Case study: how news circulates online and the changes it produces (social media, strikes, and political expression)

Untitled3bisI few months ago I took a picture and put it online. It was shared and commented on. Here is its story. 

‘Sup Internet? This text (I don’t know how to call it yet) proposes to analyze the online circulation of a particular piece of information during a controversy: a picture showing the blocked access to a university’s buildings by strikers during the 08/12/2014 strike in Brussels (Belgium). Using an Actor-Network Theory’s approach (Callon, 1986; Latour, 2005; Law, 1992 –see references at the bottom of the text-), it focuses on how the picture circulated first on social media to end up being shown on mainstream media and the effects it produced on the network. Tracing the network put in place, and stabilized during this short period of time, it was possible to gather and study a variety of contents produced online. Following then an inductive approach, a typology was created to show what types of comments were produced. Results show a network of individuals more eager to express themselves than to discuss the event.


To protest against the new federal government of Belgium’s measures, the workers’ unions decided to organize a series of strikes during the month of December 2014. On December 8th a general strike was organized in the city of Brussels and in the two regions of Brabant. Before the strike started, there were talks about possible actions at the ULB (Université Libre de Bruxelles) including the lockdown of some of the buildings and the library, within which students expected to be able to study for the next examinations (held in January 2015).

This particular sub-event of the strike in Brussels and the two Brabants, being itself a sub-event of the general strikes of Belgium in December 2014 was relayed on mainstream media, as well as on digital, community and social media. This paper proposes to combine an Actor-Network theory (ANT) approach as well as an inductive-style content and relationship analysis to trace this information’s online network and the effects this news had on the individuals in contact with it.

Using ANT to trace news networks

As such, actor-network theory is hard to quickly sum up (but I try to do it here, if you want more details), but it can in an exceedingly rough manner be defined as an approach to sociology that treats social phenomenon as network effects (Domingo, Masip, & Costera Meijer, 2014; Latour, 1996, 1999; Law & Hassard, 1999; Law, 1992; Lee & Hassard, 1999). The very nature of ANT as well as the relevance of its use within social sciences have often been the source of debates amongst scholars; firstly because researchers –even within particular fields– have employed (translated) ANT in various ways. Focusing on the field of MOS (Management and Organisation Studies), Toennesen, Molloy & Jacobs have shown four different applications of ANT in their field. They conceived a first analytical dimension of ANT invariably considered as a method (1) or a theory (2) and a second relying on its purpose as a terminology (A) or for its ontological and epistemological (B) implications. Four types of operationnalizations of ANT were then highlighted which they respectively named: simulating ANT (as a method, 1 and using its terminology, A), emulating (2.A), reasoning (2.B), and crafting (1.B) ANT (Toennesen, Molloy, & Jacobs, 2006). This paper proposes to explicitly apply ANT as a method (1) and use its terminology (A) which is consistent with Latour’s arguments of the necessity of tracing associations of actors and the use of an infra-language allowing to use the actors’ own meta-language, own vocabulary (Latour, 2005). Using ANT’s method/vocabulary as a research strategy and using an inductive approach is in this case relevant because this research aims at inductively identify relevant participants involved in the circulation of the content without prior judgment on who they are and what their activities are. Naturally, the results of this paper enable for a reflection on the ontological position of ANT and its explanatory of power as a theory.

An essential notion of ANT is the notion of actant (rather than the notion of actor) used to erase the distinction between what’s human and non-human (objects, institutions, ideals); both being endowed of agency. To understand a particular course of actions, researchers need to trace the relationships between actants evolving and creating networks –in our case “news networks”(Domingo et al., 2014)– with various degrees of stabilization. The term network is used not only to describe entities or groups of entities taking the shape of a net, but to emphasize the redistribution of action amongst actants such as humans, objects, or technologies (Latour, 2010). Regarding news networks, a particularly interesting concept is the concept Obligatory Point of Passage (OOP) used to describe entities or groups of entities that manage to shape and bend actions pleasing an hegemonic actor (Domingo & Le Cam, 2014). By tracing the evolving network of actants during a particular controversy, researchers then become able to write accounts and reflect upon the positions and actions actants.

It is true that the use of ANT in fundamental research encompasses a wide array of methodological and epistemological advantages as well as possible flaws. Nonetheless its main strengths are the fact that one can start can start with controversies to identify the actants interacting in a network –instead of imposing sociological concepts on who is part of what–, and that no pre-established sociological categories are used to describe them.

Methodological considerations: tracing the network and what was said around it

The methodology used to gather data regarding the circulation of the picture and the contents produced around it could be best described as an online controversy analysis. In order to do so, a series of qualitative research techniques were used (see Figure 1). Online media monitoring and social media monitoring were implemented to observe and gather data on mainstream and social media contents; network tracing as well as online observation was used to see of these contents were linked with one another.

The data was gathered online between the December 2nd and the 10th; comments and reactions as well as contents produced after this date were left aside. Altogether, tracing the network allowed for the detection of quantity of actors who produced 337 pieces of content (see Figure 11). The data analyzed encompasses 6 articles produced on online media (including 2 showing the picture), 31 tweets, 88 posts and comments on Facebook and 212 comments on the 6 articles (including 177 comments on the last article produced and showing the picture). The analysis does not take into account Facebook likes and favorited Tweets. It solely focuses on comments, articles written, shares, and retweets; anything possibly involving the production of additional text. It does not take into account possible mentions to other people and sub-conversation hence created.

Using the explicit relationships used on social media (mentions and tags) as well as the timestamps of contents production and a visualization software[1], it was possible to creating visual representations of the circulation of the picture online (see F. 5 – 11) useful for the analysis.

Capture d’écran 2015-06-01 à 10.28.56Figure 1. Summary of the methods of data collection and analysis.

This research strategy was put in place in order to answer three main research questions:

  • How did the picture circulate and what content was produced around it?
  • What does it say about actants in this particular network?
  • What was the role of mainstream media in informing people regarding this event?

Whose picture is it anyway? The story of the ULB lockdown

To protest against the new federal government of Belgium’s measures, worker’s unions decided to organize a series of strike during the month of December 2014. On the 8th of December a general strike was organized in the city of Brussels and in the regions of Brabants. Before the strike started, there were talks about strikes at the ULB (Université Libre de Bruxelles) including the lockdown of some of the buildings. This news (access to some of the buildings is blocked) was relayed by actants on mainstream media, as well as on digital community and social media.

The story of this controversy did not actually start on the 8th of December but a week earlier, on the 1st day of the month, when an email sent by the dean of the university to its employees was leaked. It announced that there would be no changes in salary (and in the attribution of food stamps) whether they went on strike or not, thus giving them the opportunity to do so. The next day, two articles were written (Article 01 & 02 see F. 5) were posted online[2]. Article 03 and 04[3] (F. 6 – 7) were posted on the next day, and Article 04 mentions explicitly article 01 as a source, giving an hyperlink to the article. Interestingly enough, articles 01 and 03 only generated one comment from a user, whereas Article 02 generated 33 comments.

Untitled Figure 2. Original post on Facebook.

On December 8th, at around 07:00 the strike started in Brussels and at the ULB. Rapidly, doors to buildings (including the library) and classrooms were blocked by strikers and some teachers who started their classes with a small amount of students were forced to stop. At 10:18 a picture showing the door of a building blocked by chairs, fences and various furniture was taken. A couple minutes later, the picture was posted on Facebook (F. 2) and Twitter (F. 8). Throughout out the day and the day after, the picture was shared on social media (F. 9 – 11). After a couple hours, most posts and tweets including the picture had stopped mentioning the author making it hard to trace the network in a comprehensive manner (F. 3 & 4). It was shared by private individuals and public or politically-involved figures (on Twitter) as well as by groups and pages (on Facebook) having various involvements to the ULB. The picture (without mention) ended-up on a live feed of the news website LeVif [4], and on an article by LaLibre[5] which had generated 177 comments (including 85 answers to comments) by December 10th. Since that day, no new content directly related to this issue or the picture has been produced or exchanged.

Untitled2bisUntitled3bisFigures 3 & 4. Examples of Tweets with (3) and without (4) mentions of the author. Tweeter user from figure 4 seems to be affiliated with the political party of the majority against whom the strike was intended.

These elements lead to two conclusions regarding this particular piece of information. On one hand news quickly stopped being shared and discussed online (after the 10th). The fact that the circulation of the picture stopped this quickly shows in this case ANT’s idea of the ability of controversies to create relatively (un-)stable networks and not the opposite way around.

On the other hand, the picture and what it said meant more to actants who shared it and commented on it than to who made it and than what he said about it.

Expression more than discussion

Following an inductive approach to content analysis, it was possible to separate comments and contents produced in four main categories: supporting the blockade of some buildings, against the blockade, unrelated to it, and content with an unclear meaning. It is interesting to note that most comments with an unclear meaning can be interpreted as a pun, a joke about the situation portrayed by the picture as in “very stylish this new form of aestheticism”[6] (PUI1_C01). Comments to Article 06 seem to focus on the political orientation of the students of the university more than on the strikers themselves. This might a consequence to some of the elements written in the article expending further than what the picture said. Nonetheless the same classification can still be applied.

A first important observation than can be made is that most actants want to express themselves more than actually creating a debate or a dialog. The majority of the users who shared the picture used to it to express a political stance such as in the second share from the second person who used the picture on Facebook or PU2_S2 who stated “what a group of a**h****”. After sharing the picture or commenting one of the contents, most people felt silent. Even if an actant answered a critique to a particular content posted, it was to justify itself, or re-affirm a certain position. This corresponds to what has commonly been argued by the literature on online behaviors and polarization. After analyzing online productions related to the picture, no sign of consensus had been detected within the transient network.

Another important feature of the content produced around the circulation of the picture is that the content produced around it often serves as a contextualization to express wider political opinions or to illustrate national political decisions:

PU2_C19: blocking access to education is shameful! There are exams in less than a month, this is selfish.
PU2_C20 (in response) : What is shameful is the government who tries to pass anti-social measures right before the exams when not many people can mobilize.

Most actants using the picture use the controversy as a way to influence other actants of their own digital social networks as an example of the many results of the strikes in Brussels and more generally in Belgium. Continuing to trace the network of actants involved in this controversy would imply to broaden the scope of the research to the wider controversy of the new government’s decisions and the multiple strikes that followed.

Conclusion: From ANT to the controversy and back

Through this study, we have shown that the circulation of this particular piece of news online has lived a very short life, which confirms Anderson’s observation about the diffusion of local news online (Anderson, 2010). This element also confirms ANT’s general assumption that controversies generate networks, or at least that it is useful to see things this way to analyze the circulation of news. I have argued that this controversy helped shaping a network of actants seeking to express themselves more than discussing the issue, and that it ended up on mainstream media. The picture, -considered as an actant-, seems to have acted as an Obligatory Point of Passage for individuals who wished to express themselves regarding the political situation at that moment. The picture was then re-used by mainstream media who became themselves, as stated in previous works (Domingo & Le Cam, 2014), an OPP for people to express their views. A deeper extensive research using various ANT concepts might help get a clearer knowledge on the weight of some actants (mainstream media, the picture, social networks) shaping the network. Nonetheless, the fact that a single article (Article 06) generated the production of 177 comments appears to confirm the argument that mainstream media still very much acts as an hegemonic actor when it comes to news production.

Using Actor-Network Theory –as it has been proven before- is relevant to identify actants related to a controversy without prior knowledge on who they are and without foreseeing what they might do or who(m) they will create relationships with. ANT has proven itself useful as a research strategy to access uncharted territory.

To conclude, separating the use of ANT as a methodology from its epistemological and ontological implications proves to be more difficult than it sounds; tracing the evolving network of actors and contents produced online implies that any new production is a part and an effect of the evolving network. Separating ANT in four categories (methodology, terminology, theory and ontology) might not be relevant or useful to the researcher as these are in practice very much porous. However, keeping in mind that the use of ANT to study controversies is an epistemological choice encompassing a certain way of looking at the world is of the upmost importance.

Capture d’écran 2015-06-01 à 10.45.54Figures 5 & 6. Visualizations of the evolution of the controversy on December 2.

Capture d’écran 2015-06-01 à 10.46.28Figures 7 & 8. Visualizations of the evolution of the controversy on December 8 (morning).

Capture d’écran 2015-06-01 à 10.47.12Figures 9 & 10. Visualizations of the evolution of the controversy on December 8 (Afternoon).

http___makeagif.com__media_12-18-2014_ZhU_I8Figures 11. Visualization of the actants and their relationships from December 2 to December 9 (Afternoon).


Anderson, C. W. (2010). Journalistic networks and the diffusion of local news: The brief, happy news life of the “Francisville Four.” Political Communication, 3(27), 289–309. Retrieved from

Callon, M. (1986). The sociology of an Actor-Network. In J. Law & A. Rip (Eds.), Mapping the Dynamics of Science and Technology (Macmillan.). London. Retrieved from

Domingo, D., & Le Cam, F. (2014). Journalism Beyond the Boundaries: the Collective Construction of News Narratives.

Domingo, D., Masip, P., & Costera Meijer, I. (2014). Tracing digital news networks: Towards an Integrated Framework of the Dynamics of News Production, Circulation and Use. Digital Journalism, 15(1).

Latour, B. (1996). On actor-network theory: a few clarifications plus more than a few complications. Soziale Welt, 47, 369–381. Retrieved from

Latour, B. (1999). On recalling ANT. In J. Law & J. Hassard (Eds.), Actor Network Theory and After (Blackwell., pp. 15–25). Retrieved from

Latour, B. (2005). Reassembling the Social. An introduction to Actor-Network Theory (Vol. 7, pp. 500–502). doi:10.1163/156913308X336453

Latour, B. (2010). Networks, Societies, Spheres: Reflections of an Actor‐network Theorist.

Law, J. (1992). Notes on the theory of the actor-network: Ordering, strategy, and heterogeneity. Systems Practice, 5(4), 379–393. doi:10.1007/BF01059830

Law, J., & Hassard, J. (Eds.). (1999). Actor-Network theory and after (Blackwell., p. 256).

Lee, N., & Hassard, J. (1999). Organization unbound: actor-network theory, research strategy and institutional flexibility. Organization. Retrieved from

Toennesen, C., Molloy, E., & Jacobs, C. (2006). Lost in translation: Actor network theory in organization studies. Human Relations, 1–37. Retrieved from

[1] “Gephi” is an open source network visualization software, see

[2] Article 01: L’ULB rémunérera son personnel, qu’il fasse grève ou non. (2014, December 2). Le Soir. Retrieved December 10, 2014, from

Article 02: A l’ULB, même lorsqu’on fait grève… on est payé. (2014, December 2). La Libre. Retrieved December 10, 2014, from

[3] Article 03: A l’UB, gréviste ou pas, c’est payé. (2014, December 3). L’Avenir. Retrieved December 10, 2014, from

Article 04: Vous travaillez à l’ULB et vous faites grève? Vous toucherez votre rémunération! (2014, December 3). Actualités du Droit Belge. Retrieved December 10, 2014, from

[4] Article 05 : LIVE. Grève tournante : Bruxelles et les deux Brabants paralysés. (2014, December 8). LeVif. Retrieved December 10, 2014, from

[5] Article 06: Grève Tournante: Les étudiants libéraux de l’ULB en colère. (2014, December 9). LaLibre. Retrieved December 10, 2014, from

[6] Translations made by the author (it’s me).

One thought on “Case study: how news circulates online and the changes it produces (social media, strikes, and political expression)

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s