![]()
Reading the riots: what were the police doing on Twitter?
Rob Proctera, Jeremy Crumpb, Susanne Karstedtc, Alex Vossd, and Marta Cantijoche
aManchester eResearch Centre, Manchester University, Manchester, UK; bInternet Business Solutions Group, Cisco Systems Ltd, London, UK; cCentre for Criminal Justice Studies, School of Law, Leeds University, Leeds, UK; dSchool of Computer Science, St Andrews University, St Andrews, UK; eInstitute for Social Change, Manchester University, Manchester, UK
The widespread adoption of new forms of digital communication platforms such as micro-blogging sites presents both an opportunity and a challenge for researchers interested in understanding peopleās attitudes and behaviours, especially in the context of unfolding crises and the need for government agencies such as the police to inform the public and act swiftly to ensure public order and safety. In this paper, we use a study of a recent public order crisis in England to explore how the police, other organisations and individuals used Twitter as they responded to this event.
Introduction
Like many other research communities, the social sciences are facing a data deluge that promises to revolutionise research practice, but which they are presently not equipped to exploit (Savage and Burrows 2007). Where once the main problem facing social researchers was scarcity of data, they must now cope with its abundance. A key example comes from the rapid and widespread adoption of social media, which presents both an opportunity and a challenge for social researchers.
An important area of social media research is the role different communication platforms play during unfolding crises (Barsky et al. 2006, Mendoza et al. 2010, Bruns et al. 2012). In this paper, we examine how the police, other groups and individuals used the micro-blogging platform Twitter as they responded to the August 2011 riots in England. We consider the implications for government agencies such as the police as they strive to inform the public about what is happening during crises and to act swiftly to ensure public order and safety.
The riots began as an isolated incident in Tottenham, London, on 6 August when Mark Duggan was shot dead by police. The riots quickly spread across London and to other cities in England, and gave rise to levels of looting, destruction of property and violence not seen in England for more than 30 years. Eventually, after 5 days, the riots ceased. The causes have been attributed to many different factors, including resentment by young people about policing methods and growing economic inequality (Guardian 2011, Morrell et al. 2011, Newburn 2012).
In the immediate aftermath of the riots, some politicians and media commentators were quick to blame social media, including Twitter, for their scale and extent. In a previous paper, we concluded that evidence of Twitter being used to encourage lawlessness during the riots was far outweighed by evidence that Twitter was used for more positive ends and, in particular, in the organisation of the riot clean up (Procter et al. 2013). Furthermore, we noted that the police themselves rejected the idea of closing down social media sites (www.guardian.co.uk/uk/2011/aug/12/louise-mensch-social-network-blackouts), arguing that they are a valuable tool for information gathering, for keeping the public informed and for providing advice.
However, our study of how rumours propagated through Twitter during the riots does suggest that the police (and government agencies in general) have yet to get a grip on using social media platforms like Twitter to their best advantage during such events (Procter et al. 2013).
Our aim in this paper is to examine more closely how the police, other groups and individuals used Twitter as they responded to the riots, whether the role of the former, in particular, has been effective and, if not, what lessons might be learnt.
We begin by providing an overview of Twitter as a social media platform. We then summarise previous research into the policeās use of social media. We follow this with an overview of the methods we used to analyse a large corpus of tweets collected during the riots. We then present a number of case studies to illustrate how using these tools enables us to describe in some detail how Twitter was used by the police and others. Finally, we outline possible lessons for the police and some areas for further work.
Twitter
Twitter is a micro-blogging site set up in 2006 that allows users to post messages (tweets) of up to 140 characters. Twitter can be accessed through the web, SMS and a variety of mobile devices such as smart phones and tablets. A recent estimate puts the number of UK Twitter users at 10 million. According to recent statistics released by Twitter (blog.twitter.com/2011/09/one-hundred-million-voices.html), however, a majority of users do not tweet but just read tweets sent by others.
Unlike social media platforms such as Facebook, Twitterās friendship model is directed and non-reciprocal. Users can follow whoever they like, but those they follow do not have to follow them back. When one user follows another, the latterās tweets will be visible in the formerās ātweetstreamā. It is not necessary, however, to follow another user to access tweets: by default, Twitter is an open platform; tweets are public and can be discovered through Twitter search tools. The exception is the direct message, which is private, and can be seen only by the follower to whom it is sent. Users can reference another user through the mention convention, where a user name, prefixed with ā@ā, is included anywhere in a tweet. A user, thus referenced, will see the tweet in their tweetstream. Ano...
