The Sagas of Icelanders or Íslendinga sögur are a set of stories about the struggles and conflicts among the early settlers of Iceland in the 9th, 10th and 11th centuries. They are thought to have been written in the 13th and 14th centuries by authors whose names have long since been lost. But many academics consider the texts to be gems of world literature .

An interesting question is to what extent the stories represent real events. The sagas certainly describe many events in clear and plausible detail so it is by no means far-fetched that they are reasonably accurate histories. Nevertheless, the stories may be fiction framed in a way that appears historical to the modern eye.

Today, Pádraig Mac Carron and Ralph Kenna at Coventry University in the UK throw some light on this question. These guys have analysed the Sagas of Icelanders using the same approach that computer scientists use to analyse today’s social networks.

And their conclusion is fascinating. They say that the characters in the sagas form a social network that is indistinguishable from those that form on real social networks such as Facebook and Twitter.

(Continued below…)



Network theory is a remarkable new addition to the toolbox that academics use to study literature. In recent years, computer scientists have pieced together the network of links between characters in historic texts such as the ancient Greek Iliad and Odyssey, the Anglo-Saxon Beowulf and the Irish Táin Bó Cuailnge.

In many of these texts, the characters form social networks that are remarkably similar to the ones found in real life on Facebook, Twitter and other social networks. The networks in the Iliad and Odyssey are particularly realistic.

Inspired by this work, Mac Carron and Kenna have taken the same approach with the Sagas of Icelanders, which consist of 18 tales involving more than 1500 individuals.

Each of these individuals forms a node on the network and two nodes are linked if they have a connection in the story. Mac Carron and Kenna also note whether this connection is friendly or hostile.

The resulting network is remarkably familiar. It has the famous small world character meaning that distant individuals can be linked in a much smaller number of steps than expected if the network were random.

It is structurally balanced, meaning that the pattern of hostility and friendship between characters is similar to real-world networks. In technical terms, this means that any three individuals tend not to have an odd number of hostile links. In practice, this is a reflection of the age-old notion that the enemy of an enemy is a friend.

And the sagas network also contains well-defined sub-communities, another common feature of real social networks.

Together, these features make the sagas network look remarkably realistic. “Network analysis indicates that the Íslendinga sögur comprise a highly interlinked set of narratives, the structural properties of which are not immediately distinguishable to those of real social networks,” say Mac Carron and Kenna.

Of course, this is by no means proof that the stories are based on real historical events. But the fact that the network so closely mirrors the ones we see in society today is an important discovery. As Mac Carron and Kenna conclude: “Although one cannot conclusively determine whether the saga societies are real, on the basis of network theory, we can conclude that they are realistic.”

Indeed, medieval Icelanders would not think their tales of blood feud and conflict very different from those that appear today on the likes of Facebook and Twitter.

Ref: arxiv.org/abs/1309.6134: Network Analysis of the Íslendinga sögur — the Sagas of Icelanders