The daughters the runestones forgot
Only 14 Swedish runestones were raised in memory of a daughter, against 268 for a son, and among the siblings who raise the stone for their father only one in twelve is a daughter. But where the daughters appear, they appear in their own way: the stones for them are late, and in Uppland more than one in three is a bridge. We think the daughter appears when the inheritance passed through her – and that her inheritance became a bridge for her soul.
Gullaug had the bridge made for the soul of Gillaug, her daughter, and Ulfr was married to her. Œpir carved.
So it reads on a runestone at Morby in Lagga parish in Uppland. Gillaug was grown up and married when she died, and it was her mother who built the bridge for her. It is one of very few runestones that remember a daughter. Most daughters of the Viking Age never got a stone, and most are not even named on the stone for their father. This article is about them – about why they are so few, and about what the few stones for them have in common.
The missing daughters
We read by hand all Swedish Viking-Age runic inscriptions where the Rundata translation speaks of a daughter or a maid – some fifty. Eighteen of them are memorial inscriptions for a daughter. Fourteen stand on runestones, and those are the ones we count. The other four are carved on a rock face (U 311), an earth-fast boulder (Sö 331) and grave slabs (Ög 239, Sm 94). 268 runestones were raised for a son.
The scale runs from 0 to 100 %. Source: Vitki's analysis of the Scandinavian Runic-text Data Base; Swedish Viking-Age runestones. The stones for daughters were read by hand.
Daughters were born as often as sons, and died as easily. The stones themselves tell us so. On the Hillersjö rock in Uppland (U 29), the longest family story on any runestone, it says of Geirlaug and her second husband: Then they got children. But one maid lived; she was called Inga. Of the children only one daughter survived. Yet there are nineteen times as many stones for sons as for daughters.
The daughters are missing among the living too. The 403 stones for a father where the children's names can be read name 846 children. 775 of them are sons and 71 daughters.
The scale runs from 0 to 100 %. Source: as above; the 403 stones in memory of a father where the children's names can be read. Gender from the names.
If all the children had been named, about half would be daughters. As it is, one in twelve is. Most daughters are thus silent twice over: they seldom got a stone of their own, and they are seldom named on the stone for their father.
Last in line
When daughters are named among the siblings, they usually come after their brothers. Of the 35 sibling lists that include both sons and daughters, the daughters come last in 25. If the order were random they would come last in about 14. A pattern this clear arises by chance in about one case in ten thousand. But in one half of the material the pattern is weaker (daughters last in 10 of 17 lists, against 7 expected) than in the other (15 of 18, against 7), so we do not count it as a finding. It looks as if the sons were counted first.
Sometimes it is the daughters who raise the stone – and then mostly when there is no son. At Ramsta in Södermanland (Sö 60), Auða and Inga and Erindís as the third, the mother and daughters, raised the stone for their father Sveinn and their brother Guðfastr. The brother was dead. On a stone in Linköping (Ög Fv1958;252) Sigríðr the maiden raises a stone for her father. And on Gotland, at Sjonhem (G 136), it says of a man who died at home: he had one daughter. She is called Heilvé. On 24 of the 59 stones for a father where a daughter is named, she stands without brothers.
The daughter also appears as someone's daughter. Sigríðr, Alríkr's mother, Ormr's daughter made the bridge at the Ramsund rock (Sö 101), the one with Sigurd the dragon-slayer.
Sometimes that is all we learn. At Smedby in Hammarby (U 280) Vébjôrn had the stone cut in memory of Gunni's daughter, his wife. Her own name is not on the stone.
The stones for daughters
The fourteen runestones for daughters have a few traits in common. The first is who raised them. On seven of the fourteen the mother takes part. That agrees with what holds for children in general: in Svealand the mother takes part in half of the stones for a child, against not quite one in four of the other stones – a pattern that holds in both halves of the material. Mothers mourned their daughters as they mourned their sons.
The second is time. All nine runestones for daughters in Uppland that can be dated by their style were carved after about 1050. Of all datable Uppland runestones 69 per cent are from that time, so all nine being late by chance happens in about four cases in a hundred. It is a hint, not a finding.
The third is the bridges. In Uppland four of the eleven runestones for daughters build a bridge – at Velamby (U 377), Morby (U 489), Norsta (U 861) and Gryta (U 867). Of the runestones for sons only one in fourteen does. In addition there is the carving in the rock face at Hargs bro (U 311), which also concerns a bridge for a daughter. We do not count it, since the comparison is between runestones.
The scale runs from 0 to 40 %. Source: as above; Uppland runestones whose text can be read.
A difference this large arises by chance in about one case in a hundred. In the eleventh century a bridge was a Christian deed for the soul of the dead, and several of the daughters' stones say so outright: the bridge at Morby was built for the soul of Gillaug.
Harg – four memorials for daughters
At Harg in Skånela parish there are four memorials for daughters: three runestones and a carving in the rock face by the bridge. In the rock Ingríðr had the bridge made in memory of Ingigerðr, her daughter, and her sons Ingivaldr, Ingimarr and Karl had it made in memory of their sister (U 311). And there stand three stones for one and the same girl, Jôfurfast. On the first, Gunnarr, Farulfr's son, had the landmark made for the soul of Jôfurfast, his stepdaughter, Holmdís' daughter (U 312). On the second Gunnarr and the mother Holmdís stand together (U 313), on the third Holmdís alone (U 314). No other daughter on the Swedish runestones got three stones.
Jôfurfast was Holmdís' daughter from an earlier marriage. That her stepfather too raised stones for her says something about what was at stake. At the same farm Gunnarr and Holmdís raised a stone for their own son Þórðr (U 315), and a short inscription nearby says: Þórðr owns the bridge (U 316). The Harg stones read most easily as a family's account of who owned what after the children had died – in the same way as the Hillersjö rock accounts for how the inheritance passed from Inga to her mother.
Our hypothesis: the daughter's inheritance became a bridge
Birgit Sawyer has shown that the runestones are often about inheritance: whoever raised the stone showed that she or he was an heir (Sawyer 2000). The daughters fit that picture. They appear when the inheritance passed through them – like Inga on the Hillersjö rock, like Heilvé at Sjonhem and like the daughters at Ramsta, whose brother was dead. And when they stand with brothers they stand after them, in the same order as in the medieval laws, where sons inherit before daughters and a daughter later received half of a son's share.
What is new is the bridges. We have not seen it described that the stones for daughters are so often bridges. Our hypothesis is that it has to do with inheritance. A son's inheritance stayed in the kin. A daughter's inheritance followed her when she married, and when she died with property of her own – like the married Gillaug at Morby, or Jôfurfast, who had probably inherited from her father – the family could give her share to her soul instead. The bridge was the soul-gift of the time.
A first test
If the bridges merely reflect that the daughters' stones are late – bridges became common just after 1050 – the difference should vanish when we compare late stones only. It does not. Among the late stones for sons 7 of 79 build a bridge, that is 9 per cent, against four of eleven for daughters. A difference this large arises by chance in about three cases in a hundred.
Nor is it only the mothers' custom. Women built bridges somewhat more often than men, but of the four bridges for daughters two were built by fathers: Sigþorn at Norsta and Þjalfi at Gryta. And outside Sweden we find the same thing. One of the very few Norwegian stones for a daughter, the Dynna stone in Hadeland, is also a bridge stone: Gunnvôr, Þryðríkr's daughter, made the bridge in memory of her daughter Ástríðr. She was the handiest maiden in Hadeland.
But the pattern does not hold in both halves, and so we state it as a hypothesis to be tested:
The daughter appears on the runestone when the inheritance passed through her. When she died with property of her own, her share became a bridge for her soul.
How the hypothesis can be tested further
- More stones. The hypothesis says that stones for daughters are more often bridges. With the Danish and Norwegian stones, and with new finds, the material grows.
- The laws. The medieval provincial laws say how much of an inheritance could be given to the soul, and how daughters inherited. If the hypothesis is right, the daughters' bridges should be found where the laws gave daughters property of their own.
- The families. Where the same family appears on several stones, as at Harg and Hillersjö, one can follow who inherited from whom, and whether the bridges were built when a daughter's inheritance fell free.
- The husbands. If the bridge was the daughter's own inheritance, stones for married daughters should more often be bridges than stones for unmarried ones. Gillaug was married; Jôfurfast, who got three stones but no bridge, probably was not.
Caveats
- Fourteen stones are few. A single stone more or less moves the shares by several percentage points.
- Those who raised the stones chose what was written. That a daughter is not named does not mean she did not exist, only that she was not counted in.
- The stones are dated by their style, and the dating is approximate.
- Some bridges are not mentioned in the text; a stone may stand at a ford without saying so. The shares concern what is written.
- A correlation is not a cause. The hypothesis explains the pattern, but other explanations are possible – for instance that the stones for daughters were commissioned by rich families who built bridges more often than others.
See for yourself
On the women on the runestones: She inherited from her child – the women on the runestones. On the conversion and the prayer: The prayer fell silent – but the cross remained.
The maps, curves and models come from Vitki, an open tool for runestone research that you can download and run on your own computer from GitHub. The method is described in METHODS.md.
Sources and data
- Scandinavian Runic-text Data Base (Rundata), Department of Scandinavian Languages, Uppsala University. Open Database License.
- Sawyer, B. 2000. The Viking-Age Rune-Stones: Custom and Commemoration in Early Medieval Scandinavia. Oxford: Oxford University Press.
- Gräslund, A.-S. 2006. Dating the Swedish Viking-Age rune stones on stylistic grounds. In: M. Stoklund et al. (eds), Runes and their Secrets: Studies in Runology. Copenhagen: Museum Tusculanum Press, pp. 117–139.
- Wessén, E. & Jansson, S. B. F. 1940–1958. Upplands runinskrifter (Sveriges runinskrifter 6–9). Stockholm: Royal Swedish Academy of Letters, History and Antiquities.
- Photos: Wikimedia Commons, with photographer and licence in the captions.