Skip to main content
Springer Nature - PMC COVID-19 Collection logoLink to Springer Nature - PMC COVID-19 Collection
. 2023 Jan 19;21(1):47–80. doi: 10.1057/s41253-023-00205-4

Battling to dominate the discursive terrain: how Marine Le Pen and Emmanuel Macron have framed terrorist incidents on Twitter

Miltiadis Rizakis 1,
PMCID: PMC9851577

Abstract

This article seeks to examine how the issue of terrorism has been framed by Marine Le Pen and Emmanuel Macron. Marine Le Pen has been eager to exploit these incidents since they fit neatly within her xenophobic and nationalist discourse. On the other hand, Emmanuel Macron seeks to transcend traditional political boundaries and foster unity. These different strategies in framing terrorism will be the focus of this article. In addition, the article will examine whether there are elements of populist contagion in Macron’s framing. The analysis is centred around two terrorist events that occurred in 2020: (1) Samuel Paty’s murder and (2) the Nice knife attack. The data were collected from the Twitter accounts of Le Pen and Macron and analysed via the latent Dirichlet allocation generative statistical model. The result is an in-depth analysis that showcases the different framing strategies of the two case studies regarding terrorism.

Keywords: Populism, Terrorism, Twitter, Populist contagion, Topic modelling, Framing

Introduction

France was the first European country to be targeted by state-sponsored terrorism from the Middle East and has the dubious distinction of being one of the most affected states by terrorist violence in Europe (Bartolucci 2017:439). The issues of terrorism and security have remained highly salient in the discourse of the media and political actors, and several of them have received worldwide attention, such as the November 2015 Paris attacks. In the case of the media, they share a “symbiotic” relationship with terrorism, since the end goal of terrorists is to affect the public through the coverage of their attacks by the media (Solheim 2019:206).

As for political actors, they are called to provide expert opinions on events that cause widespread shock and fear. It is here that the concept of issue ownership comes into play. Political actors will focus on issues that they own, while ignoring the ones on which their opponents enjoy a reputational advantage (Petrocik 1996:827, 828). A party is said to own an issue when it: “… has a long-term reputation of attention to the corresponding problem and is considered to be particularly able at handling that issue” (Lachat 2014:729). National Rally has been increasingly incentivised to frame the issue of terrorism within its discourse to capitalise on these events (Hutchins and Halikiopoulou 2019:68).

The party has incorporated the issues of terrorism and security within its anti-immigration agenda, claiming that its primary concern is the defence of the French Republic (Bastow 2018:23). Since the utilisation of social media opens new avenues of communication between political actors and the citizens, this article will examine how the issue of terrorism has been framed on Twitter by Marine Le Pen and Emmanuel Macron, by focusing on two terrorist incidents. The first is the murder of Samuel Paty on 16 October 2020, and the second is the Nice knife attack on 29 October 2020. These two events motivate the research questions of this article: (1) How salient was the topic of terrorism in the Twitter discourse of Marine Le Pen and Emmanuel Macron? (2) Are there elements of populist “contagion” present within the online media framing of Emmanuel Macron on the issue of terrorism?

Background

Marine Le Pen is the leader of the right-wing populist party National Rally (formerly known as National Front). NR has been transformed from a successful yet marginal party into a major power in French politics under the leadership of Marine Le Pen, who succeeded her father Jean Marie Le Pen in 2011 (Stockemer 2017:2). NR’s rhetoric is a mixture of nationalism and populism, where French society is threatened by unchecked immigration and the laxity of governmental security policies and oppressed by the directives of the European Union (Stockemer 2017:27).

After the change in leadership, Marine Le Pen and her colleagues began a process of phasing out the more extremist positions upheld by her father (Evans and Ivaldi 2018:9). Since 2011, Marine Le Pen has presented NR as a republican party and rejects any linkages to the far-right (Stockemer 2017:39). Moreover, she has abstained from making any overtly racist or anti-Semitic statements (Stockemer 2017:39). Finally, she presents her party as a defender of republican values. For example, the party has often utilised the concept of secularism to attack Islam (Stockemer 2017:39).

Le Pen performed well in the Presidential elections, but eventually lost to Emmanuel Macron and his newly formed party La République En Marche!. Macron previously held the role of deputy secretary general and economic advisor to President Francois Hollande from 2012 to 2014 (Evans and Ivaldi 2018:79, 80). Despite his prior affiliation with the Socialist Party Macron campaigned on a centrist platform, something which has not been particularly successful in France (Lachat and Michel 2020:566). His positions incorporated liberal economic policies typical of right-wing parties, culturally progressive stances like left-wing parties, and a strong support towards the European Union, as opposed to Le Pen (Lachat and Michel 2020:566).

By including a roster of former left- and right-wing politicians, as well as a sizeable contingent of unaffiliated civil society candidates, the party managed to disrupt the traditional bipolar dynamics of French party competition (Evans and Ivaldi 2017:324). The party composition and its positions allowed Macron to appeal to moderates at both sides of the spectrum (Evans and Ivaldi 2018:80). The central message of his campaign was the abandonment of outdated ideological discourses and that his focus would be the implementation of effective and practical solutions to contemporary economic and societal problems (Perrotino and Petra 2020:548).

These significant differences between the two parties and their leaders are the starting point for an analysis on their respective framing of two terrorist incidents, which occurred in the same month. The first one is the murder of Samuel Paty by a radical Islamist on 16 October 2020. Paty taught a moral and civic education course on freedom of expression and showed some of his students a caricature of the prophet Muhammad from the satirical magazine Charlie Hebdo. Almost two weeks after his freedom of speech class, Samuel Paty was attacked and murdered in a street near the school he taught. The murder triggered a wave of support towards Paty, and the issues of terrorism and freedom of speech monopolised the interest of citizens and politicians alike.

The second terrorist incident took place on 29 October 2020. The attack occurred early in the morning at Notre-Dame de Nice, a Roman Catholic basilica situated in the centre of Nice. The attacker killed three people with a knife before he was subdued by police officers that arrived at the scene. In the aftermath of the attack, Macron visited the site and called for firmness and unity while expressing his solidarity with the Catholic community. Marine Le Pen also expressed her sympathy towards the Catholic community, but also harshly criticised Macron’s government for its laxity and labelled Islam as a barbaric ideology. These two events will be the focus of this article since an important issue linked with them is the protection of values such as the freedom of speech and secularism.

Theoretical framework

Crisis exploitation

According to ‘t Hart and Tindall, crises are “… the combined products of unusual events and shared perceptions that something is seriously wrong.” (2009:6). The two main leadership challenges during a crisis are: (1) to accurately diagnose rapidly changing and often confusing circumstances and (2) to provide a persuasive account on the origins of the crisis, its overall impact, and possible solutions (‘t Hart and Tindall 2009; Boin et al. 2014). In addition, a crisis is a contested phenomenon, without any clear and distinctive boundaries, and a product of complex causality (Moffitt 2016:118).

The successful resolution of a crisis does not only entail the development of policies and efficient utilisation of available resources, but also to form persuasive narratives about: “…what is happening and what is at stake, why it is happening, how they have acted in the lead-up to the present crisis and how they propose we should deal with and learn from the crisis moving forward.” (‘t Hart and Tindall 2009:22). Simply put, it is not the events themselves that solely determine their potential impact on political actors or public policy, but their public perception and interpretation. Of course, perceptions of crisis will vary not only among communities but also within them, since they are reflections of the stakeholders’ different biases, values, positions, and responsibilities (Boin et al. 2009:84).

The only constant in an event of a crisis is that it will generate feelings of uncertainty, fear, anger, and questions about responsibility. A process of reconstruction is needed in such a situation, as everybody tries to understand how such events came to pass. To successfully combat a crisis, political actors must face two interrelated challenges: (1) to choose the appropriate policy tools, and (2) dominate political discourse by making sense out of rapidly evolving events, under extreme uncertainty, and often facing an overload of information (Hindmoor and McConnell 2015:17). This article is concerned with the second challenge, that is political symbolic in nature. Governments need to produce an authoritative account of the crisis to convince citizens, the media, and stakeholders that they are in control. As for the opposition, a crisis presents an opportunity to criticise the government and garner the public’s support.

Framing terrorist incidents

According to Entman, framing is a process that selects and highlights certain aspects of events or issues, to promote a specific interpretation, evaluation, and solution (Entman 2004:5). The article will focus on issue-specific frames, since they are pertinent to specific topics or events (de Vreese 2012:368). These frames perform at least two of the following basic functions when they cover political events, issues, and actors: (1) defining effects or conditions as problematic, (2) identifying causes, (3) conveying a moral judgement, (4) endorsing remedies or improvements (Entman 2004:5).

Discourse can shape the perception of reality both for political actors and citizens (Heath-Kelly 2019:229). As previously mentioned, crises are not neutral phenomena, and they can be mediated and performed by political actors and the media. The framing of terrorist attacks is similar, since when a particular narrative is regularly repeated by influential figures, it can become part of social and political life and gain increased validity (Jackson et al. 2011:56). As a result, several political and ideological consequences can follow: (1) a narrative may become dominant and thus enable certain policy options to be considered as viable and necessary, (2) shaping national and social identities, and (3) create a framework of understanding, through which future political events can be interpreted and understood (Jackson et al. 2011:56).

Therefore, terrorism is socially constructed via discourse, when a particular individual or group are labelled as a threat (Hulsse and Spencer 2008: 576). As stated by Richard Jackson: “Terrorism is fundamentally a social fact rather than a brute fact” (Jackson 2007:247). Most importantly, the discourse on the war on terrorism is an exercise of power, since it is meticulously constructed for a variety of goals, ranging from legitimising certain policies against terrorism, to achieving national unity (Jackson 2005:2). While terrorist violence is experienced as a brute fact by its victims, its wider cultural and political meaning are socially negotiated by a variety of actors such as the media, the authorities, and politicians (Jackson 2011:117). Therefore, this article will synergise the literatures of crisis framing and critical terrorism studies and fully showcase how terrorism is framed in a particular context regarding its causes, overall impact, blame attribution and remedial actions by the two politicians.

Framing effects are the result of the modification or activation of schemata of interpretation and can be found in how information is processed and understood, how the public talks about an issue, and how it forms political evaluations (Pan and Kosicki 2001:39). This process is multifaceted rather than linear, since the stimuli provided by political actors are not sufficient in themselves to affect the public’s perceptions and opinions. Snow and Benford have identified three key factors which affect the mobilising potency of proffered framings.

The first factor is the fit between the framing and events in the world. Can the framing be subjected to verification? (Snow and Benford 1988:208). The second factor, experiential commensurability, is dependent upon personal experience, and as a result the proposed explanations and solutions must harmonise with it (Snow and Benford 1988:208). The final factor is the degree to which proffered framings can resonate with the artefacts included within the public’s cultural heritage and are utilised to make sense of the immediate present (Snow and Benford 1988:210).

Populist contagion

The gradual rise of populist parties in Europe has brought about the thesis of the “populist zeitgeist” (Mudde 2004:551). During the last decade, rolling crises have generated a wave of public discontent that damaged the legitimacy of national governments. The “populist zeitgeist” also refers to the process of “populistization”, where populist elements are becoming increasingly present within the programs, manifestos, and discourses of mainstream parties (Manucci and Weber 2017:316). This development is referred to as “populist contagion”. The term itself is usually not clearly operationalised within the context of research on populism but refers to “…communicative or programmatic shifts of parties towards more populist discourses or policies due to changes in political surroundings” (Schworer 2021:1).

The notion of contagion has as its foundation the theory of party competition developed by Downs (1957). The electoral process is described as an electoral market where the political system is on the “supply” side and the voters on the “demand” side. Therefore, the integration of an anti-immigration policy and rhetoric within a party is viewed as a form of strategy aimed at maximising electoral gains. Meguid has presented a modified spatial model, where she adds the insight that parties may influence the salience of issues (2005:349, 350). It should be noted here that niche parties are the primary focus in this model, while National Rally has evolved to become a major opposition party. Yet still, she makes the important observation that issue ownership and the party’s issue credibility may play an important role for voting behaviour (Meguid 2005:349).

According to Meguid’s model, parties in government have three options when it comes to policy positions employed by competing niche parties. First, they can adopt an accommodative strategy by adapting and copying the policy position promoted by the party (Meguid 2005:348). The second option is to follow an adversarial strategy by increasing the policy distance between them and their competition (Meguid 2005:348). The final option is the dismissive strategy, where the party can decide to take no position on the issue addressed by their competition (Meguid 2005:349). A contagion effect will only occur if a party selects the first option, by absorbing the policy position of their opponent, and discourses perceived as being popular with the public (Schworer 2021:24). Many scholars have claimed that mainstream parties opt to utilise the accommodative strategy (Mair 2002; Mudde 2004; van Spanje 2010). The electoral success of nativist and anti-immigration parties pressures mainstream parties to adopt similar stances (Abou-Chadi 2014:423).

Social media as a source of data

The rapid growth of the number of social media users has made these platforms attractive to political parties and candidates since they can utilise them to inform, mobilise and cultivate support (Ceron 2017:8). The same applies for the citizenry since social media sites such as Twitter have enhanced the intensity and scope of information and communication exchange (Bosseta et al. 2017:53). Therefore, social media can potentially impact the agenda-building processes and affect the spread of issues among the mass media, political actors, and voters alike (Conway et al. 2015:364). In the case of crisis communications, these platforms have the potential to foster political innovation (Michailidou 2017:241). As previously mentioned, government policies and actions are as important as the dialectic process where these actions are debated and evaluated during a crisis.

In the case of political communication in the social media sphere, two characteristics have been repeatedly identified as the most important: interactivity, and to a lesser extent, political personalisation (Kruikemeier 2014:132). These characteristics are the reason for choosing Twitter as the data source for this article. Interactivity is the opportunity for reciprocal communication (two-way) (Kruikemeier 2014:132). Twitter offers the unique opportunity for citizens to directly communicate with political actors in real time and without any filtering (Parmelee and Bichard 2011:133, 135). Users can share and comment each other’s posts, as opposed to an official party site where information in the form of press statements is one-way. Second, the element of personalisation allows users to utilise Twitter for self-promotion (Kruikemeier 2014:132). Social media are an alternative way for politicians to reach-out to the electorate in a more personal manner.

In addition, while the recent literature in critical terrorism studies has begun to focus on social media, most of the online terrorism scholarship mainly examines how extremism and dissent are spread through social media channels (Downing et al. 2022:240). Most importantly, scholarship on the discourse of the two case studies has either focused on the election periods (Maurer and Diehl 2020), or in the case of Le Pen her agenda setting modus operandi (Alonso-Munoz and Casero-Ripolles 2018). Therefore, this article makes two key contributions: (1) it draws its data from a setting that has not been widely explored in critical terrorism studies and (2) it focuses on the context of crisis, which has not been explored when examining the framing of the two politicians.

Data collection and methodology

The “rtweet” package of the R programming language was utilised for the collection of the tweets. The initial number of tweets was n = 3200 for each politician, as per the restrictions set by Twitter. The next step was to narrow down the number of tweets around the period that the two incidents occurred from 16 October to 15 November 2020. The choice of the time frame allows for a better examination of the overall salience of the issue of terrorism, since the aftermath of the attacks is also included. During this period, Le Pen tweeted 115 times and Macron 130. Macron’s leadership position entails communicating with the public, and providing assurance that the situation is under control, and that remedial actions are being already planned or implemented. The data analysis section will delve deeper into the content of both politicians’ tweets.

After narrowing down the time frame, the data were analysed via the LDA generative statistical model. The analysis utilises unsupervised machine learning algorithms for processing. These algorithms attempt to uncover “hidden structures” in unlabelled data, and one of the most popular ways to accomplish this is through topic modelling (Guo et al. 2017:336). The algorithm attempts to break down the data into contributions from multiple topics shared by all the data, but to different extents (Guo et al. 2017:336). Topic modelling is usually based on latent Dirichlet allocation (LDA) (Blei et al. 2003), and it has been chosen as one of the two methodologies for examining the online framing of Macron and Le Pen on terrorism. LDA has been utilised in studies in political communication, such as measuring agenda setting in interactive political communication (Rossiter 2021), or if communication by German political parties converges to topics emphasised by populists on Facebook (Stier et al. 2017).

LDA seeks to automatically discover the main themes from a large corpus of texts (Brookes and McEnery 2019:4). The principal output of this process is lists of words which co-occur within the selected corpus (Brookes and McEnery 2019:4). Once the corpus has been uploaded, LDA searches for patterns of word co-occurrence within the texts, and if they tend to regularly co-occur it will assume that there is a relationship between them. In essence, the data are treated as arising from a generative process that includes hidden variables (Blei 2012:79). The observed variables are the words in the documents, and the hidden ones are the topic structure (Blei 2012:80). In LDA, topics are essentially content-related categories, and they are heuristically located on the intermediate level between the corpus and the documents (Maier et al. 2018:94). Figure 1 showcases the application of LDA to a corpus. For this article, the corpus is comprised of the entirety of the tweets, and each tweet is document.

Fig. 1.

Fig. 1

Application of LDA to a Corpus (Maier et al. 2018:94)

Therefore, one of the main advantages of LDA is it allows for the identification of both the topics and the context through its bag-of-words assumption (Hall et al. 2019:122). The ordering of the words in a document is ignored, since it is sufficient to describe the distribution of words in a document, to uncover its thematic structure (Maier et al. 2018:95). This methodological tool is akin to reduction techniques utilised in content analysis. Data reduction is not separate from analysis since the data is sorted, focused, and organised based on the researchers’ decisions (Milles and Huberman 1994:11). Therefore, data reduction techniques must be deliberate to an extent. For example, a way to manage or limit the large volume of data produced through qualitative research is focusing on specific themes, extracting the data coded at specified codes of interest and creating a qualitative matrix that includes both the coded text and other demographic or personal characteristics (Guest et al. 2012:129, 130). LDA effectively eliminates this process.

The topics in the output incorporate all the texts in the corpus. Each text may be strongly associated with many of the topics, just a few of them or none (Brookes and McEnery 2019:5). For each of the texts, a score is produced that shows the strength of each topic in each text in the corpus (Brookes and McEnery 2019:5). The first step in this process is to clean the data. In the case of tweets and include tokenisation (breaking down documents to term components), the removal of punctuation marks and of stop words (Maier et al. 2018:98).

Another important step when utilising LDA is the selection of the ideal number of topics (K). In general, a greater number of topics ensure that they will be more specific and narrower. However, too many topics might result into entities that cannot be easily distinguished (Maier et al. 2018:98), whereas a small number of topics may lead to large entities that incorporate a plethora of aspects that should have been separated (Evans 2014:2). The next step was to utilise the “stm” package to find the appropriate number of topics for each politician. Two different metrics were examined for this purpose: (1) semantic coherence and (2) exclusivity. The core principle of semantic coherence is that the most probable words in a topic should co-occur within the same document (Mimno et al. 2011:265, 266). This metric correlates well with human judgement on topic quality. Semantic coherence is easily achievable if there is a small number of topics that are dominated by very common words. Therefore, exclusivity acts as a counterbalance. If words with a high probability under a topic have low probabilities in other topics, then that topic is exclusive. Topics are more semantically useful when they are both cohesive and exclusive (Roberts et al. 2014:7).

According to Fig. 2, the appropriate number of topics for Macron is K = 60, with a high enough exclusivity and high semantic coherence.

Fig. 2.

Fig. 2

Number of topics (K): Emmanuel Macron

The model diagnostics for Le Pen also show that the appropriate number of topics is at K = 60, according to the semantic coherence and exclusivity metrics as shown in Fig. 3. Thus, both politicians have framed the exact number of topics in their Twitter discourse. It would have been expected that Le Pen would focus almost exclusively on the issue of terrorism, and as a result her number of topics would have been smaller. Consequently, a closer examination of the two topic models is paramount. The next step is to plot the topics in hierarchical order and examine the overall salience of terrorism in their framing.

Fig. 3.

Fig. 3

Number of topic (K): Marine Le Pen

However, examining the overall salience of the topics is the first step in the analysis. While the topics are algorithmically produced by the software, they still need to be interpreted by the researcher regarding their thematic coherence and their overall content (Brookes and McEnery 2019:5). It is for this reason that the utilisation of topic modelling will be followed by a framing analysis of the tweets on terrorism. The second step in the analysis will allow to discern whether there are elements of populist contagion in Macron’s discourse or not, by examining the diagnostic and prognostic framing of the two politicians.

Issue salience

Due to the large number of topics, the decision was made to visualise and examine only the top 20 topics in the Twitter discourse for each politician. However, two charts have been added in the “Appendix”, which include the other less salient topics. R does not include an option to label the topics, and as a result their names can only include a numerical value.

The two highest topics in Macron’s framing are related to terrorism. Macron emphasises the importance of solidarity during these challenging times and references the violence against women under radical Islam, as it can be observed from topics 28 and 52. The third highest topic is about the COVID-19 pandemic and the campaign against misinformation. The next six topics in Fig. 4 deal with terrorism exclusively. Topic 18 is about the terrorist attack that took place in Austria on 02/11/2020. Macron mentions that the entirety of his nation and Europe stand by the side of Austria in their sorrow. The concept of solidarity is again preeminent in Topic 33, since the issue of terrorism needs to be tackled by the European Council. Macron presents the fight against terrorism as a collective effort not only among the local populace but the entirety of Europe.

Fig. 4.

Fig. 4

Emmanuel Macron topics

The next topic deals with the terrorist attack perpetrated by ISIL (Islamic State of Iraq and the Levant) at the University of Kabul. What is important to note is that Macron frames terrorism as an international threat. He does not target a specific minority within the borders of his own country and thus stresses the importance for collective action at the international level. Topic 1 reiterates his desire to protect the French people against the threat of terrorism. Topic 17 again deals with the pandemic and how the spread of the virus can be controlled. The next two topics, 55 and 46, focus on the issue of terrorism. The former again references the importance of cooperation between the members of the European Council, while the later focuses on the importance of equality between the two sexes.

Topic 29 does not follow the trend observed in prior topics, since it references the author Maurice Genevoix, and the transferring of his remains to the Panthéon. However, the next two topics, 26 and 53, again reference the issue of terrorism. More specifically, Macron references the actions that he has taken to combat terrorism both at the national and international levels. Topic 42 is about the safeguarding of universal values, such as freedom of speech, and is related to Paty’s murder. Topic 34 focuses on the pandemic, and how it has negatively impacted traders, along with a reference to other countries that took measures earlier than France such as the Netherlands. Topics 39 and 44 again deal with terrorism and Macron’s commitment to protect the freedoms of the French Republic. In addition, he mentions that violence is never justified, since secularism is one of the founding pillars of the Republic. The final topic on the list references the death of Charles de Gaulle and is taken from tweets posted on 09/11/2020, 50 years after his passing.

The most salient topic for Le Pen is about the French Republic, and the need to prosecute the ones that disseminate the ideology that threatens it. Topic 18 is a criticism towards the EU due to its financing of Turkey. According to her, both the EU and Erdogan are responsible for the influx of immigrants to France and other European countries. The third most salient topic is a criticism towards the government’s plan for the second COVID wave. Therefore, the overall salience of the top 3 topics follows a similar pattern for both political actors, with terrorism and the pandemic being the main foci. Topics 38 and 11 are again focused on terrorism, with Le Pen utilising a harsh phraseology by labelling the incidents as an act of barbarism. Topic 11 is entirely comprised of hashtags, but it can be observed that Le Pen references the recent attack on 29/10/2020, and the wounding of a Greek Orthodox priest in Lyon on 31/10/2020.

Topic 35 focuses on the dissolution of organisations such as the CCIF (Collectif contre l'islamophobie en France, Collective Against Islamophobia in France), which Le Pen views as hotspots for the dissemination of terrorist ideology. Topic 36 again focuses on the pandemic and the confinement measures that have been imposed, with Le Pen labelling the entire plan as a catastrophe. Topic 30 includes criticisms towards specific problems within France such as an analysis on medicine shortages during the pandemic, and the abolition of postal voting to prevent fraud. The next topic is a criticism towards unchecked immigration and how it creates a fertile ground for radical ideologies and terrorism. The next 7 topics all focus on terrorism. For example, in Topic 29 Le Pen characterises the situation as a war that threatens the values of the Republic.

In topic 19, she attacks government officials, whose laxity caused the tragic death of Samuel Paty. Paty’s murder is again referenced in topic 27, where she also attributes blame to Erdogan, who clashed with Macron when he described Islam as a religion in crisis worldwide. Topic 5 is the commemoration of De Gaulle’s death, and how it is a reminder that France possesses both men and women ready to go into the offensive and restore the nation’s glory. Topics 12 and 26 focus on Covid, specifically the vaccination program and the detrimental effect of the pandemic on businesses. The two remaining topics, 32 and 25, focus on terrorism, and Le Pen attacks the current and previous governments for continuously compromising when it comes to the issue of Islam in schools, and petitions for the expulsion of immigrants who spread an ideology of hate.

As it can be observed from Figs. 4 and 5, the topics of terrorism and COVID largely dominated the discourse of the two politicians. Both topics were highly salient during the 1-month period, with the commemoration of De Gaulle’s death being the only deviation in this trend. However, a qualitative analysis is also important to uncover the differences in framing and examine whether there are indeed elements of populist contagion in Macron’s framing. Therefore, 1/10th of the total tweets was chosen for both politicians. Afterwards, the tweets associated with topics such as terrorism, Islamism and border controls were isolated and put through a randomiser designed with the R programming language. Although the majority of the tweets were posted on the same days, there was one exception that will be explained in the framing analysis section.

Fig. 5.

Fig. 5

Marine Le Pen topics

Tweets

Macron

16/10/2020

“But more broadly, I want to say this evening to all the teachers in France that we are with them and that the whole nation that the whole nation will be at their side today and tomorrow to protect them, to defend them, to enable them to do their profession, which is the most beautiful of them all: to make citizens free. And it is no coincidence that tonight it is a teacher that was killed by this terrorist.”

20/10/2020

“As of tomorrow, in the Council of Ministers, we will have to dissolve the de facto grouping of Cheikh Yassine. Other acts will follow. Our determination is complete.”

20/10/2020

“The words we got them. Evil we named it. We know what needs to be done. Our fellow citizens are waiting for action. The acts are there, and we will intensify them.”

25/10/2020

“Freedom, we cherish it; equality, we guarantee it; fraternity, we live it with intensity. Nothing will make us back down, ever.”

25/10/2020

“Our history is that of the struggle against tyrannies and fanaticisms. We will continue.”

25/10/2020

“We will continue. We will always stand on the side of human dignity and universal values.”

29/10/2020

“Nice, Nice. This is the third time that Islamist terrorism has struck your city, its inhabitants. I know the shock you feel, with you the whole country, and I believe it, the whole world. Let us not give in to any spirit of terror.”

29/10/2020

“Catholics, you have the support of the whole Nation. Our country is our values, that everyone can believe or not believe, that each religion can be exercised. Our determination is absolute. Actions will follow to protect all our fellow citizens.”

29/10/2020

“It is France that is under attack. I have therefore decided that our soldiers will be more mobilized in the coming hours. As part of Operation Sentinel, we will go from 3000 to 7000 soldiers.”

31/10/2020

“Contrary to what I have heard and seen a lot on social media in recent days, our country has no problem with any religion. They all practice freely! No stigmatization: France is attached to peace and to living together.”

31/10/2020

“I have seen too many lies, I want to be clear: what we are doing in France is to fight terrorism which is done in the name of Islam, not Islam. It claimed the lives of over 300 of our fellow citizens!”

31/10/2020

“They attributed me to "supporting caricatures of the Prophet that are insulting." I support being able to write, think and draw freely in my country. It is a right and it is our freedom. I realize this can be shocking and I respect that, but we have to talk about it.”

Le Pen

16/10/2020

“A professor beheaded for having presented the #CharlieHebdo cartoons: we are, in France, at this level of unbearable barbarism. Islamism is waging war on us: it is by force that we must drive it out of our country.”

20/10/2020

“Every Frenchman should read this testimony to the cynical abandonment of our republican school and of the teachers who struggle there against the terrible pressure of mad immigration and protest communalism. French, wake up! MLP”

20/10/2020

“Since terrorism is an act of war, it requires legislation of war. This exceptional legislation will comply with the rule of law, directed against terrorists, and guarantee the freedoms of citizens.”

25/10/2020

“We must move away from a certain number of articles of the" European Convention on Human Rights ", as Denmark, Ireland and Great Britain have done. Immigration, the French nation must be absolutely sovereign.”

25/10/2020

“A whole series of elected officials, and not the least, have compromised in relations with Islamist fundamentalism, with organizations that promote this Islamism, for purely electoral reasons. I have not heard a single mea culpa!”

25/10/2020

“The veil has accompanied the rise of Islamism in our country. For Islamists, the veil is a way of working for the appropriation of public space. I want it to be banned in public space.”

29/10/2020

“The dramatic acceleration of Islamist acts of war against our fellow citizens and our country is forcing a comprehensive response from our leaders aimed at eradicating Islamism from our soil.”

29/10/2020

“All our thoughts to the victims, their families, the Catholics and the people of Nice once again struck by Islamist barbarism.”

29/10/2020

“Islamism is an ideology which makes war on us. All those, associations, structures, men, women, who support this ideology, who disseminate it: they must be put out of harm's way. Through to the rule of law, through specific legislation that would be voted here in the National Assembly. But we do not have the feeling that the government has become aware of this situation.”

30/10/2020

“Is the war against Islamist ideology really being implemented? Are the measures announced by the government sufficient when we are attacked on our territory? I'm sorry to say the answer is no.”

30/10/2020

“I assume this role of politician: I want to be in power to make the decisions that will protect the French. We must know who we are fighting against, we must not be afraid of being accused of being ‘Islamophobic’!”

30/10/2020

“We must dissolve the CCIF and the UOIF, structures which have within them and develop an Islamist thought, which refused to be heard by the senatorial commission of inquiry on Islamist radicalization!”

Framing analysis

The first tweet to be analysed is a statement by Macron after Paty’s murder. His framing performs three key functions. First, he mentions the severity of the situation, and how the attack is threatening to republican values. The framing then focuses on identifying terrorism as the root of the problem and proffers a moral evaluation. However, these two framing functions do not foster division like Le Pen’s framing of the incident. He mentions the importance of education in creating free citizens irrespective of nationality or religious dogmas and states that the entire nation will stand by the teachers. The only function that is missing is a proposed solution, but this is an announcement after the attack’s immediate aftermath, and Macron must reassure the citizens that everything is under control. Le Pen’s framing of the same incident is more aggressive but performs all four functions identified by Entman. Islamism is identified as the tragedy’s cause, and the event as an act of barbarism. She describes the situation in France as a war and calls for the application of immediate and decisive force. However, her solution does not include a clear policy alternative. Her framing is an appeal to emotion, since she seeks to foster division and escalate the sense of urgency caused by the incident.

The next group of tweets were posted on 20/10/2020. Macron’s tweets define the problem, identify its causes, and present the government’s actions against radicalisation and terrorism. Moral evaluations are only present in the second tweet, which is devoid of any solutions to the problem. However, while the tweet presents the fight against terrorism as a battle of good versus evil, it does not create scapegoats by attacking Islam as opposed to Le Pen. Moreover, all the tweets showcase that the government is willing to take decisive action and safeguard the values of the Republic. Le Pen’s framing is again aggressive when defining the problem and providing moral judgements. The first tweet references an online article and harshly criticises the government for abandoning the teachers and allowing schools to become breeding grounds for the propagation of radicalism. She identifies unchecked immigration and communalism as the causes of the pathogeny. The tweet does not propose any policy alternatives and only seeks to attribute blame. Her second tweet while short performs all four framing functions. However, its structure is more like a slogan. It creates a sense of urgency by presenting the situation as a war again and calls for the implementation of exceptional measures. While the tweet is impactful, it does not provide any concrete solutions to the problem as opposed to Macron.

The tweets from 25/10/2020 follow a similar pattern to previous ones with some key differences. In the first tweet, Macron seeks to foster unity among the French people. The diagnostic component of the framing does not make an explicit mention to terrorists. However, by presenting the values of the French Republic such as freedom, fraternity, and equality he makes the juxtaposition with the terrorists seeking to undermine them. This serves as the moral evaluation component of the framing that also identifies the cause of the problem. In addition, Macron does not present any solutions, but he clearly states that the entire nation will stand united against terrorism. The other tweets are similar in their framing since the central message is the safeguarding of universal values. Furthermore, unity is presented as the antidote for an ideology of hatred and mentions that both the state and the citizens have their role to play. Macron presents himself as President for all the French irrespective of ideological or religious dogmas in accordance with the core principles of his party.

Le Pen’s framing from has remained bellicose on her group of tweets. Her first target is the EU, and the Convention of Human Rights regarding the freedom of movement. She identifies the Schengen Agreement as part of the problem and mentions the European countries that have opted out, claiming that France should follow a similar path. Furthermore, the opting out of the Schengen Agreement is linked with the country’s sovereignty, something which has been one of the key commitments of Le Pen’s party. In her second tweet, her attack is focused on both elected officials and the organisations that promote Islamic fundamentalism. The framing in this tweet includes both diagnostic components, since she identifies how severe the problem is and attributes blame to both the politicians and these organisations. Most importantly, the framing does not contain any solutions, but the element of moral evaluation is the focal point again. Le Pen accuses the politicians of callousness and greed since they made compromises with these organisations for electoral reasons. In her final tweet, she focuses on the issue of veils worn in public spaces. Her framing includes all the aforementioned functions, since she identifies the problem and labels it as an appropriation of public space. Furthermore, she links the veils with the rise of Islamic fundamentalism and proposes the practice to be banned.

Overall, certain observations can be made on the framing of Paty’s murder and its aftermath by the two politicians. Le Pen framing exemplifies the populist style of communication. Her framing is divisive, not only between French nationals and Muslims, but also between the people and an uncaring political elite. Most importantly, moral evaluations play a pivotal role in her framing. The division between “the people” and “the enemies” is also moral. Terrorists wish to sow discord and fear, while the political elites are motivated only by their personal gain. Finally, not all her tweets contain policy proposals for counteracting the issue of terrorism, placing an emphasis on the diagnostic component. Some of Macron’s tweets also lack proposed solutions. However, this omission is expected since Twitter is mostly utilised for brief and immediate communication. Furthermore, the tweets including the moral evaluation component do not seek to divide. Instead, he presents the fight against terrorism as a collective endeavour for both the state and the citizenry against a radicalised minority.

The next group of tweets were posted on 29/10/2020 in the aftermath of the attack in Nice. Macron’s first tweet provides a diagnosis of the issue, a moral evaluation, and a solution. However, it is important to note that the solution is not a policy proposal but rather a call to resist giving in to hate. Macron mentions prior attacks that have happened in Nice to showcase the gravity of the situation and mentions that their anguish is shared by everybody. He still calls for unity against the division sown by terrorists and thus combines the framing functions of moral evaluations and solutions. The next tweet is similar in content since Macron presents the attack as a clash between the entire French nation and the terrorists. Most importantly, he mentions the importance of values such as religious freedom and how the terrorists seek to undermine them. Again, the diagnostic component is intertwined with the moral evaluation to highlight the importance of unity. While Macron does not provide a detailed description of policy solutions to tackle terrorism, he promises that decisive action will immediately follow.

In the third tweet, Macron mentions that France itself is under attack. The tweet is short, but it includes all four key framing functions. He mentions how severe the situation is and presents it as an attack to the entire nation. Furthermore, the previous tweets laid the groundwork for the moral evaluation component since he mentioned the values of unity, religious freedom, and freedom of expression as the foundations of the French nation. As previously mentioned, he seeks to foster unity among the French citizens and showcase that the nation is not defined by geographic boundaries or ethnicity. For Macron, the nation is first and foremost defined by its values, and their safeguarding necessitates decisive action such as Opération Sentinelle, and the increase in security personnel. This is the only tweet that includes a tangible solution to the issue at hand, but as previously mentioned the initial period after the attack is crucial for providing the citizenry with a sense of security. Therefore, Macron first identifies the cause of the problem, recognises its severity, fosters unity through his moral evaluations and finally showcases his government’s response to the attack.

Le Pen did not tweet as frequently as Macron in the immediate aftermath of the attack, posting 5 tweets as opposed to Macron who tweeted 9 times. As previously mentioned, the leadership of the country must shoulder a heavier burden, since they need to show that they are in control of the situation. However, Le Pen could have capitalised on the immediate aftermath of the attack especially since it happened right after Paty’s murder. In her first tweet, all four framing functions are present. She identifies Islamism as the cause of the pathogeny and describes the situation as a war. And this war calls for exceptional measures since the future of France is at stake. Therefore, such measures are justified, and they should also prompt political leadership into immediate action. The next tweet performs only three of the framing functions: (1) providing a diagnosis, (2) evaluating the severity of the situation, and (3) providing a moral judgment. Le Pen offers her sympathy to the victims while mentioning the truck attack that took place on 14/07/2016. In doing so, she showcases how terrorism is a constant threat that must be dealt with. Finally, she labels the attack as an act of barbarism and thus provides a moral evaluation typical of populist communications.

The final tweet for Le Pen performs all four framing functions. First, she identifies the cause of the problem as the associations and individuals that promote an ideology of hatred. She once again frames the situation as a war caused by a bellicose ideology. This particular framing not only accentuates the severity of the situation but can also be utilised as a justification for proposing a more aggressive approach in resolving the issue. Moreover, describing the situation as war serves as a moral evaluation since the fight against terrorism is presented as a fight for survival for the French nation and its people. Most importantly, she suggests that the National Assembly must vote for legislation that will allow to dismantle this associations and incarcerate the individuals comprising them. The proposed solutions also attributes blame to the other parties, by claiming that they seem unwilling to act. Le Pen seeks to divide once more not only towards the Muslim population, but also the political system. She claims that there is a solution, but the indifference and hesitation of the governing elite is preventing it from being implemented, and thus presents herself and her party as the only ones capable of ending the crisis.

The next set of tweets were posted on different days. The reason behind this choice is that Macron did not tweet on 30/10/2020, while Le Pen was quite prolific with 10 tweets in total. However, the tweeting activity of the two politicians is reversed on 31/10/2020 with Macron tweeting 14 times and Le Pen twice. Both of her tweets did not reference the terrorist attack in Nice. Therefore, a compromise had to be made when selecting the data, to ensure that the framing of both politicians is analysed. The first tweet posted by Macron contains the diagnosis and the moral evaluation but no policy proposals. Macron wishes to reassure the public that France is a secular state and that no stigmatisation occurs. He attributes the blame to individuals that wish to sow discord and calls the citizens to be united. The moral evaluation component is significantly different from Le Pen’s since he does not seek to divide French society into two opposing camps. He repeats that the core values of the Republic override any religious dogma. However, the tweet does not contain any policy solutions. As previously mentioned, Twitter promotes a more relaxed and personalised style of communication, so it is expected that the two politicians would not frequently include overtly detailed proposals on how to fight terrorism. The second tweet’s framing follows a similar pattern. Again, no policy proposals are present even though Macron talks about the fight against terrorism.

Macron makes it clear that they are fighting against terrorism committed in the name of Islam and not Islam itself. He mentions how severe the situation is by providing the number of victims, but his moral evaluation does not include any divisionary elements. The government’s target is a radical minority that has utilised religion as a reason to commit these atrocities. The final tweet under examination again showcases the differences in the two types of framing. Macron chastises individuals that spread rumours about his support of caricatures insulting the Prophet Muhammad. Once again, his diagnosis of the issue is devoid of overgeneralisations. He does not target the entirety of the Muslim population, but rather the radicalised elements. He showcases the severity of the issue by mentioning how important it is to safeguard the freedom of expression. Furthermore, in the moral evaluation of his framing he calls for dialogue to take place. Therefore, Macron seeks to build bridges as opposed to his adversary.

Le Pen’s first tweet on the terrorist attack includes both diagnostic components as well as a moral evaluation, but no policy alternatives are present. Typical of her modus operandi, Le Pen presents the situation as a war. Moreover, both Islamism and the government’s measures are identified as the source of the problem. Therefore, the moral evaluation component includes a two-pronged attack, first against Islam and then against the government’s laxity. It would be expected for her to include some brief proposal of countermeasures to terrorism, but her framing lacks this component. Therefore, its main functions are the diagnosis of the problem and blame attribution. The next tweet again lacks a policy proposal for combatting terrorism. Typical of most of her framing, the tweet includes many elements found in the populist style of communication. The diagnostic part of the framing again places the blame on the politicians and Islam. Furthermore, in her moral evaluation she states that she has assumed the mantle of the politician to protect the French people. Once again, she utilises a Manichean dualism to cast her enemies in the worst possible light.

Furthermore, she attacks her political opponents and the media by claiming that they have labelled her and RN as Islamophobic. According to her, other political parties are mostly preoccupied with good public relations and the maximisation of votes instead of France’s safety. The third and final tweet is the only one that performs all four key framing functions since it includes a solution to the issue of terrorism. She calls for the dissolution of the Collective Against Islamophobia in France (French: Collectif contre l'islamophobie en France) and French Muslims (formerly known as Union des Organisations Islamiques de France). This tweet does not contain many populist elements, but Le Pen still makes sure to present these two organisations as dangerous and disregarding of the law.

Overall, the framing of the terrorist attack in Nice by both actors is similar to Paty’s murder. However, Macron’s tweets feature the proposed solutions section more frequently. This disparity is not unexpected. Macron is the President of the Republic, and he needs to reassure the public that measures are being taken to resolve the issue. However, the small number of proposed solutions can also be attributed to the fact that Twitter is mostly suitable for a concise style of communication. In addition, the differences identified in the framing of the two politicians are still present. Le Pen is still divisive and aggressive in her framing, whereas Macron avoids creating scapegoats. By presenting the fight against terrorism as a collective endeavour, Macron avoids utilising any of Le Pen’s framing elements. Most importantly, in Macron’s framing the nation is built upon values such as tolerance, unity and liberty. Therefore, these values supersede any type of ethnic origin or belief system.

Conclusion and discussion

The analysis has shown that there were not any significant elements of “contagion” within Macron’s discourse. The topics related to terrorism, immigration and security were more frequently utilised, but this to be expected to occur during the initial period and afterwards. As previously mentioned, when critical contingencies unfold, governments do not only have to choose the appropriate policy solutions but to also dominate political discourse by providing an authoritative account of the crisis. Therefore, public communication strategies are as vital as the policies aimed at containing and ending a crisis. If political actors are unable to convince the public and the media that they are in control, they become vulnerable to impressions of crisis mismanagement (Hindmoor and McConnell 2015:18).

Crises and leadership share a symbiotic relationship. The public experiences crises as episodes of threat and uncertainty, and the natural inclination in these events is to look to leaders to provide a solution (Boin and ‘t Hart 2003:544). Another factor that makes crisis management more complex is the fact that these events are dynamic and chaotic and are not neatly sequenced on a timescale (Boin and ‘t Hart 2003:545). Crises may wind down and then flare up again depending as much on how physical events unfold as on the framing and interpretation of these events by the public, the media and political actors (Boin and ‘t Hart 2003:545). The threat of terrorism has continuously loomed over the French Republic and has proven to be an important leadership challenge.

Effective political representation is dependent upon two interrelated processes: opinion leadership and the responsiveness of political actors. Opinion leadership, the elite-driven perspective, sees political actors as actively shaping public opinion to raise the salience of issues in which they are perceived as competent of solving (Braun et al. 2019:801). Responsiveness, or the mass-driven perspective, sees political actors as responding to the concerns of the public by adjusting issue emphases and positions (Braun et al. 2019:801). Overall, crises have the potential to promote change not only in terms of policy, but also party positions. As shown in the analysis, Macron did not alter his framing on the issue of terrorism to a significant extent, even though it could potentially provide the opportunity to garner increased political support.

The process of framing comes into play once more due to its importance in bringing issues to the political agenda. Most issues possess different dimensions and can be approached from different angles (Princen 2018:540). During the agenda-setting process, different sides of the debate will emphasise different dimensions by utilising different angles in interpreting events (Princen 2018:540). In the case of Macron, the issue of protecting both the citizens and democratic values is presented not only as important but also as a matter of tighter cooperation between the EU members. Most importantly, he avoids creating any divisions in French society. This is the key difference between Macron’s and Le Pen’s framing. Le Pen’s framing is based on divisions, just like every populist political actor. In her framing the “enemies” of the people are both the immigrants and political leaders.

Le Pen employed a more bellicose type of framing, where every Muslim was identified as a potential threat. Moreover, her policy proposals possess authoritarian undertones and remove any type of agency from the citizens, as opposed to Macron who presents the fight against terrorism as a collective effort. Emmanuel Macron avoided framing immigrants as potential threats and instead focused on the importance of improving both the security of France and the EU. Overall, Le Pen’s discourse on both terrorist incidents remained one-dimensional. In her case, it would make sense to take advantage of the French public’s fear or terrorism and Islam and maximise the resonance of her framing. Macron opted not to adopt the same strategy, even though it could prove beneficial for his electoral support.

Acknowledgements

I would like to thank Doctor Ine Goovaerts (KU Leuven) and Doctor Thomas Carl Lundberg (University of Glasgow), for their invaluable help in revising this article. I would also like to thank Doctor Johannes Gruber (Department of Communication Science- Vrije Universiteit Amsterdam) for tutoring me in R programming and text-as-data methods.

Appendix 1: Topics for Emmanuel Macron

graphic file with name 41253_2023_205_Figa_HTML.jpg

graphic file with name 41253_2023_205_Figb_HTML.jpg

graphic file with name 41253_2023_205_Figc_HTML.jpg

graphic file with name 41253_2023_205_Figd_HTML.jpg

graphic file with name 41253_2023_205_Fige_HTML.jpg

Appendix 2: Topics for Marine Le Pen

graphic file with name 41253_2023_205_Figf_HTML.jpg

graphic file with name 41253_2023_205_Figg_HTML.jpg

graphic file with name 41253_2023_205_Figh_HTML.jpg

graphic file with name 41253_2023_205_Figi_HTML.jpg

graphic file with name 41253_2023_205_Figj_HTML.jpg

Funding

The author received no financial support for the research, authorship and/or publication of this article.

Declarations

Conflict of interest

The author declares that there is no conflict of interest.

Footnotes

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

  1. Abou-Chadi Tarik. Niche Party Success and Mainstream Party Policy Shifts- How Green and Radical Right Parties Differ in their Impact. British Journal of Political Science. 2014;46(2):417–436. doi: 10.1017/S0007123414000155. [DOI] [Google Scholar]
  2. Alonso-Munoz Laura, Casero-Ripolles Andreu. Communication of European Populist Leaders on Twitter: Agenda Setting and the ‘More is Less’ Effect. Información Política y Redes Sociales (II) 2018;27(6):1193–1202. doi: 10.3145/epi.2018.nov.03. [DOI] [Google Scholar]
  3. Bartolucci Valentina. The Perils and Prospects of the French Approach to Counterterrorism. In: Romaniuk Scott, Grice Francis, Irrera Daniela, Webb Stewart., editors. The Palgrave Handbook of Global Counterterrorism Policy. Basingstoke: Palgrave Macmillan; 2017. pp. 437–438. [Google Scholar]
  4. Bastow Steve. The Front National under Marine Le Pen: A Mainstream Political Party? French Politics. 2018;16(1):19–37. doi: 10.1057/s41253-017-0052-7. [DOI] [Google Scholar]
  5. Blei DM. Probabilistic Topic Models. Communications of the ACM. 2012;55(4):77–84. doi: 10.1145/2133806.2133826. [DOI] [Google Scholar]
  6. Blei DM, Ng AYT, Jordan MI. Latent Dirichlet Allocation. Journal of Machine Learning Research. 2003;3:993–1022. doi: 10.5555/944919.944937. [DOI] [Google Scholar]
  7. Boin Arjen, Kuipers Sanneke, Overdijk Werner. Leadership in Times of Crisis: A Framework for Assessment. International Review of Public Administration. 2014;18(1):79–91. doi: 10.1080/12294659.2013.10805241. [DOI] [Google Scholar]
  8. Boin Arjen, ‘t Hart Paul. Public Leadership in Times of Crisis: Mission Impossible? Public Administration Review. 2003;63(5):544–553. doi: 10.1111/1540-6210.00318. [DOI] [Google Scholar]
  9. Boin Arjen, t’ Hart Pauk, McConnell Andrew. Crisis Exploitation: Political and Policy Impacts of Framing Contests. Journal of European Public Policy. 2009;16(1):81–106. doi: 10.1080/13501760802453221. [DOI] [Google Scholar]
  10. Bosseta M, Dutceac Segesten A, Trenz H. Engaging with European Politics through Twitter and Facebook: Participation Beyond the National? In: Barisione M, Michailidou A, editors. Social Media and European Politics. Basingstoke: Palgrave Macmillan; 2017. pp. 53–76. [Google Scholar]
  11. Braun Daniela, Adrian Popa Sebastian, Schmitt Hermann. Responding to the Crisis: Eurosceptic Parties of the Left and Right and their Changing Position Towards the European Union. European Journal of Political Research. 2019;58(3):797–819. doi: 10.1111/1475-6765.12321. [DOI] [Google Scholar]
  12. Brookes Gavin, McEnery Tony. The Utility of Topic Modelling for Discourse Studies: A Critical Evaluation. Discourse Studies. 2019;21(1):3–21. doi: 10.1177/1461445618814032. [DOI] [Google Scholar]
  13. Ceron Andrea. Intra-party Politics in 140 Characters. Party Politics. 2017;23(1):7–17. doi: 10.1177/1354068816654325. [DOI] [Google Scholar]
  14. Conway Bethany A, Kenski Kate, Wang Di. The Rise of Twitter in the Political Campaign: Searching for Intermedia Agenda-Setting Effects in the Presidential Primary. Journal of Computer-Mediated Communication. 2015;20(4):363–380. doi: 10.1111/jcc4.12124. [DOI] [Google Scholar]
  15. de Vreese Claes H. New Avenues for Framing Research. American Behavioral Scientist. 2012;56(3):365–375. doi: 10.1177/0002764211426331. [DOI] [Google Scholar]
  16. Downing Joseph, Gerwens Sarah, Dron Richard. Tweeting Terrorism: Vernacular Conceptions of Muslims and Terror in the Wake of the Manchester Bombing on Twitter. Critical Studies on Terrorism. 2022;15(2):239–266. doi: 10.1080/17539153.2021.2013450. [DOI] [Google Scholar]
  17. Downs Anthony. An Economic Theory of Democracy. New York: Harper Collins; 1957. [Google Scholar]
  18. Entman Robert M. Projections of Power: Framing News, Public Opinion and U.S Foreign Policy. Chicago: University of Chicago Press; 2004. [Google Scholar]
  19. Evans Jocelyn, Ivaldi Gilles. An Atypical ‘Honeymoon’ Election? Contextual and Strategic Opportunities in the 2017 French Legislative Elections. French Politics. 2017;15(3):322–339. doi: 10.1057/s41253-017-0040-y. [DOI] [Google Scholar]
  20. Evans Jocelyn, Ivaldi Gilles. The 2017 French Presidential Elections: A Political Reformation? Basingstoke: Palgrave Macmillan; 2018. [Google Scholar]
  21. Evans, Michael S. 2014. A computational approach to qualitative analysis in large textual datasets. PLoS One, 9(2):1–10. 10.1371/journal.pone.0087908 [DOI] [PMC free article] [PubMed]
  22. Guest Greg, MacQueen Kathleen M, Namey Emily E. Applied Thematic Analysis. Thousand Oaks: Sage Publications; 2012. [Google Scholar]
  23. Guo Yue, Barnes Stuart J, Jia Qiong. Mining Meaning from Online Ratings and Reviews: Tourist Satisfaction Analysis Using Latent Dirichlet Allocation. Tourism Management. 2017;59:467–483. doi: 10.1016/j.tourman.2016.09.009. [DOI] [Google Scholar]
  24. Hall Margaret, Logan Michael, Ligon Gina S, Derrick Douglas C. Do Machines Replicate Humans? Towards a Unified Understanding of Radicalizing Content on the Open Social Web. Policy & Internet. 2019;12(1):109–138. doi: 10.1002/poi3.223. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Heath-Kelly C. Critical Approaches to the Study of Terrorism. In: Chenoweth E, English R, Gofas A, Kalyvas SN, editors. The Oxford Handbook of Terrorism. Oxford: Oxford University Press; 2019. pp. 224–237. [Google Scholar]
  26. Hindmoor Andrew, McConnell Allan. The Contradictions and Battlegrounds of Crisis Management. In: Karyotis Georgios, Gerodimos Roman., editors. The Politics of Extreme Austerity: Greece in the Eurozone Crisis. Basingstoke: Palgrave Macmillan; 2015. pp. 15–33. [Google Scholar]
  27. Hulsse Rainer, Spencer Alexander. The Metaphor of Terror: Terrorism Studies and the Constructivist Turn. Security Dialogue. 2008;39(6):571–592. doi: 10.1177/0967010608098210. [DOI] [Google Scholar]
  28. Hutchins Rachel D, Halikiopoulou Daphne. Enemies of Liberty? Nationalism, Immigration, and the Framing of Terrorism in the Agenda of the Front National. Nations and Nationalism. 2019;26:67–84. doi: 10.1111/nana.12555. [DOI] [Google Scholar]
  29. Jackson Richard. Writing the War on Terrorism: Language, Politics and Counterterrorism. Manchester: Manchester University Press; 2005. [Google Scholar]
  30. Jackson Richard. The Core Commitments of Critical Terrorism Studies. European Political Science. 2007;6(3):244–251. doi: 10.1057/palgrave.eps.2210141. [DOI] [Google Scholar]
  31. Jackson Richard. In Defence of Terrorism: Finding a Way through a Forest of Misconceptions. Behavioral Sciences of Terrorism and Political Aggression. 2011;3(2):116–130. doi: 10.1080/19434472.2010.512148. [DOI] [Google Scholar]
  32. Jackson Richard, Jarvis Lee, Gunning Jeroen, Breen-Smyth Marie. Terrorism: A Critical Introduction. Basingstoke: Palgrave Macmillan; 2011. [Google Scholar]
  33. Kruikemeier Sanne. How Political Candidates Use Twitter and the Impact on Votes. Computers in Human Behavior. 2014;34:131–139. doi: 10.1016/j.chb.2014.01.025. [DOI] [Google Scholar]
  34. Lachat Romain. Issue Ownership and the Vote: The Effects of Associative and Competence Ownership on Issue Voting. Swiss Political Science Review. 2014;20(4):727–740. doi: 10.1111/spsr.12121. [DOI] [Google Scholar]
  35. Lachat Romain, Michel Elie. Campaigning in an Unprecedented Election: Issue Competition in the French 2017 Presidential Election. West European Politics. 2020;43(3):565–586. doi: 10.1080/01402382.2019.1655960. [DOI] [Google Scholar]
  36. Maier D, Waldherr A, Miltner P, Wiedemann G, Niekler A, Keinert A, Pfetsch B, Heyer G, Reber U, Haussler T, Schmid-Petri H, Adam S. Applying LDA Topic Modelling in Communication Research: Toward a Valid and Reliable Methodology. Communication Methods and Measures. 2018;12(2):93–118. doi: 10.1080/19312458.2018.1430754. [DOI] [Google Scholar]
  37. Mair Peter. Populist Democracy Versus Party Democracy. In: Meny Yves, Surel Yves., editors. Democracies and the Populist Challenge. Basingstoke: Palgrave MacMillan; 2002. pp. 81–98. [Google Scholar]
  38. Manucci L, Weber E. Why the Big Picture Matters: Political and Media Populism in Western Europe since the 1970s. Swiss Political Science Review. 2017;23(4):313–334. doi: 10.1111/spsr.12267. [DOI] [Google Scholar]
  39. Maurer Peter, Diehl Trevor. What Kind of Populism? Tone and Targets in the Twitter Discourse of French and American Presidential Candidates. European Journal of Communication. 2020;35(5):453–468. doi: 10.1177/0267323120909288. [DOI] [Google Scholar]
  40. Meguid Bonnie M. Competition Between Unequals: The Role of Mainstream Party Strategy in Niche Party Success. American Political Science Review. 2005;99(3):347–359. doi: 10.1017/S0003055405051701. [DOI] [Google Scholar]
  41. Michailidou Asimina. Twitter Public, Engagement and the Eurocrisis: More THAN an Echo Chamber? In: Barisione Mauro, Michailidou Asimina., editors. European Politics: Rethinking Power and Legitimacy in the Digital Era. Basingstoke: Palgrave MacMillan; 2017. pp. 241–266. [Google Scholar]
  42. Milles B. Matthew, Huberman Michael A. Qualitative Data Analysis: An Expanded Sourcebook. Thousand Oaks: Sage Publications; 1994. [Google Scholar]
  43. Mimno, David, Wallach, Hanna M., Talley, Edmund, Leenders, Miriam and McCallum, Andrew. (2011). Optimising Semantic Coherence in Topic Models. Conference on Empirical Methods in Natural Language Processing. 262–272. http://dirichlet.net/pdf/mimno11optimizing.pdf.
  44. Moffit Benjamin. The Global Rise of Populism: Performance, Political Style and Representation. California: Stanford University Press; 2016. [Google Scholar]
  45. Mudde Cas. The Populist Zeitgeist. Government and Opposition. 2004;39(4):542–563. doi: 10.1111/j.1477-7053.2004.00135.x. [DOI] [Google Scholar]
  46. Pan Zhongdang, Kosicki Gerald M. Framing as a Strategic Action in Public Deliberation. In: Reese Stephen D, Gandy Oscar H, Grant August E., editors. Framing Public Life: Perspectives on Media and Our Understanding of the Social World. New Jersey: Lawrence Erlbaum Associates; 2001. pp. 35–66. [Google Scholar]
  47. Parmelee John H, Bichard Shannon L. Politics and the Twitter Revolution: How Tweets Influence the Relationship Between Political Leaders and the Public. Plymouth: Lexington Books; 2011. [Google Scholar]
  48. Perrotino Michel, Petra Guasti. Technocratic Populism à la Française? The Roots and Mechanisms of Emmanuel Macron’s Success. Politics and Governance. 2020;8(4):545–555. doi: 10.17645/pag.v8i4.3412. [DOI] [Google Scholar]
  49. Petrocik John R. Issue Ownership in Presidential Elections, with a 1980 Case Study. American Journal of Political Science. 1996;40:825–850. doi: 10.2307/2111797. [DOI] [Google Scholar]
  50. Princen Sebastiaan. Agenda-Setting and Framing in Europe. In: Ongaro Edoardo, Van Thiel Sandra., editors. The Palgrave Handbook of Public Administration and Management in Europe. London: Palgrave Macmillan; 2018. [Google Scholar]
  51. Roberts Margaret E, Stewart Brandon M, Tingley Dustin, Lucas Christopher, Leder-Luis Jetson, Gadarian Shana Kushner, Albertson Bethany, Rand David G. Structural Topic Models for Open-Ended Survey Responses. American Journal of Political Science. 2014;58(4):1064–1082. doi: 10.1111/ajps.12103. [DOI] [Google Scholar]
  52. Rossiter Erin L. Measuring Agenda Setting in Interactive Political Communication. American Journal of Political Science. 2021 doi: 10.1111/ajps.12653. [DOI] [Google Scholar]
  53. Schower Jakob. The Growth of Populism in the Political Mainstream: The Contagion Effect of Populist Messages on Mainstream Parties’ Communication. Cham: Springer Publishing; 2021. [Google Scholar]
  54. Snow D, Benford Robert D. Ideology, Frame Resonance, and Participant Mobilization. International Social Movement Research. 1988;1:197–217. [Google Scholar]
  55. Solheim Oyvind B. Are We All Charlie? How Media Priming and Framing Affect Immigration Policy Preferences after Terrorist Attacks. West European Politics. 2019;44(2):204–228. doi: 10.1080/01402382.2019.1683791. [DOI] [Google Scholar]
  56. Stier Sebastian, et al. When Populists Become Popular: Comparing Facebook Use by the Right-Wing Movement Pegida and German Political Parties. Information, Communication & Society. 2017;20(9):1365–1388. doi: 10.1080/1369118X.2017.1328519. [DOI] [Google Scholar]
  57. Stockemer Daniel. The Front National in France: Continuity and Change Under Jean-Marie Le Pen and Marie Le Pen. Cham: Springer Publishing; 2017. [Google Scholar]
  58. ’t Hart Paul, Tindall Karen. From ‘Market Correction’ to ‘Global Catastrophe’: Framing the Economic Downturn. In: ’t art Paul, Tindall Karen., editors. Framing the Global Economic Downturn: Crisis Rhetoric and Politics of Recessions. Canberra: ANU E Press; 2009. pp. 3–20. [Google Scholar]
  59. van Spanje Joost. Contagious Parties: Anti-Immigration and Their Impact on Other Parties’ Immigration Stances in Contemporary Western Europe. Party Politics. 2010;16(5):563–586. doi: 10.1177/1354068809346002. [DOI] [Google Scholar]

Articles from French Politics are provided here courtesy of Nature Publishing Group

RESOURCES