13

LOCALIZED HATRED

The Importance of Physical Spaces within the German Far-Right Online Counterpublic on Facebook

Jonas Kaiser^*

IN 2015, AT THE HEIGHT OF THE SO-CALLED “refugee crisis,” around one million refugees arrived in Germany. And while the majority of Germans, including Germany’s chancellor Angela Merkel, welcomed the refugees, the country’s far right opposed it vehemently. For them, the arrival of Syrian, Afghan, or Iranian refugees signified nothing less than an “attack” on “their” homeland, making the ones who were in favor of welcoming the refugees “traitors.” The influx of refugees turned out to be a discursive opportunity (Koopmans and Olzak 2004), that the—until then marginalized—far right harnessed. Most prominently, the civic movement Pegida was formed; protesters—at times, more than twenty thousand people—would meet every Monday in Dresden and rally against the government, refugees, and the Lügenpresse (lying press). These protests, however, took place not only in Dresden but also across Germany. Often organized online, these protests highlighted the connection between online communication and off-line action. Against this background, I examine the role physical spaces play in the networked public sphere of the German far right on Facebook.

The refugee crisis not only led to the formation of Pegida but also sparked the success of the far-right political party Alternative für Deutschland (AfD, Alternative for Germany), which, as of late 2018, is represented in all state parliaments and the federal parliament. The rise of the country’s far right was accompanied by the rise of far-right violence. In the years 2015, 2016, and 2017, more than four thousand far-right acts of violence were officially documented (Verfassungssschutz.de 2018).¹ Off-line, far-right extremists attacked refugees and even burned down refugee shelters (ninety-two arson attacks in 2015; Diehl 2016). Online, they created websites to document alleged “refugee crimes,” filled the comment sections of news outlets with antirefugee comments and extreme speech, and created maps that showed the locations of refugee shelters, including their addresses and telephone numbers (Heimbach 2018). This happened on blogs, websites, and, of course, social media platforms—that is, integral parts of what Benkler (2006) called the “networked public sphere.” In short, the networked public sphere is the networked sum of all the spaces that allow for the formation of online publics, connected through users (both through content production and consumption), signals (most prominently, hyperlinks), or content (e.g., YouTube videos can both be on YouTube and embedded in a blog).

From a networked public sphere perspective, few online spaces are as interesting as social media platforms. They allow people to connect, communicate, and coalesce with each other regardless of their physical locations. More important, they allow people to express themselves in a way that was not possible in a semipublic or public way before.² Consequently, social media platforms are especially interesting with regard to their potential for counterpublics—that is, oppressed and/or marginalized groups—as platforms like Facebook shorten the path between fringe and center (Batorski and Grzywińska 2017; Puschmann et al. 2016).³ With its size and societal importance, Facebook, in particular, is relevant for the networked public sphere (Benkler 2006): the platform is being used by more than two billion people worldwide, of which about thirty million are German (Statista 2017). And while Facebook has private and cultural dimensions, the platform is often at the forefront regarding political communication and extreme speech—that is, vitriolic speech (Stier et al. 2017).

One aspect that is often ignored in research on the networked public sphere is the role of localities—that is, how off-line spaces affect the formation of online spaces. In this chapter, I am interested in the particular connection between extreme speech and physical spaces. More specifically, I am interested in the role that physical spaces play for the extreme right in Germany on Facebook. This aspect is being analyzed against the backdrop of extreme speech discussions being put forward by Udupa and Pohjonen (2019). The concept is fruitful for this analysis, as it highlights that extreme speech should be understood as a “spectrum of practices” and not a binary distinction. Furthermore, it emphasizes digital affordances as contributors to online communication, particularly in the context of extreme speech. Because the concept is also context dependent, it allows for inquiry into the far right’s activities on Facebook and how localities are evoked. The German far right was picked as a case study to understand the connections and potential disconnects between physical spaces—understood here as discrete references to localities (e.g., village, city, region, state, or country names) and off-line incidents like attacks on refugee homes, networks, and situations—and extreme speech.

In recent years, the German far-right party AfD, the far-right civic movement Pegida (English translation: Patriotic Europeans against the Islamization of the Occident), and the far-right extremist identitarian movement (IM; the group refers to itself internationally as Generation Identity, but IM in this context refers to the German faction of the international organization) are making use of the platform to disseminate their messages, to connect with each other, and to recruit new members.

Stier et al. (2017) show, for example, that AfD and Pegida cover distinctly different topics on Facebook in contrast to the established political parties, thus distinguishing themselves starkly. For communication, the German far right is not limiting itself to pages or groups; it also conveys distinct messages through the names of its pages. One page is, for example, named “Für Familie Volk und Heimat Multikulti und Islamisierung stoppen” (For family volk [right-wing populist description of the people] and homeland, stop multiculturalism and Islamization). Another is called “Rostock Bürgerinitiative gegen die aktuelle Asylpolitik” (Rostock civic movement against the current asylum politics). In doing so, the German far-right counterpublic is not only signaling a clear stance through its Facebook page names but also connecting its online activities with physical spaces.

The latter activity is of particular interest when it comes to the networked public sphere. In the past, publics were usually dependent on physical spaces (e.g., people could only read a certain newspaper in a certain region or town or could discuss a certain issue with the people who were at the same place at the same time). The internet allowed for communication that was uncoupled from physical space and time, forming the networked public sphere (Benkler 2006). But as Friedland argues, “we cannot really separate the types and flows of communication that take place within social networks from the social ecology in which they are embedded” (2015, 25). Similarly, Tufekci (2017) and Chadwick (2016) have highlighted that there is a complicated connection between online communication and off-line political action that goes both ways. In her research, Tufekci (2017) was able to track how important platforms like Twitter were for Turkish activists during the Tahir Square protests and how online communication and action on the ground were interlinked. Similarly, Chadwick (2016) argues that the online sphere and especially social media platforms contributed to a “hybrid” system in which intermedia agenda setting between old and new media constantly takes place. This strain of research is important in the context of civic movements and the question of where and how they will be able to have an impact (on- and/or off-line)—including those from the far right such as Pegida.

In this chapter, I am interested in the role that physical spaces (i.e., localities) play for the German far-right counterpublic in its communication on Facebook.

To be able to answer this question, I mapped the structure of the German far-right counterpublic on Facebook to then identify the communities within the far right. Based on these results, I analyzed the role that physical spaces play for the German far right. My results show that the far right is disconnected from the political mainstream and has formed a counterpublic, including alternative media outlets and different far-right actors (either voluntarily or involuntarily). I am also able to show that the German far right is connected to the international far right from Italy, France, and Hungary. Physical spaces in the German far right on Facebook usually refer to either local branches of organizations (be it AfD, Pegida, fraternities, or IM) or locality-based opposition to immigration, Islam, and/or asylum politics.

Physical Spaces in the Networked Public Sphere

While scholars often refer to the public sphere, Fraser (1990) highlighted that there is not just one public sphere. Instead, the public sphere can be best understood as consisting of numerous publics—some smaller than others and some with more power than others. But Fraser (1990) also pointed out that some publics will face discrimination or marginalization, and thus the so-called subaltern publics or counterpublics have to be considered as the least powerful publics. She also emphasized that this category includes antidemocratic groups. This, too, is true for the internet.

But to make sense of the role that physical spaces might play for networked counterpublics, we have to consider how the public sphere can be traditionally understood. For Habermas (1992/1996), “The public sphere can best be described as a network for communicating information and points of view (i.e., opinions expressing affirmative or negative attitudes); the streams of communication are, in the process, filtered and synthesized in such a way that they coalesce into bundles of topically specified public opinions” (360).

Particularly in the context of online communication, Habermas’s definition seems appropriate because it highlights the networked character of publics as a key trait of the public sphere. Benkler’s (2006) concept of the “networked public sphere,” which posits that the internet allowed for the connection of numerous publics that would be able to circumvent traditional gatekeepers and thus have a bigger impact on the agenda-setting and public opinion formation process, can also be understood in the tradition of Habermas’s definition.

In off-line definitions of the public sphere, the question of physical “spaces” was important because these spaces inherently created in- and out-groups, and online publics seemingly transcended local boundaries. But as Murray highlights in analysis of the Occupy movement, “Discursive interactions, media spectacles, and social media use are all integral to social movement success, but an understanding of their interdependence with the physical spaces of the public sphere is also essential” (2016, 16). Furthermore, Sampson puts forward that “collective action is concentrated ecologically and better explained by the density of community organizations than individual social ties or membership in traditional civic groups” (2012, 181). Indeed, this is in line with research on counterpublics and social movements before the internet and how they used bookstores or similar venues to meet up and organize (for an overview, see Nuernbergk 2013).

This, too, is true for far-right counterpublic movements like Pegida that organized online to meet off-line (e.g., Haller and Holt 2018; Nam 2017). Online references to physical spaces can, in this sense, be understood as an anchor, a direct reference that connects the off-line with the online world and thus enforces in- and out-groups based on locality. This dynamic seems interesting in the context of counterpublics because it allows for the far right to directly relate their online actions to the off-line world and vice versa (think also of the “training grounds” to which Fraser refers that were thought of as both alternative media and physical spaces). In doing so, however, the counterpublics also make themselves vulnerable to counterspeech or other forms of marginalization: in anchoring online speech to a physical space, they open themselves up to counteraction. At the same time, having clear markers that refer to the off-line world might also contribute to a counterpublic’s collective identity (Kaiser and Rauchfleisch 2019) because it specifies a clear and less abstract target for counterpublic action (e.g., wanting to enact political change in a local town might be more feasible than wanting to do so for the whole country).

Against this background, it is clear that physical spaces are an important factor in the general public sphere but even more so in the context of counterpublics. This study asks what role physical spaces play for the German far right on the social media platform Facebook.

Method

To answer this study’s overarching research question, I searched Facebook for 252 different terms that are connected to the far right (e.g., patriot, afd, pegida, flüchtlinge, etc.) using the Graph application programming interface (API). This search resulted in 26,140 unique pages. Because both users and pages can “like” other pages on Facebook, I collected the page likes as a proxy for allegiance between pages, as the “like” feature seems to be frequently used that way (e.g., the pages of political parties will like the other pages of the same political parties and their politicians; media outlets will like other affiliated media outlets of the same publishing house; local politicians will like their local sports clubs). Although this method has been used before to map the network of individual sites (e.g., Krieg, Berning and Hardon 2017), I am not aware of a similar attempt to map a country’s Facebook “landscape.” I collected the page likes of these 26,140 unique pages and, in a next step, analyzed the pages that these pages liked (liked pages from the liked pages). The method to go from liked pages to the liked pages of said pages is called a “snowball” and aims to unravel a bigger structure in a network (much like a snowball grows the longer it rolls down a hill). Unsurprisingly, what started with about 26,000 pages resulted in 2,536,575 million pages. And while I started with a focus on the German far right, most of those 2.5 million pages are not from the far right. Coca Cola is in the network. And Vogue. And Tulsa Zoo. This method, then, unveiled a map of close connections and very different topics and showed how international Facebook truly is. The data were collected in April 2018 with R and the package Rfacebook (Barberá 2017) just before Facebook closed down its API access.

To answer the research question more thoroughly, I focused on the German far-right community. I was able to identify the far-right community within the 2.5 million pages through the community detection algorithm modularity (Blondel et al. 2008). The resulting German far-right Facebook “page-like network” consists of 4,054 pages that are connected by 45,263 likes. I then identified the communities within the network through modularity (community detection algorithm; fig. 13.1) and labeled the communities based on their most prominent pages (see Kaiser and Rauchfleisch, forthcoming). And while I had planned on retrieving sample posts from each relevant far-right page, this is, unfortunately, not possible anymore because Facebook has closed the API. Consequently, in this study, I focus on an analysis of the page names.

To understand the role of physical spaces within the far-right counterpublic, I focused on the Pegida community, as it is the most prominent far-right civic movement within the German far right. As was established earlier, physical spaces are especially relevant for civic movements because they reinforce a movement’s identity through action both on- and off-line (Murray 2016). In addition, the Pegida community is the one community in the network that has references to physical spaces that go beyond their local affiliation to a party or organization. I then manually coded the 413 page names of the Pegida community based on four categories that were rooted on the earlier definition of physical spaces. “Location (general)” asked whether the page name referred to a general location (ranging from municipalities to countries or country unions like the European Union). “Location (specific)” asked whether the page name referred to a specific location—that is, to a distinct village, town, or city. I further coded whether the page name contained a distinct “message”—that is, whether the name was a message in itself (e.g., “No to the [refugee] shelter in Guben”) and not just the name of an organization (e.g., “Pegida Paris”). Furthermore, I coded whether the page names included a “target”—that is, something or someone that the page name was directed at (e.g., Angela Merkel, Islamization, refugee shelters, immigration). In a final step, I mapped the specific locations on a map of Germany with the R package ggmap (Kahle and Wickham 2016) to understand which physical spaces the Facebook pages were referring to and whether there was an underlying pattern.

Figure 13.1 The German far-right network on Facebook. Community labels were assigned based on the most prominent pages (i.e., with the most indegree) within the communities (nodes = 4,078, edges = 45,403). Source: Kaiser & Rauchfleisch (forthcoming).

Results

The network analysis of the Facebook page-like network (fig. 13.1) shows that there are several distinct communities within the German far right. Among them are far-right parties (e.g., AfD, NPD [a far-right extremist political party that can be considered neo-Nazi], Republikaner but also the Austrian FPÖ), the IM, Pegida, German and Austrian fraternities, libertarians, and far-right music, but also alternative media outlets and conspiracy theory sites. The “alternative media and conspiracy theory” community that is represented by pages such as KenFM or RT Deutsch (Russia Today’s German branch) especially seem to serve as a bridge between the different communities.

An analysis of the top 100 words of the page names (fig. 13.2) shows two aspects: on one hand, it confirms the prominence of the far-right parties AfD, NPD, and FPÖ as well as Pegida or the IM within the network; on the other hand, it also emphasizes the importance of physical spaces. Among the top 100 words are cities like Berlin, München (Munich), or Dresden. This result is not necessarily surprising. Most political parties will have local Facebook pages to flag events and meetings, to share news articles, or to give their members a virtual space to connect and communicate. The far right is no exception. Indeed, the political party communities that we see in figure 13.1 can be identified so clearly because they mostly consist of local pages that like each other, creating dense clusters. The same is true for the right-wing Burschenschaften (i.e., akin to fraternities but usually more conservative and “traditional”) that can be found in most German and Austrian university towns and cities (e.g., “Burschenschaft Arminia zu Leipzig in Dresden”). This highlights the role of physical spaces as delineators within a bigger group or organization that might also signify their importance within the overarching structure—for example, there is the general and most important AfD page, but there are also numerous local pages that, although not relevant on the federal stage, have an importance within the local context.

As a next step, I focused on the Pegida community because it can be considered both the most important and the most successful civic movement within the German far right. At its height, the weekly march in Dresden attracted more than twenty-five thousand people in 2014 (Berger, Poppe, and Schuh 2016). Although much smaller in number, Pegida still marches every week through Dresden as of late 2018. A network analysis of the Pegida community highlights that it contains two overarching subcommunities (fig. 13.3): one is Pegida and its local branches; the other one, however, is a dense network of pages that oppose refugee shelters in their villages, towns, cities, or municipalities. The latter is interesting against the background of the overarching research question. These pages are usually called “No to the refugee shelter in XYZ,” with slight deviations (e.g., replacing refugee shelter with asylum politics) and are reminiscent in their name of the refugee shelter map that was published by the far-right party NPD. Interestingly, these very local pages form a distinct community in which many pages like other similar pages. Within the overarching far-right network in figure 13.1, the anti–refugee shelter community is located in between the Pegida and the NPD communities.

When looking at the top 100 words of the Facebook page names within the Pegida community (fig. 13.4), three themes become apparent. First is the theme of opposition, which is signaled through words like no, oppose, or stop. Opposition in this context is mostly directed at the idea of establishing refugee shelters (referred to often as “Heime” or “Asylheim”) in a certain place. This general opposition is sometimes framed as opposing the “abuse of asylum” (Asylmissbrauch). The idea of abusing asylum is directed at the refugees who came into the country in 2015 (most prominently but also in the following years) and whose claim to asylum are—without evidence—being framed as illegitimate. The second theme is bound to a certain framing of citizenship that is expressed through words like our, patriotic, patriots, solidarity, or citizen’s initative (in the word cloud, the stemmed version “Bürgerin” is prominently visible; however, the word does not stand for the female version of “citizen” but for “Bürgerinitiative”; see fig. 13.4). This theme evokes the idea of the populist “true” people who stand against the elites and who know what is best for the country. The third theme that can be identified within the page names is localities. Unsurprisingly, several cities and towns such as Berlin, Stuttgart, and Heidenau, as well as countries like Hungary, Switzerland, and Austria, are among the top 100 words. As I show below, most of the countries are mentioned in the context of the national Pegida branches (e.g., Pegida France), whereas most of the more specific locations are associated with the anti–refugee shelter pages.

Figure 13.2 Word cloud based on the page names within the overall far-right community (n = 4,078; word size according to frequency).

This general analysis poses the question of just how many page names truly express a message, name a specific target to which they are opposed, and/or define a location, (i.e., a physical space). These categories are not exclusive—that is, pages could technically refer to general as well as specific locations and communicate a message as well as a target. To answer these questions, I conducted a content analysis of the 413 Facebook page names within the Pegida community (table 13.1). Based on this analysis, approximately 34 percent of the pages name a general location (e.g., Germany or Saxony), whereas approximately 43 percent refer to a specific location (e.g., Berlin, Stuttgart, Paris, or Munich). Some pages referred to both a specific location and a general one; only ninety-five pages did not refer to any location in their name. Once more, this highlights the importance of locations for the Pegida community. Indeed, it is noteworthy that more pages referred to a specific rather than a general location.

On a content level, the analysis shows that approximately 42 percent of the Facebook page names conveyed some form of message and approximately 23 percent of Facebook page names were even directed at a certain target (table 13.1). This result is in line with the top 100 words that were discussed earlier (fig. 13.4). Although some Facebook pages had names with longer messages, such as “Marburg sagt NEIN zur Überfremdung und zu Asylbetrug” (“Marburg says NO to foreign infiltration and asylum abuse”) or “Colditz sagt Nein aus Liebe zur Heimat und zum Schutz unserer Kinder” (“Colditz says no out of love of our homeland and to protect our children”), others were short, such as “Denk ich an Deutschland in der Nacht” (“Should I think about Germany at night,” a reference to a famous Heinrich Heine poem that continues, “Then I can no longer sleep, / It’s a sleeplessness I could not fight, / And with warm tears I begin to weep”⁴) or “Ein Volk hilft sich selbst” (“A volk helps itself”). In addition, some page names were also directed at a certain target. This was discussed earlier (regarding refugee shelters) but also can be seen in the examples I gave for the page names that functioned as messages. While most of the page names that were directed at a target referred to a refugee shelter, some pages referred to Germany’s chancellor Angela Merkel (“Merkel muss weg Mittwoch”; “Merkel has to go Wednesday”), the left wing (“Bündnis gegen Linksextremismus und Inländerfeindlichkeit”; “Alliance against left-wing extremism and hostility against natives”), or Islam (“Für Deutschland ohne Raute Islam und Nestbeschmutzer”; “For Germany without rhombus islam and people who foul their own nest”).⁵

Figure 13.3 Network of the Pegida community with the Pegida subcommunity in the north and the anti–refugee shelter subcommunity in the south (nodes = 413, edges = 3,288; node sizes per indegree; communities identified with modularity).

Figure 13.4 Word cloud of the page names within the Pegida community (n = 413; word size according to frequency).

Next, I imported the coded information into the Pegida network (done with the app Gephi). This allowed me to visualize the coding information in a more straightforward manner that highlights where general and specific locations were referred to and where in the network messages and targets were communicated (fig. 13.5). A visual inspection of the four different networks shows how the general locations were usually used in the context of the Pegida pages (the northern community in the network; see also fig. 13.3), whereas the specific locations were mostly used in the community against refugee shelters (the southern community). The network, however, also shows that specific locations are also present within the Pegida community. In contrast, the “message” and “target” networks highlight that barely any page in the Pegida community tried to convey a message and/or identify a target through its page name, whereas the anti–refugee shelter community consists mostly of pages that do both.⁶

Table 13.1 Coding of Facebook page names (of the Pegida community) based on four variables.

By visualizing the coded information, it becomes clear that physical spaces are important both for Pegida and the anti–refugee shelter community. The main difference is that the anti–refugee shelter community is inherently focused on specific locations—that is, the villages, towns, or cities in which refugee homes were planned or located. In that sense, these pages and their reference to specific locations serve a distinct function: to organize the local opposition in one virtual place.

In a final step, I identified all the specific locations that the Facebook page names referred to and located them on a German map (fig. 13.6; created in R with the package ggmap). Without duplicates, I was able to locate 139 villages, towns, or cities.⁷ Although most were in Germany, some were in Austria, Switzerland, Czech Republic, or France (not all were within the boundaries of the map). In doing so, I located the physical spaces to which the virtual Facebook pages referred. Next, I calculated the density within the map based on the proximity of each dot (i.e., village, town, or city) to another. This analysis shows that although the Facebook page names, in general, refer to almost any area in Germany, three areas show a higher density of Facebook pages: the area around Stuttgart (ranging from Rhineland-Palatia, Baden-Württemberg, Hesse, to Bavaria); the area around Dortmund, Cologne, and Essen in North Rhine–Westphalia; and, most prominently, East Germany. In East Germany, we can even identify two distinct hotspots: one around Berlin and one in the area of Leipzig, Dresden, and Chemnitz. The heat map shows which regions are the most outspoken against refugee shelters on German Facebook. This is interesting because refugee shelters are not only in these regions but, according to the far right’s own maps, everywhere in Germany (Sandgathe 2015).

Against the background of counterpublic theory, these findings highlight how Facebook allows the far right to coalesce and, more important, to form a collective identity that is anchored in physical spaces. From this shared identity, the far right might potentially be emboldened to spread their vitriolic speech onto other pages, as they can also use their pages to call for directed action (e.g., harassment); especially against Islam and refugees, which are core targets for the German far right.

Figure 13.5 Distribution of codes in the Pegida community network (nodes = 413; edges = 3,288) for the variables location (general), location (specific), message, and target.

Discussion

In this chapter, I examined how the German far right refers to making use of physical spaces on Facebook. To answer this question, I analyzed a Facebook page-like network based on data from April 2018. In doing so, I was able to show that the German far right uses Facebook not only to connect members with each other and to communicate but also to mobilize; indeed, the community forms its online counterpublic on Facebook. Specifically, by analyzing the Pegida community regarding the variables location (general), location (specific), message, and target, I was able to contextualize the Facebook network with off-line data.

Figure 13.6 Map and heat map of Germany and neighboring countries based on specific locations within the Facebook page names (n = 139; each dot represents a village, town, or city that was referred to by a Facebook page; the heat map shows density estimation of Facebook pages).

This analysis shows three important findings. First, three themes are prominent within the Pegida subcommunity: opposition, citizenship, and localities. While the opposition defines the target—that is, the out-group (who are we up against?)—the citizenship defines the in-group’s identity (who are we?), and the localities anchor the virtual formation of the far-right counterpublic in the physical space and offer off-line events to get involved.

Second, physical spaces are used in two distinct ways. On one hand, physical spaces (i.e., localities) function as delineators within a bigger group or organization that might also signify their importance within the overarching hierarchical structure. On the other hand, physical spaces function as anchors within collective identity formation because they offer clear in- and out-groups (see the first finding in the previous paragraph) and an immediate way to get involved, making the idea of a counterpublic more direct and meaningful—that is, you know who as well as what you oppose, but you also know where to directly get involved in the off-line world, making the link between online and off-line more immediate. In addition to these two functions, the localities may also serve a third function: recruitment. As highlighted elsewhere (Kaiser and Rauchfleisch, forthcoming), the far right uses the internet in sophisticated ways to recruit new members. The localized anti–refugee shelter pages may also serve as gateways for local citizens to the greater far-right counterpublic and their messages on Facebook.

Third, the virtual Facebook pages refer to real-life places and, as such, make a pattern visible. Although Pegida and anti–refugee shelter pages refer to almost all German regions, a density analysis highlights three particular areas in Germany where the far right is especially prominent. Most notably, eastern Germany is the region with the highest density of Facebook pages. However, this finding does not come as a surprise, given that the region is notorious for its far-right activity. Not only did Pegida originate in Dresden but also the AfD is polling as the second biggest party in the states of Saxony, Mecklenburg-West Pomerania, and Saxony-Anhalt, as well as Brandenburg and Thuringia.

This study thus highlights the importance of physical spaces within the concept of extreme speech: on one hand, these spaces allow for direct connection between on- and off-line and thus give access to pages that center around extreme speech; on the other hand, the pages in combination with focused extreme speech might manifest in real-life violence (e.g., arson attacks on refugee shelters). In looking at the intersection of off-line and online on Facebook, extreme speech connections are made visible. For counterpublic theory, this analysis highlights the importance of keeping physical spaces in mind when analyzing a counterpublic’s online activity. It is necessary to analyze the functions that physical spaces serve for counterpublics at both smaller and larger scales.

It must be noted that the results account for only a few pages within the broader German far-right spectrum, and no correlation or causality should be drawn from this analysis. Although the identification of the anti–refugee shelter pages is instructive for research into the far right, it is important to acknowledge the limitations of this study. We need more insight into general patterns of user overlap between the far-right pages to understand how successful these pages are in recruiting new members. We also need more qualitative work that outlines the role of physical spaces in the online realm against the backdrop of the public sphere and counterpublic theory.

Notes

* This work on the US and German far right was funded by the German Research Foundation (KA 4618/2-1).

1. There is a notable difference between official numbers and the numbers that observers of the far-right–like foundations or news outlets publish. The official numbers are quoted here because they can be understood as the guaranteed minimum of far-right violence. It is highly likely that the true numbers are higher.

2. This is mostly true for liberal democracies.

3. It has to be noted that neither the public sphere nor counterpublics are understood from a normative but rather a functionalist perspective in this text (see also Kaiser andand Rauchfleisch 2019). Nancy Fraser argues, for example, that “that subaltern counterpublics are [not] always necessarily virtuous; some of them, alas, are explicitly anti-democratic and anti-egalitarian; and even those with democratic and egalitarian intentions are not always above practicing their own modes of informal exclusion and marginalization” (1990, 67).

4. Poem available at http://www.heinrich-heine.net/haupt.htm.

5. “Rhombus” is a reference to the so-called Merkel rhombus, which describes the hand gesture that Merkel does in the shape of a rhombus.

6. Although Pegida is in itself an abbreviation of a targeted message against Islam, the pages were only coded as such if the name was spelled out, as it cannot necessarily be assumed that everyone knows what the abbreviation Pegida stands for.

7. Because some page names referred to towns with names that exist multiple times in Germany or to overly specific places (e.g., one page name referred to a military base), I had to double check which place they were referring to. To do so, I visited the Facebook pages, looked at the Facebook page’s information (sometimes it would refer to an associated website that specified the town in its imprint), the pages that the page had liked (e.g., one page liked the village’s football club) or the Waybackmachine (sometimes the web archive will have a prior version with more information).

References

Barberá, Pablo. 2017. Package “Rfacebook.” https://cran.r-project.org/web/packages/Rfacebook/Rfacebook.pdf.

Batorski, Dominik, and Ilona Grzywińska. 2017. “Three Dimensions of the Public Sphere on Facebook.” Information, Communication and Society 21 (3): 356–374. doi:10.1080/1369118X.2017.1281329.

Benkler, Yochai. 2006. The Wealth of Networks: How Social Production Transforms Markets and Freedom. New Haven, CT: Yale University Press.

Berger, Roger, Stephan Poppe, and Mathias Schuh. 2016. Zur Problematik der Zählung von Demonstrationsteilnehmern (Everything counts in large amounts). In Pegida—Rechtspopulismus zwischen Fremdenangst und »Wende«-Enttäuschung? Analysen im Überblick, edited by Karl-Siegbert Rehberg, Franziska Kunz, and Tino Schlinzig, 113–132. Bielefeld, Germany: Verlag. doi:10.14361/9783839436585-009.

Blondel, Vincent D., Jean-Loup Guillaume, Renaud Lambiotte, and Etienne Lefebvre. 2008. “Fast Unfolding of Communities in Large Networks.” Journal of Statistical Mechanics: Theory and Experiment 2008(10): P10008. doi:10.1088/1742-5468/2008/10/P10008.

Chadwick, Andrew. 2016. The Hybrid Media System: Politics and Power. New York: Oxford University Press.

Diehl, Jörg. 2016. “Gewaltwelle: BKA zählt mehr als tausend Attacken auf Flüchtlingsheime.” Spiegel Online, January 28. http://www.spiegel.de/politik/deutschland/fluechtlingsheime-bundeskriminalamt-zaehlt-mehr-als-1000-attacken-a-1074448.html.

Fraser, Nancy. 1990. “Rethinking the Public Sphere: A Contribution to the Critique of Actually Existing Democracy.” Social Text 25/26:56–80. doi:10.2307/466240.

Friedland, Lew A. 2015. “Networks in Place.” American Behavioral Scientist 60 (1): 24–42. doi:10.1177/0002764215601710.

Habermas, Jürgen. 1996. Between Facts and Norms: Contributions to a Discourse Theory of Law and Democracy. Cambridge, UK: Polity. (Original work published 1992.)

Haller, André, and Kristoffer Holt. 2018. “Paradoxical Populism: How Pegida Relates to Mainstream and Alternative Media.” Information, Communication, and Society 22 (12): 1665–1680. doi:10.1080/1369118x.2018.1449882.

Heimbach, Tobias. 2018. “NPD veröffentlicht auf Google Maps Karte mit Asylunterkünften.” Welt.de, February 5, 2018. https://www.welt.de/politik/deutschland/article173227076/NPD-veroeffentlicht-auf-Google-Maps-Karte-mit-Asylunterkuenften.html.

Kahle, David, and Hadley Wickham. 2016. Package “ggmap.” https://cran.r-project.org/web/packages/ggmap/ggmap.pdf.

Kaiser, Jonas, and Adrian Rauchfleisch. 2019. “Integrating Concepts of Counterpublics into Generalized Public Sphere Frameworks: Contemporary Transformations in Radical Forms.” Javnost—The Public 26 (3): 241–257. doi:10.1080/13183222.2018.1558676.

———. Forthcoming. Like, Subscribe, Radicalize: The Internet in the Hands of the Far-Right. New York, NY: Oxford University Press.

Koopmans, Ruud, and Susan Olzak. 2004. “Discursive Opportunities and the Evolution of Right-Wing Violence in Germany.” American Journal of Sociology 10 (1): 198–230. doi:10.1086/386271.

Krieg, Lisa, Jenny Moritz Berning, and Anita Hardon. 2017. “Anthropology with Algorithms? An Exploration of Online Drug Knowledge Using Digital Methods.” Medicine Anthropology Theory 4 (3): 21. doi:10.17157/mat.4.3.458.

Murray, Billie. 2016. “The Sphere, the Screen, and the Square: “Locating” Occupy in the Public Sphere.” Communication Theory 26 (4): 450–468. doi:10.1111/comt.12101.

Nam, Sang-hui. 2017. “Spontane Mobilisierung und der Wandel kollektiver Formationen im Internet. Eine Fallstudie zur PEGIDA-Bewegung.” Forum: Qualitative Social Research 18 (1): 1–31.

Nuernbergk, Christian. 2013. Anschlusskommunikation in der Netzwerköffentlichkeit: Ein inhalts- und netzwerkanalytischer Vergleich der Kommunikation im “Social Web” zum G8-Gipfel von Heiligendamm. Baden-Baden, Germany: Nomos.

Puschmann, Cornelius, Julian Ausserhofer, Noura Maan, and Markus Hametner. 2016. Information Laundering and Counter-Publics: The News Sources of Islamophobic Groups on Twitter. Paper presented at the International AAAI Conference on Web and Social Media, Cologne, Germany. http://www.aaai.org/ocs/index.php/ICWSM/ICWSM16/paper/view/13224.

Sampson, Robert. 2012. Great American City: Chicago and the Enduring Neighborhood Effect. Chicago: University of Chicago Press.

Sandgathe, Laura. 2015. “#BrauneKarte: Rassistische Karte zeigt Flüchtlingsheime in Deutschland.” RP Online, July 17, 2015. https://rp-online.de/panorama/deutschland/braune-karte-google-maps-zeigt-fluechtlingsunterkuenfte-in-deutschland_aid-18710611.

Statista. 2017. “Anzahl der Nutzer von Facebook und Instagram in Deutschland im Jahr 2017 (in Millionen).” Statista—Das Statistik-Portal. https://de.statista.com/statistik/daten/studie/503046/umfrage/anzahl-der-nutzer-von-facebook-und-instagram-in-deutschland/.

Stier, Sebastian, Lisa Posch, Arnim Bleier, and Markus Strohmaier. 2017. “When Populists Become Popular: Comparing Facebook Use by the Right-Wing Movement Pegida and German Political Parties.” Information, Communication, and Society, 20, 1365–1388. doi:10.1080/1369118X.2017.1328519.

Tufekci, Zeynep. 2017. Twitter and Tear Gas: The Power and Fragility of Networked Protest. New Haven, CT: Yale University Press.

Udupa, Sahana, and Matti Pohjonen. 2019. “Extreme Speech and Global Digital Media Cultures: Introduction.” International Journal of Communication 13:3049–3067.

Verfassungsschutz.de. 2018. Rechtsextremistische Straf- und Gewalttaten. https://www.verfassungsschutz.de/de/arbeitsfelder/af-rechtsextremismus/zahlen-und-fakten-rechtsextremismus/rechtsextremistische-straf-und-gewalttaten-2017.