In and out of the Mass Media System: Crisis Microblogging in a Social System Theory Perspective

The large diffusion of social media platforms, togheter with the diffusion of mobile connectivity, has generated, during the last few years, a growing amount of real time production of user generated content. This broad phenomenon is even more interesting when it comes to crisis related communication. During crisis events traditional mass media communications  might be slower less efficient and less accurate than user-generated information. This is why, over the last few years, we are seeing a growing number of research projects aiming at using these data for rescue operations or other civil protection activities.We are facing this kind of data from a different perspective. Within this paper we will claim that – within specific circumstances – user generated communication will act surrogating the traditional Mass Media System in its function of self observation of the Social System (Luhmann 2000) by applying the same operational selection between information and non-information. In addition we will claim that, as soon as the initial conditions disappear, user generated communication will evolve into a new – and still largely unexplored social function.

To support our thesis we will use Twitter data collected during the first five hours after the earthquake that struck Emilia Romagna region in Italy on May 20th 2012 monitoring the #terremoto hashtag. By focusing on the first 5 hours of the Twitter stream we have been able to detect the early user-led phase of the phenomenon, showing which type of users has been the first to fill the information gap and, by then, what happened until the early morning when traditional media came on stage. We will show how in this time span it is possible to observe how specific social network analysis metrics (Bruns, Stieglitz 2012) evolve from a specific mass media like structure toward a more conversational structure.