Thematic analysis of in-group and out-group debates in an online right-wing extremist community

Abstract

The resurgence of Right-Wing Extremism (RWE) has become a cause for concern in Western countries during the 21st century. An increase in white nationalist sentiment in recent years has provided impetus to understand the contemporary drivers of far-right ideology. This study examines in-group and out-group debates on Stormfront.org, which is the largest and oldest online community dedicated to white nationalism and extreme right-wing political views. It used a dataset of approximately 1m posts collected from the open Opposing Views forum on Stormfront from the period 2001 to 2020 to create a corpus of over 195m words for thematic analysis. A Natural Language Processing (NLP) model was used to analyze the corpus, and a supervised phrase mining algorithm was used to identify key topics in the debate. The study finds that key issues being debated between in-group and out-group members of the Stormfront online community relate to perceptions of white identity, African American identity, racial issues, conservative political issues, and the history and politics of the United States. These findings highlight the issues of mutual importance to in-group and out-group members and identify opportunities for further research into the ideology of online RWE communities.

Keywords

Right-Wing extremism thematic analysis political discourse online community Stormfront

Introduction

The resurgence of nationalist politics and Right-Wing Extremism (RWE) in the West has become a significant cause for concern in the 21st century (Fahey and Simi, 2019; Ravndal, 2018). There has been a particular rise in far-right hate groups in the US (Chermak et al., 2013) and Europe (Falter and Schumann, 1988; Koehler, 2016). Following a wave of violent attacks motivated by right-wing grievances in the 2000s and 2010s, RWE has come under greater scrutiny as a potential source of emerging threats to domestic security (Stevenson, 2019). This threat has included the threat or justification of violence against minority groups, particularly vulnerable groups, and social behaviors that are inconsistent with highly conservative worldviews often promoted in far-right movements (Bjoergo, 2003; Farago et al., 2019).

Amidst debates about tensions between liberties, such as freedom of speech and political expression, and protections from hate-speech and violence, the line between political and social movements and extremists movements has become blurred in popular commentary (Adamczyk et al., 2014; Lowe, 2020). RWE has been defined variously as a fringe or extreme version of the political views espoused by far-right social movements (Carter, 2018; Perliger, 2020). It is generally associated with a broad spectrum of conservative political perspectives shared with contemporary far-right movements, such as the New Right and the Alt Right. These movements share overlapping political perspectives, including nationalist, traditionalist, anti-immigration, anti-abortion, and fundamentalist Christian views (Blee and Creasap, 2010).

An important distinction between the far-right and RWE is the willingness of extremists to support or adopt illiberal means accomplish goals that are largely incompatible with an existing societal context. The key features of this are support for nationalism, a strong or authoritarian state, racism and xenophobia, opposition to liberal democracy, appeals to populism, and anti-establishment rhetoric (Carter, 2018).

The rise of far-right social movements in contemporary societies is often depicted as reactionary. In one sense, it represents a backlash against globalization and migration. In another sense, it represents a response to widespread physical and digital content that promotes racist, authoritarian, misogynistic, and homophobic values. In both cases, the growth of far-right and RWE content has been enabled by Information and Communications Technology (ICT) in the 21st century (Gaudette et al., 2020). RWE community members, supporters, and sympathizers all have unfettered access to the kinds of ICT that deliver triggering news to them around the clock. Meanwhile, promoters of RWE materials have a previously unmatched ability to deliver content to their audiences in online communities. Members of RWE movements are often highly active on social media platforms (Froio and Ganesh, 2019; Wahlstrom and Tornberg, 2019) and adept at creating discourses and discourse communities (Hajer and Laws, 2006; Wodak, 2015).

Previous research has examined the evolution of RWE, including contemporary manifestations (Babilunga, 2013), and has charted the “ecosystem” of online right-wing communities (Baele et al., 2020b). Two studies have also explored the link between online and offline behaviors in online RWE communities, finding that there is a direct connection between real-world events and online engagement with extreme content (Baele et al., 2020a; Bliuc et al., 2019). An important focus for research about online RWE communities has been the forum site Stormfront.org, which is the largest, most viewed, and longest running right-wing site in the world. Stormfront is an online community that presents itself as a home for non-violent white nationalist social and political views (Brown, 2009; Caren et al., 2012; Thompson, 2001). Previous research on Stormfront has examined qualitative individual experiences of white nationalist rhetoric (De Koster and Houtman, 2008; Hartzell, 2020), patterns in quantitative data about positing behaviors (Scrivens, 2020), and politicized discourses used in pseudo-rational online discussions (Meddaugh and Kay, 2009).

This study focused on the content of online discussions posted on Stormfront. It takes a corpus linguistics approach to surveying the Opposing Views forum on Stormfront in order to identify key themes in the discussions that in-group members and out-group members regularly engage in. This study presents a snapshot of key words and phrases used in roughly one million posts over a period of almost 20 years. The next section explains the methodology, data, and methods used in the study. The following section presents the results of the study, focusing on a qualitative analysis of the top 100 most frequent phrases in the Opposing Views corpus. The final section concludes that a small number of themes, all of which are consistent with far-right political views, dominate debates between in-group and out-group members on Stormfront.

Research Design

This study aimed to better understand the qualitative content of themes being discussed during interactions between members of an online community in-group, comprised largely of individuals with right-wing or white nationalist views, and an out-group composed of individuals who do not identify as in-group members. It is important to note that the in-group and out-group distinctions used in the study relate to self-identified membership of the Stormfront community and its values, rather than to any offline community or identity that may be the subject of debate. This is due to the relatively anonymous nature of online posting and the impracticality of verifying the identity of individuals or the veracity of their posts.

The study used a corpus linguistics methodology and mixed quantitative and qualitative methods to identify trends in the data and then illustrate key qualitative themes. This is similar to methods used to study other violent extremist discourses (Brookes and McEnery, 2020). The methodology is corpus-driven as it applies computer-aided text analysis methods to empirical data to identify frequencies, clusters, collocates, and concordances for further examination (Crawford and Csomay, 2016). The methodology is a good fit for the corpus size and scale as it would be impractical to attempt to understand the volume of data using a corpus-based approach in the absence of previous studies having identified an existing thematic or linguistic structure in the data. It is also a good fit for the research design, which uses a content analysis approach to identify themes in textual data (Pollach, 2012).

Stormfront was identified as the most suitable source of corpus data for this study because it is the most well-known online community that espouses far-right, white nationalist, and RWE values (Bowman-Grieve, 2009). Stormfront is also unique in the RWE online ecosystem. It is the oldest online RWE community, having been in operation continuously for over 25 years and is widely recognized as the largest in the world, with the highest number of registered members, visitors, and page views (Baele et al., 2020b; Bliuc et al., 2019). Within Stormfront, there are different sections for different kinds of forums. Many are visible to anyone but require membership to post; some are open to anyone to post in. These open forums include an “Opposing Views” discussion which invites non-members to engage in civil debate with Stormfront members.

Ethical concerns related to data collection are particularly salient for politically divisive subjects, such as fringe or extremist political views. In order to minimize any potential harm, data were only collected from public forums where authors had voluntarily contributed in full knowledge that their contributions would be accessible to anyone via the internet. In addition, User IDs, which are not publicly linked to real identities, were anonymized in all data retained for the study. This is consistent with the general principle that public acts that are deliberate and intended to be observed by others do not require consent so long as individuals are not identified (Rodham and Gavin, 2006).

The Opposing Views forum was selected as a large sample of text-based discourse with the stated purpose of interaction between in-group and out-group members. Based on the structure of the website, eight variables were identified for collection via an automated web-scraping program. The records included the following data: user ID (anonymized), user join date, number of user posts, thread ID, thread name, post date, post ID, and post content. Data were collected for the period 01 September 2001 to 20 June 2020. Although Stormfront existed prior to 2001, its current servers do not store any earlier data.

A total of 1,004,628 records were collected from the Opposing Views forum. After a manual review, 5870 records were removed from the final data set due to incomplete or corrupted data. The remaining 998,758 records, containing a corpus of 196.6 million words, were retained for further analysis. For the purpose of this study, the term data set refers to all variables collected and the term corpus refers to the total words included in the post content for each record, but not the data included in the other fields.

The dataset has high validity as the corpus is large (n = 196.6 m), was collected in a consistent format over a span of almost 20 years, and was recorded objectively as text. However, it is important to note that user identities were not verified, and it is possible that some accounts claiming to be non-members may have been members using alternate accounts. This is an inherent limitation of similar studies, but it is mitigated somewhat by the relatively anonymous nature of online forums and the kind of communication they facilitate.

A total of 911 significant N-grams were identified. These phrases had a minimum frequency of 1232 and minimum term frequency - inverse document frequency (TF-IDF) of 12,442.5. The top 100 phrases were extracted for further analysis. After an interpretive review of the 100 most prevalent phrases, five thematic groupings and one miscellaneous category were created. The top 100 phrases were then manually coded in one of the six thematic groups shown in Table 1.

Table 1.

Thematic groups for extracted phrases.

Theme	Explanation
Whiteness	Content related to white Anglo-Saxon
Whiteness	racial identity and perceptions
African Americans	Content related to African American
African Americans	racial issues
Political issues	Content related to political issues
	perceived as important to politically
	conservative communities
Race issues	Content related to racial friction in
Race issues	society and between communities
United States	Content related to the US and its
United States	politics and history
Miscellaneous	Content related to other issues

These six themes are intended to be descriptive, rather than prescriptive, and serve to highlight the key issues being discussed in the Opposing Views forum.

Results

The results are presented in two stages. The first set of results present word frequency statistics in a visual form. This serves to demonstrate the consistency between the tokenized word frequencies and the N-grams identified through phrase mining. The second set of results presents the key phrases extracted from the data set and examines each category identified through thematic analysis of the 100 most common phrases in the Opposing Views forum corpus.

Word frequencies

The word frequency results are visualized as a word cloud in Figure 1. These individual words depict a counterintuitive picture of the key issues discussed on the forum. Despite perceptions that white nationalist content on Stormfront focuses primarily on racially motivated hate-speech and bigotry (Meddaugh and Kay, 2009), the most common words relate to whiteness and white identity.

Figure 1.

Wordcloud of token frequency.

The most immediate observation drawn from the word frequency analysis was that in-group and out-group debates in the Opposing Views forum focused heavily on white identity. The words white and people were by far the most common individual tokens in the corpus. In this visualization, all individual words are presented separately to avoid any potential misunderstanding of the specific uses of certain terms, such as nationalist or nationalism which may have distinct meaning in the context of the communication in which they were used. When combined with lemmatized word forms, some terms, for example, white, whites, and whiteness, create even more significant word frequencies.

Another significant observation was that the word people was highly significant in its own right. This suggests that the corpus skews toward words about people, with the most likely explanation being that people were discussed in reference to racial and societal issues. This is consistent with other high-frequency words in the corpus, many of which refer to racial, religious, or social groups. The assumption that the high frequency of the word people was related to politicized groups was tested in the phrase mining phase of research.

Extracted phrases

A supervised phrase mining algorithm identified 911 significant N-grams in the Opposing Views corpus. The highest 100 phrases were selected for thematic analysis. The top 100 phrases were distributed into themes as per Table 2.

Table 2.

Thematic grouping of top 100 extracted phrases.

Theme	N	Freq	N Cases	% cases
Whiteness	31	691,880	488,956	48.96
African American	10	141,673	100,053	10.02
Political issues	14	114,673	86,137	8.63
Race issues	6	106,076	61,633	6.16
United States	4	68,068	46,991	4.70
Miscellaneous	35	317,878	249,502	24.98

The most significant observation from the thematic grouping of key phrases was that the theme Whiteness was significantly over-represented in both frequency and proportion of cases. While 31 phrases were associated with the Whiteness theme, the frequency of those phrases was 48% of the total frequency of the top 100 phrases, and they occurred in 49% of total records in the data set.

While the remaining themes are distributed more proportionately than Whiteness, they have some unique features. The African American is the only theme that is directly proportionate across number of phrases and all frequency measures. The Political Issues and Race Issues themes have very close phrase frequencies, despite having fourteen and six phrases, respectively, and having a 25% disparity in the proportion of cases they appear in. The United States theme has the smallest frequency overall, but it appears in almost 5% of all posts despite containing only four phrases. Finally, the miscellaneous theme accounts for the largest number of phrases, but these phrases are under-represented in both the phrase and case frequencies.

Whiteness

The most frequently discussed theme in the Opposing Views forum over 20 years was Whiteness. This spanned a range of key phrases, the most common of which was the phrase white people, which was the highest frequency and TF-IDF phrase from the corpus.

Table 3 shows the frequency, case distribution, and TF-IDF of each of the extracted phrases for the Whiteness theme. It depicts a corpus that is heavily skewed toward white racial identity. Twenty eight of the thirty one phrases included in the Whiteness theme begin with the word white in a racial context and link that concept to another term.

Table 3.

Whiteness theme phrases.

Phrase	Freq	N cases	% cases	TF-IDF
White people	132,232	82,972	8.31	142,880.5
White race	75,927	51,555	5.16	97,732.4
White nationalists	44,389	31,639	3.17	66,549.7
White nationalism	42,374	29,254	2.93	64,971.0
Anti-white	39,463	26,752	2.68	62,039.9
White man	38,238	28,300	2.83	59,180.0
White nationalist	35,826	27,929	2.80	55,652.3
White women	28,167	17,468	1.75	49,495.5
White men	25,717	17,003	1.70	45,491.7
White person	24,073	19,736	1.98	41,025.2
White countries	21,653	14,621	1.46	39,722.1
White nations	18,960	13,547	1.36	35,410.0
White nation	16,268	12,410	1.24	31,001.7
Pro-white	13,208	9740	0.98	26,560.0
White culture	11,795	8280	0.83	24,550.5
White pride	11,669	8721	0.87	24,025.2
White country	11,412	8703	0.87	23,506.3
White woman	11,403	9045	0.91	23,296.9
White Americans	10,755	8116	0.81	22,479.2
White children	9164	7480	0.75	19,478.6
Anti-whites	7641	5782	0.58	17,095.8
White society	7423	6183	0.62	16,391.9
White population	7366	6135	0.61	16,291.0
White guy	6824	5687	0.57	15,317.0
White supremacist	6112	5023	0.50	14,048.4
White genocide	5979	4566	0.46	13,990.4
White male	5743	4707	0.47	13,362.3
White males	5725	4199	0.42	13,604.4
White supremacy	5670	4426	0.44	13,344.1
White supremacists	5382	4390	0.44	12,685.4
White European	5322	4587	0.46	12,442.5

The most significant group of terms referred to people either individually or collectively. Twelve of the thirty one phrases referred to people, men, women, or children in some way. Another eight referred to countries, nations, or culture in some way. The remaining 11 phrases primarily referred to nationalism, supremacy, and pride, or opposition to those things.

African Americans

The second most frequently discussed theme in the Opposing Views forum was African Americans. This is the most expected theme given Stormfront’s reputation (Brown, 2009). Table 4 shows the 10 phrases included in the African American theme.

Table 4.

African American theme phrases.

Phrase	Freq	N cases	% cases	TF-IDF
Black people	46,186	30,006	3.00	70,306.7
Black man	22,608	16,484	1.65	40,296.4
Black men	13,879	8494	0.85	28,734.4
Black person	11,614	9703	0.97	23,373.8
African American	10,551	7773	0.78	22,250.7
African Americans	9604	7019	0.70	20,679.2
Black women	9118	5872	0.59	20,339.3
Black race	6924	5755	0.58	15,505.7
Black guy	5809	4760	0.48	13,487.6
Black community	5380	4187	0.42	12,791.3

Phrases in the African American theme focused almost exclusively on people and largely on individuals. Six phrases referred directly to individuals using terms like person, man, and woman, and these accounted for 77% of the term frequency within the theme. Another three referred to people collectively, either as African American(s) or as the black community. These accounted for 18% of the term frequency. The final phrase was black race, which could also be used to refer to people collectively or could be used in another context. This phrase was underrepresented in the theme and accounted for 5% of frequencies.

Political issues

The third most frequently discussed theme in the Opposing Views forum was Political Issues. This theme includes a broad range of issues, as shown in Table 5.

Table 5.

Political issue theme phrases.

Phrase	Freq	N cases	% cases	TF-IDF
Affirmative action	19,075	13,551	1.36	35,622.4
Civil rights	12,559	8701	0.87	25,870.2
Opposing views	9457	7931	0.79	19,861.0
Political correctness	9377	7072	0.71	20,159.8
Middle class	7944	5950	0.60	17,674.9
Point of view	7914	6999	0.70	17,050.1
Politically correct	6669	5914	0.59	14,855.7
Free speech	6530	4888	0.49	15,086.4
Hate crime	6460	4188	0.42	15,358.4
Social construct	6438	4443	0.44	15,140.8
National socialism	5947	4078	0.41	14,207.5
Working class	5750	3960	0.40	13,810.1
Freedom of speech	5445	4291	0.43	12,887.8
Illegal immigration	5331	4171	0.42	12,683.6

The kinds of issues discussed in the Opposing Views forum are consistent with far-right ideology and the views generally espoused in the Stormfront community. Of the 14 phrases included in the Political Issues theme, 9 relate to political perspectives, such as civil rights, political correctness, hate crime, and freedom of speech. These make up 69% of the frequency of phrases within the theme, which is consistent with their proportion. Three of the five remaining phrases relate to socialism and economic class, which together represent 10% of phrase frequency within the theme. The last two are illegal immigration, which accounts for 5% of frequencies and affirmative action, which is the single most significant phrase in the theme, accounting for 17% of frequencies for all 14 phrases.

Race issues

The fourth most frequently discussed theme in the Opposing Views forum was Race Issues. Despite having fewer than half as many phrases, the Race Issues theme has virtually the same frequency of phrases in the corpus as the Political Issues theme. As shown in Table 6, there is less diversity between the phrases included in the Race issues theme, but each phrase has a higher frequency than the majority of phrases included in the Political Issues theme.

Table 6.

Race issue theme phrases.

Phrase	Freq	N cases	% cases	TF-IDF
Anti-racist	23,064	14,799	1.48	42,189.3
Race mixing	22,915	16,804	1.68	40,652.3
Anti-racists	22,071	11,010	1.10	43,207.8
Mixed race	13,624	9919	0.99	27,288.8
Anti-racism	13,559	6966	0.70	29,239.7
Race problem	10,843	2135	0.21	28,951.5

One of the most noticeable features of the Racial Issues theme is that it is split evenly between phrases that discuss race and phrases that begin with the word anti. Frequency is also closely divided between the two groupings. The three phrases that refer to anti-racist(s) and anti-racism account for 55% of frequency within the theme. Meanwhile, the three phrases that refer to race mixing, mixed race, and race problem account for the remaining 45% of frequency. This suggests that race issues may be one of the issues that is most evenly debated between in-group and out-group participants with differing perspectives.

United States

The fifth most frequently discussed theme in the Opposing Views forum was the United States. This is the smallest of the cohesive thematic groupings and contained four phrases, as shown in Table 7.

Table 7.

United States theme phrases.

Phrase	Freq	N cases	% cases	TF-IDF
United States	42,021	27,603	2.76	65,489.9
Civil war	9749	7281	0.73	20,836.2
North America	9575	7610	0.76	20,280.6
Founding fathers	6713	4497	0.45	15,752.3

The four phrases included in the United States theme are split evenly between two which refer to the country geographically and two which refer to its political history. However, geographical references account for 76% of the frequency of the theme, and historical references account for only 24%. This is consistent with a tendency in conservative political perspectives to prefer nostalgic world views (Stefaniak et al., 2021). The inclusion of the phrase civil war is also politically charged within the context of far-right politics in the US.

Miscellaneous

The remainder of the top 100 most frequently discussed phrases in the Opposing Views forum was combined into the Miscellaneous theme. This theme is not as cohesive as others as it serves as a container for the 35 phrases that did not cohere into clear thematic groupings. The individual phrases are shown in Table 8.

Table 8.

Miscellaneous theme phrases.

Phrase	Freq	N cases	% cases	TF-IDF
Years ago	31,433	25,560	2.56	50,038.2
South Africa	19,591	13,277	1.33	36,759.8
Middle East	18,109	14,018	1.40	33,551.9
Skin color	14,328	10,671	1.07	28,244.1
Human beings	10,655	8579	0.86	22,013.5
Native Americans	10,389	7645	0.77	21,984
World war	10,319	7650	0.77	21,832.9
Vast majority	10,155	9037	0.90	20,751.1
Long time	9777	9277	0.93	19,867.4
High school	9621	7686	0.77	20,336.5
Human race	9107	7592	0.76	19,298.7
Jewish people	8849	6826	0.68	19,160.7
Thousands of years	8670	7713	0.77	18,313.1
Group of people	8511	7679	0.77	17,993.6
Anti-Semitism	8215	5724	0.57	18,416.1
Human rights	7517	5215	0.52	17,155.3
East Asians	7478	5116	0.51	17,128.6
Blue eyes	7421	5444	0.55	16,797.7
Soviet union	7394	5223	0.52	16,869.7
Common sense	7188	6230	0.62	15,849.3
Ethnic groups	7079	5794	0.58	15,832.1
Good thing	7047	6473	0.65	15,421.3
Western civilization	6897	5261	0.53	15,714.1
Native American	6880	5384	0.54	15,606.3
Ethnic group	6862	5741	0.57	15,374.1
Gas chambers	6821	3892	0.39	16,433.7
Rest of the world	6131	5485	0.55	13,857.8
Real world	6059	5196	0.52	13,837.5
Middle eastern	5983	4933	0.49	13,798.9
Human nature	5781	4261	0.43	13,700.7
Anti-Semitic	5747	4368	0.44	13,558.2
Hip hop	5608	3669	0.37	13,655
Crime rates	5531	4429	0.44	13,015.3
Crime rate	5386	4156	0.42	12,822.9
Double standard	5339	4298	0.43	12,633.1

There are three important observations that can be drawn from the Miscellaneous phrases from the Opposing Views forum. The first is that references to time are common, including the most significant phrase, years ago, as well as long time and thousands of years. Interestingly, two of these three phrases hint at nostalgia in a similar way as the United States theme did. The second observation is that there are numerous references to geography in the Miscellaneous theme. These geographic references include a country, South Africa, a region, the Middle East, the former Soviet Union, and an ambiguous reference to the rest of the world. The third observation is that a number of phrases refer to groups of people in a way that is similar to the Whiteness and African American themes. These observations are aligned with other themes discussed above and are generally consistent with far-right political perspectives.

Conclusions

This study analyzed the content of public in-group and out-group discussions on Stormfront in order to identify key themes in the discourse between self-identified members and on-members of the largest online RWE community in the world. It used a corpus linguistics approach to examine the content of a dataset of approximately 1 m posts including a corpus of over 195 m words collected from the Opposing Views forum between September 2001 and June 2020. A mixed-method content analysis identified six key themes in the most frequently discussed phrases. Further examination of these themes and phrases revealed three important observations about the data.

The first observation was that the discussion between in-group and out-group members is heavily centered on white racial identity. The concept of Whiteness accounted for nearly half of all instances of top 100 phrases in the corpus, and it appears in nearly half of all cases in the entire data set. Despite presenting itself as a white nationalist forum, the Opposing Views forum discussed whiteness and threats to white culture far more than nationalism, although wherever nationalism was mentioned it was linked to whiteness.

The second observation was that the corpus is highly focused on people, either as groups or individual representatives of groups. This was evidenced in both the word frequency and phrase mining methods used. The high frequency of the individual word people was assumed to be related to politicized groups in the first stage of analysis. This was validated through the phrase mining, which showed that numerous N-grams containing references to people were linked to racial, religious, or social groups.

The third observation was that common political and miscellaneous phrases indicated a disposition toward nostalgic framing. This manifested both geographically and temporally in the data and is consistent with conservative politics generally and far-right political views specifically.

The implications of these findings are useful for researchers interested in fringe or extremist political views, in-group/out-group discourse, RWE discourses, or communication on the Stormfront site in particular. This study has provided an initial snapshot of the content of one public corpus on Stormfront. This can inform researchers interested in further studies of similar data available on Stormfront. It could also inform other forms of research into the themes and content in online RWE communities’ discourses.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by Rabdan Academy.

Correction (June 2025):

The article has been updated with correct dataverse link in the supplementary material section. For more details, please see the correction notice: .

ORCID iD

John Hardy

Supplemental Material

Supplemental material for this article is available online.

The files can be found at .

References

Adamczyk

Gruenewald

Chermak

, et al. (2014) The relationship between hate groups and far-right ideological violence. Journal of Contemporary Criminal Justice 30(3): 310–332. DOI: 10.1177/1043986214536659

Babilunga

(2013) When the clarion call sounds: right-wing extremism and marginals in the past and present. Anthropology and Archeology of Eurasia 52(1): 57–73. DOI: 10.2753/AAE1061-1959520104

Baele

Brace

Coan

(2020a) The ‘tarrant effect’: what impact did far-right attacks have on the 8chan forum? Behavioral Sciences of Terrorism and Political Aggression 1: 1–23. DOI: 10.1080/19434472.2020.1862274

Baele

Brace

Coan

(2020b) Uncovering the far-right online ecosystem: an analytical framework and research agenda. Studies in Conflict and Terrorism 1: 1–21. DOI: 10.1080/1057610X.2020.1862895

Bjoergo

(2003) Violence against ethnic and religious minorities. In: Heitmeyer

Hagan

(eds.) International Handbook of Violence Research. Dordrecht: Springer Netherlands, pp. 785–799.

Blee

Creasap

(2010) Conservative and right-wing movements. Annual Review of Sociology 36(1): 269–286. DOI: 10.1146/annurev.soc.012809.102602

Bliuc

Betts

Vergani

, et al. (2019) Collective identity changes in far-right online communities: the role of offline intergroup conflict. New Media and Society 21(8): 1770–1786. DOI: 10.1177/1461444819831779

Bowman-Grieve

(2009) Exploring “stormfront”: a virtual community of the radical right. Studies in Conflict and Terrorism 32(11): 989–1007. DOI: 10.1080/10576100903259951

Brookes

McEnery

(2020) Correlation, collocation and cohesion: a corpus-based critical analysis of violent jihadist discourse. Discourse and Society 31(4): 351–373. DOI: 10.1177/0957926520903528

10.

Brown

(2009) White supremacist discourse on the internet and the construction of whiteness ideology. Howard Journal of Communications 20(2): 189–208. DOI: 10.1080/10646170902869544

11.

Caren

Jowers

Gaby

(2012) A social movement online community: Stormfront and the white nationalist movement. In: Earl

Rohlinger

(eds) Media, Movements, and Political Change. Bingley and UK. Emerald Group Publishing, pp. 163–193.

12.

Carter

(2018) Right-wing extremism/radicalism: reconstructing the concept. Journal of Political Ideologies 23(2): 157–182. DOI: 10.1080/13569317.2018.1451227

13.

Chermak

Freilich

Suttmoeller

(2013) The organizational dynamics of far-right hate groups in the United States: comparing violent to nonviolent organizations. Studies in Conflict and Terrorism 36(3): 193–218. DOI: 10.1080/1057610X.2013.755912

14.

Crawford

Csomay

(2016) Doing Corpus Linguistics. New York: Routledge.

15.

De Koster

Houtman

(2008) Stormfront is like a second home to me. Information, Communication and Society 11(8): 1155–1176. DOI: 10.1080/13691180802266665

16.

Fahey

Simi

(2019) Pathways to violent extremism: a qualitative comparative analysis of the us far-right. Dynamics of Asymmetric Conflict 12(1): 42–66. DOI: 10.1080/17467586.2018.1551558

17.

Falter

Schumann

(1988) Affinity towards right–wing extremism in western europe. West European Politics 11(2): 96–110. DOI: 10.1080/01402388808424684

18.

Farago

Kende

Kreko

(2019) Justification of intergroup violence – the role of right-wing authoritarianism and propensity for radical action. Dynamics of Asymmetric Conflict 12(2): 113–128. DOI: 10.1080/17467586.2019.1576916

19.

Froio

Ganesh

(2019) The transnationalisation of far right discourse on twitter. European Societies 21(4): 513–539. DOI: 10.1080/14616696.2018.1494295

20.

Gaudette

Scrivens

Venkatesh

(2020) The role of the internet in facilitating violent extremism: insights from former right-wing extremists. Terrorism and Political Violence 34: 1339–1356. DOI: 10.1080/09546553.2020.1784147

21.

Hajer

Laws

(2006) Ordering through discourse. In: Moran

Rein

Goodin

(eds) The Oxford handbook of public policy. Oxford: Oxford University Press, pp. 251–268.

22.

Hartzell

(2020) Whiteness feels good here: interrogating white nationalist rhetoric on stormfront. Communication and Critical/Cultural Studies 17(2): 129–148. DOI: 10.1080/14791420.2020.1745858

23.

Koehler

(2016) Right-wing extremism and terrorism in europe current developments and issues for the future. PRISM 6(2): 84–105.

24.

Lowe

(2020) Far-right extremism: is it legitimate freedom of expression, hate crime, or terrorism? Terrorism and Political Violence 34: 1433–1453. DOI: 10.1080/09546553.2020.1789111

25.

Meddaugh

Kay

(2009) Hate speech or “reasonable racism?” the other in Stormfront. Journal of Mass Media Ethics 24(4): 251–268. DOI: 10.1080/08900520903320936

26.

Perliger

(2020) American Zealots: Inside Right-Wing Domestic Terrorism. New York: Columbia University Press.

27.

Pollach

(2012) Taming textual data: the contribution of corpus linguistics to computer-aided text analysis. Organizational Research Methods 15(2): 263–287. DOI: 10.1177/1094428111417451

28.

Ravndal

(2018) Right-wing terrorism and militancy in the nordic countries: a comparative case study. Terrorism and Political Violence 30(5): 772–792. DOI: 10.1080/09546553.2018.1445888

29.

Rodham

Gavin

(2006) The ethics of using the internet to collect qualitative research data. Research Ethics Review 2: 92–97. DOI: 10.1177/174701610600200303

30.

Scrivens

(2020) Exploring radical right-wing posting behaviors online. Deviant Behavior 42: 1470–1484. DOI: 10.1080/01639625.2020.1756391

31.

Stefaniak

Wohl

MJA

Sedikides

, et al. (2021) Different pasts for different political folk: political orientation predicts collective nostalgia content. Frontiers in Political Science 3: 15.

32.

Stevenson

(2019) Right-wing extremism and the terrorist threat. Survival 61(1): 233–244. DOI: 10.1080/00396338.2019.1568059

33.

Thompson

(2001) Watching the Stormfront: white nationalists and the building of community in cyberspace. Social Analysis 45(1): 32–52.

34.

Wahlstrom

Tornberg

(2019) Social media mechanisms for right-wing political violence in the 21st century: discursive opportunities, group dynamics, and co-ordination. Terrorism and Political Violence 33: 766–787. DOI: 10.1080/09546553.2019.1586676

35.

Wodak

(2015) The Politics of Fear: What Right-Wing Populist Discourses Mean. Los Angeles: Sage.