Riots: going viral

Police line 12 August 2011

Online social media such as Twitter have been blamed for exacerbating the riots, by making it possible to quickly spread false rumours and incite others to violence. But this is far from a new phenomenon, explains Helen Margetts, Professor of Society and the Internet at the Oxford Internet Institute.

"Riots such as this have taken place before and in other countries, both before and after the advent of the Internet and social media, and the idea can spread across geographical locations through traditional media perfectly well. For loosely co-ordinated action like this, a simple network structure will suffice, with clusters of people in specific geographical areas - and then just one 'tie' between clusters, such as a TV report or image of burning buildings and rioters on the street, will allow it to spread to another area," she says.

Similar network spreading has been observed with viruses or certain types of animal behaviour.

The 'loose' nature of online communities has been highlighted in comments, where many people with little or no connection to each other can be mobilised quickly. This type of environment was well suited to the rioters' goals, Professor Margetts points out.

"This was not a collective action around public goods, as in the revolutions of the Arab Spring. It was around private goods, and the value of having more people there was to provide critical mass in terms of discouraging the police from intervening and preventing them from being able to do so forcefully - creating a general atmosphere of chaos and excitement. At a certain point, more people would have been counter-productive, running the risk of crowding out the possibility of obtaining the private goods. So in some ways, rioters would have wanted other participants that they didn't know, as much as people that they did know. The looser the ties the better, as long as some of them turned up."

In the ESRC-funded research project The Internet, Political Science and Public Policy: Re-examining Collective Action, Governance and Citizen-Government Interactions in the Digital Era Professor Margetts is studying online behaviour and the effects of information environments on mobilisation – although in a political context.

She concedes that social media does reduce co-ordination costs in collective action and were used by some people in the riots, but points out that closed Blackberry message networks seem to have been much more important in alerting potential rioters to where the action was going to be.

"At the same time, they would not have wanted the police to be alerted to what was going on, so the closed messenger networks were more useful - as well as being what the people involved habitually use to communicate with each other. Twitter, in particular as an open network, would not be very efficient for organising something like this, because any reduction of co-ordination costs for the rioters would result in an equal gain for the police," she adds.

"It has indeed been argued that Twitter was used to confuse the police about which areas rioting was going to take place. But this just highlights the need for police agencies to tool up and strengthen their analytic capacity to use data from Twitter and other publicly available social media to garner intelligence, just as they do habitually for many other sources."