The old lands: Sotholm – the runestones that did not pray

Imagine coming to Tören from the sea, a thousand years ago. It is not a peninsula yet. The water runs all the way round the land, and the strait at Tälje in the west is still open. From the boat you see the rock first: bare, ice-smoothed slabs rising out of the water, pines holding on in the cracks, and behind them the forest climbing towards Hanveden, the high middle of the island. In there lie bogs and dark forest lakes between the ridges, and people are few and far between. Even in the Middle Ages that forest was called desolate.

But follow one of the long fissure valleys inland, and the forest opens. In the valley floor lies clay, and on the clay lie the farms, each in its own clearing, with names ending in -sta and -by. In the smaller valleys the cattle graze; those places will be called -sätra and -boda. Imagine the smoke from the houses, the smell of dung and hay, the dogs barking when someone comes along the path. Above the farms, on the slopes, lie the grave fields where the old ones rest under round mounds. At Kumla, on a point in Lake Drevviken, lie some seventy mounds, the largest sixteen metres across, plain to see for anyone rowing past. In a mound at Vendelsö the claws of a bear have been found – perhaps a bearskin that went with the dead on the pyre.

The roads run along the gravel ridges, dry paths through the wetland, and Drevviken ties the settlements together summer and winter: by boat when the lake is open, across the ice when it has frozen. This is no poor island. At Norrby a sword has ended up in the muddy shore, a Frankish sword with the smith's name, Ingelri, inlaid in the blade. On the same farm's land someone has buried their silver: nearly two hundred coins and coin fragments, German and English, struck for kings such as Æthelred and Cnut the Great, and as the youngest a coin of Harthacnut, struck in Lund between 1035 and 1042. That silver has seen things on the other side of the North Sea.

And now they are raising stones. Somewhere in the district iron rings against stone. On the crest of an esker, where four roads meet, stands a new stone with a serpent winding round its edge and a cross at the top. Anyone coming up from the Haninge countryside meets it first. The runes read:

Anundr(?)/Øyndr(?) ... ...undr(?) ræistu [s]tæin þenna at Viking, faður sinn, sun Stora.

Anund (?) and … raised this stone in memory of Viking, their father, Store's son.

That is all. The cross speaks, but the words do not pray. There is no Guð hialpi and hans, "may God help his soul", with which so many stones around Lake Mälaren end. And so it is with almost every stone on this side of the island.

Enlarge ⤢

Two districts on one island

Tören was divided into two districts, called Upper and Outer Tör in the medieval charters: the hundred of Svartlösa in the west and north, and the hundred of Sotholm in the east and south. No one needed to draw the border between them. It followed the high, desolate Hanveden and its spurs, ran through Kaggfjärden in the south-west and through Drevviken to Erstaviken in the north-east. Outer Tör faced outwards, towards the sea. At the end of the Middle Ages the hundred gathered for its assembly on a high islet at the inner end of Landfjärden, on the land of Häringe, and it is that place, Sotholm, that gave the hundred its name. Whether the assembly met there already in the Viking Age we do not know.

Enlarge ⤢

It was Outer Tör that raised the most runestones on the island: Sorunda alone has 21 inscriptions, seven of them cut straight into the bedrock. Furthest south, at Vansta in Ösmo, the carver Amunde cut one of the Ingvar stones, in memory of Tosti who died in Ingvar's retinue in the east. It is the people here – Store and his son Viking, Hjälmvid, Farbjörn, Hallsten – and their stones that the rest of this text is about.

The stones that did not pray

Walk among the stones of Outer Tör, and much is familiar from the rest of the Mälaren valley. Someone has had a stone raised for a father or a son, the carver has cut a serpent round the edge, and at the top there is often a cross – on more than half of the stones, as often as in the rest of Södermanland. The people here had chosen their side. Roger Wikell, who has studied the settlements around Lake Drevviken, calls the runestones "offensive Christian markers" and notes that Amunde's stones "often [have] large crosses". A farm without stones, Gudö, he describes as "a heathen glade in a forest of Christian runestones".

But read the words. In the rest of Södermanland about one stone in five ends with a prayer: Guð hialpi and hans, may God help his soul. Here almost none does. Of the forty-three stones in the hundred that can be read, a single one prays. It is not because the stones are older than their neighbours' – they were raised in the same decades. In the years early in the eleventh century when most of them were carved, about one stone in four in the rest of the province prayed. Here none.

The image preached. The words did not pray.

It has been seen before

We were not the first to notice. When Kaj Janzon and Sigurd Rahmqvist described Tören in Det medeltida Sverige in 2002, they wrote that "the runestones of Södertörn have very stereotyped texts" and that "a prayer for the soul occurs in only two cases". What we have been able to do is count again and see that the difference is too large to be a coincidence.

And it probably holds for the whole island. In Upper Tör, on the other side of Hanveden, hardly any stone prays either: two of twenty-two. Nor do the stones praise the dead – "a good farmer", "a capable man" stands on one stone in five in the rest of Södermanland but on almost none here. The stones of Tören say who raised the stone and for whom, and then they fall silent.

One small word did set the island's two halves apart. In Upper Tör people often wrote the word for 'this' in a form of their own, þennsa, as they did elsewhere south of Lake Mälaren. In Outer Tör almost never.

The district's carvers

Most of the stones here with a known carver were made by three men: Amunde, Halvdan and a carver usually called Björn. Two of them put their names on the stones:

Hialmviðr ræisti stæin at Þorgaut, sun sinn. Amundi risti runaR.

Hjälmvid raised the stone in memory of Torgöt, his son. Amunde carved the runes.

Farbiorn/Frøybiorn let haggva stæin at Haulf, sun si[nn]. Halfdan hiogg runaR.

Farbjörn had the stone cut in memory of Håulv, his son. Halvdan cut the runes.

None of them ever wrote a prayer, not even on the stones they cut outside the hundred. So was it the carvers and not the district? The answer is not so simple. Among the stones that other hands cut in Sotholm there is also only a single prayer. The district had its carvers, and the carvers had the district's way of writing. They cannot be told apart, and perhaps they are the same thing: in Outer Tör the stone got the cross but not the prayer, and so it stayed for several decades.

The exception that came out of the ground

The only stone with a prayer in Sotholm was found in 2011, at Småhamra in Österhaninge:

Gunnarr ok UlfR ok SighialmR letu ræisa stæin at Halfdan, faður sinn. Guð hialpi and hans.

Gunnar and Ulv and Sighjälm had the stone raised in memory of Halvdan, their father. May God help his soul.

No one knows who carved it. That the exception turned up so late is a reminder that the ground still hides stones, and that the picture can change with the next find.

A land of its own

The sea around the island and the forest in its middle drew the borders. The people of Outer Tör had their carvers and their own way of remembering the dead: the cross on the stone, the names of those who raised it and the name of the dead – but no prayer. We do not know why. It may have been the carvers' school, where the formula was never taught. It may have been the families, who wanted the cross but not the prayer. It may have been a priest or a missionary who taught differently just here. But whoever stands by Sö 269 stands by a stone that left the prayer out, in a district that was a land of its own.

How we know

Behind the text lies a comparison of every hundred with at least fifteen runestones, in which Sotholm was one of the few that differed clearly from its province. The difference in prayers holds when the stones are split into two groups parish by parish, when only stones in their original place are counted, and when the stones of the main carvers are removed. What we cannot decide is whether the habit belonged to the district or to the carvers. That it held for the whole of Tören is likely but not certain, since the stones of Upper Tör are too few. The plans and calculations are open on GitHub, linked below.

Sources

• Janzon, Kaj & Rahmqvist, Sigurd 2002. Det medeltida Sverige 2. Södermanland 1. Tören: Svartlösa, Sotholm. Swedish National Heritage Board. (Quotations translated by us.)
• Wikell, Roger 2011. Att resa runsten eller inte – det är frågan. Tankar kring två vikingamiljöer i norra Haninge. Situne Dei 2011, pp. 98–105. Sigtuna Museum. (Quotations translated by us.)
• 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.
• The locked analysis plan, the stones and Vitki's calculations are open on GitHub (analysplaner/2026-11-02-sotholm.md and 2026-11-02-gamla-granser-nya-monster.md).
• Geological Survey of Sweden (SGU), shore-displacement model for AD 1000 (the map).
• Image: Wikimedia Commons, with the photographer in the caption.