Our Deep History Our Deep History A museum of unexplained ancient sites and relics
Canada rooms

Keele Range, northern Yukon, above the Bluefish River, Canada

Bluefish Caves

about 24,000 years (a cut-marked horse jaw, calibrated from 19,650 radiocarbon years)

The oldest marked bone in the 2017 paper is a horse jaw from Cave II. Radiocarbon 19,650 plus or minus 130, calibrated to about 24,000 years. The drawing is not a photograph of the cave. Diagram: Our Deep History.

Three small caves above a Yukon river held on to a horse jaw that someone seems to have cut while freeing the tongue. Radiocarbon on that jaw reads 19,650 years, plus or minus 130, which calibrates to about 24,000. If the mark and the layer both hold, people were in the far northwest during the deepest cold of the last ice age.

Why this is extraordinary

Between 1977 and 1987 a dig worked three little caves in the Keele Range of the northern Yukon, about 54 kilometres southwest of Old Crow and 250 metres above the Bluefish River. They open at the foot of a limestone ridge, and none of them is a hall. The largest is a cavity of less than 30 cubic metres. The crew went down in thin spits through loess, the wind-blown dust that sits on the bedrock and is covered by later humus. Out of that dust came a great quantity of animal bone, a modest set of stone tools including microblades, and the argument that people had been here during the last glacial maximum, the coldest stretch of the last ice age, when many archaeologists thought the way into the Americas was still shut.

The dates sat uncomfortably for decades. They were early, the caves are tiny, and it was fair to ask whether the marks were butchery or the work of carnivores, roots, and trampling, and whether the objects and the dates even shared a layer. In 2017 a taphonomic restudy, a careful reading of how each bone was damaged, went back through the collection. Of the thousands of fragments, fifteen were judged confidently cut by people, and twenty more only probably. That is less than one percent of the bone. Six of the confident pieces were dated by accelerator radiocarbon. They run from 10,490 plus or minus 55 radiocarbon years to 19,650 plus or minus 130, a span that calibrates to about 12,000 to 24,000 calendar years ago.

The oldest is a horse mandible from Cave II. According to the notes it lay in the basal loess, 142 centimetres below datum. On the inner side, under the molars, in the place where a tongue is cut free, the marks are straight and parallel, scored inside the way a stone edge scores. A caribou pelvis from the same cave falls in the same cold centuries. The paper reads the pair as people in eastern Beringia, the Yukon and Alaska side of the old land bridge, during the glacial maximum. A population waiting out the cold there is already an idea in the genetics, and a cave of this age would give that idea a place to stand. The catch is the one this argument has always carried. Exact depths for most of the dated bones could not be rebuilt from the archive, and the marked pieces are few. A last-glacial presence this far east is still ahead of the sites many colleagues treat as the secure threshold, nearer 14,000 years ago. A horse's jaw with a butcher's marks is either the earliest archaeological foothold in North America, or a reminder of how much weight a handful of bones are being asked to carry.

Why it matters

The peopling of the Americas is usually told as a walk that starts once the ice loosens, around 14,000 years ago and later. Bluefish asks whether a smaller presence had already sat in Beringia through the cold. One jaw cannot settle a continent. It can keep the question from being closed.

The pin is the caves, at 67 degrees 9 minutes north and 140 degrees 45 minutes west. No open photograph of the cavities is in the set used here, so the figure is the date and nothing else. Sites that more researchers accept, a different chapter after the coldest peak, include White Sands and Monte Verde.

Photo tour

What is still unexplained

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

Observed in the restudy Fifteen bones out of thousands Confident cut marks are recorded on fifteen specimens, from horse, caribou, wapiti, and possibly sheep and bison, plus a bird bone that had already been published. The placement is read as skinning, taking a limb apart, and cutting meat. Twenty further bones are only probable. A signal that small is either precious, because an ice-age visit would have been brief, or fragile, because a few misread marks would carry the whole date.
Open A date that is only as good as the archive In the notes, the horse jaw sits 142 centimetres into the basal loess, and its radiocarbon age agrees with that position. For the other dated pieces, the exact spit could not be recovered from the records that survive. Loess can lie tidy, and in a cave this small it can also slump. A mark dated to 24,000 years is a mark on a bone of that age. It is a human presence of that age only if the mark was made then, and in that layer.
The disagreement Ahead of the sites people trust Many archaeologists still place the first solid American sites after the glacial maximum, nearer 14,000 calendar years ago, which is where repeated camps, tools in place, and dates agree with one another. Bluefish has tools in the loess and, with the restudy, a taphonomic argument. It does not yet have a dozen similar caves for company. Being the oldest is the reading the paper offers, and it is the reading a single site always struggles to keep.

Location and dating

Location
Bluefish Caves, in the northern Yukon, at the base of a limestone ridge above the Bluefish River, about 54 kilometres southwest of Old Crow.
Period label
Current consensus — open to revision Cut-marked bone from about 24,000 to 12,000 calendar years ago in the 2017 series, held in loess of the last glacial maximum and of the years after it. Current consensus is still split. The marks and the dates are published. Whether they rewrite the peopling of the Americas is not settled.
How this dating was done
Accelerator radiocarbon dates the collagen in the bone, the organic fraction that once was the animal. Calibration then stretches or squeezes those radiocarbon years onto calendar years, which is why 19,650 becomes about 24,000 and is not a second measurement. The human part is the cut, read from its shape rather than from a chemical test for a hand.
What could change it
Date again a marked bone whose depth is secure, and the early end gets thicker. Show that the "cuts" match a known natural scar, and the jaw drops out of the argument. Another cave with the same age and the same marks would matter more than either of those.
Published ranges, drawn as bars. A bar is a claim about a span, not a point.
Six cut-marked bonesContested
Often-cited later threshold
24000 BCE17500 BCE12000 BCE
  • Six cut-marked bones. The 2017 series runs from 19,650 plus or minus 130 radiocarbon years at the old end, calibrated to about 24,000, down to about 12,000.
  • Often-cited later threshold. About 14,000 calendar years, the threshold where the paper says most archaeological evidence still prefers to begin. Not a date measured in these caves.

What is actually there

Open questions

Question Status
Are the marks butchery? For fifteen bones the 2017 study says yes, and it shows why carnivores and weathering do not fit those particular scars.
Were people in the Yukon at the glacial maximum? Open. This collection is the case for yes, and the case is thin enough that the argument has not closed.
Does one cave date a continent? No. If the layer holds, it dates a visit, and it offers that visit to the larger argument.

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

The secure threshold is still later

Whatever happened inside these caves, the archaeological record most researchers treat as secure still begins after the glacial maximum, nearer 14,000 calendar years ago, in the centuries when sites repeat.

What supports this

  • The 2017 paper itself says this later threshold is where the weight of evidence has sat.
  • One cave, a handful of marks, and an archive with holes in it are a thin foundation for a continent.

What pushes back

  • Putting the jaw aside because it is early goes in a circle if the marks are real.
  • Genetics has its own reason, separate from these caves, to expect people in Beringia during the cold.

What would settle it

This is the cautious consensus. It is a poor reason to avoid the bone, and a good reason to let the bone carry only the weight it can.

Emerging narratives

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

An emerging narrative

People in the Yukon during the last glacial maximum

Cut-marked bones from Bluefish Caves are read as people in eastern Beringia by about 24,000 calendar years ago, earlier than the sites usually taken as the start of the American record.

What supports this

  • Fifteen bones carry marks the restudy gives to stone tools, including a horse jaw dated to 19,650 plus or minus 130 radiocarbon years.
  • That jaw's recorded depth is in the basal loess, the place an early date belongs.
  • Microblades and other tools were excavated from the same dust.
  • Genetics already pictures a population paused in Beringia, and a cave of this age would be archaeological company for the idea.

What pushes back

  • Confident marks make up a tiny fraction of the bone, and the depths of most dated pieces are lost.
  • For years the site was doubted for a precise reason. The anthropogenic signature was unclear, and so was the stratigraphy.
  • One locality at 24,000 still sits ahead of the repeated camps nearer 14,000.

What would settle it

The restudy is stronger than the old objection that nobody had looked hard at the marks. It is not yet the settled opening chapter of the Americas. The paper says oldest. The field has not all agreed.

Should the current view be reevaluated?

Open to revision

The cut marks and the radiocarbon series are a real restudy of a collection people had disputed. Whether they move the start of the human Americas from about 14,000 years to about 24,000 is still open. The sample is small, and the archive is incomplete.

What drives this

  • A horse jaw of about 24,000 calibrated years, the marks on it read as butchery.
  • Fifteen confident specimens, and not a camp with its tools still in place.
  • A field that still sets the secure record later than this.

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.

People in the Yukon during the last glacial maximum Unsure The secure threshold is still later

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.

People were in the Yukon at the glacial maximum

The pause in Beringia would have a place, and sites later on, south of the ice, would be a second movement rather than the first arrival. A model of an empty America until 14,000 would be too simple.

Why it is still worth a test

Depths restored for the other marked bones, together with a second northern site of the same age and the same kind of mark.

Sources

  1. Earliest human presence in North America dated to the Last Glacial Maximum PLOS ONE, 2017. The caves, the fifteen cut marks, and the radiocarbon series.

Image credits

← Canada World map