The right-leaning be memeing: Extremist uses of Internet memes and insights for CVE design

This article constitutes a first step in understanding how Internet memes are used in extreme-right online milieus and formulating appropriate policy responses. First it looks at existing literature on memes as communication units. Secondly, it looks at the particular ways in which transnational extreme-right groups use Internet memes. Thirdly, it discusses the applicability of these memes to resilience-building projects targeting the extreme right. The article’s conclusions discourage the use of Internet memes by state and security actors, while highlighting positive uses by grassroots organisations. Some notes on further necessary research conclude the piece.


Introduction
In the last two decades we have seen an increase in extremist right-wing activities. These activities may be connected to institutionalized politics, via the electoral success of right-wing political parties, or they may relate to violent or non-violent protest incidents -even, in the most extreme cases, to terrorist attacks. Indeed, scholars nowadays speak of a new wave of right-wing extremism, characterized by the emergence of an underground subculture of racist and frequently violent young extremists with their own symbols, myths, and language ). This new wave is international, and native to the digital environment (Europol, 2020); its users have digitally driven ecosystems, favoring the reinforcement of shared terms and facilitating collaboration between different extreme-right communities.
While this wave of extreme-right mobilization has had relatively limited consequences in terms of 'body count', its danger stems from its capacity to encourage the growth of xenophobic tendencies in public opinion [1]. Digital platforms, whether social media or Internet forums, have "empowered the extreme right to directly broadcast their content globally to niche and mainstream audiences alike" [2].
Recent European Union counter-terrorism efforts, articulated around countering violent extremism (CVE), have sought to address right-wing radicalization and the accompanying social polarization. Increasingly, EU member states are coming together to fight extreme-right violence under anti-terrorism legislation (European Council, 2019;European Council ECTC, 2019). This approach, however, continues to frame the problem in terms of security, thereby missing one of the most important developments of the extreme right today: the solidification of a mainstream 'extremist' culture online (Huey, 2015).
This paper aims to set out current knowledge on the use of Internet memes by online extreme-right milieus, in order to establish a theoretical baseline for incorporating Internet memes into future CVE programmes. It therefore reviews the most relevant literature from communication studies and extreme-right studies, drawing from over 20 articles and books published in the last five years.
The present literature review attempts to fill a gap in research and policy on appropriate CVE responses to the cultural and social aspects of online right-wing milieus. It chooses to focus on Internet memes because they are a key element of the online communications and identity formation processes of young people today , and play a role (however unclear) in online radicalization (Meleagrou-Hitchens and Kaderbhai, 2017).
Below, I will first give a brief review of literature on right-wing milieus and the Internet. Secondly, I outline the main characteristics of Internet memes as cultural artefacts. Thirdly, I present a review of extreme right-wing uses of Internet memes, and their purposes. Fourthly, I use this knowledge to outline basic lines of action and theoretical underpinnings for incorporating Internet memes into resilience work.

Theoretical framework
My first step for this review is to define and theoretically contextualize the concepts I will be exploring, namely: extreme right-wing milieus, extremist online culture and CVE.
In the first place, there is a broad consensus in the field that right-wing extremism primarily describes an ideology in one form or another [3]. A succinct review of the literature reveals that there is a high degree of consensus among the definitions of 'far right' put forward by different scholars [4]. Extreme right-wing groups are primarily characterized by their populist, nativist, and authoritarian ideology (Mudde, 1995). The defining properties of right-wing extremism include "authoritarianism, anti-democracy, and exclusionary and/or holistic nationalisms", while "xenophobia, racism and populism are accompanying characteristics" [5].
The extreme-right political milieu is not a monolithic or holistic entity, however. Rather, it is composed of "a host of heterogeneous strands of diffuse political activities", with divergent justificatory narratives [6]. Indeed, both ideologically and geographically, manifestations of the extreme right vary greatly (Klein and Muis, 2019;Ravndal, et al., 2020). In this paper I will focus on research and trends pertaining to western EU countries, mainly Germany, France, Italy and the United Kingdom. The paper's insights are gathered from research on the online extreme-right communities in these countries.
Importantly, online networks are a key novel element in the activities of extreme milieus. With their capacity to generate collective identities, while also facilitating the exchange of resources and information (Della Porta and Wagemann, 2012b), they both create a feeling of inter-group solidarity and introduce shared goals and enemies, thereby strengthening even further the cooperation links between different groups. Studies on violent radicalization stress that isolated individuals can find a common identity via extreme-right Web sites, which foster the comforting sense that they are not alone, but part of a community [7] (De Koster and Houtman, 2008).
Secondly, I would like to define further the concept of 'online extremist culture'. This paper approaches far-right and right-wing extremism as not just a political space, but also a visual and cultural one. As such, although the focus may be on the online facet of this culture, this comes with the awareness that physical spaces designed for activities such as concerts or sports, for example, play an important role in sustaining extremist communities in Europe (Bouchard and Thomas, 2015).
Following this approach, individual extreme-right milieus are nested within a larger 'online ecosystem' of the extreme right. This online ecosystem is "vast, heterogeneous and multifaceted" [8]. While the selfproclaimed Islamic State is usually credited as the most technologically savvy extremist group at present, " the width and sophistication (not the quantitative production) of several far-right strategic communities is comparably impressive" [9]. Their activity has long remained under the radar of scholars and policymakers, despite the right wing's extensive use of social media platforms to disseminate their ideas, mobilize, and organize (Della Porta and Wagemann, 2012b).
Indeed, the digital world has had a fundamental impact on the spread of extreme-right communities, online and off-line, by enabling them to create a veritable cultural online identity. Mullhall (2018) observes a "rise of far right social media personalities who, despite not being part of traditional activist organisations or parties, now have the ability to reach unprecedented numbers of people". While many of the extreme right's classical ideas, such as xenophobia and misogyny, persist, the symbolism has been updated and fitted out with new images and catchphrases more easily recognizable by young people. The symbols and icons of the extreme right have shifted recent years, and have been re-coded in urban clothing, hipster emblems, and humorous memes [10].
In fact, these cultural artefacts ("hipster emblems", videos, forum boards and even Internet memes) should be regarded as both "exploitative" and "constitutive" objects, shaping identity, desires, engagement, a sense of belonging and even participation in extremist milieus [11]. It is in the context of this online cultural dimension of the extreme right that Internet memes are particularly relevant. Online communities are "primarily responsible for the rapid creation, evolution, appropriation, and circulation of far-right memes in the contemporary era" [12].
Finally, to anticipate the discussion of findings, we shall briefly introduce CVECountering Violent Extremismas the predominant approach in the EUs prevention of terrorism programmes. CVE is defined as efforts to influence the individual and/or environmental factors that are believed to create the conditions in which violent extremism can flourish, using social or educative tools rather than explicitly security-driven or punitive measures [13]. Most of these initiatives include or are centered on the notion of resilience building, which argues that (attraction to) extremism can be prevented by developing key emotional capacities or skills. Making individuals and groups strong, critical and flexible prevents them from being drawn to violent extremist ideologies or groups (Grossman, et al., 2020).
We emphasize building a theoretical focus for resilience because, thanks to institutional preferences and widespread support among policy-makers and some sectors of academia (McNeil-Willson, et al., 2019), resilience is likely to be the dominant paradigm in future strategies to prevent right-wing extremism (RWE). The concept has been discussed extensively in the context of jihadi terrorism but, so far, countries with little or no history of countering extremism face major challenges when it comes to formulating appropriate CVE strategies (Hardy, 2019). Germany and Nordic countries such as Norway and Denmark stand out as having a track record of devising CVE policies specifically aimed at the extreme right. Even for them, however, Internet memes and other relatively novel aspects of the online sphere remain significant obstacles on the road to building effective resilience to extreme-right ideas among their populations (Weilnböck and Kossack, 2019).

Literature review
In this literature review I shall address the most prominent features of the extreme-right's presence in social media, as well as presenting what little research exists on its use of Internet memes.
On the one hand, most existing research on online extreme-right milieus has looked at how institutional political actors have used images and other visuals to transmit propaganda and mobilize their supporters to action in mainstream social media sites. This research has shed valuable light on the emotional power of memes (Klein, 2018;Klein and Muis, 2019) and their capacity to cloak downright discriminatory attitudes in humour (Bhat and Klein, 2020;Klein, 2020a), and even to promote a mainstreaming of violence (Mudde, 2010).
To begin with, content that arouses high emotions is more likely to go viral (Berger, 2016;Klein, 2020a;Pauwels and Schils, 2016), attracting attention and increasing mobilization potential. Other research also points to the "indignation effect", whereby people online are more inclined to share what upsets them than what makes them happy (Larsson, 2018). Further research differentiates between types of Internet memes and emotional responses, suggesting that shocking or humorous messages generate the most engagement, but information-based messages are likely to be shared most often (Gough, et al., 2017).
Secondly, recent work on online right-wing milieus shows how they "normalize access to far-right ideas and also normalize the ideas themselves" [14]. This leads to the mainstreaming of exclusionist ideas of citizenship and democracy by exploiting citizens' discontent and fear of the unknown (Klein and Muis, 2019). Here the word 'mainstreaming' describes a "confluence of processes that together cultivate sympathy among large portions of the general public for social attitudes that would otherwise be considered unacceptable" [15].
Thirdly, authors note how, with the expansion of online communication, the Internet has become a natural medium and a catalyst space for extreme-right propaganda, making "digital space a central site for the current resurgence of far-right influence" [16]. In the digital sphere, online memes offer the extreme right an especially attractive medium it can use to mainstream violence against its chosen targets. Internet memes allow it to "appropriate Antonio Gramsci's idea of the struggle over cultural hegemony -i.e., the production of consenting ideas in civil society" [17].
On the other hand, however, recent research on Internet memes has opened up promising avenues of enquiry on the role of humour in extreme-right online communications and communities. The humour here is of a specific kind, characterized by detached, cynical irony; it has been said to be symptomatic of present-day youth communication and culture and has been dubbed 'memetic irony' or 'the logic of lulz' by the few scholars working in the field. The term 'lulz' -a derivation of 'lol' or 'laughing out loud'denotes a detached, dissociated sense of schadenfreude, also known as 'Internet trolling'. Comparable research on other brands of extremism, such as jihadism, considers how pro-Islamic terrorism is turned "into something hip and trendy among online audiences, drawing on humour to reinforce ideological messages" (Huey, 2015).
We should finish this section by mentioning that some research has examined institutional/state uses of Internet memes in health-care awareness campaigns (Gough, et al., 2017), institutional public relations (Wood, 2020) and even political activism. On the latter topic, one large-scale content analysis (Penney, 2017) of political uses of Internet memes focused on Bernie Sanders' 2016 campaign. It found that, in the run-up to the Democratic primary election, Sanders' rival received three times more coverage in mainstream broadcast and print media. However, while the "official" Sanders campaign organizers built applications and distributed tightly controlled social media messages, this was complemented by "unofficial" grassroots networks that circulated more "informal and culturally oriented appeals", also supporting Sanders. The Sanders case study thus illustrates how, outside of extreme-right groups, political actors and their supporters can tactically deploy Internet memes, and it will serve as a basis for the CVE recommendations at the end of the paper.

Internet memes and extreme-right uses: Main lines of research
While not enough research has yet been done on how the extreme right uses Internet memes, we do have a strong body of evidence showing the nature of these memes and their creation, reception and distribution.
Here we draw on this knowledge to outline the main characteristics of Internet memes, their functioning and the core element of humour. We then move on to examine how, where and why online extremist milieus use Internet memes, based on the existing evidence from Germany, France, the U.K., and Italy.

(a). Internet memes: Nature and functions
What is an Internet meme? The term refers to "units of popular culture that are circulated, imitated, and transformed by Internet users, creating a shared cultural experience" [18]. In Internet culture, memes are images, videos, and/or text, usually of a humorous nature, that are copied and spread with slight variations on the Internet (Seiffert-Brockmann, et al., 2018). In social media and other online platforms the Internet meme is recognized by users as a valuable product, symbolising a new kind of digital cultural production and exchange in a society (Shifman, 2014b). Moreover the competitive structure of social media pushes users to share popular and humorous images further, as a way of increasing their own engagement [19].
Aesthetic considerations are important in the popularity of Internet memes. The wittiness of the 'caption' or punchline, the meme's relatability and the immediacy of its message are all pondered together with its artistic value, and all affect the reception and potential redistribution of the meme in question by the user . Indeed, visuals have become a more and more dominant medium for transferring cultural and political messages (Highfield and Leaver, 2016). Internet memes often present a simplistic message conveying "one uncomplicated idea or slogan" [20]. It is logical, then, that they should constitute the most efficient format for propaganda in modern society, which relies increasingly on audio-visual, rather than textual, communication (Manjoo, 2018).
Humour has been highlighted as a key factor in the success of Internet memes as vehicles of information in the online sphere. Internet memes rely on a quintessentially twenty-first century, 'millennial' kind of humour: 'memetic irony' or 'the logic of lulz' are some of the literature's names for this phenomenon. Humour in Internet memes starts with the selection of a well-known [21] image as a visual template; it becomes an 'image macro' or 'stock character macro'. From an established template, participants add unique text to make a joke (Milner, 2013). Macro humour derives from what Shifman (2012) calls 'incongruity': a clash between expectation and experience.
As we have seen, the term 'lulz' -'laughing out loud' -denotes a detached and dissociated sense of schadenfreude, also known as 'Internet trolling'. Internet 'trolls' use this humour and antagonism to provoke angry responses and shift the content and tone of a conversation . Memetic irony thus favours personal distance and critique. However, trolling is 'equal-opportunity laughter': while it often works at the expense of core identity categories like race and gender, it can also be employed to 'troll' those categories themselves.
Finally, Internet memes -the choice of images and the discourses that surround them -thoroughly reflect the hegemonic discourses and social dynamics of the collectives creating them. Some authors argue -and this research agrees -that Internet memes constitute a formal 'media lingua franca', pliable and legible by ideologically and demographically dispersed collectives .

(b). Extreme-right online milieus and Internet memes: How, where, and why?
Now that we have explored the nature of Internet memes in some depth, we can move on to consider how, where, and why online extremists use them. We should preface this section by pointing out the lack of quantitative and qualitative data on the use of Internet memes by extreme right-wing online milieus in general. Studies do exist, but either they focus on terrorist organizations or political parties or they constitute individual studies on how social media, or the Internet more broadly, are used by particular groups. This section therefore moves away from individual or localized uses to identify general traits or trends in the Western European extreme-right milieu more broadly As previously stated, humour seems to be a key component of Internet memes. The extreme right, and especially this particularly digital milieu, fully exploits humour for several purposes. First, humour serves as a vehicle for disseminating political messages more effectively (Klein, 2019), thanks to the engaging tone it adds to visually attractive Internet memes. In social media, "memetic content is repeatedly deployed to promote extremist narratives under the guise of pop-cultural aesthetics, humour, and irony, thus lowering the barrier for participation" [22].
While some memes explicitly promote violence or extremist narratives, others are softer or more 'contextual' [23], meaning that they take on extremist connotations only when situated within a broader extremist context. Interpreting memes such as these requires some level of digital literacy and a knowledge of the extreme rightindeed, the more extreme Internet memes may perhaps be designed solely for collective participatory platforms like 4chan or Reddit, which have well-established right-wing networks familiar with the codes and eager to keep non-believers out. Conversely, softer memes are meant for mainstream social media platforms, which tend to ban or remove obviously violent, graphic or threatening content, and they target a wider audience .
This leads us to the second and perhaps most important way in which Internet memes are exploited by extreme-right groups: to facilitate the whitewashing of violence. Existing research points to this, arguing that the memes are a key aspect of online culture for the far right as they may lower the barrier for participation in extreme ideologies by using humour to mask overtly racist narratives (Marwick and Lewis, 2017).
In the case of whitewashing, aesthetic considerations are also central to the success of their messaging, "combining a variety of stylistic and aesthetic strategies and visual tools to appeal to multiple audiences, and still convey messages in line with their core ideological far-right beliefs" [24]. The cartoonish style and heavy-handed surrealism of some memes, for example, is intentionally ironic and trivialising, yet it also provides users with a degree of inherent deniability (Goldenberg and Finkelstein, 2020).
Furthermore, initial research suggests that the modern pop-cultural aesthetic deployed attracts a younger generation of digital natives, who are drawn in initially by the visual culture and then slowly become more tolerant of radical and extreme ideologies (Klein, 2020b). "Through this gradual indoctrination, a seemingly infinite number of racist, misogynist, and bigoted worldviews are made more palatable" (Marwick and Lewis, 2017). This gradual indoctrination is concerning, as the continuous exposure of Internet users to funny, vaguely racist, homophobic, or misogynistic Internet memes makes such ideas much easier to accept.
The question of 'where' is more relevant than it might perhaps appear. One reason for the extreme-right's hugely successful and varied use of Internet memes is their knowledge of different Internet ecological niches and communities. This has been confirmed by research on online right-wing milieus which distinguishes extreme-right behaviour on different platforms, from the big social media sites like Facebook, YouTube, Twitter, and Instagram to board sites like Reddit (Alava, et al., 2017;Berger, 2016;Ribeiro, et al., 2020).
All in all, the presence of the extreme right is most pervasive on imageboard sites, the so-called "chans"a multiplicity of anonymous forums (whose origins can be traced back to the early 2000s in Japan). These imageboard sites, such as 4chan and 8chan, revolve around highly controversial content nested in thematic discussion boards [25]. Here, image is king, and Internet memes are created, reformulated, distributed, and generally given a 'test run' before being distributed to other imageboards and social media sites (Milner, 2013).
The extreme right often relies on '-chan' sites for communication, and they are where appeals for violence against 'enemies' can frequently be found. Extreme-right and mainstream users of these sites also demonstrate widespread community support for acts of violence and violent revolt against the state [26]. Partly in response to recent right-wing terrorist attacks, increased attention is being paid to the significance of manifestos and their intertextuality with online extreme-right culture (Ware, 2020), as well as to the interplay between '-chan culture' glorification of violence, online radicalization processes, and extremeright symbols .
While social media and imageboards are important extreme-right fora, access to mainstream media is seen as a coveted prize, whether it be in traditional format, such as TV or newspapers, or through Facebook or Google News aggregators. Indeed, for the originators of such Internet memes, the fact that their 'work' is featured beyond the Web sites where they emerged is in itself a major success (Ludemann, 2018). Not only does it boost the ego and the online reputation of the original posters: often it also sows confusion among conventional commentators, who are rarely versed in Internet culture. Most importantly, exposure contributes to the spread of the worldviews behind those memes [27].
In terms of reach, "the alt-right manages to reach mainstream audiences through intentionally outrageous claims, via media either circulating similar ideas or attacking them", and the result is that its ideas spread to previously closed-off mainstream publics (Liyanage, 2020). The potency of memes as propaganda tools is also directly linked to amplification: collective storytelling is reinforced and perpetuated even when it is denounced, as, in online social and traditional media channels, "content spreads memetically whether participants share something to signal support, disgust, or anything on the spectrum in between" [28].
This leads us to our last question: why do extreme-right milieus use Internet memes? Propaganda and the dissemination of messages are indeed part of it. However, these memes play a role in far subtler processes, notably identity building, in/out group dynamics, and enemy identification.
First, identity-wise, some argue that cyberspace is helping to create a strong extreme right-wing collective identity and a sense of belonging on a global scale via a process of networking, sharing (of not just facts and information, but also of values, symbols, and fears), discussion, recruitment, and event organization, in similar ways to religious extremist movements (Kimmel and Ferber, 2000;Liyanage, 2020;Séraphin, et al., 2017;Weimann, 2015). Internet memes act here as a way of signalling belonging, since, by sharing or understanding particular memes, users signal their awareness of various right-wing norms, thereby creating a community accessible only to those with the appropriate codes .
For example, the production of old symbols and their reformulation into new shapes (Miller-Idriss, 2019) likely contributes to intra-group identity processes whereby 'new' or young extreme-right individuals can distinguish themselves from other extreme-right groups, or from waves within the same group. A certain level of gatekeeping can also be inferred from the way in which current Internet memes feature current pop culture references which are probably obscure or incomprehensible to older generations (Huey, 2015). From this use we can gather that, among other things, Internet memes enable individuals to signal belonging to specific groups (Bhat and Klein, 2020). These in/out group dynamics function with regard both to mainstream audiences, who would not be able to 'read' these Internet memes, and to other extremeright groups, from which the individuals wish to be distinguished ).
Last but not least, Internet memes, whether explicitly violent or not, contribute to the identification of a common enemy, thereby bolstering identity formation processes. Unsurprisingly, extreme-right milieus rely on well-established prejudices against their targets, with Internet 'trolls' acting on behalf of online rightwing milieus disproportionately targeting minorities and women (Milner, 2013). Historical stereotypes and recurring myths about immigration, homosexuals, and feminists are widespread in conservative circles, and have simply found a better propagation mode in the extreme right's nooks and crannies of the Internet (Neiwert, 2017). Similarly, existing crises that play upon those prejudices (terrorist attacks, health crises, immigration waves, etc.) are promptly capitalized on and exploited by far-right groups (Bliuc, et al., 2020).
Enemy identification is key to understanding how extreme-right groups, while divergent in ideology, composition, and activity, often share the same targets: women, gays, and ethnic or cultural minorities (Klein, 2019). Common enemy imagery also reinforces identity creation and group belonging across different fringe groups that are commonly gathered under the umbrella term 'extreme right'. For example, 'incels' (short for 'involuntary celibates'), a misogynistic group, "often blame feminism, interracial relationships, and other 'symptoms of cultural Marxism' for their lack of social and sexual success. Meanwhile, white supremacists have co-opted some of the 'incel' ideas and lexicon" [29].

Internet memes and preventing right-wing extremism
This paper will close by discussing its main findings regarding the applicability of memes in future CVE strategies. Before that, however, we introduce some problems encountered throughout the paper, as well as outlining desirable further research in the field.
First, we must acknowledge the relative uncertainty of any claims relating to Internet memes, as scholarly research, though increasing, is still insufficient, and few research projects have long-term funding. State officials and policy-makers are also likely to lack knowledge of the functioning, occurrence, and relevance of memes. Online radicalization and online extremist activity, likewise, are still new areas of research hampered by a severe lack of empirical findings, further problematizing any observations made so far.
Secondly, we must acknowledge an inherent tension in even considering how to adapt Internet memes for CVE campaigns. As seen above, memes are basically highly effective communication units, popular among highly digitalized individuals. Current approaches to CVE, on the other handespecially those addressing right-wing extremismtend to be hyper-localized in geographical terms, in their efforts to focus on creating resilient individuals or communities and counter the influence of extremist propaganda and narratives. This conceptualization of CVE clashes fundamentally with the uses of Internet memes. In stark contrast to CVE practices, which are emphatically local in theory and practice, intolerance, hate speech, and right-wing extremism are not restricted by national borders, but operate transnationally.
Thirdly, there are the potentially massive counterproductive effects for state and civil actors engaging in a 'meme war' with groups that are largely transnational and digitally competent. The biggest obstacles to state and social actors wishing to use memes are cultural articulation and legitimacy. Memes often reference pop culture and draw humorous inspiration from problems common to a certain audience (Wiggins, 2019). The sense of legitimacy conferred by the user origin of Internet memes is also key, as there is a widespread (if incorrect) perception that memes are made 'by users, for users': they carry an assumption of egalitarian origin (Chateau, 2020;Wiggins, 2019).
Furthermore, as we have explored, memes have the potential to arouse anger in people (Klein, 2020a), to the point where their sense of frustration at an institution could become outright rejection either of the institution itself or of the prevention project being conducted (Wood, 2020). A project using memes would have to devote significant human resources to developing successful cultural narratives that fitted in with the format and thematic references of the target audience while also remaining humorous enough. In fact, existing state initiatives to mobilize Internet memes to their advantage have encountered significant production problems due to their lack of knowledge of the Internet vernacular, both textually and visually (Gault, 2021). Moreover, on the Internet, such campaigns are often greeted with mockery (Wood, 2020).
Regarding the dissemination of Internet memes, there is one other development to watch closely: the pending ratification and entry into force of the EU's Directive on Regulating Extremist Content Online (European Commission, 2018). In contrast to CVE approaches, blocking propaganda means a return to traditional security-oriented approaches to extremism, it constitutes a suppression, rather than a contestation or re-channelling, of extremist ideas. The regulation has been prompted by current developments, such as the online distribution of terrorist attack footage and rampant Nazi imagery (European Commission, 2018); nevertheless, existing research and many civil society groups question the effectiveness of this method (Kuczerawy, 2018).
Internet memes will likely be flagged under this measure, but it is unclear how automated filters, or even human actors without a thorough knowledge of the extremist milieu, could identify or distinguish between the hundreds of memes used every day, in a variety of contexts, with more or less explicit violence, over multiple platforms. Fuzzy areas of law and respect for civil rights are likely to emerge as issues here, as has already happened in Spain in the case of rap music, irony, and terrorism (Nortes, 2017) [30].
To conclude, we will now discuss the main findings regarding future uses of Internet memes as part of CVE campaigns. We should start by noting that scholars and state officials are behind in this race, as "far-right media strategists are aware of the dual nature of memes (...). Humor and satire are key to contemporizing hate messages and distorting public discourse, but also to veiling the ideological roots of Nazi symbols and to [circumventing] censorship" [31].
First, while states might have a hard time deploying the humour necessary for using Internet memes successfully, individual actors and grass-roots organizations can be far more skillful at this: their youth gives them a better grip on the topics and frames best used when creating memes. At the same time, their separation from state and security institutions grants them a degree of legitimacy similar to that of individual, anonymous users. The example of Bernie Sanders, mentioned in the literature review, illustrates perfectly how a grassroots use of Internet memes can be combined with traditional, official-led approaches to produce better engagement and outcomes (Penney, 2017).
It is thus crucial for aspiring users of Internet memes to define who creates them, for what audience and with what goal. They always run the risk of inciting further mockery, rejection or anger directed against the organising state or NGO, which they represent. This is important, and we would do well not to trivialize or minimize far-right youth anger, as both "cultural and emotional rationales for extremist engagement" should be taken equally seriously [32]. Indeed this advice echoes that of established CVE/EXIT [33] projects in the EU, such as EXIT Europe or U-Turn, which highlight the importance of taking young people's grievances seriously, improving their relationship with state authorities and conducting programmes in the most localized way possible so as to avoid the stigmatization of innocent communities and/or individuals. The same approach should be taken online, when using Internet memes as vehicles for countering online extremist activity.
Secondly, and despite the above warning, there are some hopeful notes in the existing approaches to Internet memes in institutional work. While directly CVE-related Internet memes might not be a good idea, there are other possibilities. One of the most promising examples, given the importance of education-related activities in CVE, is the use of Internet memes as tools in digital literacy education (Reddy, et al., 2020).
Indeed, primary prevention targeting schoolchildren and adolescents could benefit from an NGO-led approach to the topics of extremism and political polarization. This may be the only avenue for truly countering extremist narratives with memes, as it would allow trustworthy, non-state actors to reach the target audience using familiar, emotionally and culturally powerful images in a controlled discussion environment -the classroom. Input from a mixture of NGOs and educators could help students debate and digest current extremist movements, a phenomenon they are already well aware of. At the same time, working with Internet memes in a supervised environment, in the context of political extremism, could be an excellent way to train students to identify and mistrust political propaganda. It would thus simultaneously enact primary preventive practices and promote digital literacy -a crucial policy goal in today's increasingly digital information ecology.
Thirdly, another strategy for incorporating Internet memes into CVE is to focus on awareness-raising and outreach campaigns to de-radicalize already radicalized individuals, with a special focus on youth. This means creating campaigns that use Internet memes to promote disengagement and EXIT work among extremist communities, and that offer alternative, non-extremist right-wing channels where they can voice their grievances and be heard.
In the context of countering online radicalization, "the creative subversion of memes can provide excellent opportunities for sparking tiny counter-revolutions aimed at making extremist content far less 'cool'" [34]. Disengagement and EXIT projects often have a hybrid nature, focusing both on delegitimatization and on the fostering of other, positive traits (Mérida Merenciano, et al., 2017;Ritzmann, et al., 2019). Popular formats include youth leadership programmes; engagement within and across communities; and youth sports and cultural events (BRaVE, 2020;Jaeger, 2016). In disengagement work, there is little space for Internet memes. Nevertheless, outreach campaigns are a necessary first step; further down the line, 'disengagement' memes can even foster a sense of identity and camaraderie among former extremists, who often struggle to regain a sense of self after exiting an extremist community (Reinares, 2011).
Finally, we should remember that Internet memes are, above all, powerful communicators when the aim is "to craft timely and sticky messages". They have great potential to act as sentiment shapers among the citizenry, something especially important in 'information ecologies' where governments keep a tight rein on information flows and censorship. Whether individually created or factory produced, Internet memes sometimes allow democratic citizen feedback, contentious opinions, and subversive challenges to filter through to mainstream audiences. As avenues for grassroots activism they are just as crucial as their sites of extremist activity. Depending on how it is used, the much-mistrusted 'logic of lulz', which favours distanced irony and critique, may facilitate vibrant, agonistic discussion rather than disenfranchising antagonism (Milner, 2013).

Conclusions and further research
This paper has examined different aspects of how Internet memes are used by online right-wing milieus, the secondary goal being then to devise ways in which CVE campaigns can adopt these memes too. They are crucial to understanding the online communication and processes of belonging of young people today. Being a highly effective communication format, they are likely to remain a key feature of the online sphere: through research, therefore, we must find socially positive and constructive roles for Internet memes in modern society.
First, we have introduced memes as communication units, native to the digital cultural ecology and containing a mixture of visual and textual cues. Internet memes often rely on a detached sense of humour, known as memetic irony or 'the logic of lulz', and social media and mainstream media alike act as amplifiers of extreme right-wing memes and content.
Secondly, in our examination of online extreme-right milieus and their use of memes we have seen that humour fulfills several roles: it facilitates the acceptance of hard/controversial content (whitewashing of violence); it promotes identity and group belonging (signalling group belonging and acting as community gatekeeper); and finally, it can spark heated debate while maintaining the ironically detached air favoured in online communications. It is important to remember that differences between those uses are not readily perceivable by external and/or uneducated actors.
While we still have much to learn about Internet memes, we do know that their creation and dissemination can be a goal in itself. In other words, the use of coded symbols by online milieus can be understood as an everyday form of youth resistance "to perceived societal pressure or the unattainability of their future goals" [35]. Only policies and research that earnestly take these needs into consideration will be able to harness the power of Internet memes for CVE.
Thirdly, with a view to potential CVE approaches, we have outlined ways in which Internet memes are and are not a good fit for CVE projects. Individual and grassroots uses of Internet memes shed light on how humour and visual culture can be used to explore and negotiate issues of race, gender and group belonging. Finally, the role of Internet memes as catalysts of political debate needs to be framed as helpful in handling the problems of social polarization and political extremism(s). This article ends by encouraging further research that will provide a better and more nuanced understanding of how audiences receive and interact with Internet memes on different platforms. We also need a more accurate picture -age, gender, ethnicity, education -of the sections of the extreme-right milieu (in itself quite vast) that are most likely to use Internet memes, and why. The location and type of content deployed are also of interest for future research and policy work alike.
Finally, research in this area will have to consider developments brought about by the forthcoming EU legislation, which is likely to stir national sentiment and, presumably, alter how these online, transnational, extreme-right milieus use Internet memes.

About the author
Inés Bolaños Somoano is a Ph.D. researcher at the European University Institute in Florence and a visiting fellow at the Institute for Security and Global Affairs (ISGA) in The Hague. Her dissertation examines the European Union's governance of prevention of terrorism, its development and the role of institutional actors and practitioners in the implementation of Preventing and Countering Violent Extremism (P/CVE) programmes in Europe. Her research favours interviews, participant observation and policy analysis. Inés is also active as a lecturer in terrorism studies.