Hvalseyjarfjord, near Qaqortoq, Greenland

Hvalsey Church

About 985 to the mid-1400s (a wedding in writing, then a radiocarbon tail)

A stone church still stands above a southern Greenland fjord. The last document from the Norse farms is a wedding held here in the early fifteenth century. People did not simply vanish on that day. The ground keeps them a little longer, and then the farms are empty.

Why this is extraordinary

The sagas put Norse farms in southwestern Greenland from about 985, after an exile from Iceland. Two clusters grew up. The Eastern Settlement, in the far south, held on until the middle of the fifteenth century. The Western Settlement, farther up the west coast, was already empty in the fourteenth, and one sea-level study uses about 1350 for that earlier end. The Eastern Settlement, in the same study, ran to roughly five hundred sites and an average population of about two thousand, between 59.75 and 61.5 degrees north.

The last written page is a wedding at Hvalsey Church, in the early fifteenth century. Radiocarbon from the settlement continues for about another half century after that page. The church in the photograph is the largest Norse ruin still standing in the area, and since 2017 it has sat inside the Kujataa World Heritage listing, Norse and Inuit farming at the edge of the ice. The walls are a stack of flat stones. The roof is gone. Nothing in the stone says why the parish stopped.

What ended it has been a list for a long time. Cold, as the Medieval Warm Period (about 900 to 1250 in that sea-level paper) gave way to the Little Ice Age (about 1250 to 1900). Drying, which a lake record beside a former farm finds without a large temperature drop. A rising local sea, one to three metres between 1000 and 1450 across the farms in the model, enough to take pasture at the water. A break with Europe, when the ivory and hide trade thinned. Hostility with Inuit neighbours, present in written accounts and not matched by a battlefield in the ground. The church is where the written record stops. It is not, by itself, the cause. Bone collagen from 27 Norse skeletons, not from Hvalsey's own yard, is read as a shift from about 20 percent marine food to about 80 percent across the settlements, herds giving way to seals. The endpoints are a choice, and the uncertainty is about 10 percent. Walrus ivory is the export a 2015 study and a 2018 review treat as the good that paid for iron and cloth. A later collapse of that trade is a cause that can arrive on a ship, without a failed hayfield. The 2018 paper has the mid-fjord church sites, Hvalsey among them, still busy in the fourteenth century while outer farms emptied, and it treats grazing erosion as real but more limited than a story of collapse by overgrazing. The catch on that busy century is that the word thrive is a survey pattern cited to a thesis, not a new trench under this floor.

Why it matters

If cold closed the settlement, this is a climate story with a date. If the sea took the hay meadows, or the rains failed, or the ships stopped, the same empty church is a different failure. The walls do not choose.

Hvalsey is the pin. Brattahlíð, farther up the same Eastern Settlement, is the farm the sea-level study uses for lost hectares, and it is in the second photograph, not a second pin.

Photo tour

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

What is still unexplained

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

Open A wedding, then a half century Writing ends with the Hvalsey wedding in the early fifteenth century. Radiocarbon from Norse occupation runs about fifty years past that ceremony. A document can stop because the writer left, or because the ship that carried letters stopped, while a household was still burning turf. The reverse is also open. A late charcoal date can be driftwood, or a hearth used once by someone passing.

Open Dry grass, a higher shore, and a temperature that barely moves A lake core next to a former Eastern Settlement farm finds no substantial temperature change through the settlement centuries. It finds a drying trend from about 950 that peaks in the sixteenth century. Winter hay is the hinge. Livestock had to be fed indoors, and drier summers grow less grass. Separately, ice growing on the nearby ice sheet is modelled to have lifted local sea level, by crustal sinking and by the extra pull of the ice on the ocean, by roughly one to three metres from 1000 to 1450. An older estimate quoted in that work gives Brattahlíð about 50 hectares of home land lost, and about 200 hectares gone at the head of Tunulliarfik. Drying and flooding can both be true. Neither is a letter saying we are leaving.

Open A conflict that the texts have and the ruins do not Later writings, including an account from about the 1340s, describe the Western Settlement overrun. Saga pages describe hostile meetings with people called Skrælingjar. Inuit things do turn up in Norse farms in the fourteenth century, and Inuit camps appear in the outer fjords. A review of that record says there is no robust archaeological trace of a battle, and that low-level violence would be hard to see. Contact is not the same as a conquest. Absence of a battlefield is not a proof of peace.

Location and dating

Location: Hvalsey Church, Hvalseyjarfjord, near Qaqortoq. The pin is 60.829 north, 45.783 west. The Eastern Settlement spreads through the fjords from about Cape Farewell north toward the ice. Brattahlíð, at Qassiarsuk, is the other famous farm in that cluster.
Period label: Current consensus — open to revision Norse settlement from about 985. Western Settlement gone by about 1350. Eastern Settlement largely abandoned by the early to mid fifteenth century, with radiocarbon running about a half century past the Hvalsey wedding. Inuit farming in the same district is part of the World Heritage listing and is not the date of this church.
How this dating was done: The start is the saga chronology, about 985. The written end is the wedding record. The tail is radiocarbon on occupation. The climate readings are lake sediments and a geophysical sea-level model, not a date on a named stone of the church.
What could change it: A sequence of dates from inside Hvalsey itself, from the floor and from under the wall, would show whether this building was still in use in that radiocarbon tail. A counted loss of hectares at Hvalsey, rather than the Brattahlíð estimate, would show whether this fjord drowned the same way.

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

Settlement centuriesContested

Modelled sea-level windowContested

985 CE1250 CE1450 CE

Settlement centuries. From the saga start to the mid-fifteenth-century end used for the Eastern Settlement, with the wedding early in that last century.
Modelled sea-level window. The years over which the model gives about 1 to 3 metres of local rise at the farms.

What is actually there

The Eastern Settlement is given as about 500 sites and an average of about 2,000 people.
The Hvalsey wedding is the last written evidence, early in the fifteenth century.
Radiocarbon continues for about half a century after that wedding.
Modelled local sea-level rise from 1000 to 1450 is about 1 to 3 metres, varying by farm.
The lake record finds drying from about 950, peaking in the sixteenth century, without a large temperature drop.
No published solstice or star azimuth for the church was found for this room.

Open questions

Question	Status
Did the parish end because the hay failed, the shore rose, the ships stopped, or people were driven out?	Open. Drying, local sea level, trade, and conflict each have a published piece. None has a document of departure.
Does the half century of radiocarbon belong to Hvalsey, or to other farms?	Open. The tail is a settlement result. It is not yet a date from under this floor.

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

Several stresses, no single eviction

The farms were left in the fifteenth century because hay, sea level, trade, and isolation stacked up. No one cause has to win. The church is the last address in writing, not the weapon or the drought.

What supports this

A lake core finds drying through the settlement without a large temperature fall, which hits winter fodder.
Collagen from 27 skeletons, not from this churchyard, shifts from about 20 percent marine food to about 80 percent. The endpoints are a choice, and the uncertainty is about 10 percent.
Ivory that paid for iron and cloth can fail on a ship. A 2018 review keeps mid-fjord churches, Hvalsey among them, busy in the fourteenth century, and treats overgrazing as more limited than a collapse story. "Thrive" is cited to a thesis.
The sea-level model gives about 1 to 3 metres of local rise from 1000 to 1450, and drowned features are reported in the fjords.
The wedding is early fifteenth century and the radiocarbon tail is short, which fits a rundown rather than a single night.

What pushes back

A stack of stresses can describe any abandonment. It does not say which one tipped Hvalsey.
Written accounts of violence are left outside the stack, and they are real texts.
The model’s ice growth is fitted to a salt-marsh record at Nanortalik, not measured at the church door.

What would settle it

It is the reading that keeps every measurement and refuses a single villain. It is also the reading that is hardest to falsify, which is its weakness.

Emerging narratives

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

An emerging narrative

The Little Ice Age closed the farms

Temperature falls after about 1250 made pastoral farming impossible, and the Eastern Settlement emptied as the cold deepened.

What supports this

The sea-level paper uses a Medieval Warm Period of about 900 to 1250 and a Little Ice Age of about 1250 to 1900, and ice in southern Greenland did readvance.
Outer-fjord farms were given up earlier than inner ones, which fits a climate gradient.

What pushes back

The lake beside a farm finds no substantial temperature change in the settlement centuries. Drying is the signal, not cold.
People stayed through the first centuries of that cold. A slow ice age does not date a mid-fifteenth-century end by itself.

What would settle it

Cold was the textbook line. The lake result is why this room does not leave it in the centre. It remains a contributor to the ice that lifted the local sea.

An emerging narrative

Inuit expansion ended the Norse parishes

Thule Inuit arrival in the Norse fjords in the fourteenth century, and hostile contact recorded in writing, pushed the settlements out.

What supports this

Inuit material is found in Norse sites, and Inuit activity is placed in the outer fjords from the fourteenth century.
A written account from around the 1340s says the Western Settlement was overrun.

What pushes back

A review finds no robust archaeological evidence of a battle, and says a fifteenth-century war is a conjecture.
Trade and copying of tools look like contact. Contact has more than one mood.

What would settle it

The texts earn it a full hearing. The missing battlefield is why it is not the current narrative.

Should the current view be reevaluated?

Open to revision

The abandonment is real, and the wedding-plus-radiocarbon bracket is the best end date we have. Which stress closed Hvalsey is not settled. Drying and local sea level are measurements. Conflict is a text without a grave of battle. Cold, as a temperature collapse, is the reading the lake does not support.

What drives this

A written end and a radiocarbon tail that do not fall on the same year.
A drying record and a 1 to 3 metre local sea-level rise.
Written hostility without a published battle site.

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.

Several stresses, no single eviction Unsure The Little Ice Age closed the farms

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.

Local sea level and drying made the farms unviable

The Eastern Settlement was lost to a shoreline that rose beside a growing ice sheet, and to hay that failed in a drier summer, while the temperature story was the smaller part.

Why it is still worth a test

A hectare count for Hvalsey’s own shore, and dates from under the church floor.

Conflict ended the parishes

The mid-fifteenth-century silence is a violent end, and the Western Settlement’s earlier loss is the first chapter of the same event.

Why it is still worth a test

A published trauma study of the latest Norse burials, and a map of Inuit and Norse dates in the same fjord.

Sources

Sea-level rise in Southwest Greenland as a contributor to Viking abandonment Proceedings of the National Academy of Sciences
Prolonged drying trend coincident with the demise of Norse settlement in southern Greenland Science Advances
Kujataa Greenland, Norse and Inuit Farming at the Edge of the Ice Cap UNESCO
Change of diet of the Greenland Vikings Arneborg and others, Radiocarbon, 1999. Twenty-seven skeletons, about 20 percent marine food to about 80 percent. Not from Hvalsey's yard.
Cultural adaptation and Norse Greenland Dugmore and others, PNAS, 2012
Was it for walrus? Frei and others, World Archaeology, 2015. Ivory as the export that paid for iron and cloth.
Disequilibrium, adaptation, and the Norse settlement Jackson and others, Human Ecology, 2018. Mid-fjord churches busy in the fourteenth century. Grazing erosion treated as limited. "Thrive" is cited to a thesis.

Image credits

Hvalsey Church, walls still standing, roof gone, on the grass above the fjord. The doorway in this frame is in the near gable. No compass bearing is printed on the photograph. . Number 57, CC0 1.0. Shown unedited.
Brattahlíð, not Hvalsey. Stone footings in the grass, and a modern red house on the hill. This is the farm for which a quoted estimate counts about 50 hectares of home land lost to the rising shore. . PederM, CC BY-SA 2.5. Shown unedited.

← Greenland World map