do not yet know how or into whom.
The DNA says the lineage persisted.
The DNA also says it may not have survived.
Both can be true, and it is the kind
of answer that makes everyone uncomfortable.
Because it will not tell you where to stand.
Somewhere, in a location the public is not told,
small bones are kept under lock.
In a lab, a branching diagram glows on a screen,
with a short line near the base labeled,
Ancient Beringians.
Two very different rooms.
One story.
And a question hanging between them that no instrument has answered.
So think about that trench again.
It has been backfilled now.
Earth over the hearth.
Earth over the pit.
Whatever the soil still holds, it holds quietly.
The two infants were laid in that pit facing the river,
a river that has not stopped running since.
Braided channels, pale silt, loess bluffs,
a summer light that does not let go of the sky.
They were buried with spears they never lifted.
They carried a family tree nobody knew existed.
They lay beneath a fire that had burned another child,
whose story still cannot be settled.
And somewhere along that same river, maybe right now,
stands someone whose ancestors knew exactly
who they were. Or maybe no one does.
The science cannot tell you which. Today, in a place the public
is not told, two small girls,
eleven and a half thousand years old, still carry a question the
whole continent is trying to answer.
Who were they? And who are they to us?
The bluff keeps its secrets. The ground has been closed.
And out past the backfilled trench,
the river keeps running toward
the place where they once lay, facing it.