The old lands: Roden – the coast that counted in ships

The coast of Uppland, a thousand years ago, does not look as it later will. The sea reaches far into the land, the bays cut deep between the hills, and what will later become river valleys are still inlets where a boat can glide in among the alders. Everywhere there are points and narrow necks of land, islets and skerries. Furthest north the land has only just risen from the sea; there the settlement is young, the farms new, and the shore meadows grow outwards with every generation. Where the Dalälven river falls into the sea at Älvkarleby, people catch salmon – when the charters begin to speak, in the 1180s, the pope protects the cathedral's fisheries there and the king gives part of the fishery to a convent.

Imagine the smell of seaweed and tar, the sound of oars against the thole-pins, the gulls over the shallows. In the south, in the archipelago of Värmdö, the sea road runs between the islets; still in the medieval Danish sailing description the routes end at Runmarö and Vindö. This is Roden, the coastal land. Here the water is the road.

The farm by the bay

The farms lie where a bay meets a valley, and their names speak of the water. Farms with names in -vik, 'bay', are more than three times as common here as in the rest of Uppland, and -näs, 'point', more than twice as common, while the inland -sta and -by are half as common or less. Along the shores the surveyors have found four times as many boat landings and harbour remains, counted against everything else, as in the rest of Uppland. They come from all periods, but they show where the boats were hauled up. Imagine a boat hauled up below every farm, as much a matter of course as the cow in the byre.

The dead are buried on the slopes above the shores. On the grave fields here, mounds outnumber the low stone settings more often than in the folklands inland, and the standing stones are fewer. It is not just different ways of surveying: the differences remain when we compare only with neighbours in the same county.

Furthest north, in Hökhuvud, the land is hilly, with large stretches of pine forest between the farms. The settlement there was fully cleared only during the Viking Age and the early Middle Ages, and the grave fields lie beside most of the villages. The roads between the villages are so old that the ancient monuments follow them. By one of them, beside a small grave field south of Hanunda, stands a stone that three brothers had raised for their father. Besides the runes it carries a rider and a bird:

Arnniutr ok Vigniutr ok Signiutr þæiR letu ræisa stæin þennsa æft[i]R ..., faður sinn. Þorfastr risti runaR.

Arnnjut and Vignjut and Signjut had this stone raised in memory of …, their father. Torfast carved the runes.

Later in the Middle Ages a small trading place grows up by a bay in the north, Hamar, where fishermen and farmers trade goods with one another. It is easy to imagine the same exchange along the whole coast long before: what the soil did not give, the sea did.

A land of ship districts

The three folklands inland – Tiundaland, Attundaland and Fjädrundaland – were divided into hundreds. Roden was divided into ship districts, skeppslag, and it shows in the law. When the Law of Uppland was written down around 1300, a dispute in the folklands went from the hundred's assembly to the folkland's assembly and finally to the king; in Roden it went from the assembly of one ship district to that of two, then to that of six, and finally to king or duke. The levy for the king's fleet, too, was reckoned in Roden by ship districts and not by hundreds. The coast counted in ships.

How old that order is we do not know. The historian Mats G. Larsson argues that the ship districts in Roden are "in most cases probably a younger phenomenon" – the Roden of the law is the Middle Ages', and the land where the stones were raised may have been organised differently. Nor is it certain how large Roden was, or whether it was a land of its own or part of the folklands. Göran Dahlbäck and Gunnar T. Westin write in Det medeltida Sverige that "the question of Roden's territorial extent and its fiscal and judicial division is disputed". In the six-year tithe of 1314 Roden appears as a heading under Tiundaland.

Enlarge ⤢

Jarlabanke's Täby

Further south, where the coast breaks up towards the mouth of Lake Mälaren, lies Täby. A great man lives here, Jarlabanke, who has stones raised at a bridge he has built and has it written that he alone owns the whole village:

Iarlabanki let ræisa stæina þessa at sik kvikvan, ok bro þessa gærði fyr and sina, ok æinn atti allan Tæby. Guð hialpi and hans.

Jarlabanke had these stones raised in memory of himself while alive, and made this bridge for his soul, and he alone owned all of Täby. May God help his soul.

Täby was counted in the ship district of Danderyd. But on a stone at Vallentuna church, in Attundaland, Jarlabanke writes that he "made this assembly place" and "alone owned all of this hundred". The border between the coastal land and the folkland ran through the middle of his world. An assembly place was something one made, and at Bällsta in Vallentuna, too, Ulf's sons wrote on a stone that they "made here an assembly place".

In Jarlabanke's district women raise stones for those who never came home. Guðlaug for her son Holme, who died in Lombardy. Åstrid for her husband Östen:

Æstriðr let ræisa stæina þessa at Øystæin, bonda sinn, es sotti IorsaliR ok ændaðis upp i Grikkium.

Åstrid had these stones raised in memory of Östen, her husband, who went to Jerusalem and died up in Greece.

Enlarge ⤢

Stones that tell of death far away – in Lombardy, in Greece, in the east – are more common in Roden than in the rest of Uppland. But nearly all of them stand in Täby. That is the history of one district: a family and its neighbours who went out into the world and raised stones when they did not come home.

Fewer crosses

Walk among Roden's stones, and after a while one thing stands out. The cross, which in the folklands inland sits on six stones in ten, here sits on only four. Prayers for the dead are rarer still: in the rest of Uppland about one stone in five prays, here hardly one in twelve. It is not down to particular carvers – even when Fot and Öpir, Roden's most prolific, are left out, the crosses are fewer.

The natural thought is that it is the sea, that people by the shore lived differently from people inland, whatever their land was called. Roden's stones do stand closer to the water: half of them within a hundred metres or so of the shoreline of AD 1000. But in the rest of Uppland it makes no difference whether a stone stands by the water or far from it – the cross is just as common. And when we take the closeness of the water into account, the difference for Roden remains, as large as before. It is the land, not the shore.

A land that faced the sea

Why did Roden's stones carry fewer crosses? We do not know. But there is a clue. Roden's stones seldom stand near the old central places, Husby and Tuna, where royal power had its estates – only one in five, against half in the folklands. Perhaps the cross reached Uppland along the roads that ran through the king's estates, and the coastal land, which faced the sea and was governed in its own way, took it up more slowly. It is a thought, not a result.

Whoever rows past Roden's shores sees something other than differences between stones, though: the boats hauled up on land, the mounds on the slope above, the smoke from the farms at the far end of the bays. A land where the water was the road and where people counted in ships.

This is the second part of the series The old lands. The first took us to Sotholm on the island of Tören, where the settlements lay in forest clearings and the stones carried crosses but did not pray.

How we know

The landscape and the life along the coast rest on Det medeltida Sverige, the county board's descriptions of the districts and the register of ancient monuments; the names and the graves we have counted from the register and the maps, and the boat landings may be from any period. The ship districts and the trading place are known from the Middle Ages, not from the Viking Age. The crosses we have measured: Roden's runestones were compared with the rest of Uppland's, with time held constant, and the test that the closeness of the sea does not explain the difference was fixed before we counted. The difference holds when the stones are split into two groups parish by parish and when the stones of the main carvers are removed, but not when only stones in their original place are counted, and a large share of Roden's stones stand in Täby, whose place in Roden is itself disputed. The plans and calculations are open on GitHub, linked below.

Sources

• Dahlbäck, Göran, Jansson, Bengt & Westin, Gunnar T. 1972. Det medeltida Sverige 1. Uppland 1. Norra Roden. Royal Swedish Academy of Letters, History and Antiquities. (Quotations translated by us.)
• Janzon, Kaj & Rahmqvist, Sigurd 2002. Det medeltida Sverige 2. Södermanland 1. Tören: Svartlösa, Sotholm. Swedish National Heritage Board (the Danish sailing description).
• Larsson, Mats G. 1993. Kung Eriks nyordning. Uppland 1993, pp. 105–134. (Quotation translated by us.)
• County Administrative Board of Uppsala 2018. Areas of national interest for cultural heritage: Bygden norr om Hökhuvuds kyrka (C18), Östhammars stad (C19).
• Swedish National Heritage Board, register of ancient monuments (Kulturmiljöregistret), extract 2026.
• Geological Survey of Sweden (SGU), shore-displacement model for AD 1000.
• Scandinavian Runic-text Data Base (Rundata), Department of Scandinavian Languages, Uppsala University. Open Database License.
• Institute for Language and Folklore, place-name register (PlaceNameService), parishes and hundreds. CC BY. Farm names from OpenStreetMap (ODbL).
• The locked analysis plans, the data and Vitki's calculations are open on GitHub (analysplaner/2026-11-02-roden.md and 2026-11-02-landprofiler.md).
• Image: Wikimedia Commons, with the photographer in the caption.