The role of news outlets in spreading information about current events and highlighting specific societal issues has always been crucial. With the transition from traditional news reporting and television broadcasts to a presence on social media platforms, a growing number of individuals are now turning to social media for their daily news consumption (Geers, 2020).

This shift in news dissemination highlights the importance of investigating the relationship between social media and the ensuing discourse on news websites. Online discussions can influence public opinion. Therefore, it is important to understand how the discussion unfolds as well as the role that social media platforms play in shaping the discussions.

In the case of Seymour Hersh, the investigative journalist who owes his mythical status to a series of articles about the war crimes committed by the US military in South Vietnam in the 1960s, he claims the United States blew up the pipelines (Hersh, 2023). In doing so, he intervenes in the political discussion that is already going on. His writing, after it was captured by the mainstream media, influenced the discourse around the Nord Stream pipelines.

While studies may explain the emergence of the discussion surrounding the Nord Stream pipelines and highlight the potential consequences, little research has been conducted on the similarities and differences between discussions on prominent platforms such as Twitter and Facebook, as well as how the affordances of these platforms may influence them. As a result, it is critical to analyse Facebook and Twitter discussions regarding the Nord Stream explosions to determine whether and which affordances contribute to the development of a certain discourse. To accomplish so, this study answers the following research question:

RQ: How do platform affordances influence social media users’ discussions of foreign policy issues?

To answer this question, three sub-questions were examined: A. How did social media users respond to claims that the United States had blown up the Nord Stream pipeline? B. How did the discourse vary between Facebook and Twitter? C. How did the discourse vary between the United States and Europe?

Varis (2016) used digital ethnography to get extensive insights on these topics. This research dives into the growth of digital technology, contributing to scientific and societal concerns about digital media and foreign policy. The hybrid media system (see Chadwick, 2017) has guaranteed that talks about foreign policy have grown in popularity on social media platforms and in the news, suggesting that it is still a major societal problem.

Media Evolution

The literature review of this study highlighted that while the role of news platforms in disseminating information on topics such as foreign policy issues has always been crucial, the role of news platforms has changed since the advent of the new paradigm hybrid media system. The way individuals consume and produce news has changed radically since the advent of the Internet, and, more in particular, since the rise of social media platforms.

There has been a transition from a one-way communication flow to a cross-media communication flow (Chadwick, 2017). As a result, for example, disinformation or fake news is spread more easily. Moreover, news consumers consume news through various sources. Because the hybrid media system has influenced the relationship between politics, media, and power, it logically influences how the news consumer practices and experiences news consumption as well.

What social media platforms like Twitter and Facebook facilitate is that the news consumer can express their opinion and vision on issues online. It is the affordances of such a platform that ensure this. High-level affordances include the capabilities and limitations a platform offers, while low-level affordances include the literal buttons and features (Boyd et al., 2010). This study showed that while there are a few similarities between Facebook and Twitter, the underlying infrastructure of both platforms differs. For example, Facebook has a more one-to-one character in which mutual friendships are central, and Twitter is characterized by being a microblogging platform in which a one-to-all culture prevails and one-sided 'relationships' arise.

Platform Comparison

To compare the conversations about the Nord Stream pipelines on Twitter and Facebook, this qualitative study was carried out following the principles of digital ethnography. Digital ethnography has emerged as a methodological approach that enables researchers to investigate social practices, culture, and identity in online contexts (Varis, 2016). Using these principles, the differences and correlations between the conversations under online coverage of Nord Stream were mapped. This provided the opportunity to make a comparison between the two platforms and their affordances – and to consider their influence.

To perform useful research, a manual search was done for the social media channels where the conversation around the Nord Stream pipelines was most active. This created a divide between Facebook and Twitter. These platforms are an attractive study issue since their capabilities and affordances differ (Bucher & Helmond, 2018; Kirschner & Karpinski, 2010). The news outlets whose coverage of the Nord Stream was analysed were likewise chosen manually. Here, too, the available data from the accessible reports was analysed. To show contextual disparities, it was deliberate to choose news providers from various corners of the world. A news item is required to satisfy the following requirements:

• The article must be from a recognized news institution
• The article must be posted via a verified news profile
• The article should be about Nord Stream
• The article should be about Hersh's article
• The article must have at least 10 comments
• The news outlets must have a profile on both Twitter and Facebook
• There may not be more than one news outlet from the same country

The news outlets that were identified are the Washington Post (US), The Times (UK), Le Figaro (FR), and DIE ZEIT (GER). After de posts were analyzed and the data was mapped, it was decided to look at different categories. First, a sentiment analysis was performed manually. If an opinion was clearly expressed in the message, it was determined whether a message had positive or negative content. If no opinion was included, the purport was labeled neutral. It was also displayed when the sentiment was undefined. When this first analysis was done, the engagement level of the user comments was looked at. If someone had less than five likes or responses to their comment, the engagement level was low. Over five it was considered medium and over 15 was high.

The next step that was taken was to decide the tone of voice of the messages. The categories distinguished here were: Sarcastic/ironic, serious, provocative, angry, joke/funny, sad, annoyed, flabbergasted, confused/wondering, neutral, and undefined. In addition, it was examined whether the comment showed whether people agreed with Hersh's claim or not. This was represented by agree, disagree, neutral, and undefined. Finally, the context was looked at again and it was briefly summarized so that patterns could be identified.

In determining the results, it was decided to exclude responses that have nothing to do with the subject or that are indefinable. After all, these do not contribute to the political discussion that takes place online. The results have been compared and the different categories have been taken into account to study whether, how, and, if so, which platform affordances contribute to the development of political discussions online.

Social Media Insights

All data together showed that messages with a negative sentiment predominate on both Facebook and Twitter (figure 1). The percentage of negative messages is higher on Twitter than on Facebook. In addition, the results show that Twitter users are less likely to express positive sentiments than Facebook users on the subject of Nord Stream explosions. It could be argued that due to the anonymous nature of Twitter, people are more inclined towards positive ideologies on topics such as war.

After all, Twitter's structure concerns one-to-many communication that does not require mutual acceptance, while Facebook's structure is more geared towards friendly bubbles in which people deliberately enter into connections. However, the data shows that this is not necessarily the case. Whether or not one agreed with Hersh's claim, Facebook users had less difficulty expressing their opinions and emotions with a positive approach. This concerned both users from Europe and the United States.

The percentage of 'neutral/undefined' messages, on the other hand, is higher on Twitter. Layering and infrastructure are likely to be blamed for this. The data showed that the comments that fall under 'neutral' or 'undefined' are often comments that are part of a discussion. For example, these comments consisted of short questions or comments, and no sentiment was expressed in the comment. Facebook's stratification only goes three layers deep.

A discussion under a Facebook post is visible in its entirety as soon as the comments are clicked open. No comments are hidden on Facebook. The layering of Twitter is infinite. Dialogues branch below the original post. This contributes to the anonymous nature of Twitter, where a user can have conversations without a reader being guaranteed to read those conversations. It has been shown that this makes discussions on Twitter more sustainable and well-argued. Because more discussions and dialogues take place, the percentage of neutral/undefined messages on Twitter is higher as a result.

When the figures of the emotion analysis are put side by side, it is seen that there are no extreme outliers in the expression of emotions in the comments on either Facebook or Twitter (figure 2). On all four platforms on both Facebook and Twitter, the percentage of users who seriously raised the issue is the highest. It does become clear, however, that people on Twitter are more inclined to adopt a sarcastic/ironic or otherwise 'funny' tone in their reporting. This may be explained by the high-level affordances that define Twitter as a microblogging platform where throwing out a short 'Tweet' is the norm. Users may be more inclined to blurt something out and thus make fun of serious subjects.

The percentages of provocative comments are striking. The numbers show a clear difference between the United States and Europe. With 9.8% on Facebook and 8.8% on Twitter, the most provocative messages appear below the Washington Post, with a difference of 1%. This may also have to do with the way of reporting that causes discussion. Because the United States is identified by Hersh as the perpetrator of the Nord Stream explosions, they are directly interested in the matter. This is probably why it was decided not to post anything about Seymour Hersh's research article initially.

On the one hand, this ensures that some of the users have already heard about the issue through other channels and have formed an opinion. On the other hand, people lose confidence in the newspaper and in the country due to the lack of transparency. These are both things that can cause division. The communicative affordances (Hutchby, 2001) and the social affordances (Schrok, 2015) of both platforms allow such discussions to unfold under one post. It is striking that, particularly in France and Germany, such reports are virtually non-existent. Since both platforms in Europe do not differ in affordances, the explanation must nevertheless be sought in whether or not a country has an interest or in the way in which the news item is presented.

When analyzing opinion figures across different news outlets (refer to Figure 3), a notable trend emerged. The proportion of neutral messages was notably higher across all news outlets on Facebook. Conversely, on Twitter, this trend was observed primarily at The Times. Interestingly, Twitter users exhibited a greater tendency to express disagreement with Hersh's claim. This phenomenon can be attributed to the anonymous nature of Twitter, which encourages users to voice opinions that may diverge from the mainstream. In contrast, users on Facebook may opt for a safer approach.

The higher percentage of dissenting opinions on Twitter, particularly evident at The Times (43%), could be linked to sustained discussions under posts. Twitter's infrastructure facilitates branching dialogues, fostering an environment where dissenting voices are more readily expressed. This anonymity fosters a greater inclination for users to voice dissenting opinions, as reflected in the higher percentage of disagreement with Hersh's claim across all news outlets on Twitter compared to Facebook.

Finally, there is a difference when looking at the number of posts expressing their agreement with Hersh's claim. Except for The Times, the percentages on Twitter are higher than on Facebook. The biggest differences can be seen when looking at Le Figaro and Die Zeit. Where Le Figaro reported factually, Die Zeit did not.

What stood out in the comment section of both news outlets, however, was that there was criticism of the country, the news outlet, or the journalist who wrote the article. At Die Zeit in particular, the criticism was often directed at the journalist personally. In both comment sections, however, it emerges that people have distrust towards their country, the newspaper, or the journalist. Here too, it can feel safer to express that in an environment that is characterized by its anonymous character.

Conclusion

By studying how social media users responded to claims that the United States had blown up the Nord Stream pipeline, the analysis of social media data showed that user behavior and sentiments differ between Facebook and Twitter regarding the Nord Stream explosions. Although negative sentiments prevail on both platforms, Twitter users appear less likely to express positive opinions. The anonymity of Twitter may encourage more provocative and dissenting opinions, while Facebook's structure might result in more neutral or safer comments.

The discussions on Twitter were arguably more sustainable, likely due to the platform's layered structure. Although Facebook allows longer messages to be posted, sustainable and substantive discussions were more difficult to get going. No evidence was found that placing emotions on a comment on Facebook influenced the behavior, sentiment, or frequency of user discussions. Additionally, it was not proven that more likes on Twitter lead to more reactions to a comment.

When delving into how the discourse varied between Facebook and Twitter, it stood out that users on Twitter were more likely to post a comment expressing an opinion, while users on Facebook more often remained neutral in expressing an opinion. This confirms the anonymous character of Twitter where communication is not about mutual one-to-one communication, but about one-to-all communication where shared interests are central instead of mutual friendships.

What was also striking, however, was that when looking at the comments in which users for example personally attacked each other, although it could differ per case where it occurred more often, they were found on both Facebook and Twitter. The anonymous nature of Twitter therefore does not automatically mean that such comments only occur mostly on Twitter. Although the varieties of the discourse on the platforms can be explained by the high-level affordances of the platforms in some cases, the data showed that this line cannot simply be extended to all cases.

A striking discovery arose while comparing discourse between the United States and Europe: consumers, notably in the United States and France, are sceptical of their respective governments. The divergence is like this distrust: users responding to the Washington Post frequently blame the current US president for the Ukraine-Russia conflict and the Nord Stream explosions, whereas users engaging with French news outlets criticise the information provided by the French government and local news sources about the Ukraine conflict. Furthermore, extensive conversation ensued in the comments section of Die Zeit's articles, exposing pushback from users on both Facebook and Twitter to the outlet's reporting practices, demonstrating the influence of the hybrid media system.

Although this study has shown that platform affordances can influence social media users’ discussions of foreign policy issues, it also emerges that there is not always a clear line to be discovered. Twitter's high-level affordances allow users to engage in dialogues that are not easily traceable. This allows the user to dare to express their opinion sooner or more freely. Social control is less prevalent on Twitter, due to the one-to-many character based on shared interests instead of shared friendship. Facebook's high-level affordances ensure that social control is there.

In addition, discussions can be followed at a glance. As a result, the comments on the platform more often remain neutral in terms of opinion forming. This shows that although low-level affordances mainly contribute to the dissemination and visibility of the original messages from the news outlets themselves, it was mainly the high-level affordances that caused the difference in the discourse.

As said, there is not always a clear line to be discovered. After all, data showed that personal attacks, for example, occurred on both platforms. In addition, it is not only platform affordances that determine the direction of foreign policy discussions. For example, the interest that a user has in the issue can be of influence when it comes to discussions on social media. Moreover, how a news outlet shapes the message has of big influence on how the discourse under the message is formed. It has become clear that news outlets can no longer post a news item the way they want, because the readers of the news item, due to the hybrid media system and the cross-media news gathering, no longer simply assume what a news outlet writes.

References

Boyd, d. (2010). Social network sites as networked publics: Affordances, dynamics, and implications. In Z. Papacharissi (Ed.), A Networked Self: Identity, community, and culture on social network sites. Routledge.

Bucher, T., & Helmond, A. (2018). The Affordances of Social Media Platforms. In J. Burgess, A. E. Marwick, & T. Poell (Eds.), The SAGE handbook of social media. Sage reference.

Chadwick, A. (2017). The Hybrid Media System. Oxford University Press eBooks. 

Geers, S. (2020). News Consumption across Media Platforms and Content. Public Opinion Quarterly, 84(S1), 332–354. 

Hersh, S. (2023, February 8). How America Took Out The Nord Stream Pipeline. Seymour Hersh.  

Hutchby, I. (2001). Technologies, texts and affordances. Sociology, 35(2), 441-456.

Kirschner, P. A., & Karpinski, A. C. (2010). Facebook and academic performance. Computers in Human Behavior, 26(6), 1237-1245. 

Schrock, A. R. (2015). Communicative affordances of mobile media: Portability, availability, locatability, and multimediality. International Journal of Communication, 9, 909-926.

Varis, P. (2016). Digital ethnography. In A. Georgakopoulou, & T. Spilioti (Eds.), The Routledge Handbook of Language and Digital Communication (pp. 55-68). Routledge.