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

Avella, Washington County, Pennsylvania, United States

Meadowcroft Rockshelter

Hearths from about 16,000 radiocarbon years ago (the coal question is the catch)

The rockshelter at Meadowcroft. The cliff and the dry ground under it are the ancient camp. The roof, the deck, and the stairs are how you visit it now. Lee Paxton, CC BY-SA 4.0. Shown unedited.

A groundhog and a patient farmer opened one of the deepest arguments in American archaeology. Under a ledge above Cross Creek, hearths date to about 16,000 radiocarbon years, which the site's museum reads as about 19,000 calendar years. The objection is not the stone tools. It is coal in the hill, which can make charcoal look older than the fire.

Why this is extraordinary

In 1955 a farmer on his family's land near Avella, in the hills of southwestern Pennsylvania, looked into a groundhog hole under a sandstone ledge and saw a stone tool. He covered the hole and waited nearly twenty years for an excavation that would not wreck the layers. Digging began in 1973. What came out was a campsite used again and again, from historic times down through the Woodland and the Archaic into a Paleoindian level, and then, if the oldest charcoal holds, into a time before Clovis.

So how old is the bottom? The National Historic Landmark nomination is plain about which samples are which. Five dates on hearths or fire floors in the lowest cultural layer run from 16,175 radiocarbon years, plus or minus 975, down to 13,240, plus or minus 1,010. The museum that keeps the site takes 16,000 radiocarbon years and, once the calibration to calendar years is applied, reads a human presence about 19,000 years ago. Three still older assays, including a piece of bark at 19,600 plus or minus 2,400, come from material that might be a cut basket and might be a forest fire. Those are not the hearths.

The catch has a name, and the name is coal. The country rock here includes bituminous coal, which is dead carbon, so old that it adds age to anything it touches. Critics argued that coal dust, or coal dissolved in groundwater, had seeped into the charcoal and made the deep dates too old. The excavators answered that the samples were looked at under a microscope, that coal particles would have shown, and that the dates sit in order with the layers, the plants, and the tools. A small blade kit under Early Archaic hearths is a real stratigraphy. Whether the carbon in those hearths is only wood is the argument that has never quite sat down.

A separate objection is that the plants and animals look too temperate for a glacial-age camp. The 1980 reply exists because of that objection, and the excavators answered that the dates agree with the valley’s history. A mismatch can be a refuge, a misidentified bone, or a bad date. No species from that list is named here.

What this changed

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

Older textbook view
For a long time the first widespread culture south of the ice was Clovis, and anything older had to survive a contamination argument before it could move the book.
New evidence
Hearth dates from 16,175 ± 975 to 13,240 ± 1,010 radiocarbon years, under a long sequence of later camps, with a museum calibration of the early end near 19,000 calendar years.
How it is dated
Radiocarbon on charcoal from fire features. The coal critique is about what else might be in that charcoal.
What moved
Meadowcroft became the eastern type site for a pre-Clovis camp. It did not end the argument. It moved the argument into the chemistry of the samples.
What is still disputed
Particulate coal, dissolved dead carbon, and three older assays that are not clearly from hearths. The order of the layers is the excavators' answer.

Why it matters

If the hearths are as old as the carbon says, people were camped in western Pennsylvania thousands of years before the Clovis horizon, and the White Sands tracks are not the only early clock in the east of the argument. If coal donated the extra age, the tools may still be pre-Clovis, or they may sit closer to 13,000 radiocarbon years, and the famous bottom of the shelter gets younger without disappearing. That is the difference between a revolution and a careful early camp.

Radiocarbon years are what the laboratory measures. Calendar years are that number translated through a calibration curve, because the amount of radiocarbon in the air has not stayed constant. Clovis is the widespread fluted-point horizon near 13,000 calendar years ago. A hearth is a fire people laid. Bark in the dirt may be a basket, or a tree.

Photo tour

1 photograph, unedited. On a phone, swipe the row. Choose one to look closely.

The rockshelter at Meadowcroft. The cliff and the dry ground under it are the ancient camp. The roof, the deck, and the stairs are how you visit it now.Lee Paxton, CC BY-SA 4.0

What is still unexplained

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

The critique that defined the site Coal in the water, or not Bituminous coal has no radiocarbon left. A little of it in a charcoal sample pulls the age backward, toward too old. Critics asked for tests that would catch coal particles and coal-derived humus, and they noted that ordinary laboratory cleaning does not always see coal. The reply from the dig is that the Smithsonian laboratory examined the samples, that local coal has no cell structure and would have been obvious next to wood, and that you would need a large fraction of dead carbon to drag an 11,000-year fire all the way to 19,000. The hearth series, with its wide errors of about a thousand years, is the set tied to fire features. The bark at 19,600 ± 2,400 is the sample everyone should handle gently. It might be cut, or it might be a wildfire, and the nomination says so.
The part that does not depend on the oldest number A sequence you can climb Above the debated layer sit Early Archaic hearths and a long stack of later camps, Archaic, Woodland, and historic. The tools change as you come up. That stack is why a single stray old date is a poor description of Meadowcroft. Even a reader who will not grant 19,000 calendar years still has a shelter used, on the nomination's cultural dates, from somewhere in the 16,000-radiocarbon-year range upward. Where inside that range you stop believing is the coal question, not a question about whether anyone came.

Location and dating

Location
Meadowcroft Rockshelter, 401 Meadowcroft Road, Avella, Washington County, Pennsylvania, under a ledge above Cross Creek. A National Historic Landmark. The wooden cover in the photograph is modern.
Period label
Current consensus — open to revision Hearths from about 16,175 to about 13,240 radiocarbon years ago, read by the site museum as reaching about 19,000 calendar years at the early end. Later camps continue into historic times. The bark dates near 19,000 to 21,000 radiocarbon years are not securely cultural.
How this dating was done
Radiocarbon on charcoal from hearths, and on bark of uncertain origin. Stratigraphy is what puts those fires under Early Archaic ones. Calibration turns radiocarbon years into calendar years. Coal, if it is in the sample, is invisible age.
What could change it
A new assay of hearth charcoal from the lowest layer, cleaned by a method meant to see coal, that either repeats the 16,000-radiocarbon-year reading or moves it.
Published ranges, drawn as bars. A bar is a claim about a span, not a point.
Cultural hearthsContested
Older, uncertain samplesContested
19100 BCE15000 BCE11300 BCE
  • Cultural hearths. 16,175 ± 975 to 13,240 ± 1,010 radiocarbon years before present, on fire features. The museum reads the early end as about 19,000 calendar years.
  • Older, uncertain samples. Includes bark at 19,600 ± 2,400 radiocarbon years. The nomination says the cultural origin is uncertain.

What is actually there

Open questions

Question Status
Is the carbon in the deepest hearths only wood? Open. That is the coal argument. The layers above it are not the dispute.
If the hearths hold, what route reached this creek so early? Open. A Pennsylvania camp does not draw the map. It only refuses a late start.

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 camp by about 16,000 radiocarbon years ago

People lit fires under the ledge well before Clovis. The hearth charcoal is wood, the dates sit in stratigraphic order, and the museum's calendar reading of about 19,000 years is a fair translation of the early end. The bark dates are a separate, weaker claim.

What supports this

  • Five dates on fire features from 16,175 ± 975 to 13,240 ± 1,010 radiocarbon years.
  • A long sequence of later occupations above them, in order.
  • Microscopic checks that the excavators say found no coal particles.
  • A National Historic Landmark nomination that treats the pre-Clovis material as the reason the place matters.

What pushes back

  • Coal is in the hill, and dissolved dead carbon need not look like a lump of coal under a lens.
  • The errors on the hearth dates are about a thousand years wide.
  • The famous 19,600-year bark sample is not securely a tool or a basket.

What would settle it

The centre of the room is the hearth series, not the bark. Coal remains a live way for those hearths to be too old, which is why the bar is marked contested.

Emerging narratives

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

An emerging narrative

Dead carbon made the bottom look early

Groundwater or coal dust aged the charcoal. The tools and fires are real and later, closer to other Paleoindian camps, and the 19,000-calendar-year reading is an artifact of contamination.

What supports this

  • Bituminous coal is a documented contaminant of radiocarbon samples, and ordinary cleaning can miss it.
  • The oldest assays are the ones with the least secure link to a human act.
  • A shelter this famous had to survive a second look at its carbon, and the second look is not finished.

What pushes back

  • Dragging a date from about 11,000 to about 19,000 radiocarbon years takes a lot of dead carbon, which is hard to hide in a small sample.
  • The dates form a sequence. Random contamination would be noisier.
  • Plants, animals, and tools were part of the excavators' reply, not only the carbon.

What would settle it

It is the right objection, and it has not carried the field. It would carry it if a coal-specific redate moved the hearths and left the upper sequence alone.

Should the current view be reevaluated?

Open to revision

The shelter, the long sequence, and a pre-Clovis hearth series are the finding the landmark status rests on. The calendar figure of about 19,000 years is the museum's calibration of about 16,000 radiocarbon years, and the coal critique is why that figure is not closed. The bark at 19,600 radiocarbon years should not be asked to carry the camp.

What drives this

  • Hearth dates from 16,175 ± 975 to 13,240 ± 1,010 radiocarbon years.
  • A calibration to about 19,000 calendar years at the early end.
  • A coal-contamination critique and a microscopic reply.
  • Older bark that may not be cultural.

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 camp by about 16,000 radiocarbon years ago Unsure Dead carbon made the bottom look early

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.

Coal aged the deepest charcoal

The shelter remains an unusually long camp sequence. The headline age drops toward the younger hearth dates or further, and Pennsylvania loses its claim to a 19,000-year start. White Sands and other early sites would no longer have this ledge as a neighbour of the same age.

Why it is still worth a test

A coal-specific preparation of new charcoal from a lowest hearth.

The bark basket is real and 19,600 radiocarbon years old

The camp starts even earlier than the hearths, and a perishable container sits at the bottom of eastern prehistory. That reading needs the cut marks to convince a second pair of eyes.

Why it is still worth a test

The bark itself, republished with the reasons it is cut rather than split by fire.

Sources

  1. Meadowcroft Rockshelter National Historic Landmark nomination National Park Service
  2. How do we know people used the Meadowcroft Rockshelter 19,000 years ago? Heinz History Center
  3. Comments on the Meadowcroft radiocarbon chronology and coal contaminants American Antiquity
  4. The 1980 reply on the plants and the animals American Antiquity, 1980. The reply exists because the plants and animals were said to look too temperate. The excavators answered that the dates agree with the valley’s history. No species is named here.

Image credits

← United States World map