Abstract
Incels (involuntary celibates) are a group of people, linked to online misogyny and violent acts of terrorism, who mobilize around their inability to form romantic and/or sexual relationships. They have been shown to display signs of a violent extremist ideology. We conceptualize the ideology promoted by incels as misogynist and by bringing together different theories of gender and the gender order to formulate how the hetero-patriarchal and cisgenderist understanding of gender becomes an extremist worldview. We call this gender-based extremism misogynist extremism because misogyny is the most obviously violent structure of hetero-patriarchal gender order. Then, drawing on radicalization research and the social network analysis paradigm, we answer the research question: what are the communication patterns (network connections and actor attributes) that predict misogynist extremism among incels? We conduct our analysis on publicly visible posts from the forum incels.is, creating an undirected, unweighted network and then answering our research question using the auto-logistic actor attribute model to understand what individual attributes and network configurations predict user extremism. This study finds that extremists online form closed all-extremist communication triads. Consequently, they are significantly less likely to start new threads in the forum, suggesting that bonding social capital plays a more important role in an individual user’s extremism than bridging social capital.
Importance of studying misogyny as an extremist ideology: an introduction
Misogyny and antifeminism are often factors in terrorism regardless of ideology (Gentry, 2022), yet only recently have they been noticed by terrorism and extremism experts (O’Hanlon et al. 2024; Perliger et al. 2022). Notably, one’s sexual life (or its absence) serves as a powerful catalyst for extremism among young men (Hoffman et al. 2020) and a perceived threat to manhood; the subsequent quest to reaffirm masculinity through violent acts is recognized as a shared factor among terrorists of any ideology (Ferber and Kimmel, 2008; Kimmel, 2013). We argue that looking at patterns surrounding violence motivated by extreme misogyny and acknowledging this ideology for what it truly is—gender-based extremism (Berger, 2018: 34)—is crucial. The role of misogyny and antifeminism in radicalizing young men leads us to label this form of gender-based extremism as misogynist extremism. In naming it so, we aim to highlight the direct anti-women agenda (Tranchese and Sugiura, 2021), misogyny and antifeminism that is central to this ideology. In contrast, there is no comparable evidence of violent political mobilization in the name of misandry, even when considering cases like femcels (Kay, 2022). However, despite ample similarities in the misogynist movements with far-right, this alone fails to capture the full potential for extremist mobilization fueled by hetero-patriarchal cisgenderistFootnote1 masculine frustration. Scholars emphasize that overlooking the radicalizing power of misogyny results in the systematic erasure of violence committed in its name (e.g., Hoffman et al. 2020; Tranchese and Sugiura, 2021; Gentry, 2022; Green et al. 2023; Miller, 2024).
Most of the mobilization around misogyny as an ideology takes place on the internet, in the manosphere (Ging, 2017), a diverse group of forums, communities and web pages loosely structured around the idea that boys and men are victimized in contemporary culture due to feminism (ibid). Our research draws on literature that recognizes misogynist hate speech online and traces it to the activities of the manosphere (e.g., Jane, 2018; Ribeiro et al. 2021). For this research however, we focus on incels, because (1) they are the only manosphere group to which terrorism has been linked (Tomkinson et al. 2020); (2) because of the clear dehumanizing language they use, they are claimed to hold the most radical ideology of the manosphere (Baele et al. 2019; Tranchese and Sugiura, 2021); and (3) their ideology has been studied from terrorism and radicalization perspectives (Miller, 2024; Moskalenko et al. 2022; Green et al. 2023).
There is a growing body of literature that recognizes the key role the hetero-patriarchal cisgenderist gender order plays in this ideology, although it has not been named this way. Calls to use theoretical frameworks of hegemonic masculinity (Connell and Messerschmidt, 2005) have been made, as this concept from masculinity studies seems to be central to understanding the ideology (Aiolfi et al. 2024) and the power that misogyny has in extremism (Ferber and Kimmel, 2008; Kelly et al. 2024). We answer this call by considering hegemonic masculinity in the context of gender order (Connell, 2002) and the coloniality of gender (Lugones, 2007) while employing the gender theories of trans scholars (Bornstein, 1994; Butler, 1988) to describe how misogyny is the first and most obvious violent structure of this gender order.
We connect this theoretical grounding with the radicalization research paradigm (Akram and Nasar, 2023; Alava et al. 2017; Della Porta, 2018; Gentry, 2022) to study cognitive radicalization within the misogynist extremist online community. By doing so, we contribute to the radicalization literature on incels and, more broadly, radicalization within online forums. We introduce a new operational definition of misogynist extremism, the need for which has been named by O’Hanlon et al. (2024), that is firmly rooted in intersectional feminism and highlights the understanding of gender among trans scholars. We chose a methodology that brings forth the connective nature of incel ideology—social network analysis—researching radicalization in a novel way through online communication between community members. Researching the radicalizing network as highlighted by the 3N radicalization approachFootnote2 (Kruglanski et al. 2019) is especially crucial when researching misogynist extremism, the networked nature of which has been emphasized for years (Ribeiro et al. 2021; Marwick and Caplan, 2018). This perspective has been declared necessary by several studies (e.g., Akram and Nasar, 2023; Alava et al. 2017), yet to this day scarcely researched (Lara-Cabrera et al. 2017; Renström et al. 2020, Bacaksizlar Turbic and Galesic, 2023).
Incels: a case study in misogynist extremism
Incels mobilize around their inability to find a sexual and/or romantic partner. They see “society as fundamentally hierarchized along sex and attractiveness lines” (Baele et al. 2019: 1). This hierarchy is believed to favor women and exclude unattractive men. Their name itself, incels—an abbreviation of “involuntary celibates”—shows this lack of sexual relationships is central to their self-identification. Due to the patriarchal notion that they are entitled to women and their bodies, incels perceive their lack of romantic success not only as personal frustration but also as moral wrongdoing inflicted on them by women and feminism. They believe that they have been wronged by society for its indifference to their situation and lack of political will to change it (Cottee, 2021).
Incels are part of the manosphere, which is tied together by the general idea that men are victimized, a concept embraced as “the red pill” (Ging, 2017). Incels elaborate on this idea with “the black pill”, recognized as central to the radicalization process in this community (Green et al. 2023).
The blackpill reframes the perceived reality of Incels to include hopelessness that reframes the struggle of Incels from a problem to be solved (rejection by women), to a social restratification (a return to the perceived strict patriarchy of the midcentury), to a holy war (a recognition that society will not change and violent reprisal is the only acceptable response). (Green et al. 2023: 17)
It is the hopelessness of the black pill why incels have been named the most radical group in the manosphere. While other groups like Pick-Up Artists and Men Going Their Own Way are advocating for certain strategies that should elevate the suffering supposedly inflicted on them by feminists, incels believe their situation to be immutable (Baele et al. 2019; Perliger et al. 2022).
This hopelessness is argued to lead to violence (e.g., Baele et al. 2019; Cottee, 2021; Green et al. 2023) on three levels: personal, interpersonal and societal (Aiolfi et al. 2024). Current radicalization measures of incel ideology focus on the approval of societal violence (Moskalenko et al. 2022; Moskalenko et al. 2022); however, within incel forums, there is not a broad consensus on societal terror (Aiolfi et al. 2024). Therefore, feminist scholars have been calling for recognition of misogynistic, hetero-patriarchal and cisgenderist violent language’s power as it leads not only to terrorism but to other forms of violence against women (Chang, 2022; Cottee, 2021; Jane, 2018; Kelly et al, 2024; Tranchese and Sugiura, 2021).
Relational nature of extremism online
According to Berger (2018: 79), extremism is “a belief that an in-group’s success or survival can never be separated from the need for hostile action against an out-group”. Using Berger’s definition of extremism and Neumann’s definition of radicalization (“a process whereby people become extremists”; Neumann, 2013: 874), we define radicalization as a process by which people adopt a belief that the success of their in-group is inseparable from hostile actions against out-group(s).
In radicalization research, there is a distinction between cognitive (developing extreme beliefs) and behavioral (engagement in violent behaviors) radicalization (Whittaker, 2023). Borum (2011: 26) writes that three essential measurements of violent radicalization are “(1) developing antipathy toward a target group; (2) creating justifications and mandates for violent action; [and] (3) eliminating social and psychological barriers that might inhibit violent action”. These three measurements can be separated into cognitive and behavioral radicalization. Borum’s first two measurements are dimensions of cognitive radicalization, whereas the third transforms this cognitive radicalization into violent action (behavioral radicalization).
A central concept for the relational approach to radicalization is social capital (Saal, 2021). Following the broadly accepted distinction between bridging and bonding social capital (Putnam, 2001), Saal notes that bonding social capital seems crucial for radical networks, as it builds cohesive group identity while promoting antagonism towards outsiders (Saal, 2021, p. 45). On the other hand, this clique-like characteristic of bonding social capital constrains the network size and limits the potential to recruit new members, allocating this task to organizational hubs that hold significant bridging social capital (Saal, 2021, p. 46).
Researchers have long recognized the importance of networks in radicalization (Kruglanski et al. 2019) and, recently, have been expanding this view to online communities as well. Research is showing that active engagement in online extremist communities is a better predictor of radicalization than pure extremist content consumption (e.g., Akram and Nasar, 2023; Alava et al. 2017; Renström et al. 2020). Even though incels and other terrorists radicalized online are often labeled “lone wolfs”, this label is misleading. Scholars agree that these actors may be physically alone, but they have a network of like-minded people with them, on their phones (Neumann, 2012; Miller, 2024). This means that for every behaviorally radicalized lone-wolf terrorist, there is a network of cognitively radicalized people online.
Concerned with cognitive radicalization (Whittaker, 2023), we aimed to answer the research question, “What are the communication patterns (network connections and actor attributes) that predict misogynist extremism among incels?” In so doing, we developed a measurement of extreme misogyny, adopting Borum’s (2011) first two measurements: developing antipathy toward a target group and creating justifications and mandates for violent action. The need for research on misogynist extremism with an explicit definition of the concept was named in a scoping review of the concept (O’Hanlon et al. 2024). Our operational definition of cognitive misogynist extremism captures misogynist dehumanization, gender essentialism and the approval of violence, which are the key pillars of misogynist extremism (Green et al. 2023; Miller, 2024; Tranchese and Sugiura, 2021). The following section thus explains how misogynist extremism emerges from a hetero-patriarchal, cisgenderist understanding of gender relations, that is, the Eurocentric gender order (Connell, 2002; Lugones, 2007).
Conceptualization of misogynist extremism
The Eurocentric gender order, shaped by a binary, cisgenderist and hetero-patriarchal understanding of gender, produces hegemonic masculinities (Connell and Messerschmidt, 2005) and other gendered identities. It establishes, structures and hierarchizes power dynamics between “men”, “women” and their diverse “others”. The Eurocentric gender order gained prominence through European imperialism and globalization and perpetuates, organizes, embodies and legitimates masculine dominance in the world while delegitimizing any other (e.g., indigenous) understanding of gender roles (Connell, 2002; Lugones, 2007).
Transgender scholars highlight how the naturalization of biological sexual dimorphism, linked to discrete gender and attraction to the opposite sex/gender, is a cultural construct serving the interests of the dominant, Eurocentric culture (Butler, 1988; Bornstein, 1994; Spade, 2003). Both gender and compulsory heterosexuality are mundane, ritualized forms of social legitimization: “performative accomplishment[s] compelled by social sanction and taboo” (Butler, 1988: 520). These performances include being perceived as cisgender and heterosexual (Spade, 2003). Gender and sexuality function as regulatory tools with punitive consequences. Misogyny, as Bornstein writes (1994: 106), is the “most obvious violent structure” of gender, maintaining the binary gender narrative by forcing people to be one or the other, the oppressor or the oppressed.
Out of the gender order that maintains this binary, cisgenderist and hetero-patriarchal understanding of gender, misogynist extremism can easily arise, as misogyny is a violent structure of gender. Misogyny is “an unreasonable fear or hatred of women, […] a prejudice that is symbolically exchanged [and] shared” (Gilmore, 2010: 9). Employing this definition of misogyny and Berger’s (20018) definition of extremism, misogynist extremism is a gender essentialist belief that anyone assigned female at birth falls into the category of women and that anyone assigned male at birth should be wary of them. It segregates society into two impervious genders based on the sex assigned at birth. Through misogynistic beliefs, it categorizes these genders as an in-group and an out-group. Then, again, through misogynistic beliefs, it creates a narrative of threat from the out-group, which needs to be resolved through hostile action. This means that misogynist extremism is inherently cisgenderist and hetero-patriarchal. Apart from misogynistic terrorism (Gentry, 2022), it often appears as online harassment: gendered hate speech, doxxing, revenge porn, social shaming, intimidation and threatening (Jane, 2018; Kay, 2022; Marwick and Caplan, 2018).
Hypotheses
Active engagement in online extremist communities has been shown to be a better predictor for individual-level radicalization than mere content consumption (e.g., Akram and Nasar, 2023; Alava et al. 2017; Marwick et al. 2022; Renström et al. 2020; Ribeiro et al. 2020). Nevertheless, it has been pointed out that not every active user is an extremist (Marwick et al. 2022; Whittaker, 2023). We grapple with this by hypothesizing that more senior (i.e., active and loyal) users are more likely to be extremists (H1).
A central concept for the relational approach to radicalization is social capital (Saal, 2021). The significance of social capital, having been shown in real-life radicalization networks, leaves the question of how this mechanism translates to online radicalization. Cottee (2021), argues that “black-pilled” members hold elite status among incels, as they are believed to see the world as it is, without illusions or wishful thinking—the black pill is the final stage of the incel radicalization pipeline (Green et al. 2023). This would suggest that extremist users hold significant bridging social capital in an online radical network. Therefore, we expect regular users to communicate with one extremist user but not to the same extent as with each other: actor k-stars (H2). On the other hand, extremists are supposed to be prone to developing cohesive cliques of high-bonding social capital (Saal, 2021). To test which of these is true in the online radicalization network of incels, we hypothesize that non-radical users tend to connect to more than one extremist user, resulting in a triangle of communication with two extremist users in it: a partner attribute triangle (H3). This is a finding that suggests a compromise between bridging and bonding social capital. Further, we expect extremist users to communicate with each other, resulting in a triangle of communication between all extremist users: a partner-partner attribute triangle (H4), representing the communication network among radicalized users (Della Porta, 2018) and containing high levels of bonding social capital (Saal, 2021). All of these attributes are defined and visualized in Appendix 5.
Cottee (2021) observed the use of highly group-specific language, nicknames and slurs as a demonstration of group membership. Baele et al. (2019) also support this, claiming that the use of a group-specific language determines the boundaries of the incel community. In another paper, Baele et al. (2024) use the frequency of highly group-specific language as a radicalization measure of the forum as a whole. Therefore, we expect that extremism is related to the use of highly group-specific language, as they are both used as a confirmation of group membership (H5).
We expect to find prejudices towards other groups among extremist users and fewer prejudices among regular users as Zick (2015) suggests that prejudices are connected. On incel forums, ample racism and homophobia have been found (Jaki et al. 2019). There has not been much research done on the role of homophobia on the incel worldview; however, our earlier study (Coufal et al. 2025) suggests that homophobia in the forum is used to police masculinity and femininity. Similarly, the notion of “homophobic homoeroticism” has been found in the forum (Held, 2023). Racist beauty ideals in contemporary culture have been pointed to as an integral part of recruitment to the incel ideology (Cunningham Rogers, 2024; Gheorghe, 2024) and internalized racism as an important step in the “black pill” radicalization pathway for people of color (Chang, 2022; Green et al. 2023). Based on these findings, we expect users who show homophobia and racism to be more likely to be misogynist extremists (H6).
These expectations lead to several hypotheses:
H1: More senior users are more likely to be misogynist extremists.
H2: Users who form k-star network formations are more likely to be misogynist extremists.
H3: Users who are a part of partner attribute triangle formation are more likely to be misogynist extremists.
H4: Users who form partner-partner attribute triangle formations are more likely to be misogynist extremists.
H5: Users who use group-specific language are more likely to be misogynist extremists.
H6: Users who show prejudice towards other groups are more likely to be misogynist extremists.
H6a: Users who show homophobia are more likely to be misogynist extremists.
H6b: Users who show racism are more likely to be misogynist extremists.
Methods
Social network analysis is uniquely suited to researching the structural communication patterns of extremist communities. Scholars emphasize the significance of analyzing social contexts and relationships which contribute to disseminating extremist ideologies rather than solely concentrating on individual-level factors (Borum, 2011; Della Porta, 2018). Zick (2015) highlights group dynamics as crucial to individual radicalization, claiming it is more important than demographic predispositions and personality profiles. In social groups, members form a collective identity, and integration into the collective can often be through radicalizing beliefs.
Data
We approached our research questions by constructing a network based on the interpersonal communication of the largest incel forum: incel.is. Incels.is is an online forum, the content of which is visible without a log-in. However, to post, one must make an account. The forum is organized into threads to which users can reply or they can start a new thread. Additionally, users can reply to any post in a thread—this does not start a new thread. In that manner, incels.is works very similarly to Reddit, which comes naturally as the incel community was one of the main reasons for Reddit’s increased platform moderation in 2018 and the subsequent quarantine and ban of major incel subreddits (Copland, 2020; Ribeiro et al. 2021). This increased moderation was one of the reasons incels.is was founded as a separate platform.
The analysis is based on all the posts visible on the forum, the largest of its kind, in November 2022 (Wedel, 2023). For each post, we know the author, the time of posting, the content of the post, and whether the post starts a new thread, answers a previous post or is simply posted in a thread without directly referring to a previous post in the thread. For the analysis, we marked every reply to a thread as a reply to the author that started the thread. If a user was replying to a particular post in the thread, this was recorded as a reply to that post’s originator and not to the user starting the thread. Additionally, since users can reply to more than one post in a thread, which was also encoded, it resulted in multiple replies. Based on this reply information, we constructed an undirected, unweighted reply network (Gaisbauer et al. 2021) of this forum, which represents the social structure of the forum in terms of public user-level communication. Through the thread organization of the forum, users starting a thread naturally have more edges than users only replying to threads.
Analysis
To test our hypotheses, we employed an auto-logistic actor attribute model (ALAAM) that predicts the extremism of a user based on their structural position within the network (Parker et al. 2021). ALAAMs are intended to investigate the extent to which a pattern of social relations among individuals (e.g., proximity to others and connection to people possessing specific attributes may explain a particular dichotomous attribute (Robins et al. 2001), misogynist extremism in this study. These models allow for understanding “how individual behavior may be constrained by position in a social network and by the behavior of other actors in the network” (Daraganova and Robins, 2012: 102). In ALAAMs, social relations with other actors, who hold specific attributes, are considered explanatory variables of a particular actor attribute. The idea of this model is that individual attributes are “potentially dependent on and may potentially influence the attributes of others” (Daraganova and Robins, 2012: 104). A more detailed explanation of this method is available in Appendix 1.
The analysis was conducted using the ALAAMEE package for Python and the open-source software GNU Parallel (Stivala et al. 2024; Tange, 2018). The ALAAMEE package enables the analysis of large-N networks because it utilizes the Equilibrium Expectation (EE) algorithm, which allows model-fitting onto these (Stivala et al. 2024; more on this package and the EE algorithm in Appendix 1). GNU Parallel was used for parallel runs of the EE algorithm. The model defined in this study ran for approximately 200 h on 16 processors, with 40,000 iterations of the EE algorithm. The results of the model were therefore computed from 640,000 estimates. Estimates, standard errors, t-ratios and the significance of estimates were computed based on the ALAMMEE package’s definitions. This package estimates significance only on a true/false level, using the t-ratio threshold and a normal distribution of standard errors.
Measurement
Based on the data described above, we established text measures and network measures as variables for the ALAAM. The content of each post was used to establish dictionary measures based on van der Vegt et al. (2021), Farrell et al. (2019), a dictionary derived from Incel WikiFootnote3 (Incel Wiki, 2024) and previous research (Coufal et al. 2025). These measures were used to operationalize misogynist extremism on a post level and, subsequently, on the user level. We will first explain the operationalization of misogynist extremism, which was responsible for the labeling of each post and, consequently, each user as the target variable in our model. We will then provide the context of the operationalization of the independent variables in our model.
Operationalizing misogynist extremism
We flagged a post as misogynist extremism if it contained misogynist extremist language, that is, language that seeks to dehumanize women, marking them as an out-group (dimension I) and using the language of violence (dimension II). To measure our operational definition of misogynist extremism on both post and user levels, we developed a dictionary for each of the two dimensions that our definition is based on: (a) the operational definition of radicalization by Borum (2011) and (b) the cisgenderist, hetero-patriarchal worldview that is central to incel ideology (e.g., Kelly et al, 2024; Tranchese and Sugiura, 2021).
For the first dimension (radicalization, i.e., in-group/out-group demarcation), we considered the language used by incels for women: incels talk about women and females interchangeably, assuming that anyone assigned female at birth falls into a category of women and holds negative traits which they ascribe to this gender. In the incel discussions, the word female served as an etymological base for the derogatory terms “femoid” and “foid”, which are a combination of female and android, suggesting that females do not have human qualities. Baele et al. (2024: 397) named these “dehumanizing outgroup labels”, and Chang (2022) analyzed their usage to mark women as a “monstrous-feminine”, a popular misogynist imagination ascribing negative traits to women. In incel discourse, these terms are often used interchangeably with two other dehumanizing terms: “holes” and “roasties”, referring to female genitalia. These four slurs for women are offensive, derogatory, misogynistic and gender essentialist. They dehumanize people assigned female at birth and, in a gender-essentialist way, equate them with women’s gender and reduce women to female genitalia. They demonstrate the hetero-patriarchal, cisgenderist understanding of gender present in incel ideology and are tightly linked with in-group/out-group rhetoric (Coufal et al. 2025).
The second dimension of our misogynist extremism operationalization, the language of violence, is measured using two established dictionaries: the Grievance Dictionary (van der Vegt et al. 2021) and the Misogynist Lexicon (Farrell et al. 2019). We select these two dictionaries because they have been successfully used to predict violence from extremist texts (van der Vegt et al. 2021) and to measure misogyny on the incel forum (Ribeiro et al. 2021). Out of these two established dictionaries, we use only selected categories because they are related to the measurements of radicalization by Borum (2011):
Grievance Dictionary: deadline, hate, murder, soldier, suicide, violence, weaponry
Misogynist Lexicon: physical violence, sexual violence
We combined these dictionaries because of the unique nature of misogynist extremism, which is strongly linked to gendered and sexual violence (Baele et al. 2019) but shares similarities with other extremist worldviews (Perliger et al. 2022). Therefore, combining the two dictionaries allows us to capture both the specifically gendered and sexualized violence (Misogynist Lexicon) of misogynist extremism as well as the overall notion of extremist violence (Grievance Dictionary).
We defined a post as misogynist extremist if it contains at least one dehumanizing slur (“femoid”, “foid”, “holes” and “roasties”) and one term of violence (from either dictionary). Combining the language of violence with dehumanizing, gender-essentialist terms mostly connected with in-group/out-group rhetoric (Baele et al. 2024)Footnote4 allowed us to capture the hereto-patriarchal, cisgenderist, as well as violent nature of misogynist extremism. The full dictionaries for both dimensions of our operational definition of misogynist extremism are listed in Appendix 2. Our operationalization of misogynist extremism led to 5% of forum posts being flagged as misogynist extremists. We did not differentiate between different frequencies of term usage since the usage of any number of terms must consequently result in the qualitative assessment that a post expresses misogynist extremist opinion.
This gave us a post-level classification of misogynist extremism, but we still needed to transpose that to the user level to answer our research question. We defined a user’s extremism and subsequent radicalization level via consistent misogynist extremist content production as a proxy (Neumann, 2012). To identify users who consistently create misogynist extremist content over time, we developed a binary variable (needed to comply with the specifications of ALAAM as a logistic regression model) indicating whether a user is a consistent content creator. This variable was set to 1 if the user met both of the following criteria:
Minimum volume threshold: at least 35% of a user’s content in a given month is misogynist extremist.
Activity frequency threshold: the user must meet the minimum volume threshold in at least half of the months this user is active.
This operational definition is likely a conservative estimate of misogynist extremists. Both thresholds were selected using the elbow method, graphs for which can be found in Appendix 3.
Operationalizing independent variables
As independent variables, we operationalized individual user characteristics and the characteristics of the network structures in which users are embedded. Following our hypotheses, the independent variables were forum seniority, three language-specific attributes (i.e., use of highly group-specific language, homophobia, racism) and three network attributes (i.e., partner triangle, partner-partner triangle and k-stars formation—all defined in Appendix 5).
Forum seniority was measured through incel.is’s native user seniority rank, spanning from 0 to 12. The same rules requiring activity on the forum and time since registration are applied when ranking users to each higher level, making this measure a combined ordinal scale of activity and loyalty. The only exception is the twelfth rank, the administrator. With administrators being commonly very dedicated members of their respective communities across internet forums, the administrator rank can thus be considered naturally the final rank for a forum’s ranking system. We assume the same is true of incels.is and have kept the three administrators in the dataset.
Prejudice toward other groups is measured based on dictionary measures of homophobia and racism taken from Farrell’s et al. (2019) Misogynist Lexicon. Incel slang is measured as a dictionary measure of incel-specific language sourced from a signpost on Incel Wiki. This signpost is displayed at the bottom of each Incel Wiki page (Incel Wiki, 2024). Though its intent is not explained anywhere on the wiki page, the authors’ knowledge of incel jargon allows us to estimate that it includes the most searched terms of incel slang. From this signpost, single-word terms and two-word terms have been selected to form a dictionary of incel slang. We measure all three variables—homophobia, racism and incel slang—as the proportion of posts of a user with at least one word from the dictionary out of all posts of said user. We approach these dictionary measures consistently with how they have been used in Farrell et al. (2019), claiming that pure presence of slurs from the dictionaries would allow for a post to be qualitatively assessed as containing homophobia, racism or incel slang respectively. Complete dictionaries used for measures of homophobia, racism and incel slang are reported in Appendix 4.
Network attributes in this model are partner attribute triangle, partner-partner attribute triangle and geometrically weighted k-star as described and implemented by Stivala et al. (2024). These three network measures are described in Appendix 5.
Results and discussion
The undirected, unweighted reply network of the incel.is forum has 11,796 nodes and 1,119,848 edges. Self-edges have been removedFootnote5. The graph has a density of 0.016 and a transitivity of 0.317. The mean degree of this graph is 189.87, and the maximal degree is 3580. This means that on average, users have 189 connections to other users which are a combination of replies to posts of the user of concern or replies of the user of concern to posts of other users. When comparing users marked as misogynist extremists with regular users of the forum, we have 436 extremist users compared to 11,360 regular users. The degree distribution of misogynist extremists vs. regular users (see Table 1), shows us that regular users have, on average, around 50 more connections with other users than misogynist extremists have and that the most connected misogynist extremists have two-thirds the connections of the most connected regular user.
Table 1 Degree distribution for users marked as misogynist extremists and regular users.
Full size table
From the visualization in Fig. 1, we can see that extremist users, as well as all of our non-network measures, are relatively evenly distributed across the network. In other words, there are no distinct clusters of communication that disproportionately represent any particular attribute. Incel slang and racist slurs appear with similar frequency throughout the network, while homophobia is less prevalent. Although previous research has found homophobia to be widespread in the forum (Jaki et al. 2019; Coufal et al. 2025), the proportion of homophobic posts among each user's total posts is relatively small. Finally, the visualization shows that misogynist extremist users tend to be relatively new or junior members, which aligns with findings from earlier research (Wedel and Coufal, 2025).
Fig. 1: Visualization of non-network user level attributes.
figure 1
Blue dots represent non-extremist users, and red dots represent extremist users. The size of the dot represents the attribute value (10 minimum, 110 maximum, 60 baseline size). The first panel displays a baseline visualization of the user base with all dots having equal size. The second panel displays a visualization of the user base, with the size of the dot representing the intensity of homophobia in the posts of that user. The third panel displays a visualization of the user base, with the size of the dot representing the intensity of incel slang in the post of that user. The fourth panel displays a visualization of the user base, with the size of the dot representing the intensity of racism in the post of that user. The fifth panel displays a visualization of the user base, with the size of the dot representing the seniority (user rank) of that user.
Full size image
The results of our model of misogynist extremism as an attribute of users on the incel.is forum are displayed in Table 2. We can see that all of the hypotheses are significant. Hypothesis H1, stating that more senior users on the forum are more likely to be misogynist extremists, is significant, but the estimate is negative. This finding contrasts with the consensus in the literature that engagement with extremist communities is a decent predictor of user radicalization (e.g., Akram and Nasar, 2023; Alava et al. 2017; Marwick et al. 2022; Renström et al. 2020; Ribeiro et al. 2020). Nevertheless, it has been pointed out that not every active user is an extremist (Marwick et al. 2022; Whittaker, 2023). Our findings further comply with novel evidence that for incels.is more senior members seem to be less active in extremist threads than more novice users (Wedel and Coufal, 2025).
Table 2 Results of auto-logistic actor attribute model of misogynist extremism on the incel.is forum.
Full size table
Hypotheses H2–H4 are concerned with bridging and bonding social capital (Putnam, 2001). H2, stating that users who form k-star network formations are more likely to be misogynist extremists, suggests that extremist users hold significant bridging social capital. With a significant but negative estimate, we found the opposite. This result brings essential understanding to the forum dynamics and directly relates to the forum’s organizational structure. The forum is organized into threads, to which users can reply but also respond to each other within that thread. Our result suggests that misogynist extremists are less likely to be the users who create these threads. Instead, they perhaps just contribute to threads created by other users.
Hypothesis H3 assumes that regular forum members connect to more than one extremist user, suggesting a compromise network formation between bridging and bonding social capital. The estimate for H3 was significant and negative. On the other hand, hypothesis H4 assumes that extremist users tend to connect among themselves, suggesting bonding social capital, and we found moderate support for this expectation. This indicates that extremist users form closed all-extremist communications triads and that regular users are less likely to be part of a closed triad of communication with two extremist users. This result is in line with the popular claim about face-to-face radicalization, which is said to happen through tightly knit friendship groups (Della Porta, 2018; Saal, 2021). Radicalization scholars agree that so-called lone-wolfs, that is, terrorists radicalized online, may be physically alone but have a network of like-minded people on their phones (Neumann, 2012; Miller, 2024). We found moderate evidence that misogynist extremists form closed triads of all-extremist communication. In the social network analysis paradigm, this is interpreted as trust among a group (Scott, 2006). Therefore, we can conclude that bonding social capital is a factor in online radicalization through forum communication, with bridging social capital likely being “outsourced” to the forum structure and perhaps the networking tendency of extremist forums on the internet in general (Brace et al. 2024; Ribeiro et al. 2021).
The most unexpected finding of our study was in H5 and 6—users who use group-specific (H5), homophobic (H6a) and racist (H6b) language are more likely to be misogynist extremists. We found these measures to be statistically significant, but negative. This suggests that the in-group slang of incels, homophobic and racist slurs are used by misogynist extremists significantly less than by regular users. Cottee (2021) stated that group-specific language is likely to demonstrate in-grouper status, and Baele et al. (2019) suggested that in-group slang keeps the boundaries between in-group and out-group in place. Racism is amply present in the forum, as found by Jaki et al. (2019) as well as in our data (85% of users have used a racist slur at least once). Regarding homophobia, there is very little research done exploring its role in incel ideology. Our previous research (Coufal et al. 2025) showed that homophobia is amply present on the forum and used to police correct expressions of femininity and masculinity.
It is therefore worth further investigation as to what the exact role racism and homophobia plays in misogynist extremist ideology. For racism, it could be that neocolonial beauty ideals affect the early stages of radicalization process (Chang, 2022; Green et al. 2023), as it plays an important role in the black pill theory of how the world functions (Gheorghe, 2024). As homophobia is used to police correct expressions of gender and sexuality (Coufal et al. 2025, Held, 2023), it could also play a role in early stages of radicalization, and as a user becomes more radicalized, they stop caring about other out-groups and focus on the main enemy—women. As all of the language measures have effect in the opposite direction than what was expected based on the literature, this warrants further study. It could be that the users who use highly misogynistic violent language (as per our operational definition of misogynist extremism) engage mostly in “black pill” theorizing regarding their everyday life and don’t care too much about other outgroups. More in-depth, perhaps qualitative analysis of these users should be conducted.
The limitation of this research is that the radicalization model is cross-sectional and defined on an undirected network. Nevertheless, the user-level extremism model, measuring cognitive radicalization (Borum, 2011; Whittaker, 2023) through the peer-effect radicalization paradigm, has brought some crucial insights in understanding the communication patterns of misogynist extremism. Future work should focus on theory-derived expectations on directed networks, such as whether extremist users tend to be senders or receivers of connections, contagion reciprocity and so on (Stivala et al. 2024). Another significant future development would be integrating temporal elements into the model to measure how extremism spreads throughout the network over time. Another limitation of this model is that it does not include centrality measures. This should be further investigated and would warrant a paper of its own.
Conclusion
This article answered the research question: what are the communication patterns (network connections and actor attributes) that predict misogynist extremism among incels? Through our theoretical conceptualization of misogynist extremism, we extend the concept of gender-based extremism (Berger, 2018: 34). We theorize misogynist extremism through different theories of gender and the gender order (e.g., Butler, 1988; Bornstein, 1994; Connell, 2002; Lugones, 2007; Spade, 2003;) specifically to highlight how hetero-patriarchal and cisgenderist understandings of gender allow misogyny to become an extremist worldview.
The studied case of misogynist extremism in this paper is incels (involuntary celibates) who are linked to online misogyny and violent acts of terrorism (Gentry, 2022; Marwick and Caplan, 2018; Tomkinson et al. 2020). We analyzed the public communication on the incels.is forum, creating an undirected, unweighted reply network (Gaisbauer et al. 2021), using the auto-logistic actor attribute model (Daraganova and Robins, 2012; Stivala et al. 2024). We position our research in the cognitive peer-effect radicalization paradigm and incorporate the feminist critique of the current state of radicalization research.
This study found that active and loyal users on the forum are less likely to be misogynist extremists. Further we found that extremist users do not hold significant bridging social capital (Putnam, 2001). This implies that these extremists are less likely to create threads but instead contribute to existing ones. On the other hand, we found that bonding social capital is crucial for online radicalization. This suggests that both online and “offline” radicalization work in a similar manner through friendship networks (Saal, 2021; Della Porta, 2018). Surprisingly, we found that misogynist extremists are less likely to use group-specific language and the notorious presence of racism and homophobia on this forum does not seem to be linked to the consistent creation of misogynist extremist content.
Data availability
The data is available at Wedel (2023) GESIS data archive: 10.7802/2485.
Notes
Cisgenderism and heterosexism “grant privileges to heterosexuals whose behaviors and appearance match the gender roles assigned to them, and can oppress lesbian, gay and bisexual individuals, as well as anyone displaying non-cisgender behavior, identity or experiences such as transgender individuals” (Hutton et al. 2020).
The 3N radicalization approach posits that radicalization towards extremism occurs when an individual experiences the need for significance, encounters and embraces an ideological narrative that legitimizes violence as the path to achieving that significance and has that narrative then validated and reinforced by a supportive network (Kruglanski et al. 2019).
Incel Wiki describes itself as “an encyclopedia about the manosphere at large, particularly incels, in the English language. It is meant to be a repository of academia, folk theories, memes, people, and art associated with involuntary celibates” (Incel Wiki, 2024).
Our approach is different from the dictionary of Baele et al. (2024), which measures the extremity of incel language, because this dictionary does not distinguish between general incel slang and misogynist extremist language. They consider any incel slang to be a sign of extremity and, in this way, measured the radicalization of the forum as a whole. Because of hypothesis H5, testing if group-specific slang predicts user-level extremism, we cannot use this dictionary.
Because it is a requirement of ALAMEE package utilized to compute our model.
References
Aiolfi I, Palena N, Ó Ciardha C, Caso L (2024) The incel phenomenon: a systematic scoping review. Curr Psychol 43(32):26264–26278. https://doi.org/10.1007/
Akram M, Nasar A (2023) Systematic review of radicalization through social media. Ege Acad Rev 23(2):279–296
Alava, S, Frau-Meigs, D, & Hassan, G (2017) Youth and violent extremism on social media: mapping the research. UNESCO Publishing
Bacaksizlar Turbic, NG, & Galesic, M (2023) Group threat, political extremity, and collective dynamics in online discussions. Scientific Reports, 13(1), 2206. https://doi.org/10.1038/
Baele S, Brace L, Ging D (2024) A diachronic cross-platforms analysis of violent extremist language in the incel online ecosystem. Terror Polit Violence 36(3):382–405. https://doi.org/10.1080/
Baele SJ, Brace L & Coan TG (2019) From “Incel” to “Saint”: Analyzing the violent worldview behind the 2018 Toronto attack. Terror Polit Violence, 1–25. https://doi.org/10.1080/
Berger JM (2018) Extremism. The MIT Press. https://doi.org/10.7551/
Bornstein K (1994) The First Question. In: Gender outlaw: on men, women, and the rest of us. Routledge, pp 103–113
Borum R (2011) Radicalization into violent extremism I: a review of social science theories. J Strateg Secur 4(4):7–36
Brace L, Baele SJ, Ging D (2024) Where do ‘mixed, unclear, and unstable ideologies come from? A data-driven answer centred on the incelosphere. J Polic Intell Count Terror 19(2):103–124. https://doi.org/10.1080/
Butler J (1988) Performative acts and gender constitution: an essay in phenomenology and feminist theory. Theatre J 40(4):519–531. JSTOR https://doi.org/10.2307/
Chang W (2022) The monstrous-feminine in the incel imagination: Investigating the representation of women as “femoids” on /r/Braincels. Fem Media Stud 22(2):254–270. https://doi.org/10.1080/
Connell RW, Messerschmidt JW (2005) Hegemonic masculinity: rethinking the concept. Gend Soc 19(6):829–859. https://doi.org/10.1177/
Connell RW (2002) Globalization, imperialism, and masculinities. In: Handbook of studies on men & masculinities. SAGE Publications, Inc. https://doi.org/10.4135/
Copland, S (2020) Reddit quarantined: Can changing platform affordances reduce hateful material online? Internet Policy Rev 9(4). https://doi.org/10.14763/2020.
Cottee S (2021) Incel (E)motives: resentment, Shame and Revenge. Stud Confl Terror 44(2):93–114. https://doi.org/10.1080/
Coufal, L, Saresma, T, & Kosonen, H (2025) Fight with words: Misogynist extremism on an incel discussion forum. NORMA, 1–19. https://doi.org/10.1080/
Cunningham Rogers A (2024) Racialization on incels.is: Racial Status and Status Threat. Toronto Metropolitan University. https://doi.org/10.32920/
Daraganova, G, & Robins, G (2012) Autologistic Actor Attribute Models. In D Lusher, G Robins, & J Koskinen (Eds.), Exponential Random Graph Models for Social Networks: Theory, Methods, and Applications (pp. 102–114). Cambridge University Press; Cambridge Core. https://doi.org/10.1017/
Farrell, T, Fernandez, M, Novotny, J, & Alani, H (2019) Exploring Misogyny across the Manosphere in Reddit. Proceedings of the 10th ACM Conference on Web Science, 87–96. https://doi.org/10.1145/
Ferber AL, Kimmel MS (2008) The Gendered Face of Terrorism. Sociol Compass 2(3):870–887. https://doi.org/10.1111/j.
Gaisbauer F, Pournaki A, Banisch S, Olbrich E (2021) Ideological differences in engagement in public debate on Twitter. PLOS ONE 16(3):e0249241. https://doi.org/10.1371/
Gentry CE (2022) Misogynistic terrorism: it has always been here. Crit Stud Terror 15(1):209–224. https://doi.org/10.1080/
Gheorghe RM (2024) “Just Be White (JBW)”: incels, race and the violence of whiteness. Affilia 39(1):59–77. https://doi.org/10.1177/
Gilmore DD (2010) Misogyny: the male malady. University of Pennsylvania Press, Incorporated. https://books.google.cz/books?
Ging, D (2017) Alphas, Betas, and Incels: Theorizing the Masculinities of the Manosphere. Men and Masculinities, 22(4), 638–657. https://doi.org/10.1177/
Green R, Fowler K & Palombi A (2023) The Black pill pipeline: a process-tracing analysis of the Incel’s continuum of violent radicalization. Sexualities https://doi.org/10.1177/
Held S (2023) incels://cheeks/jaws: on fragile masculinity, fatal body ideals, homophobic homoeroticism and National Socialist aesthetics revisited. Fash Style Pop Cult 10(1):147–166. https://doi.org/10.1386/fspc_
Hoffman B, Ware J, Shapiro E (2020) Assessing the threat of incel violence. Stud Confl Terror 43(7):565–587. https://doi.org/10.1080/
Hutton V, Sisko S, Barnett M, Fotheringham F, Hutton V & O’Loughlin K (2020) Heterosexism and cisgenderism. In: Multicultural responsiveness in counselling and psychology: working with Australian populations. Springer International Publishing. https://books.google.cz/books?
Incel Wik (2024) https://incels.wiki/w/Meme
Jaki S, De Smedt T, Gwóźdź M, Panchal R, Rossa A & De Pauw G (2019) Online hatred of women in the Incels.me forum: linguistic analysis and automatic detection. J Lang Aggress Conflict 7(2):240–268 https://doi.org/10.1075/jlac.
Jane EA (2018) Systemic misogyny exposed: translating rapeglish from the manosphere with a random rape threat generator. Int J Cult Stud 21(6):661–680. https://doi.org/10.1177/
Kay JB (2022) Abject desires in the age of anger: Incels, femcels and the gender politics of unfuckability. In: Harrod M, Leonard, S & Negra, D eds. Imagining “We” in the Age of “I”. Routledge, pp 29–46
Kelly M, Rothermel A-K, Sugiura L (2024) Victim, violent, vulnerable: a feminist response to the incel radicalisation scale. Perspect Terror 18(1):91–119
Kimmel MS (2013) Angry white men: American masculinity at the end of an era. Nation Books
Kruglanski AW, Bélanger JJ & Gunaratna R (2019) The three pillars of radicalization: needs, narratives, and networks. Oxford University Press. https://doi.org/10.1093/oso/
Lara-Cabrera R, Pardo AG, Benouaret K, Faci N, Benslimane D, Camacho D (2017) Measuring the radicalisation risk in social networks. IEEE Access 5:10892–10900
Lugones M (2007) Heterosexualism and the colonial/modern gender system. Hypatia 22(1):186–209. JSTOR
Marwick AE, Caplan R (2018) Drinking male tears: Language, the manosphere, and networked harassment. Feminist Media Stud 18(4):543–559. https://doi.org/10.1080/
Marwick A, Clancy B & Furl K (2022) Far-right online radicalization: a review of the literature. The Bulletin of Technology & Public Life
Miller S (2024) Supreme Gentlemen: the path of radicalization for the incel community’s lone wolves. Terror Polit Violence 36(6):818–833. https://doi.org/10.1080/
Moskalenko S, González JF-G, Kates N, Morton J (2022) Incel ideology, radicalization and mental health: a survey study. J Intell Confl Warf 4(3):1–29. https://doi.org/10.21810/jicw.
Moskalenko S, Kates N, Fernández-Garayzábal González J & Bloom M (2022) Predictors of radical intentions among incels: a Survey of 54 self-identified incels. J Online Trust Safety, 1(3). https://doi.org/10.54501/jots.
Neumann P (2012) Countering online radicalization in America. Bipartisan Policy Center
Neumann PR (2013) The trouble with radicalization. International Affairs, 89(4), 873–893. https://doi.org/10.1111/1468-
O’Hanlon R, Altice FL, Lee RK.-W, LaViolette J, Mark G, Papakyriakopoulos O, Saha K, De Choudhury M, & Kumar N (2024) Misogynistic Extremism: A Scoping Review. Trauma, Violence, & Abuse, 25(2), 1219–1234. https://doi.org/10.1177/
Parker A, Pallotti F, Lomi A (2021) New network models for the analysis of social contagion in organizations: an introduction to autologistic actor attribute models. Organ Res Methods 0(0):10944281211005167. https://doi.org/10.1177/
Perliger A, Stevens C & Leidig E (2022) Mapping the ideological landscape of extreme misogyny. International Centre for Counter-Terrorism. https://read-me.org/s/mapping-
Della Porta D (2018) Radicalization: A Relational Perspective. Annu Rev Political Sci 21(1):461–474. https://doi.org/10.1146/
Putnam RD (2001) Bowling Alone: The Collapse and Revival of American Community. New York: Simon & Schuster. ISBN 0-7432-0304-6
Renström EA, Bäck H, Knapton HM (2020) Exploring a pathway to radicalization: the effects of social exclusion and rejection sensitivity. Group Process Intergroup Relat 23(8):1204–1229. https://doi.org/10.1177/
Ribeiro MH, Blackburn J, Bradlyn B, De Cristofaro E, Stringhini G, Long S, Greenberg S, Zannettou S (2021) The evolution of the manosphere across the Web. Proc Int AAAI Conf Web Soc Media 15(1):196–207. https://doi.org/10.1609/icwsm.
Ribeiro MH, Ottoni R, West R, Almeida VAF & Jr W. M. (2020) Auditing radicalization pathways on YouTube. Proceedings of the 2020 conference on fairness, accountability, and transparency. abs/1908.08313, pp 131–141. https://doi.org/10.1145/
Robins G, Pattison P, Elliott P (2001) Network models for social influence processes. Psychometrika 66(2):161–189. https://doi.org/10.1007/
Saal, J (2021) Theory: The Dark Social Capital of Religious Radicals. In J. Saal, The Dark Social Capital of Religious Radicals (pp. 13–67). Springer Fachmedien Wiesbaden. https://doi.org/10.1007/978-3-
Scott J (2006) Social network analysis: a handbook. Sage Publications, Inc
Spade D (2003) Resisting Medicine/Remodeling Gender. Berkeley Women’s Law J 18:15–37
Stivala A, Wang P & Lomi A (2024) ALAAMEE: open-source software for fitting autologistic actor attribute models. https://arxiv.org/abs/2404.
Tange O (2018) Gnu Parallel 2018. Zenodo. https://doi.org/10.5281/
Tomkinson S, Harper T, Attwell K (2020) Confronting Incel: exploring possible policy responses to misogynistic violent extremism. Aust J Political Sci 55(2):152–169. https://doi.org/10.1080/
Tranchese A & Sugiura L (2021) “I Don’t Hate All Women, Just Those Stuck-Up Bitches”: how incels and mainstream pornography speak the same extreme language of misogyny. Violence Against Women https://doi.org/10.1177/
van der Vegt, I, Mozes, M, Kleinberg, B, & Gill, P (2021) The Grievance Dictionary: Understanding threatening language use. Behavior Research Methods, 53(5), 2105–2119. https://doi.org/10.3758/
Wedel L, Coufal L (2025) Radicalization or relief: divergent impacts of the COVID-19 lockdown on incels by seniority. N Media Soc. https://doi.org/10.1177/
Wedel L (2023) Incels.is forum dataset. https://doi.org/10.7802/2485
Whittaker J (2023) Online radicalisation: what we know. https://doi.org/10.13140/RG.2.
Zick A (Director) (2015) Group-focused enmity and radicalisation [video recording]. Tom Lantos Institute. Youtube. https://www.youtube.com/watch?
These authors contributed equally: Linda Coufal, Lion Wedel.
Authors and Affiliations
Charles University, Faculty of Social Sciences, Prague, Czech Republic
Linda Coufal
Weizenbaum Institute, Berlin, Germany
Lion Wedel
Corresponding author
Correspondence to Linda Coufal.
Ethics declarations
Competing interests
The authors declare no competing interests.
Ethical approval
Ethical approval was not required for this study, nor was informed consent. The study included solely publicly available online content and our subsequent analysis of it does not allow for identification of concrete users nor concrete people.
To see more documents/articles regarding this group/organization/subject click here.