An Item Received Without a Provenance

A cracked sheet of vellum covered in close rows of runic characters, pinned flat under a glass plate on a dark table between two brass magnifiers, a pair of calipers, an inkwell, an oil lamp and a paper accession tag on a string
Accession 4471, received without a covering letter. The hide is real. The Bureau has so far said nothing about the road.

The object is a single sheet of calfskin, roughly a forearm across, cracked along four folds and pinned under glass since it came to us. It carries about ninety characters in close-set rows and one continuous line running corner to corner. Our correspondent describes it as a sailing direction. He does not say to where, in the covering note, because there is no covering note; the description arrived separately, six weeks later, and says the destination is Hel.

The Bureau has no quarrel with the calfskin. What it would like to draw attention to, before anything else is said, is the shape of the drawing. There is no coastline on it. There is no island, no sounding, no compass rose, no scale, and nothing that could be called a landmark. There is a road. The road goes down, and the ninety characters tell you what is on it, in order, the way a man tells you what is on a road he has walked.

A People Who Did Not Draw the Sea

A large old parchment sea chart of northern Europe and the Atlantic, worn and folded down the centre, with a lettered cartouche in the top left corner, decorated borders and compass roses set in the corners
A chart of the northern seas: coastline, cartouche, compass rose. Everything accession 4471 does not have, and everything no Norse hand is known to have drawn.

That absence is not a defect in the document. It is the document’s one plausible feature, because the Norse did not draw the sea. Norse ships were navigated by tides, by elapsed sailing times, and by knowing what a particular headland looks like in bad light. The sagas have a word for what happened when that failed — hafvilla, bewildered — and they use it often enough to be worth noticing. There was no magnetic compass in the north in this period. There was no chart. The wooden disc from Greenland that was once read as a sun compass has slits spaced too unevenly to have functioned as an accurate one.

The famous alternative is the sunstone, and it is thinner than its reputation. Exactly one medieval text, Rauðúlfs þáttr, describes the thing in use. Thorkild Ramskou proposed in 1967 that it was birefringent Iceland spar; a stone did come up in 2002 from an Elizabethan wreck off Alderney, which sank in 1592, some six centuries after the fact. Experiments in 2011 got direction to within five degrees and then to within one; a replication attempt in 2014 got nothing, and neither party could say why. That the Viking Age steered by such a stone is filed in the literature as a hypothesis. It has never been filed as anything else.

Set against that, consider how much else survives. The Oseberg ship came out of a Norwegian mound in 1904 with her strakes, her oar ports, her carved sleighs and a four-wheeled cart, and dendrochronology puts the burial in the autumn of 834. The Gokstad ship, dug in 1880, was built from timber felled around 890 and is 23.80 metres long with thirty-two oars; a replica of her crossed from Bergen to Chicago in 1893 to prove it could be done. At L’Anse aux Meadows there are eight sod buildings, a forge, eight hundred Norse objects, and three tree-ring samples that put Norse hands on the wood in the year 1021 exactly. The Norse left us their ships, their houses, their nails and their butternuts.

They did not leave a chart. The nearest thing is the Skálholt map, which the schoolmaster Sigurd Stefánsson drew in 1570 — five and a half centuries after L’Anse aux Meadows — and even that is gone, surviving only in a copy a bishop made of it in 1690. Everything earlier is a road told out loud.

The Saga, As It Was Copied Out for Us

The ninety characters have been rendered into English verse by the correspondent, at some cost to their brevity and, one suspects, at some profit to their content. The Bureau prints his rendering as he supplied it, in nine stanzas, and reserves comment for the sections that follow. Where his account and the surviving Norse account differ, the difference is noted; where his account supplies a detail the surviving account does not, that is noted too, being the more interesting case.

They spread the hide across the thwart.
The runes came up like weed through frost.
No harbour marked, no headland named —
one road, and what the road would cost.

A worn tan hide spread across a longship's timbers among coils of rope, one unbroken straight line inked across it from the lower left to the upper right, with rows of angular runic characters above and below the line and no other marking of any kind

“It is not a chart of anywhere,” the assessor wrote. “It is a chart of one direction.”

They put out on the ebb at dusk.
No priest was asked, no bird let fly.
The fjord shut like a door behind.
One star stayed open in the sky.

Two longships under torchlight in a narrow fjord at night, one under a rune-painted sail with her crew aboard, snow-covered walls of rock closing in on both sides and a single bright light burning through the cloud overhead

“We were three days out before anyone asked what the second half of the road was for.”

Nine days together showed no sun.
The steersman raised, against the grey,
a stone gone milky as an eye,
read two lights in it, named the way.

A weathered hand held up against a flat grey sky, pinching a small clear crystal between finger and thumb, with the sail, rigging and wet hull of a longship behind it and snow in the air

“The stone is the only instrument aboard, and the only one nobody else can read.” — steersman's deposition, undated.

The Descent, Checked Against Snorri

A lone bearded traveller with a long staff walking a causeway of rune-carved stone slabs through a dead forest, the huge trunks banded with gilded knotwork and a carved pillar standing in the mist ahead
“Down, and to the north.” That is the whole of the sailing direction the Norse sources preserve. The rest of the road is the correspondent's.

Everything we securely know about the road to Hel comes from one chapter of one book written by one man more than two hundred years after the conversion. Snorri Sturluson compiled the Prose Edda in Iceland around 1220, and chapter 49 of its Gylfaginning contains the whole of it: Baldr is killed with a shaft of mistletoe, and Hermóðr rides Sleipnir nine nights through valleys so deep and dark that he saw nothing at all, to ask for him back.

At the bottom he comes to a river called Gjöll, the Noisy, and to Gjallarbrú, which Snorri says is thatched with glittering gold. It is kept by a maiden, Móðguðr — Furious Battler — who tells him that five companies of dead crossed the bridge the day before and shook it less than he does by himself, and that he is the wrong colour for a dead man. She then gives him the only sailing direction in the whole corpus: down, and to the north.

Our correspondent’s stanzas four to six run alongside that account for most of their length and then depart from it in a single, specific, and rather awkward particular.

Below the keel a grey coil ran
the length of every day they sailed.
An eye came open in the cloud
and watched them, and the daylight failed.

The coil of an immense sea serpent moving through dark water beneath a watching eye in the clouds

“It kept station under us for nine days and never once drew breath.” Marginal note in a later hand: neither, it appears, did the copyist.

The river came up loud and black
with ice and blades inside its roar.
The bridge above was thatched in gold.
A woman held the farther shore.

A narrow bridge whose deck is thatched with beaten gold, spanning black water strewn with drifting ice below sheer dark cliffs, a single pale figure standing at the far end holding a torch

“You are the wrong colour for this bridge,” said the watchwoman. She is recorded as saying it to all of them.

The chart showed neither gold nor thatch —
it drew the span in rib and bone.
Five companies had crossed by noon
and made less noise than one alone.

A bridge built of great bones spanning a chasm between dark mountains, with figures crossing it

The chart draws the span in bone. Snorri draws it in gold. Only one of the two has ever claimed to be a document.

The Gate, the Hound, and the Long Table

A vast roofless hall of knotwork-carved stone arches and pillars standing in fog, with small armed figures walking up a stony streambed toward a carved doorway at the head of a flight of steps
Éljúðnir, by report. The sources give the hall a name, a door and a table, and then stop.

Past the bridge the sources thin out to a handful of nouns. There is a gate, which Hermóðr does not open but jumps, Sleipnir clearing it without touching. There is a hall, Éljúðnir. There is a hound — Garmr, blood-stained, bound at Gnipahellir, whose howling in Völuspá is the sign that the end has started and who kills Týr at Ragnarök and dies of it. Scholars have long argued about whether he is Fenrir under another name, which tells you how much there is to go on.

And there is the condition. Hel will give Baldr up if every thing in the world, living and dead, weeps for him. Everything does, except one giantess called Þökk, sitting in a cave, who declines. That is the whole of the underworld’s administrative procedure as it has come down to us: one clause, one refusal, no appeal.

Our correspondent’s last three stanzas cover the gate, the hall and the return. The Bureau notes that the return is the portion for which the Norse sources provide no precedent whatsoever, Hermóðr having come back empty-handed and nobody else having come back at all.

The road ran down and did not climb.
A gate stood shut in the cliff’s face.
A thing got up upon its chain
and did not bark, and kept its place.

An arched gate of stone closed with an iron grille, set into a cliff at the end of a frozen road, a shaggy grey hound standing before it on a heavy chain among scattered bones, with one torch burning further down the road

“It did not bark. That was the part we did not care for.”

Within: a hall of frost and beam,
a long board set, and no one fed.
One bowl. One seat. One woman there
who did not rise, and knew them dead.

A pale woman in a spiked headdress seated on a high carved chair in a frost-covered stone hall lit by wall torches, one hand on the end of a long empty trestle table set with a single wooden bowl, mist lying across the floor

“The board is always set,” she said. “It is never cleared.” One bowl, the Bureau observes, for the hall of the entire dead.

Of those who took the road, two turned.
They cut the account in stone, and then
they reddened every stroke of it,
and never put to sea again.

A standing grey runestone in a grassy coastal field at sunset, its interlaced serpent band and rune row cut fresh and stained deep red, a bearded man kneeling beside it with a wooden mallet and iron chisel, a turf house and the sea behind

“Cut it all down, and redden it, and let it stand where the ships go by.” No such stone appears in any catalogue.

The Precedent, Which Is Recent

A hand holding a sharpened stylus over a large parchment map of coastlines, rivers and lettered place names unrolled across a wooden table, a rolled chart lying at the top edge
The parchment was fifteenth-century. The ink was not. Between those two facts sits a hand, and the Bureau has yet to identify one.

The Bureau has been here before, and so has everyone else. In 1957 a map surfaced with a book dealer, bound in with a medieval text. Paul Mellon bought it, gave it to Yale at the end of 1964, and Yale published it on 11 October 1965, the day before Columbus Day, at which point the Norse had reached America on parchment as well as in fact. The Vinland Map showed a named island out in the western ocean, and it showed it a good half-century before Columbus.

The parchment was genuine: radiocarbon put it at 1423 to 1445. The ink was not. Yale sent the map to Walter McCrone in 1972, and his team found the yellowish lines to contain anatase titanium dioxide, a pigment made for pale colours since the 1920s and before that not made at all in that form. Raman spectroscopy confirmed it in July 2002, and Yale’s own specialists called the map a modern forgery in 2018. In 2021 the Beinecke’s curator of early books dropped the hedging altogether: “The Vinland Map is a fake. There is no reasonable doubt here.” It is still in the Beinecke, where it is now an exhibit about itself.

The instructive part is not the forgery. It is the appetite. A map of Norse America was wanted very badly for a very long time, and in due course a map of Norse America turned up. The Bureau sets that beside its own accession 4471 and declines, for the moment, to draw the obvious line between them, on the grounds that the reader can draw it unassisted.

The Bureau’s Position

One last thing is out of proportion, and it is the reason this file is open at all. The Norse were not a people who failed to write things down. Some three thousand runestones survive out of roughly six thousand runic inscriptions in Scandinavia — between seventeen hundred and twenty-five hundred in Sweden, as many as twelve hundred in the province of Uppland by itself. They were cut in the sixteen characters of the Younger Futhark, mostly between 950 and 1100, and they were painted in earth colours, white and red lead dominating. They record inheritances, bridges, dead sons and the occasional debt. About one in ten announces a journey abroad, and a death on it.

Not one of them is a chart. Across three thousand stones, two intact ships, a Newfoundland settlement and a literature, the Norse never once drew a coastline. What they did instead was tell you the road: this headland, then that sound, then the current that sets north of the skerry. Accession 4471 is a road told in exactly that manner, which is either the strongest argument in its favour or an unusually well-briefed piece of work.

The Bureau wrote to the correspondent asking where the hide had been between the eleventh century and last March. The reply came promptly and consisted of a single line, which he states is the closing line of the inscription and which the Bureau has been unable to match to any of the ninety characters on the sheet: it was kept by the one who came back second, and he was not asked either.

Plates: the Remainder of the File

Twenty-five sheets that answer to no stanza and no paragraph. The Bureau prints them in the order they were received and attaches no weight to any of them.

A longship under a painted sail running before a heavy green swell with more ships astern between steep cliffs, and a helmeted man watching from the rocks in the foreground
A crowded longship under a sail painted with a red device, four smaller ships following her through a running sea, sunlight breaking through cloud
Three long-haired Norsemen leaning over a great rune-drawn chart spread flat across a table by candlelight in a timber hall
A large parchment chart unrolled on a plank table under a single candle, ringed devices and knotwork drawn across it, weights and a stylus at its edges
A bearded man in furs bent over a huge open book on a stone table, its pages lit from within, torchlit ships and mountains behind him
A long stone bridge on timber piers crossing a mountain gorge toward a torchlit gatehouse, with a weathered skull-shaped crag overhanging the right of the frame
A stone bridge with rope handrails crossing a valley between jagged peaks, several small figures walking it and braziers burning on the far side
An arched stone bridge over a river between mountains, armed figures crossing it, torches on the far bank and a great skull lying among the rocks
A parchment sheet on a table drawn with a knotwork serpent coiled in concentric rings, a rolled scroll, a candle and brass vessels set around it
An open vellum book drawn across both leaves with two interlaced knotwork serpents and runic roundels, no coastline or scale anywhere on it
A bearded warrior in a horned helmet standing with a long axe in a frozen ravine of ice and black rock under moonlit peaks
A white-haired man in a dark cloak holding a stylus up to a huge curling chart propped on a frame in a moonlit rocky place
Two warriors in horned helmets crouched over rune-marked boards on a cave floor with a small fire between them and a skull at the left
A wide frozen valley of black rock and meltwater with a handful of small figures walking away toward ice-blue falls
A man in a horned helmet with a long axe leading a file of shield-bearing companions across a snow-covered plain of black rock
A warrior smoothing a large chart flat across a stone table while others stand behind him with torches, candles and runic caskets at the edge of the frame
A giant horned figure with a burning axe striding out of a fire-lit gorge, a smaller horned warrior at its feet and a flaming rune-disc in the air
A colossal armoured giant with molten cracks across its body roaring above a single small swordsman on a burning plain
Two warriors standing before a huge curling chart that is burning along one edge in a dark cavern
A winged horned figure with a great axe standing on a field of lava and broken rock
A circular table map of pale land in blue water inside a runic border, a small lettered placard laid on it, candles and a scroll alongside
A lone warrior in a horned helmet seen from behind with his sword lowered, standing at the edge of black water in a frozen basin
A huge shaggy horned beast standing in fog before a single armed man with a shield, torch posts burning behind them
A colossal horned figure looming beyond a carved gate at the end of a ruined colonnade, with armed figures standing before it
A longship under a striped sail in heavy night mist between steep fjord walls, her crew bent over a chart by torchlight and one star showing above