The Narrative Foundations of Radical and Deradicalizing Online Discursive Spaces: A Comparison of the Cases of Generation Islam and Jamal al-Khatib in Germany

Abstract

Radical/extremist Islamist actors use social media to disseminate uncompromising stories of monist religious political orders and identities. As a reaction, counter-movements to online Islamist radicalism/extremism emerged in Western societies (and beyond), while uncertainty about effective outcomes remains widespread. In a bid to understand how inclusionary and exclusionary discursive spaces are created, we ask: How do some Muslim actors create discursive spaces open to self-reflection, pluralism and liberal-democratic principles, while others construct illiberal, particularistic and non/anti-democratic spaces? To respond to this question, we compare two contrasting storytellers, one who agitates for exclusionary Islamist radicalism/extremism (Generation Islam) and one who offers inclusionary prevention and deradicalization work against that (Jamal al-Khatib). We draw on novel narrative approaches to the Discourse Historical Approach (DHA) in Critical Discourse Studies (CDS), via which we compare text-level and context-level narratives disseminated about three Muslim-related crises: the racist terrorist attacks/genocide to represent the national, European and global level. Our two-layered, DHA-inspired narrative analysis illustrates that, at the level of text, narrative persuasion varies between both contrasting actors. While Jamal al-Khatib disseminates persuasive stories, Generation Islam is much less invested in narrative persuasion; it seems to address an already convinced audience. These two text-level strategies reveal their meaning in two antagonistic narrative genres: Jamal al-Khatib’s “self-reflexive savior” creates an inclusionary discursive space represented in a self-ironic narrative genre, while Generation Islam’s ”crusading savior” manufactures an exclusionary discursive space represented in a romance featuring a nostalgic return to the particularistic Islamic umma.

1. Introduction

Islamist radicalization and extremism is a challenge to Western societies. While precise causal relationships and their directions between offline and online radicalization remain unclear thus far (cf. Somoano and McNeil-Willson 2022), it is fair to state that, in one way or another, as cause, means or factor, online mobilization plays a substantial role in Islamist radicalization. Recent research further stresses the crucial role played by the performative dimension (e.g., storytelling and visual aesthetics) of radical and extremist online content (Pieslak and Lahoud 2018). However, our contemporary knowledge is still very limited when it comes to understanding how Islamist radical and extremist discursive spaces materialize and how they compare to discursive spaces which seek to protect individuals from Islamist radicalization or aim to deradicalize them.

Thus, in a bid to theorize how inclusionary and exclusionary discursive spaces are created, we ask: How do Muslim actors create discursive spaces that are open and closed to self-reflection and liberal-democratic principles in Muslim-minority Western European contexts, i.e., in Germany?

Discursive inclusion and exclusion is one of Critical Theory’s key questions. When and how does critique succeed in revealing social inequalities and transform them, given social relations of domination and/or exclusion? Second generation critical theorists, such as Jürgen Habermas (1998) and Seyla Benhabib (1996), problematized the relationship between the construction of the nation-state, deliberative procedures and the exclusion of ‘the Other’. It is against this background that Critical Discourse Studies (CDS), in particular, the Discourse Historical Approach (DHA), emerged as an approach to analyze the context and forms of expressions in which “critiques” of social inequalities emerge (Reisigl and Wodak 2009). What renders the DHA particular is its focus on all semiotics that relate to societal knowledge structures. With that narrow focus, the DHA draws on pre-existing knowledge structures (topoi), intertextuality, interdiscursivity, argumentation and pluri-perspectivity in the analysis to reveal the content, strategies and empirical realization of discourse, i.e., of discursive intention. However, not every discursive intervention fulfils and/or fosters liberal-democratic principles, that is, some intervene for the (egalitarian) inclusion of “new” or dissimilar perspectives, while others intervene for egalitarianism within strictly defined boundaries. The latter promote particularistic closure. Far-right discursive interventions, for instance, push for illiberal democratic or anti-democratic polities (Mudde 2019); they restrict the set of legitimate interlocutors in discourse with reference to ancestry-based signifiers (Özvatan 2020). Thus, to assess the normative content of discursive interventions, i.e., to distinguishes true “critique” from banal reactionary criticism, we draw on a more elaborate approach which highlights the role of narrative. Narratives mark the ontological and epistemological foundation of social identities (Somers 1994), the coherent wholes which unveil deep-level structures of meaning associated with particularistic or pluralistic social identities (i.e., subjectivities).

In order to substantiate these theoretical mechanisms, we compare two Muslim actors who are said to speak to those two wings of the particularism-pluralism spectrum. On the rather Muslim-particularistic end, we selected the accounts of Generation Islam (GI) and Realität Islam (RI), while on the rather Muslim-pluralist end, we rest our analysis on Jamal al-Khatib’s (JaK) account. The latter is a Muslim actor who offers counter-radicalization content by delivering authentic alternative narratives to jihadist and Islamist-extremist online propaganda (Ali and Reicher 2020; Baron 2021), while Generation Islam & Realität Islam are considered to be part of the online presence of Hizb ut-Tahrir (Bundesamt für Verfassungsschutz 2020; Möller et al. 2021; Pankhurst 2016), a radical Islamist group founded in 1953, in East Jerusalem, by Palestinian politician and theologian Taqi ad-Din an-Nabahani.

It is important to note that we do not claim to (re)analyze whether these actors are truly radical/extremist or deradicalizer/preventer; instead, we analyze how they create discursive spaces in which openness and closure towards “new” or dissimilar perspectives are fostered.

Methodologically, we follow Forchtner’s (2021) narrative intervention in the DHA. In opposition to Reisigl’s (2021) contention that narrative is omnipresent and conceptually too blurry, we follow Forchtner and Özvatan’s (2022) approach which argues that in order to respond to narrative’s conceptual blurriness, one must refrain from an entire dismissal of narrative and, instead, conceptualize a more sophisticated strategy of its integration. They offer a novel technique for narrative’s conceptual and methodological integration by distinguishing between a deeper-level communicative context, the “storied context” (Forchtner and Özvatan 2022) or “narrative context” (Özvatan et al. 2022), and the text-level realization of communicative content—the latter (text-level communication) reveals its meaning in the context of the former (narrative context). In this article, we contribute to this conversation by offering a two-layered, DHA-inspired combination of narrative analysis: deep-level (i.e., narrative context) narrative genre analysis and text-level narrative persuasion analysis.

In so doing, we reveal the social mechanisms behind supply-side discursive communication by Muslim actors who perform particularistic, illiberal, anti-pluralist and, at times, anti-democratic discourses on the one hand, and those who perform universalist, liberal, pluralist and democratic discourses on “the Other”. In fact, our two-layered, DHA-inspired narrative analysis shows that both sides’ communicative supply varies in terms of both text-level narrative persuasion and narrative genre.

The Introduction includes a literature review of online Islamist radicalism and extremism, prevention and deradicalization and presents a novel approach to them: the theorization and methodological application of narrative therein. After that, we disclose our methodological approach, i.e., the novel two-level narrative analysis. In Section 3, we present our cases and data, and the findings of our two-level narrative analysis of online content by GI and JaK. This is followed by a chapter in which we discuss our findings vis-à-vis both actors and narrative levels (i.e., text and context). We close with a conclusion and future outlook on the forming of inclusionary and exclusionary discursive spaces in the online communication of Muslim actors in the West.

1.1. Research on Islamist Online Radicalization and Extremism

Conceptions of Islamism and Political Islam as terms and phenomena vary greatly in depth and focus and frequently lack clear definitions and delimitations; looking at a variety of approaches shall thus lead to more clarity. Tibi describes Islamism as “[…] collective memories of past Islamic glory to claim a return of history” (Tibi 2012, p. x); Mandaville (2014) explains it as politics that aim to establish a political order following Islamic law. Ayoob points to the necessity of speaking about multiple “Islamisms” (Ayoob 2004, p. 1) and sees the decontextualization of Islam as a central characteristic of the phenomenon (Ayoob 2005). Roy describes post-Islamism as “[…] the reaffirmation of the autonomy of the political, of the struggle for power, of the logic of national or ethnic interests, of the precedence of politics over religion” (Roy 2004, p. 4). It is within the scholarly discourse about Islamism and post-Islamism that we need to understand the role of “Islam” in “Islamist radicalization”. That is, one must elucidate “Islamisms” in the societal and discursive contexts in which they appear.

With regard to current understandings of radicalization and (violent) extremism (VE), Kruglanski et al. (2017) define extremism as “[…] motivated deviance from general behavioral norms […]” and add that extremism “[…] is assumed to stem from a shift from a balanced satisfaction of basic human needs afforded by moderation to a motivational imbalance wherein a given need dominates the others” (2017: 1). Beelmann (2020) suggests a ”Social-Developmental Model of Radicalization” and adds that “[…] political or religious extremism […] cannot be viewed as a form of political opinion or a religious orientation, but, above all, as an expression of deviant social development” (Beelmann 2020, p. 5)—following the line of Kruglanski et al. (2017); Gaspar et al. (2020, p. 1) propose understanding radicalization as a broad phenomenon that ranges “[…] from radicalization without violence to radicalization into violence and radicalization within violence”. Moghaddam (2022) slightly adapts his prominent “The staircase to terrorism” model and, among other points, emphasizes that “[…] an interactional process […]” seems to be an essential element of radicalization. Khalil et al. add to the discourse on modeling radicalization processes. Their “Attitudes-Behaviors Corrective (ABC) Model” seeks to conceptionally separate sympathy for violent behavior from involvement in it (Khalil et al. 2022, p. 425). The supremacy of one’s own group over others based inter alia on religion is a crucial element of extremist groups, according to Pearson et al. (2020, p. 31). Köhler’s study (Köhler 2021) on side-switching renounces earlier superficial attempts at “profiling” and instead engages in a close analysis of the personal storylines of individuals. Lösel et al. (2020) provide renewed insights on radicalization processes by looking at risk and protective factors that may lead into radicalization or that may protect against it. Altier et al. (2022) analyze whether specific roles taken on by individuals in extremist groups may correlate with the likelihood for these individuals to later disengage with or leave the groups. Gender perspectives add insights into the role of women in extremist ecosystems (e.g., Rothermel and Shepherd 2022). The importance of online worlds for radicalization and VE is well-known.

Baaken and Schlegel (2017) point to the specific influence of members of the ecosystem in online recruiting process and thereby also highlight the particularities of the internet for radicalization processes. Whittaker (2022) argues that treating online and offline radicalization and behavior as two distinct and separated spheres could mean that research misses their complex interconnectedness. Meleagrou-Hitchens and Kaderbhai (2017), among other aspects, point to the necessity of nuanced approaches based on data harvested from social media that would facilitate the close analysis of online behavior. Krona’s (2020) of the groups’ strategies on Telegram is one example for an analysis of how content and technical features of a specific platform are interconnected, and of how official IS channels and Pro-IS supporter channels collaborate. Conway (2017) explains in detail how future studies must critically engage with the exact role played by online worlds in radicalization processes.

While categorizations such as primary, secondary and tertiary prevention have been frequently used in theory and practice, terms associated with P/CVE1 often lack intersubjectively shared definitions. Baaken et al. (2020), against the background of applied P/CVE work in Germany, present a novel approach which neatly conceptualizes “deradicalization”. Along these lines, Silke et al. (2021) clarify differences between disengagement and deradicalization and Johansson et al. (2022) discuss evaluation processes in P/CVE work in multi-agency settings.

Both P/CVE work, as well as activities by extremist groups/individuals, have largely taken place online in recent years; thus the “gamification of extremism” found its way into scholarly debates. Lakhani (2021) introduces several platforms exploited by VE groups and the responses by the gaming and tech industry. Lakhani and Wiedlitzka (2022) focus on the Christchurch assailant in 2019 and his live streaming of the attack. Somoano and McNeil-Willson (2022) point to difficulties in “linking online behaviour [sic] to offline violence”. There are ongoing debates on whether VE online content should be deleted and if so how; with Tech against Terrorism and the Global Internet Forum to Counter Terrorism (GIFCT) taking leading roles on these questions.

Contemporary research on Islamism is divided into different foci: (1) global sociopolitical contexts and (2) the individual/group level of radicalization/VE. Looking at the global context, Hamid (2022) and Adraoui (2018) offer insights into current Islamist politics and ideas; Tibi (2002) provides critical evaluations of the securitization discourse. Despite slightly different foci, both Jensen and Larsen (2021) and Sheikh and Svensson (2022) discuss how individual and group behaviors may be connected in Islamist radicalization and deradicalization processes—and thereby also highlight the potential challenges of such approaches.

In P/CVE work in the field of religiously motivated extremism, critical evaluations of the role of Salafism in radicalization processes (Blanc and Roy 2021), and the continued development of theoretical frameworks for (sociopsychological) distancing work (Von Berg 2022), broaden the discourse. In recent years, and as the prosecution of European Foreign Fighters progresses, the attention paid to women in Jihadist organizations has increased (Koller 2021)—also leading to the emergence of gender-sensitive approaches in P/CVE work in the European context (RAN 2022).

The introduction of counter-/alternative narratives emerged as a promising way to support individuals vulnerable to religiously motivated extremism. Narratives play an ever bigger role in the radicalization process of Islamist extremism (e.g., Ingram 2017; Schmid 2014; cf. Schlegel 2021, p. 195; OIIP 2018; Junk et al. 2020; Briggs and Feve 2013). Thus, it seems crucial to consider narrative as a key site for radicalism/extremism and its prevention/countering.

1.2. Narratives and Deradicalization—“Studying the Brain without the Neurons”

We follow Ebner (2018, p. 21) who argues that “studying extremism without studying stories is like studying the brain without studying the neurons”. In ontological terms, we recognize the fundamental role of narrative (Somers 1994; van Laer et al. 2014; Kvernbekk and Frimannsson 2013; Hinyard and Kreuter 2007; Green and Brock 2002) and, more specifically, of narrative genre (Frye 1957; White 1973) in providing narratively structured contexts (lifeworlds) in which discursive realizations “make sense” (Forchtner and Özvatan 2022).

Following pledges for a “narrative turn” in the social sciences (Czarniawska 2004; see also Schlegel 2021; Nünning 2012) and Fisher’s (1984) postulate of the homo narrans, we understand the Lebenswelt to be strMuctured narratively. Consequently, to represent sociological knowledge production, one must reconstruct what we define as “narrative context”. That is, the narrative configuration of protagonists, events and emplotment within which text-level communicative content becomes meaningful. To distinguish between narrative’s two levels—the level of concrete text and the level of narrative context—Meretoja (2022) conceptualizes “explicit” and ”implicit” narratives. Explicit narratives entail concrete text-level realizations. They are “narrative artifacts”, hence “stories told by someone to someone in a concrete textual or other material form” (Meretoja 2022, p. 37). In contrast, implicit narratives entail more fundamental models of sense-making which underlie explicit narratives “but may not be anywhere available in a material form” (ibid.). As such, implicit narratives must be “constructed by interpreters of explicit narratives, which carry implicit narratives within them” (ibid.).

We extend Meretoja’s compelling distinction by developing a Habermas-inspired model for communicative action which theorizes “critique” as a narratively guided mechanism operating on both these levels of communication. In a narratively structured world, even highly rational ideal types of communication and discourse ethics are produced and evaluated through the lens of the narratively structured Lebenswelt, i.e., the narrative context. Moreover, we follow Frye (1957) and White (1973) who argue that the narrative context itself is structured into four narrative genres which preconfigure the realization of concrete text-level communicative content (cf. Forchtner and Özvatan 2022).

1.3. From Narrative Genre to Narrative Persuasion: Radicalism/Extremism and De/Counter-Radicalization in Light of Two-Layered Narrative Analysis

Forchtner et al. (2020) have demonstrated that, in Western societies, these four narrative genres—romance, comedy, tragedy and irony (Frye 1957)—can constrain (romance and comedy) or trigger (tragedy and irony) the inclusion of “Others” by delimiting or expanding the set of legitimate actors, perspectives and arguments in a discourse. Tragic and ironic emplotments allow for emancipatory and inclusionary critique by calling naive presumptions of the in-group’s purity into question, they induce mechanisms for inclusionary social reorganization. In contrast, romantic and comic emplotments stifle emancipatory and inclusionary critique by portraying an idealized, undivided in-group threatened by a troubling “Other”, set in an unambiguous, black-and-white plot. In the latter genre, argumentative exchange flows within the boundaries delimited by the “good” people (in-group), while (counter)arguments by othered interlocutors are preemptively discarded by the genre’s us–them moral dichotomy (see Forchtner et al. 2020, p. 207). In romances, “the good guys are clearly good from the outset: they wear the white hats while the bad guys wear the black hats” (Wagner-Pacifici 1986, p. 282). Consequently, tragic and ironic emplotments foster the enactment of Habermasian discourse presuppositions which allow for text-level communication in new formations of social relations beyond pre-existing boundaries, while comedy and romance impede such enactment and manifest the social boundaries between “the good us” and “the bad them”.

Text-level narrative persuasion can emerge, in both narrative contexts, as a means to persuade others within (1) romantic and comic employments, as well as (2) within tragic and ironic emplotments. Narrative persuasion, as such, describes the act of storytelling so as to convince others to join one’s ideas and convictions, which succeeds if the target audience is absorbed by the story and is reflecting its narrative message (Hamby et al. 2018). A text has this persuasive potential if it meets the basic definition of a narrative: “a story that raises unanswered questions, presents unresolved conflicts, or depicts not yet completed activity; characters may encounter and then resolve a crisis or crises. A story line, with a beginning, middle, and end, is identifiable” (Green and Brock 2000, p. 701).

For the time being, there are numerous different approaches in narrative persuasion which follow different logics, are based on different forms of empirical evidence, and are fragmented across disciplinary fields (cf. Green et al. 2020; Hamby et al. 2018, p. 2; Hinyard and Kreuter 2007, pp. 783–84; Tukachinsky 2014, p. 2). Hamby et al. (2018) developed an integrated perspective on narrative persuasion and identified two different characteristics implicitly or explicitly present in all approaches: the first one focuses on the idea already raised that people are absorbed in a story once they’ve started to read it. They call this process “deictic shift”—it can also be found under the terms “involvement” (Moyer-Gusé 2008; Moyer-Gusé and Dale 2017; Quintero Johnson and Sangalang 2016), “transportation” (Green and Brock 2002; van Laer et al. 2014) and “engagement” (Busselle and Bilandzic 2008). The second one emphasizes an internal process where the individual connects the meaning of the story to real-world beliefs which is labelled as a “deictic return” or “reflection”. In this paper, however, we do not research the impact of narrative persuasion attempts on audiences; instead, we analyze the degrees to which criteria for narrative engagement are met by the online content offered. In a nutshell, we reconstruct the supply-side for narrative persuasion by particularism- and pluralism-oriented Muslim accounts and identify the narrative genre in and through which their text-level communication becomes meaningful.

2. Methododological Approach: Two-Level Narrative Analysis

Following our rationale to reveal the context- and text-level dynamics of narrative persuasion in radicalization and counter-radicalization content by Muslim actors, we conducted a two-layered, DHA-inspired narrative analysis of relevant actors’ online content. In the following, we present the methods of both layers individually.

2.1. Narrative Persuasion Analysis

To explore the mechanisms of narrative persuasion, we used a mixed-methods design. The selection of data relevant for an in-depth narrative analysis, both context-level narrative genre and text-level narrative persuasion analysis, followed a combination of multiple steps of data collection and analysis.

First, relevant actors for the comparison were identified. We collected Instagram data from Jamal al-Khatib, Generation Islam and Realität Islam. The former is a fictional character, based on the impulse of an imprisoned young Muslim adult who has renounced his sympathy for the so-called Islamic state and wanted to hold off other people from joining them. Created by young Muslim adults and professionals, and based on authentic stories of these youngsters, Jamal al-Khatib offers counter-radicalization content by delivering alternative narratives to jihadist and Islamist-extremist online propaganda by applying Online-Streetwork and narrative biography methods (Ali and Reicher 2020). The latter two are considered part of the online presence of Hizb ut-Tahrir (Bundesamt für Verfassungsschutz 2020; Möller et al. 2021), a radical Islamist group founded in 1953, in East-Jerusalem, by Palestinian politician and theologian Taqi ad-Din an-Nabahani (An-Nabhani and Abdul-Qadim 2022).

Second, we quantified the Instagram posts of JaK, GI and RI covering Muslim grievances for one year (namely, 2021) to reconstruct the annual average and style of Muslim grievances coverage. For this purpose, we counted mentions of injustices against Muslims on a national and international scale for each actor in 2021. The latter year was selected because data were easily accessible for that time period. Instagram was selected because it is a prominent social media channel used by the senders and offers an additional layer of analysis by allowing content creators to share non-video content as well.

Third, based on insights we drew from the quantitative data, we sampled three-layered data for qualitative analysis. That is, we selected data referring to three historical events of Muslim grievances, each tied to a national (i.e., the terrorist anti-migrant and anti-Muslim attack in Hanau), European (i.e., the Srebrenica genocide) or international (i.e., the terrorist anti-Muslim attack in Christchurch) level. We selected those events for these three layers because they are theoretically and empirically highly significant experiences in Muslim memory politics of the past decade. We focused on Instagram posts which we retrieved in October 2022 and did not gather any Instagram stories since they are only available for 24 h. Instagram posts include video, image and text data. For the text-level narrative analysis, we drew on the image or video post as well as the text below the respective post which mirrors the political statement by the uploading Instagram account.

The Instagram posts were analyzed based on three aspects: First, whether the posts are a narrative or not, based on the definition of narrative as stated above, minimally including a beginning, middle part and ending (Green and Brock 2000). Second, identified narratives were categorized according to their persuasive potential. The following list points out the most relevant criteria2 as identified in previous research, which enhance engagement as a necessary condition for persuasion:

• Psychological similarity refers to similarities between the receiver and the main character in the story concerning their attitudes and personality traits. The idea that psychological similarity leads to a stronger engagement is proven by several previous studies (Turner and Oakes 1997; Cohen and Perse 2003; Eyal and Rubin 2003; Andsager et al. 2006);
• Familiarity entails the idea that the receiver knows something about the narrative world, such as a shared cultural background, knowledge of institutions or themes, as well as personal experience of a certain topic. Again, several studies have proved that familiarity leads to stronger engagement (Green 2004; Howard 1991; Larsen and László 1990);
• Regarding the narrative perspective (third or first person), first-person narratives are more involving if the main character and the audience belong to the same group (Winterbottom et al. 2008; Kaufman and Libby 2012). Second, independent of whether the main or the antagonizing character holds a similar attitude, the audience always identifies more with the main character (de Graaf et al. 2012; Hoeken and Fikkers 2014);
• Imagery will make the receiver draw a certain conclusion by using strong images which are directly linked to the quality of the story. Novelists or filmmakers can produce better results than unskilled individuals or researchers (Green and Brock 2000, 2002; Green 2004; Mazzocco and Green 2011);
• Arc of suspense can be described as a temporal sequence having a beginning, middle/climax and an end. Narratives start by raising unanswered questions or unresolved conflicts which lead to some kind of suspense. This opening is followed by a clear climax where the dramatic intensity is modulated, and the receiver of the story feels tense because he is concerned about the fate of the character(s) in the story. This tension leads to a need to want to know the ending of the story (Escalas 1998; Green and Brock 2002; Mazzocco and Green 2011);
• Character evolving with intent refers to the idea that the character in a story is changing his attitude throughout the story towards the desired one. Alternatively, the message can be integrated as part of the story which influences later events. Intertwining the persuasive part with the overall story proves to be promising (Bandura 2004; Sabido 2004; Frischlich et al. 2017);
• An epilogue is added to ensure that the hidden persuasive message is understood correctly (Hinyard and Kreuter 2007; Moyer-Gusé and Dale 2017). This explicit appeal does not lead to reactance in prior studies (Moyer-Gusé et al. 2012).

Third, particularly relevant statements are singled out to reconstruct text-level interventions by the actors under scrutiny.

2.2. Narrative Genre Analysis

We focused on the following criteria to reconstruct an actor’s narrative genre: (1) the plot movement, i.e., the way events are selectively and causally linked, (2) the im/purity of the hero/ine, i.e., the degree of idealization of the hero/ine and (3) the degree of idealization in story resolutions, i.e., idealized story resolutions reduce the need for debating alternative steps, while non-idealized facilitate discursive spaces for the communication of real-world ambiguities and complexities. Moreover, we followed Wagner-Pacifici’s (1986) postulate that a story’s resolution signifies its plot because the storyteller’s aim to make a moral point is driven by the need to control the story’s ending. Consequently, we identified story resolution in order to reconstruct the narrative genre in general.

The joint application of these genre-related criteria is presented in the model in Figure 1. The relationships are as follows: The story resolution determines the need to debate further steps and it creates heroic figures. These heroic figures project positive or negative self-images. If the purity of the hero/ine is high and the story resolution reduces the need for debate, the certainty about social boundaries rises. Such constellations are represented by romantic and comic genres. If both are low (impurity of the hero/ine and “open” story resolution), uncertainty about social boundaries is fostered. Such constellations represent tragic and ironic genres.

Juxtaposing text-level narrative persuasion and context-level narrative genre analysis, we show the text-level persuasive potential of both actors’ communication and, on the level of context, we explore their inclusionary and exclusionary potentials. That is, we reveal if and how a communicative text fulfills criteria for being defined as a narrative and, if so, we evaluate its persuasive power. This text-level evidence is then integrated into the analysis of narrative genre as a key site for the identification of an actor’s mobilization for an inclusive or exclusive Muslim identity.

3. Results

In the following, we present two empirical steps via which we reach the conclusion for narrative persuasion analysis (text level). After that, we present both actors’ (JaK and GI) narrative genres to reconstruct the structural context in which these narrative persuasion strategies reveal their meaning.

3.1. Step 1: Quantification of Instagram Posts by JaK, GI and RI

Table 1. Relative Mentions of Muslim grievances in 2021.

Reference	JaK	GI	RI
National (Europe)	2	88	233
International	3	131	51
Overall Anti-Muslim	5	219	284
Overall Posts	60	471	392
Relative Mentions	8.3%	46.5%	72.5%

displays the numbers of Instagram posts created by JaK, GI and RI, with reference to any event or topic described as anti-Muslim in 2021. 2021 was chosen for this first step since it is the most recent complete year the three actors were active. We separated these anti-Muslim topics or events with regard to the national and international dimension. The descriptive data shows that only 8.3 per cent of JaK’s Instagram posts cover anti-Muslim themes, while the relative shares amount to 46.5 per cent in the case of GI and 72.5 per cent in the case of RI. Further, our data collection shows that huge differences in terms of the absolute numbers of Instagram posts are apparent.

3.2. Step 2: Narrative Persuasion Analysis

In Step Two, we shift from focusing on a certain year to focusing on three events.

Table 2. Numbers of narratives among total numbers of posts.

Account	Hanau		Srebrenica		Christchurch		Total
	Total	Narrative	Total	Narrative	Total	Narrative	Total	Narrative
JaK	1	0	7	1	2	2	10	3
GI	12	4	11	0	66	6	89	10

illustrates the quantities of Instagram posts by Jamal al-Khatib and Generation Islam that cover Muslim grievances vis-à-vis key multi-layered events, such as Hanau (national, Germany), Srebrenica (Europe) and Christchurch (international), as well as the numbers of narrative posts among all posts according to the definition of “narrative” in Section 1.3.

Compared to the total number of relevant posts per account as shown in Table 2, 30 per cent of the posts by JaK include text-level narration, whereas this holds for only 11 per cent of GI posts. One explanation for this relative difference may be disbalances in key resources, such as time constraints. Creating a persuasive narrative is both time- and cost-intensive. Whereas GI reacts immediately to events like the terrorist attacks in Hanau and Christchurch, disseminating a core message or narrative (see below), JaK invests resources in the preparation of balanced, mindful and complex posts. The first finding, thus, is that the core narrative (genre, see below) of GI is more adaptable to different settings, while JaK’s core narrative necessitates putting thinking and reflection into the preparation of online content. This materializes in unequal temporalities.

Table 3. Persuasive potential of narratives.

Account	Hanau	Srebrenica	Christchurch	Average
JaK	6.0	-	5.5	5.8
GI	-	3.5	3.0	3.3

sums up the persuasive potential of the narratives shared by the actors, based on the seven criteria as described in Section 2.1. Whereas Jamal’s narrative persuasion scores an average potential of 5.8 out of 7, Generation Islam’s potential for persuasiveness regarding their narratives is only 3.3. The numbers are the result of the detailed discussion in the following subchapters. It shows empirically that JaK is more invested in narrative persuasion than GI is.

Our qualitative analysis further illustrates why both actors’ modes of narrative persuasion differ: While JaK seeks to offer counter-narratives, which include the ambition to persuade others to buy into “new” or different, i.e., less Islamist extremist, worldviews, GI’s communicative content speaks predominantly to an already persuaded audience which preconfigures its text-level content. That is, GI does not emerge as a discursive space to persuade people for extremist content, but mainly one in which extremist ideologies are taken for granted.

In the following, we present the results of the three-layered narrative persuasion analysis. We specifically highlight the role of pervasive potential (engagement) and persuasive messages in our analysis of JaK’s and GI’s online content related to the remembrance and coverage of violent events in Hanau, Srebrenica, and Christchurch.3

3.2.1. Hanau—Communicating the National Context of Muslim Grievances

Generation Islam and Jamal al-Khatib present different strategies of narrative persuasion in the national case of Hanau. The latter explicitly highlights the diverse (not explicitly Muslim) protests, which indicate extensive solidarity with Muslim experiences within German society. The former, in contrast, does not mention any solidarity and/or protest event, but instead addresses feelings of powerlessness to mobilize a particularistic Muslim identity for a better future in a (Western) world, where nobody but Muslims will solidarize with Muslims’ experiences of violence and oppression.

GI’s Text-Level Narrative Persuasion Strategy of Hanau

In a total of 12 posts, different framings and narratives related to Hanau were created by GI. Five of these were posted immediately after the terrorist attack (within the following nine days) and the remaining posts were released on Hanau’s first anniversary.

The first post after the attack published one day after the attack, summarizes GI’s Hanau content remarkably well: “Will Germany wake up now or is it too late? One week after #Teutonico was uncovered, now in #Hanau an #attack with 11 dead has been committed by terrorist #TobiasRathjen. Would such conditions be possible without the permanent #Islam-baiting by #media and #politics against #Muslims and #migrants?” (Generation Islam, 20 February 2019).

Interestingly, there is no grievance offered to the audience. Nothing implies any empathy with the victims or their relatives. One could even argue that the lives of these predominantly Muslim victims are by no means prioritized here but rather instrumentalized for an Islamist extremist political agenda opposing “German” elites. The underlying and mobilizing message is presented by a rhetorical question asking whether events like this would have happened without the constant scapegoating of Islam and Muslims by politicians and the media. Another example for that is present in the latest post on 19 February 2022, which reads: “The real reason for HANAU”. It criticizes the fact that the media and politicians still continue their alleged fight against Muslims and Islam, or, in the words of GI: “The perpetrator did not carry out his attacks in a vacuum. No, he was radicalized by the same institutions as the person who throws stones at a mosque or wants a Muslim woman to return to her home country” (Generation Islam, 19 February 2022). This post visualizes a Muslim man named Bilal who “reveals” the “true reasons” behind the Hanau attack, highlighting anti-Muslim attacks by prominent mainstream political actors within the German Bundestag, or—in a bid to show that this is not an exclusively German issue—by the French premier, Macron. As such, GI indicates that it does not come as a surprise when—after politics and the media constantly portray Muslims as dangerous—some actors and groups develop a defensive reflex and decide to counter that “danger” (Braddock 2020).

The four narratives were posted immediately after the initial reaction of GI and mainly consist of videos of Muslim men sharing their stories: a survivor (20 February 2019), a survivor who converted (21 February 2019) and a grieving father (twice; 23 February & 28 March 2019). These posts are analyzed regarding their persuasive potential, based on the criteria in Section 2.1: All posts meet the criteria for “similarity” and ”familiarity”, since the protagonists emerge as Muslim individuals who understand themselves as part of Umma, the collective, supra-national community of Muslims. The survivor, as well as the convert, report emotionally how the assassin entered, murdered and left. This is narrated from a first-person perspective, which means the videos fulfil the criteria “arc of suspense” and “perspective”. The post of the convert survivor is presented as a specifically authentic voice to circulate GI’s core message: by becoming the set of Muslims, you enter the group of the morally “good” people. Hereby, this post also fulfils the criteria of “character evolving with intent” as well as “epilogue”. The other videos do not meet these two criteria. The remaining criterion “imagery” is met by none of the posts.

JaK’s Text-Level Narrative Persuasion Strategy of Hanau

In relative terms, Jamal al-Khatib was substantially less visible in the communication of Hanau. As the post suggests, the attack came while the team was in Sarajevo for JaK’s “Srebrenica” campaign. One of the posts (Jamal al-Khatib, 21 February 2020) shows Jamal in front of the Srebrenica graveyard and includes the hashtag “Hanau”.

JaK’s core persuasive message is perfectly exemplified by the following quote: “Also here in #Sarajevo we commemorate the victims of racism and right-wing extremism. Our thoughts are with their relatives” (Jamal al-Khatib, 21 February 2020). This persuasive message manufactures an umbrella concept of victimhood which seeks to persuade Muslim identity audiences to bridge their particular experiences of victimhood from extremist-right terrorist attacks with potential allies who might relate to these feelings associated with that perpetrator.

In so doing, the suffering is abstracted to a larger context which is not exclusively defined by a Muslim identity. The focal point is shifted to the concrete perpetrator whose action is not presented as limited to anti-Muslim motives, but as part of a larger racist, extremist-right ideology. As such, Jamal presents a Muslim identity as part of a larger set of targets defined by “out-groups” of a white extremist-right ideology, which allows for intersectional alliances and solidarity among those marginalized and attacked by the respective ideology.

Unfortunately, the persuasiveness of the content from JaK regarding Hanau cannot be analyzed, since the given post does not fulfil the text-level criteria for “narrative”. Interestingly, JaK was not invested in persuading the audience in any way with reference to Hanau.

In sum, in the national case of Hanau, JaK tackles feelings of powerlessness (in the West) presented by Islamist radical/extremist actors like GI. JaK makes the protagonist Jamal point to a broader set of victimhood by extremist-right attackers beyond a Muslim identity and, thereby, lays the ground for pluralist alliances among Muslim and non-Muslim actors.

3.2.2. Srebrenica—Communicating the European Context of Muslim Grievances

The case of “Srebrenica”, which is considered to be one of the worst post-Second World War genocides where Muslim Bosnians were systematically killed by the Serbian army, was very salient for both JaK and GI. As opposed to the two other terrorist events, the posts refer to an event in the past which means both accounts emerge as actors of memory politics. In the case of communicating Srebrenica, JaK relied much more on narrative persuasion than GI.

GI’s Text-Level Narrative Persuasion Strategy of Srebrenica

GI published a total of 11 posts covering the massacre, of which one was produced in 2016, one in July 2019 and three in the following year. None of the posts satisfy the threshold criteria for narrative and, thus, cannot by analyzed regarding their persuasive potential. However, they do spread persuasive messages. The first one is presented as having learned a “lesson”, which is that the Western world will never care about Muslim lives, thus summing up what the reader has to take away from these posts. First expressed in the third post on 11 July 2020, the main message is clearly at the very end: “What remains is the bitter lesson that we Muslims must learn from Srebrenica so that such terrible atrocities are never repeated. The West will never be interested in and committed to the protection of Muslims” (Generation Islam, 11 July 2020).

This view, namely that “the West” basically does not care about Muslim lives, is repeated over the years in different posts, as seen in Post No.7 from 22 July 2021, where the same sentence can be found, this time with the addition that “the West cannot be trusted”, and another in their 10th Post in July 2022.

The second recurring narrative depicts “the West”/European countries as silent bystanders during the massacre. While a case can be made for the passivity of European countries and the fatal errors of the UN Blue Helmets, it was still the same UN that initiated an investigation into the massacre and the International Court of Justice that sentenced the people involved in the massacre to prison. The extent of the massacre was only made clear to the world, including “the West”, after investigations and eyewitness reports. Even Muslim countries are depicted as the “vassals of western rulers” (Generation Islam, 11 January 2022), a trope not only found in many posts by Generation Islam, but is also in line with the writings of the group’s founder, An-Nabahani, who criticised Muslim countries for not being “Islamic” and lacking Islamic jurisprudence.

JaK’s Text-Level Narrative Persuasion Strategy of Srebrenica

The account of Jamal al-Khatib published seven posts in 2020 in a designated video campaign on “Srebrenica”. Jamal is the narrator in all of these videos, except for the video of the march for peace and of “Alhamdulillah’s” (no narrator, only music), and they are all between 30 and 90 s long (except for the video on genocide). The first is a trailer which mirrors an artistic introduction to the story post that follows.

The introduced video shares a dua (prayer) which wishes for Srebrenica to never happen again. In the subsequent video, footage of the March for Peace, organized annually since 2005 as a commemoration event, shows the crowd and posters like “never forget Srebrenica”. In the fourth video, which is a comparatively long one (8 min), Jamal informs the viewer about the political background in former Yugoslavia and describes the unexpected path from peaceful multiethnic harmony to the violent Srebrenica genocide. This is interesting, insofar as Jamal points to the destructive side of human behavior with his reconstruction of history. Jamal concludes with a call for intervention if people separate humans into closed groups, thus dehumanizing particular social groups, and pledges a search for commonalities as opposed to distinctive characteristics.

The next video focuses on courageous historical individuals who refused orders and stood up to authority. Their heroic deeds are stressed and bridged to the message that history teaches us that people act in solidarity irrespective of religion, nationality or other traits. The subsequent script presents Sarajevo as a multiethnic city in which mosques, churches and synagogues coexist historically. The final post shows Jamal with “Alhamdulillah” (thanks to Allah), and a thank you note in the caption which emphasizes the importance of doing good.

The post about the Srebrenica genocide fulfils the criteria for narrative and persuasive potential, which is analyzed according to the criteria in Section 2.1: Jamal is again the protagonist who here too satisfies the criteria for “similarity” and “familiarity”. As opposed to the terror video addressing Christchurch, Jamal is not disclosing his own story, but acts as a third-person narrator; consequently, the criterion “perspective” is not met. This video too is produced professionally, meeting the criterion “imagery”. Regarding the ”arc of suspense”, the first part of the video where Jamal talks about the Srebrenica genocide does have a clear beginning with open questions, followed by a climax and an end. What follows afterwards can be interpreted as a long ”epilogue”, clarifying the message. The persuasive message of this post is representative of the message of the whole campaign, aiming for peaceful coexistence as the lowest common denominator and hoping that humans will take a stance against inhuman treatment.

3.2.3. Christchurch—Communicating the International Context of Muslim Grievances

The terrorist attack in Christchurch is instrumentalized to agitate against politics and the media in Germany by GI, while JaK integrates this event into rethinking his past convictions and learning inclusionary lessons from them. As regards the communication of Christchurch, JaK relies much more on narrative persuasion than GI.

GI’s Text-Level Narrative Persuasion Strategy of Christchurch

Here again, GI’s posts outnumber those by JaK with a total of 66 posts. In the first three weeks after the attack, GI shared 14 posts containing mainly information about the victims, the survivors including stories of hero/ines who tried to stop the assassin, as well as grieving relatives. Only six out of 66 posts fulfil our defined criteria for narrative and were scrutinized further for narrative persuasion, analyzed according to the criteria in Section 2.1. Out of these six narratives, one is posted twice and another is posted three times, resulting in three actual narratives: The story of a man who confronted the murderer in an attempt to stop him (Abdul Aziz Wahabzada), the story of a mother picking up the car of her son who was killed in the attacks and the testimony of a mother (Maysoon Salama) whose son was assassinated as well. All of them meet the criteria for “similarity” and “familiarity”, since the main characters in all these cases center around a Muslim identity which is shared by the target audience. None of the narratives are shot professionally (criterion “imagery”), nor do they meet the two criteria “character evolving with intent” and “epilogue”, since none of the main characters evolve with a clear-cut speech act which clarifies the persuasive message of GI. As opposed to the two stories told by mothers, the posts featuring Wahabazada’s story meet the criterion “arc of suspense”. On the other hand, the Wahabazada post does not meet the criterion “perspective”, but the stories told by mothers do.

On 5 April, GI started a campaign using the hashtag #51Wochen51Helden, which inaugurated a series of postings for 51 weeks for Friday prayers, including information on one individual who was killed during the attack. All of these postings contain the following text: “The killer was radicalized in the West. He was not part of an occupying army. Hate speech and lies about Islam, which also broke out in a large part of society here in Germany, have made him the monster he is now. The media and politicians never tire of infecting society with hatred and hate speech even after the events in Christchurch” (Generation Islam, 5 April 2019).

After the said campaign, old, related content was reposted annually. In addition, there were two posts accompanying the trial of the attacker in summer 2020, sharing the statements of relatives of two of the victims. One post shows a part of a sketch by the comedian Isaac Butterfield, who makes fun of the victims by stating that the saddest thing about the attack was that hundreds of people that night could not make it home from a nightclub because all the cab drivers were dead. GI comments: “Imagine if this was a terrorist attack on a #synagogue in which 52 Jews were killed. How would the joke go down then? The outrage would result in his shows being banned for #antisemitism and he would have to apologize publicly. But because they are Muslims, it is even accepted with applause. The extent of #Islamophobia in the West is appalling.” (Generation Islam, 3 August 2020). Actually, Butterfield was criticized for his inhuman joke and has made jokes about Nazi gas chambers and 9/11 attacks, as well as other attacks in the past. Consequently, GI uses this joke to add to the world view that Muslims are under attack by the West.

JaK’s Text-Level Narrative Persuasion Strategy of Christchurch

JaK’s two published posts referred to the events of the terrorist attack in Christchurch, New Zealand, while JaK showed no immediate public reaction after the attack. Given that the first Instagram post appeared two months later, we assume that the account was inactive at the time of the attack.

Both posts have an indirect connection to Christchurch, as they introduce a video featuring the protagonist who discusses the topic of terror in general. The massacre in Christchurch is but one of the incidents raised in the video. Other events cover 9/11 or Charlie Hebdo, with the latter event being particularly salient.

In one of the video posts, the main character is reflecting on the aftermath of Charlie Hebdo. Teachers seemingly judging young Muslim students for the terrorist attacks are mirrored there, followed by a speech act by Jamal, in which he wonders why the victims of the Charlie Hebdo attack victims are prioritized vis-à-vis the many victims of Western invasions in Muslim-majority countries. He recalls an event where the principal called for a minute’s silence for the victims of the Charlie Hebdo attack, which he did not want to attend, for which he has been publicly reprimanded. Jamal continues by retelling the circumstances of the Christchurch attack, via which he stresses that he realized how the coverage of the terror attack differed from what he was used to and that, for the first time, it had finally discussed the causes of terrorism in class, which did not transpire before. He continues to point out the importance of discussing the roots of violence against everyone, rather than trying to classify the deservingness of particular victims. Jamal finishes his monologue with lessons he learned from not participating in the minute’s silence. He concedes that he should have joined the mythical act of showing respect for the victims—and that he should then have asked for permission to remind others of many other victims of terrorist attacks elsewhere.

Both posts fulfil the criteria for narrative and, thus, their persuasiveness must be investigated according to the criteria in Section 2.1. The main video discussing terror meets all the criteria. The main character shares similarities with the target audience based on a certain “Muslim” attitude towards the discussion of terror and attacks in the Western world, as well as a Muslim identity reflected by a style of clothing. The video at hand is produced professionally; for instance, it uses strong images to mirror the tensions between Jamal and his classmates and teachers. It further includes a straightforward arch of suspense, that is, Jamal is overwhelmed at the beginning by the injustice felt, which led to an inappropriate reaction, which is then followed by a self-distancing which results in a pluralistic world view. This self-ironic turn in the story undermines the potential for a simplistic and romantic narrative of “us against them” and, consequently, embraces ambivalence and pluralism.

4. Discussion

4.1. What Do We Know When We Know This? Comparing the Narrative Genres of GI & JaK

Our two-layered narrative analysis of social media content by GI and JaK reveals that the former creates discursive spaces in which top-down narrative persuasion is hardly performed, and that the discursive space itself is shaped by a romantic-leaning narrative which configures an essentializing Muslim identity in a perennial and irreconcilable black-and-white scheme vis-à-vis the West. With regard to JaK’s content, we show that JaK is strongly invested in narrative persuasion which reveals its meaning within a self-ironic-leaning narrative genre which accommodates emotional distancing from an overemphasis on a Muslim identity and the supposed uniqueness of Muslim victimhood. GI, on the other hand, narrates an ideal romantic world, “the Nostalgic Return to an Islamic Umma”, via the configuration of essentialized binary social roles: “the good us” versus “the bad them”.

4.2. GI’s “Crusading Savior” and a Nostalgic Return to Islamic Umma

The representation of both Muslim and “Western” identities is overstretched and essentializing, and both constructed groups emerge as irreconcilable in the romantic plot presented (for the definition of the romantic genre see Section 2.2). “The West” appears as a morally inferior perpetrator that (supposedly) has perennially forged sinister plans against Muslims, which implies that, especially for Muslims in Western societies, they live in extraordinary danger.

GI’s text-level narratives further feed into the political ideology of Hizb ut-Tahrir, which rests on the promise of salvation in Islamic Umma. Here, we find what Wagner–Pacifici defined as the storyteller’s desire for controlling the ending of a story, because it determines its moral implications and social roles. Empirically, this story resolution is (implicitly mirrored) in a post from January 2021: “Under the rule of the Ottoman Empire, it was possible for all peoples and religious communities to live together without falling into nationalism or racism” (Generation Islam, 11 January 2021). The story resolution presented displays a return to the nostalgically idealized Ottoman Empire, which appears as a proxy for the Islamic caliphate.4 Thus, the present-day downturn (exemplified by Hanau, Srebrenica and Hanau) and continuous “Western threats” call for a united Muslim people in an Islamic caliphate to overcome this suppression and live in peaceful harmony.

In more detail, our analysis shows that, on the particularism-leaning, radical/extremist side of Muslim accounts, Generation Islam recounts an idealized black-and-white narrative of Islam and Muslims being perennially oppressed and perpetrated by “the West”. The only solution to overcome this and liberate them from this everlasting threat is presented as a festive social transformation into an Islamic caliphate, which perfectly aligns with the meta narrative of Hizb ut-Tahrir. The idea of the caliphate, however, is not constantly and explicitly propagated but, rather, emerges implicitly as an ideal response to the violence against and oppression and marginalization of Muslims in Germany. In other words, this action point is an implicit common theme, the core future promise of salvation following liberation from Western oppression. It is based on this romantic portrayal that closed social boundaries are drawn around a Muslim identity vis-à-vis multiple and intersectional facets of social identity.

Finally, we can identify a common logic when comparing the group posts with regard to the three attacks. In both cases, the focus was not placed on the underlying ideology, be it Serbian nationalism, Islamophobia or right-wing extremism (e.g., Christchurch), but rather abstracted to “the West” and Germany which is depicted as the ”true” root of evil.

4.3. JaK’s ”Self-Reflexive Savior” and Continuous Self-Distancing and Bridging

On the pluralism-leaning side, our analysis shows that Jamal al-Khatib tells a predominantly self-ironic story which parodies GI’s romantic story of a nostalgic return to Islamic Umma. Here too, Muslim-centering grievances are presented as legitimate reasons for Muslim critique in Germany. This criticism of prevalent social relations, however, is not instrumentalized for an idealized salvation to come once “we” overcome the hurdles posed by “our” enemy, as it would in a romantic plot. Instead, JaK inscribes self-irony into his narrative, which means that failures are not exclusively externalized to “them”, as ”our” ones are equally conceded. As such, JaK’s “self-reflexive savior” includes no “own” story resolution beyond the parody of romantic portrayals of ”us” and ”them,“ but this parodying resolution is crucial for the narrative configuration of inclusive social boundaries.

It is against this background of the “self-reflexive savior’s” decentering that the Muslim experience blurs boundaries towards other subjects’ experiences of violence and marginalization in German society. Thus, it allows for the inclusion of pluralism, i.e., of alternative interpretations of events and topics, and for the formation of manifold alliances under the umbrella of shared experiences of suffering from far-right discourses in Germany. In a nutshell, the self-reflexive hero is the backbone to facilitate the imagination of inclusive social relations beyond a Muslim identity.

5. Conclusions

In this article we asked how Muslim actors create discursive spaces open and closed to self-reflection and liberal-democratic principles in a Muslim-minority Western European context, i.e., in Germany. Our analysis shows that, at the level of text, narrative persuasion varies between both contrasting actors, while the pluralism-leaning Jamal al-Khatib disseminates persuasive stories, Generation Islam is much less invested in narrative persuasion. These two text-level strategies reveal their meaning in two radically antagonistic constructions of saviors at the level of narrative genre: Jamal al-Khatib’s “self-reflexive savior” creates an inclusionary discursive space represented in a self-ironic narrative genre, while Generation Islam’s ”crusading savior” manufactures an exclusionary discursive space represented as a romance and a nostalgic return to Islamic umma.

We thus show that Generation Islam creates radical online discursive spaces in which morally “good” Muslims and their sinister ”Western” perpetrators are dichotomized in a romantic plot. It is in the context of this romantic plot, which facilitates radical and extremist Islamist interpretations of an Islamic Umma, that ambivalence and self-doubt are eradicated and, in turn, the idealization of the morally superior self and inferior other are fostered. As such, and in fact, the anti-Muslim experience is overemphasized and singled out and, in turn, organized solidarity by non-Muslims is blind-sided and disregarded. Against the background of evil anti-Muslim politics (in Western societies), such a romantic emplotment fosters text-level arguments for moral superiority of Muslims, and justifies and legitimizes violent and non-violent reactions by the crusading savior and his/her followers.

Contrary to that, Jamal al-Khatib creates deradicalizing and preventive online discursive spaces in which self-doubt is fostered by the self-reflexive savior. The self-ironic genre facilitates text-level arguments for perspective-taking between the Muslim and non-Muslim experience, even vis-à-vis clearly anti-Muslim terrorist and genocidal attacks such as Srebrenica, Hanau and Christchurch. As such, Jamal al-Khatib’s communication of the three anti-Muslims cases proves that self-reflexive savior facilitates cognitive structures for bridging perspectives, and fosters ‘post-migrant alliances’ (Foroutan 2019) between Muslims and other marginalized perspectives in contemporary Germany.

Future research should include intersectional approaches to the making of inclusionary and exclusionary discursive spaces. The study of masculinities and femininities constructed in and by Islamist extremism is not novel (cf. Cultures Interactive e.V. 2015; Fritzsche 2018); however, the application of our novel analytical approach promises new comparative insights and results.

Other research projects could shift their focus away from the sender to the receiving end, in order to analyze the extent to which text-level persuasion resonates with audiences. An in-depth analysis of the social-media accounts of Jamal al-Khatib, including the discussion of user reactions to narratives, is planned (Walter forthcoming). Furthermore, TikTok could be put into focus, as it is emerging as a central vehicle for propagandistic religious and political speech and—although used intensely by Islamist extremist activists—it has not been thoroughly studied yet, let alone included in P/CVE-related ambitions Ali (2020, 2021, 2022). We conclude with the self-ironic postulate that we need to do better to counter extremists of all sorts.