Fabricating Twitter: Social media narratives, political dissidence, and false information in Iran

Project Information

FIonPT

Grant agreement ID: 101029945

DOI

10.3030/101029945

Project closed

EC signature date 28 April 2021

Start date 1 October 2021

End date 30 September 2023

Funded under

• EXCELLENT SCIENCE - Marie Skłodowska-Curie Actions

Total cost

€ 186 167,04

EU contribution

€ 186 167,04

Coordinated by

UNIVERSITAT WIEN
Austria

Periodic Reporting for period 1 - FIonPT (Fabricating Twitter: Social media narratives, political dissidence, and false information in Iran)

Reporting period: 2021-10-01 to 2023-09-30

Summary of the context and overall objectives of the project

False information has become a major concern in contemporary societies. With the rapid growth and popularity of social media communication, bad actors, e.g. non-democratic forces, employ them to share disinformation and toxitize user-generated discourses. Such suppressive actions are of paramount significance in authoritarian countries like Iran. In Western democracies, most people have access to a variety of communication channels. As a result, social media is only another tool at their disposal. In non-democratic societies, however, social media like Twitter are the citizens’ only chance to seek sensitive information and raise their voices beyond regime restrictions and censor machines. Therefore, making these spaces toxic and overwhelming them with false messages and inaccurate information deprives people of the sole chance to create more democratic systems. Thus, it is critical to investigate through which discursive mechanisms and to which extent authoritarian regimes share inaccurate information on social media.
FIonPT investigates how false information is shared and circulated in the Iranian Twittersphere. It aims to identify the networks and practices of false information in which Twitter activism is suppressed from the inside in Iran. While authoritarian regimes employ hard measures like internet blockage and online surveillance to stop people from using social media, they also develop proactive tactics like sharing false information. Therefore, investigating their tactics could inform people in such countries and help them overcome such suppressive strategies. It could eventually help them to fight with regime-sponsored disinformation campaigns. This research could be of interest even from a Western perspective. Many of these suppressive tactics have found their way to social media in Western countries in recent years. In addition, non-democratic countries like Russia usually orchestrate disinformation campaigns to mislead Western citizens. Therefore, understanding in which ways and through which mechanisms bad actors work in online spaces could also weaponize Western citizens with the knowledge they require to compete with them. FIonPT produces this knowledge with a focus on an understudied context: Iran.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

FIonPT started with a literature review and collection of data. First, the scholar reviewed the existing literature in the field to inform the project and indicate the gaps that the project aims to address. Then, research data was collected from Twitter. The research dataset included 926,257,362 Persian tweets. I followed a mixed-method approach to analyze data and investigate the actors, networks, and practices of false information on Persian Twitter. The first step was identifying the main communities in Persian Twitter. I used Social Network Analysis to identify the dominant communities in the network. Having identified five communities that represent mainly political parties in Iran’s political sphere, I identified the most influential users in each community. The following steps are concentrated on those users and their tweets. I followed a human-driven qualitative coding process to code and analyze the data. All users had been coded regarding their career, political orientation, and identity. Their tweets have also been coded. Therefore, tweet attributes like type (e.g. comment) and tone (e.g. critical), in addition to the type of information they shared, have been coded. A significant step in data analysis was bot identification. Social bots are automated programs that are active and connect with other users on social media. They are usually used in malicious campaigns to share false information. Thus, it was of significant interest to see the volume of social bots on Persian Twitter and the extent to which they played a role in sharing false information in comparison with non-bot users. The research dataset was coded discursively based on critical discourse analysis (CDA). CDA postulates that language is inherently political and in an inseparable connection with power and ideology. It is a method that helps us to go beyond the simplistic and obvious meaning of words to see how they become units of meaning-making that frame and translate our everyday experiences.
The result of FIonPT has been presented at high-profile conferences like the International Communication Association conference, the Association on Internet Researchers conference, and the European Consortium on Political Research conference. In addition, the research findings have been submitted, and some have already appeared in top-tier journals like American Behavioral Scientist, Social Media and Society, and Information, Communication, and Society. Furthermore, I have discussed the research findings with prominent scholars in our field on occasions when I was invited to speak at the Milton Wolf Seminar in Vienna, The Complexity Science Hub in Vienna, and The Weizenbaum Institute for the Networked Society in Berlin.

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

The existing literature on false information on social media has extensively concentrated on Western countries and employed computational methods. FIonPT pushes forward this line of inquiry by focusing on a non-Western country and combining discourse analysis with computational methods. Methodologically, FIonPT developed a new approach to conduct CDA on Twitter. The combination of SNA, CDA and human-driven qualitative interpretation in several rounds enable researchers to investigate the micro and macro dynamics of meaning-making on social media at the same time. This approach maximizes the benefits of both camps and minimizes their risks simultaneously.
In addition, FIonPT enhances our understanding of the ways through which an authoritarian regime like Iran employs social media, in particular Twitter, to suppress dissident discourses and narratives. The project delineates the actors and networks and the strategies they developed to share false information. Results both identified different camps on Persian Twitter and their role in sharing various types of false information, e.g. disinformation. The discursive understanding of Twitter suppression in Iran, which FIonPT offered, could be a valuable source for Iranians who are trying to establish a more democratic political system in their own country. In also paves the way for other researchers to have an understanding of social media activism and suppression in understudied contexts. As a result, more comparative and collaborative projects between researchers from different societies could be shaped.