Contact FutureLearn for Support
Skip main navigation
We use cookies to give you a better experience, if that’s ok you can close this message and carry on browsing. For more info read our cookies policy.
We use cookies to give you a better experience. Carry on browsing if you're happy with this, or read our cookies policy for more information.

How knowledge of networks can shed light on old questions

In the following five Steps, we look at how researchers have used network analytics in their work, to build models of how communities behave, explore social theories such as Homophily, calculate higher level properties like Trust, and even to use social networks as predictors of events in the real world.

We start with looking at how knowledge of networks can shed light on old questions. Dr David Millard introduced this topic in the video in the Step ‘What are Networks’. This Step provides further background information and highlights the key network properties that the researchers used in their work.

Universal properties of Mythological Networks

“Universal properties of Mythological Networks” by MacCarron & Kenna1 (2012) describes the development of a quantitative method for the analysis of mythological epics through the application of network theory - they use many of the same properties that have been introduced in this course. By analysing the extent to which the network properties of the epics Iliad, Beowulf and Táin Bó Cuailnge (T’ain) follow real world social network properties the objective was to give a measure of historicity (the state or fact of being historically authentic) for use in comparative mythology.

To enable analysis of possible network features of the narratives, two types of relationship in the networks were identified. A ‘friendly’ relationship was said to exist if the characters knew, were related to, had spoken to or met each other and if conflict existed between the characters, the relationship was described as ‘hostile’.

To further aid analysis, a weighting was assigned to a relationship based upon the number of encounters between the characters. Regarding each of the characters in the networks as a node and the relations between them as links or edges, a number of statistical tools describing network features were utilised to afford comparison to real world social networks.

Through statistical analysis of the mythological network features, a similarity between real world social networks and each narrative’s ‘friendly’ network was suggested to differing degrees.

Network features

Connectivity: each of the narratives were found to exhibit small world like properties with mean path lengths and clustering coefficients similar to those of random networks with the same average degree and size. The hostile networks exhibited much smaller mean path lengths and limited clustering when compared with appropriate random networks.

Degree: for each of narratives the ‘friendly’ network degree distribution followed that found in a real world social network. Targeting the most connected 5% of the characters i.e. removing them from the network, the networks became disconnected whilst random removal of characters had little effect upon the network.

Assortativity: MacCarron and Kenna define assortivity as the tendency for similar nodes to be connected (as in the maxim “birds of a feather flock together”).

Of the three epics the Iliad was found to be assortative with both Beowulf and the T’ain appearing disassortative to differing degrees.

The Iliad was found to exhibit properties closest to those of a real world network. The mild disassortiveness of the Beowulf narrative appeared to be overcome with removal of the main character from the network. Though the original network is unrealistic according to measure of assortativity with the main character removed the remaining network exhibits a greater degree of assortativity which the authors interpret this as supporting claims for historicity. By extending the premise of character removal it was found that the T’ain network exhibited assortiveness by removal of the six most connected characters.


In short, using methods from network science the authors compared network properties exhibited by the networks of mythological epics.

The findings indicate that, due to similarities with real world social network properties, each of the myths may be able to lay claim to a degree of historicity. Addressing issues of assortivity by removal of particular characters, it is suggested that each mythological epic may reflect networks which existed in real societies.

What do you think are the limits of this approach? If the network within a text has properties similar to a real world network do you think that is strong evidence of historicity (i.e. that it is describing a network that really did exist)? And what do these limits tell us about using analysis techniques on current digital social networks?

  1. MacCarron, P. and Kenna, R. Universal properties of mythological networks. Eprint arXiv: 1205.4324 2012 [online]. Available at: (accessed 20th April 2016) 

Share this article:

This article is from the free online course:

The Power of Social Media

University of Southampton

Course highlights Get a taste of this course before you join:

  • Welcome to the course
    Welcome to the course

    This course explains how by understanding networks better we can exploit social media. Watch Dr David Millard & Dr Lisa Harris tell you more.

  • What are networks?
    What are networks?

    Watch Dr David Millard explore what we mean by a network, and look at some of the examples of networks around us and their characteristics.

  • Recruitment in a digital world
    Recruitment in a digital world

    Watch Dr Lisa Harris, Nic Fair and Sarah Hewitt discuss the role of social media in recruitment from the perspective of employers & potential employee