TL/DR: I published an article in the Journal of Cyber Policy explaining how you can visualise cyber diplomacy voting records at the UN with a network model. It applies a community-detection algorithm to group countries that have voted in a similar way and discusses which of these communities might form the ‘middle ground countries’ in cyber diplomacy. The paper is behind a paywall but you can get a free copy by clicking the link below.
A new way to visualise UN voting records
The dynamics of cyber diplomacy negotiations and voting records at the UN are complex and dynamic. And with each vote it has become harder and less useful to display voting records with colour coded maps. There are simply too many different voting permutations to record.
In an effort to help diplomats and stakeholders follow and engage in the UN process I proposed a new approach to visualising voting records in my 2021 post: Cyber diplomacy: a new way to visualise UN voting records.
The solution I presented was to display UN members as a network where the position of each country was determined by similarity of voting history. The more similar any two countries voting records were on cyber diplomacy issues the closer they would be to each other in the network. And conversely, the more times that countries voted differently from each other the farther away they would be in the network.
At the end of 2024, the Journal of Cyber Policy published Network Modelling As A Tool For Cyber Diplomacy explaining the approach I first shared in this blog. The paper describes the methodology behind the modelling and develops it one step further by using community detection algorithms to identify groups of like-voting countries within the network.
Who are the middle ground countries in cyber diplomacy?
Of particular interest when detecting communities within the cyber diplomacy network is the question: which are the middle ground countries? My article explores that question using an algorithm and some simpler vote comparison methods.
The community detection algorithm identified seven communities in the cyber diplomacy network. Two of these were the communities at either end of the network that voted either identically to the United States or to Russia. Between these two poles were four communities which are strong candidates to be described as the “middle ground”. A final community detected by the algorithm was a mixed one sitting at the border of the middle ground and the group that voted consistently with Russia.
Why four communities in the middle ground and not one?
A community-detection algorithm can helpfully reveal the complexity and diversity of the middle ground countries. Within the middle ground are both countries that vote regularly (but in ways that keep them ‘on the fence’) and countries that rarely vote. There are also countries that have moved between these two positions – either becoming more or less engaged over time. These differences are reflected in national voting records and that in turn allows them to be represented in the communities of similar voting countries that the algorithm detects. A table listing the members of each community is provided in the article.
What can I use network mapping for as a diplomat?
I’ve already hinted at one of the reasons diplomats should be interested in network maps: they can help draw attention to countries that are less engaged in the UN cyber diplomacy processes, as evidenced by them not using their votes. Obviously the process will be a healthier and more inclusive one if all countries are encouraged and supported to participate.
A second use of network modelling is to help identify changes in country voting behaviour. Applying the community detection algorithm after each round of voting allows you to see where countries have shifted from one community to another. And when communicating changes in voting behaviour to colleagues, one can illustrate the shift by showing the movement of a country’s node within the network graph.
What am I using the network modelling for in my own research?
The next step in my own research is to explore the differences between the algorithm-detected communities in the cyber diplomacy network. In particular, whether there are differences between the interventions and influence of the middle ground countries on the one hand and the ‘polar camps’ of which the United States and Russia are members on the other.
My initial finding is that there are differences in which topics the middle ground countries do and don’t mention in their interventions and that this has potential practical applications. It will come as no surprise to those following the negotiations that the middle ground talk about capacity building more frequently than the polar camps. But I have been struck by the frequency with which they discussed emerging threats and how little they discussed a new UN treaty. More on this to follow I hope.
I’m also interested in the ways in which gender has affected the negotiations. Not all communities have an equal balance of male and female representatives taking the microphone. And this invites an important question: what difference, if any, has the Women In International Security And Cyberspace Fellowship programme made to the negotiations?
How to access the article
I’m grateful to the Cyber Policy Journal for providing a platform for the paper, but unfortunately it’s the nature of the journal publishing world that it’s now behind a paywall. That is no fault of the journal.
The good news is that my author agreement permits me to email you a copy of the paper if you request one. So to receive a copy just click the button below and enter the email address I should send it to. Please check your junk/spam folder and let me know in the comments if it doesn’t arrive.
Leave a Reply