Introduction
QAnon is an online movement that centers around a conspiracy theory suggesting that the United States is controlled by a secret cabal of satanic and pedophilic ruling elites, primarily comprised of democratic lawmakers, celebrities, and “deep state” bureaucrats (Moskalenko and McCauley, Reference Moskalenko and McCauley2021). Mainly originating from the forum 4chan, it has spread significantly over the years to more mainstream fora such as Twitter and has influenced political outcomes in the United States. Most notable among these political outcomes are QAnon's electoral support of former President Donald J. Trump as the solution to the “deep state” conspiracy, the discourse surrounding the Biden family and Hilary Clinton, the January 6th Capitol Insurrection, and the election of QAnon supporters like Representative Marjorie Taylor Greene to U.S. Congress.
A recent survey by the Public Religion Research Institute 2021 found that a sizeable portion of Americans agree with core tenets of the QAnon movement. For example, it estimated that 15% of Americans agree that “the government, media, and financial worlds in the U.S. are controlled by a group of Satan-worshipping pedophiles who run a global child sex trafficking operation.” Further, 20% believe that “there is a storm coming soon that will sweep away the elites in power and restore the rightful leaders” (Public Religion Research Institute, 2021). With the rise in reactionary conservative influence in American politics, QAnon has acquired a cultural salience that promises to become only more consequential with time.
Increasingly, contemporary politics in America seem to be proceeding without reference to boundaries that have customarily divided “the religious” from “the political” (Schmitt, Reference Schmitt1976; Margolis, Reference Margolis2018). The prominence of terms like “Christian nationalism” (Martí, Reference Martí, Grace and Penny2020; Whitehead and Perry, Reference Whitehead and Perry2020; Armaly et al., Reference Armaly, Buckley and Enders2022) and “Christofascism” (Sölle, Reference Sölle1982; Foertsch and Pieper, Reference Foertsch and Pieper2023) suggests as much—namely, that religious and political affiliative spaces have become increasingly interpenetrated in ways that generate synthetic reactionary identities that encompass both categories. With the remarkable capacity for assemblage that conspiratorial culture evinces (Hofstadter, Reference Hofstadter2008; DeCanio, Reference DeCanio2011; van der Linden et al., Reference van der Linden, Panagopoulos, Azevedo and Jost2021), QAnon can arguably be seen as both a symptom and a cause of this process, being at once both an ascendant political force (with more than 80 supporters vying for nominations in the 2020 Congressional primary races [Enders et al., Reference Enders, Uscinski, Klofstad, Wuchty, Seelig, Funchion, Murthi, Premaratne and Stoler2022]) and a potential religious phenomenon (being labeled a “cult” in mass media and even some academic treatments [Cohen, Reference Cohen2022; Hughes, Reference Hughes2022]).
There has been little academic work on this synthesis as it relates to QAnon, however, with some important exceptions (Beverley Reference Beverley2020; Armaly et al., Reference Armaly, Buckley and Enders2022; Enders et al., Reference Enders, Uscinski, Klofstad, Wuchty, Seelig, Funchion, Murthi, Premaratne and Stoler2022; Joosse and Zelinsky, Reference Joosse and Zelinsky2022; Young and Boucher, Reference Young and Boucher2022; Bromley and Richardson, Reference Bromley, Richardson and Miller2023; Richardson, Reference Richardson and Miller2023). Cultural analyses of the topic have largely been relegated to the journalistic realm, many of which conform to patterns of denigration that will be familiar to anyone who follows popular coverage of emergent religious and political phenomena. In this article, we explore the boundary of religion and politics in the case of QAnon through mixed method sociological inquiry. Quantitative analyses rely on data from wave 6 of the Baylor Religion Survey, while qualitative data are drawn from notable QAnon internet servers. Ultimately, we argue that QAnon operates much like contemporary Christian nationalist (Whitehead and Perry, Reference Whitehead and Perry2020) or Christofascist (Sölle, Reference Sölle1982; Foertsch and Pieper, Reference Foertsch and Pieper2023) movements in that it is best characterized primarily as a political movement which has coopted elements of religion, rather than an entity that is originally religious with an adopted political dimension. In short, QAnon is not a “cult” (Cohen, Reference Cohen2022; Hughes, Reference Hughes2022). It is a political movement that makes use of religious rhetoric and symbols to construct the us/them boundaries on which populism thrives (Laclau, Reference Laclau2005; Arato and Cohen, Reference Arato and Cohen2017; Joosse and Zelinsky, Reference Joosse and Zelinsky2022; Young and Boucher, Reference Young and Boucher2022).
QAnon, politics, and religion
What is religious affiliation, and what is political affiliation? How do they correspond? How do they differ? Such themes have been a mainstay for investigation within the contemporary sociology of religion. Despite the prevalence of QAnon symbols during the Capitol Insurrection and within reactionary conservative politics generally, QAnon has received only a few academic treatments thus far. In order to address our central question, we must foreground our empirical analysis with an examination of the existing literatures on QAnon, politics, and religion.
Much of the emerging literature about QAnon can be characterized by a meta-dialogue between those who would view it primarily as a religious movement (Cohen, Reference Cohen2022; Hughes, Reference Hughes2022), and those who would view it more as a political phenomenon (Amarasingam and Argentino, Reference Amarasingam and Argentino2020; Beverley, Reference Beverley2020; Moskalenko and McCauley, Reference Moskalenko and McCauley2021; Young and Boucher, Reference Young and Boucher2022; Bromley and Richardson, Reference Bromley, Richardson and Miller2023). Falling into the latter category are those who view it as a form of political extremism (Enders et al., Reference Enders, Uscinski, Klofstad, Wuchty, Seelig, Funchion, Murthi, Premaratne and Stoler2022), those who consider it a security threat (Amarasingam and Argentino, Reference Amarasingam and Argentino2020; Moskalenko and McCauley, Reference Moskalenko and McCauley2021), and those who, agreeing that QAnon poses a threat to societal security, call for strategies of “deradicalization” to be implemented toward the movement (Garry et al., Reference Garry, Walther, Mohamed and Mohammed2021). Sources differ on the level of threat QAnon poses to political institutions, however. Moskalenko and McCauley (Reference Moskalenko and McCauley2021), for example, argue that QAnon should not be labeled as a terrorist organization and predict that overly zealous governmental interdictions would only serve to exacerbate the threat it poses. By contrast, Amarasingam and Argentino (Reference Amarasingam and Argentino2020) point to evidence of past QAnon inspired terrorist acts as a means of suggesting that QAnon is a growing security threat (see also Garry et al., Reference Garry, Walther, Mohamed and Mohammed2021).
Several articles on QAnon have focused on its processes of ideological diffusion (de Zeeuw et al., Reference de Zeeuw, Hagen, Peeters and Jokubauskaite2020; Bleakley, Reference Bleakley2021; Tollefson, Reference Tollefson2021). These sources often link the spread of QAnon beliefs to the influence of President Trump (Tollefson, Reference Tollefson2021) or focus on network effects through social media (de Zeeuw et al., Reference de Zeeuw, Hagen, Peeters and Jokubauskaite2020; Bleakley, Reference Bleakley2021). This latter literature discusses the “normiefication” (or normalization) of fringe QAnon discourse via mainstream platforms like YouTube, Reddit, and Twitter (now known as “X”)—outlets from which many of the original QAnon content producers have been banned, but where QAnon ideology nevertheless continues to spread (de Zeeuw et al., Reference de Zeeuw, Hagen, Peeters and Jokubauskaite2020; Bleakley, Reference Bleakley2021).
The above analyses of QAnon's communicative and mobilizing processes have been accompanied more recently by categorical assessments, and this is where discussions about QAnon's religious qualities have come to the fore. Building upon some of the initial online network propositions, for example, Cohen (Reference Cohen2022) has described QAnon as a “digital cult.” Examining QAnon's online network effects, Hughes (Reference Hughes2022) similarly argues that religion plays a greater role in the dissemination of QAnon than it did with previous secular conspiracy movements, such as the 9/11 Truth movement.
There has been resistance to such religious classifications, however. Bromley and Richardson (Reference Bromley, Richardson and Miller2023), for example, argue that the QAnon conspiracy narrative is a specific type of “danger narrative” that, while it may contain quasi-religious themes, is more closely related to political concerns about the deterritorialization–reterritorialization (Deleuze and Guattari, Reference Deleuze and Guattari1983; Foucault, Reference Foucault1991) of the population, which is ultimately a consequence of globalization. Young and Boucher (Reference Young and Boucher2022) similarly describe QAnon not as a religious movement, but rather as a political conspiracy story generated by social inequality:
[I]t is worth noting that [QAnon's] capacity to narratively assimilate these concerns [undemocratic beliefs and critique of the current social arrangements] strongly depends on a background of rising social inequality. This is especially characterized by declining mental health services, revelations about institutionalized child abuse, histories of state deceptions and surveillance, and a significant disconnect between political parties and the popular classes. Something is rotten, in other words, in the state of Denmark… (4)
In this view, then, QAnon's causative grievance is a decline in status for the typically empowered—white middle-aged men—which itself is mostly due to structural forces of globalization (Armaly et al., Reference Armaly, Buckley and Enders2022). Anger has been described as “the essential political emotion” (Lyman, Reference Lyman1981, 61), and other researchers have examined the impact of anger on populism (Magni, Reference Magni2017, Rico et al., Reference Rico, Guinjoan and Anduiza2017, Reference Rico, Guinjoan and Anduiza2020). The effect of a person's sense of self and control on populism has had limited exploration to date. However, some recent scholarship identifies uncertain sense of self being associated with increased likelihood of association/identification with groups who have similarly ambiguous identities (Hogg and Gøtzsche-Astrup, Reference Hogg and Gøtzsche-Astrup2021).
This “cultural backlash” toward status anxiety that Young and Boucher (Reference Young and Boucher2022, 15) describe can thus best be summarized as a form of reactionary populism, a negative ideology embodying the “paranoid style” (Hofstadter, Reference Hofstadter2008; DeCanio, Reference DeCanio2011; van der Linden et al., Reference van der Linden, Panagopoulos, Azevedo and Jost2021) that is a common thread found throughout American political history generally. New religious movement scholar James A. Beverley thus criticizes the tendency to depict QAnon as a religious cult and argues rather that it should be understood as a byproduct of extreme political polarization (Beverley, Reference Beverley2020, 140).
Literature on populism (O'Donnell et al., Reference O'Donnell, Schmitter and Whitehead1986; Linz and Stepan, Reference Linz and Stepan1994; Mouffe, Reference Mouffe1999; Laclau, Reference Laclau2005; Linz, Reference Linz2009; Arato and Cohen, Reference Arato and Cohen2017; Revelli, Reference Revelli2017; Martí, Reference Martí, Grace and Penny2020) suggests that this “paranoid style” (Hofstadter, Reference Hofstadter2008; DeCanio, Reference DeCanio2011; van der Linden et al., Reference van der Linden, Panagopoulos, Azevedo and Jost2021) is channeled into rhetoric that generalizes “the people” as a signifier (Laclau, Reference Laclau2005), thus constructing us/them boundaries (Schmitt, Reference Schmitt1976). One salient boundary is religion (Arato and Cohen, Reference Arato and Cohen2017), which provides an easy dichotomy between believers and non-believers (Williams, Reference Williams and Smidt2003, 181). This forms the basis of populist rhetoric based on culturally Christian signifiers (Martí, Reference Martí, Grace and Penny2020).
It is possible that QAnon operates much like Christian nationalism or Christofascism, in which political and religious affiliations are blurred to the point where any distinction between them becomes difficult. Nevertheless, in such cases, affiliative sequences can still offer clues as to whether religion or politics is dominant or primary in the relationship. In her book From Politics to the Pews for example, Margolis (Reference Margolis2018) convincingly argues that, due to recent trends in polarization, an individual is more likely to affiliate religiously after they have adopted a conservative political identity. This pathway—from political to religious affiliation—is also advanced within the Christian nationalism literature (Martí, Reference Martí, Grace and Penny2020; Whitehead and Perry, Reference Whitehead and Perry2020; Armaly et al., Reference Armaly, Buckley and Enders2022), which suggests a dangerous blurring of the ideotypical institutions of religion and politics. These authors suggest that Christian nationalism is ultimately not originally religious, but rather that it is a recapitulation of a familiar form of American nativism. Armaly et al. (Reference Armaly, Buckley and Enders2022) make this connection clear in the case of QAnon by linking it to Christian nationalism and approval of political violence. They find that approval of political violence like the Capitol Insurrection is linked to Christian nationalist beliefs, perceived victimhood, white identity, evangelicalism, and support for QAnon, suggesting that wide support of Christian nationalism is channeled by political elite for specific political goals (Armaly et al., Reference Armaly, Buckley and Enders2022). Foertsch and Pieper (Reference Foertsch and Pieper2023) corroborate this argument, suggesting that reactionary political conservatives have strategically coopted religious symbols for reactionary conservative causes.
Thus, informed by the above discussions involving the sociology of religion and politics, we propose one exploratory proposition, which we seek to confirm or deny through our analysis.
Proposition: QAnon can be considered a blend of both religion and politics (Sölle, Reference Sölle1982; Margolis, Reference Margolis2018; Whitehead and Perry, Reference Whitehead and Perry2020; Armaly et al., Reference Armaly, Buckley and Enders2022; Foertsch and Pieper, Reference Foertsch and Pieper2023).
We intend to test this proposition against the ideas that QAnon is exclusively a political movement fueled by reactionary anti-elitism (Amarasingam and Argentino, Reference Amarasingam and Argentino2020; Beverley, Reference Beverley2020; Moskalenko and McCauley, Reference Moskalenko and McCauley2021; Young and Boucher, Reference Young and Boucher2022; Bromley and Richardson, Reference Bromley, Richardson and Miller2023), or that it is exclusively a religious “cult” (Cohen, Reference Cohen2022; Hughes, Reference Hughes2022).
Data and methods
Like stereotypical ideation generally, preconceived notions about QAnon have often stood in the way of social scientific comprehension of the subject. Journalistic accounts have followed predictable patterns of interpretation (Young and Boucher, Reference Young and Boucher2022)—as Beverley (Reference Beverley2020, 23) has pointed out, there is a “tendency to treat Anons as the ‘deplorables’ of 2020….. [T]he mainstream press loves to publish long-winded essays holding up QAnon as evidence that Trump supporters are conspiracy-addled rubes.”
For the purposes of the study, and following Beverley's (Reference Beverley2020) lead, we adopt a position of methodological agnosticism with respect to the core claims of QAnon, choosing to accord such beliefs with the same epistemological status that we would grant for tenets of any ideological movement. QAnon's core tenets contribute to and arise from “definitions of the situation” that are operative for their adherents, and our present aim is to ascertain the sociopolitical importance of these interpretive frames, without entering into moral evaluations of them or claiming to determine their ultimate truth value in an objective sense.
There is some indication that QAnon adherents are not dogmatically bound to the particular claims that circulate within their movement and that their ideological alignment is in this sense resistant to empirical “disconfirmation” (Weiser, Reference Weiser1974; Bader, Reference Bader1999). As Young and Boucher (Reference Young and Boucher2022, 3) note for example, believers in QAnon, “relate to the ‘truthfulness’ of the narrative, that way it captures in their estimation ‘what typically happens,’ not the truth of individual propositions” (emphasis added).
To recognize this perspective on QAnon's truth-claims is to credit QAnon adherents with the critical-evaluative ability to be aware of their disenfranchisement (DeCanio, Reference DeCanio2011; Beverley, Reference Beverley2020; Richardson, Reference Richardson and Miller2023) while revealing ways in which their ideologies can remain immune to disconfirmation or debunking (Weiser, Reference Weiser1974; Bader, Reference Bader1999). The promise of such an approach is that patterns of adherence to beliefs themselves can be revealing of the societal–structural arrangements that give rise to them.
Because QAnon is still very much in its infancy, this research is necessarily exploratory in nature. For this reason, we have chosen a relatively diverse strategy for empirical capture, with two stages to our mixed research design: the first, quantitative; the second, qualitative. We will describe these in turn.
Quantitative data
Quantitative data were gathered with wave 6 of the Baylor Religion Survey, a national survey of American adults living in all 50 states and the District of Columbia. The survey was conducted between January and March, 2021. In total, 11,000 respondents were selected for contact at random, and the final response rate was 11.3%, or 1,248 respondents. Survey respondents could choose an online or paper-based completion method, with English and Spanish options. The average age of survey respondents was 49 years old, with an approximately 52 to 47% male-to-female ratio in the sample.Footnote 1
The survey measured both explicit self-identification with QAnon membership as well as the prevalence of its core belief in the sample population (i.e., that the Democratic Party is controlled by a global network of pedophile groomers). We also included support for Donald Trump and Joe Biden,Footnote 2 and a class-based analysis that suggests that QAnon might be underpinned by feelings of resentment and anger among disenfranchised individuals (DeCanio, Reference DeCanio2011; Armaly et al., Reference Armaly, Buckley and Enders2022; on ressentiment, Nietzsche, Reference Nietzsche1998; Rico et al., Reference Rico, Guinjoan and Anduiza2017; Joosse and Zelinsky, Reference Joosse and Zelinsky2022). We have chosen to test for nativist beliefs (in line with expectations from literature [Schmitt, Reference Schmitt1976; Arato and Cohen, Reference Arato and Cohen2017; Whitehead and Perry, Reference Whitehead and Perry2020; Armaly et al., Reference Armaly, Buckley and Enders2022; Young and Boucher, Reference Young and Boucher2022; Foertsch and Pieper, Reference Foertsch and Pieper2023]), self-reported income,Footnote 3 and overt emotional states of anger and sense of control over one's life (on status anxiety, see Hofstadter, Reference Hofstadter2008; van der Linden et al., Reference van der Linden, Panagopoulos, Azevedo and Jost2021) in order to see how these factors might be propitious for Q belief, whether explicit or implicit.
To validate our proposition that QAnon should be considered a blend of the political and religious, we also check for biblical literalism (in line with Cohen, Reference Cohen2022; Hughes, Reference Hughes2022). We use biblical literalism over other religiosity scales or variables because it captures a core element of perceived “fundamentalism,” which informs conduct in a way that religious service attendance and measures of religious salience do not. Particularly, it includes unchurched believers who believe in Christian tenets but who do not engage with traditional church structures (for more on the use of biblical literalism as a proxy for religiosity, see Manning [Reference Manning2015]; see Sullivan [Reference Sullivan2012] for a discussion of unchurched Christianity).
Independent variables
Nativism: We treat nativism as a scale variable predicated upon participant responses to the following questions:
33e: Do you favor or oppose the following? Sending all unauthorized immigrants back to their home countries? (1 = Strongly Oppose 2 = Oppose, 3 = Favor, 4 = Strongly Favor).
35g: Rate whether you agree or disagree with the following statements. The world would be a better place if people from other countries were more like Americans. (1 = Strongly Disagree, 2 = Disagree, 3 = Undecided, 4 = Somewhat Agree, 5 = Strongly Agree).
35h: Rate whether you agree or disagree with the following statements. Generally speaking, the United States is a better country than most other countries. (1 = Strongly Disagree, 2 = Disagree, 3 = Undecided, 4 = Somewhat Agree, 5 = Strongly Agree).
After accounting for missing responses, we rescaled the variable by adding the total Likert scale values and dividing by the total variable count, taking the floor of the result to produce the following new scale variable (0 = Non-Nativist, 1 = Low Nativism, 2 = Moderate Nativism, 3 = High Nativism) (Cronbach's alpha = 0.67).
Anger: Anger was measured based on participant responses to Question 4G: In the past WEEK, about how often have you had the following feelings? I felt angry. (1 = Never, 2 = Hardly ever, 3 = Some of the time, 4 = Most of the Time). This variable was recoded to account for missing/undecided responses.
Sense of Self: Feelings of no control was treated as a scale variable (Cronbach's alpha = 0.65Footnote 4) based on participant responses to questions 5A–D, recoded to account for missing/undecided responses:
5A: Please rate the extent to which you agree or disagree with the following statements. I have little control over the things that happen to me. (1 = Strong Disagree, 2 = Disagree, 3 = Agree, 4 = Strongly Agree).
5B: Please rate the extent to which you agree or disagree with the following statements. There is really no way I can solve some of the problems I have. (1 = Strong Disagree, 2 = Disagree, 3 = Agree, 4 = Strongly Agree).
5C: Please rate the extent to which you agree or disagree with the following statements. I often feel helpless in dealing with problems of life. (1 = Strong Disagree, 2 = Disagree, 3 = Agree, 4 = Strongly Agree).
5D: Please rate the extent to which you agree or disagree with the following statements. I can do just about anything I really set my mind to. (1 = Strong Disagree, 2 = Disagree, 3 = Agree, 4 = Strongly Agree). This question was reverse coded for consistency with the other questions.
Perceived Income/Social Disadvantage: Perceived income or relative perceived social disadvantage was measured as participant's response to Question 20: Which of the following best describes your (your household's) ability to get along on your (its) income? (1 = Always have money left over, 2 = Have enough with a little extra sometimes, 3 = Have just enough but no more, 4 = Can't make ends meet).
Biblical Literalism: Biblical literalism was a dichotomous variable originally based on participant responses to Question 50: Which one statement comes closest to your personal beliefs about the Bible? The variable was treated as a dichotomous variable between those who said the Bible should be interpreted word for word/taken literally on all subjects versus those who did not. Excluded were individuals who omitted responses or described themselves as uncertain.
Trump Support: Trump support was based on participant responses to Question 28: Whom did you want to win the Presidency? (1 = Trump, 2 = Biden, 3 = Other).
Dependent variables
Q Support: Support for QAnon was based on participant responses to Question 34G: Do you support or oppose these social movements? -QAnon (1 = Strongly Oppose, 2 = Oppose, 3 = Support, 4 = Strongly Support). Due to cell size and degree of support being relatively unimportant, the variable was recoded as a dichotomous variable to measure whether or not an individual self-identified as a Q Supporter. Excluded were respondents who refused the question and/or were unsure.Footnote 5
Democrat/Elite Pedophile Collusion: Belief in pedophile democratic elites was based on participant responses to Question 36D: Please indicate your level of agreement with the following statements: Top democrats are involved in elite child sex-trafficking rings. (1 = Strongly Disagree, 2 = Disagree, 3 = Neither agree nor disagree, 4 = Agree, 5 = Strongly Agree). Missing responses were excluded.
Control variables
To ensure rigor of the results, race, gender, and urban proximity were controlled.
Models 1–3 measured QAnon support through binary logistic regression. Models 4–6 measured belief in Democrat/elite pedophile collusion through ordered logistic regression. To test our hypothesis that both QAnon support/belief in elite pedophile collusion are a blend of religious and political fervor, we utilized a nested modeling strategy. Models 1 and 4 used only our religious variable (biblical literalism) and controls, models 2 and 5 used only our political variables (Presidential Support, Nativism, Anger, Sense of Self) and controls. Lastly, models 4 and 6 tested the combination of our religious and political variables. All regression models and tabulation/significance test validations used the survey weights provided to account for variances in sample observation, strata, and methodology. The reference category used in all models was white women with a bachelor's degree who reported supporting Biden, lived in the suburbs, always had money left over, and were not biblical literalists (Table 1).
Descriptive statistics
* indicates the reference category.
Qualitative data
Our qualitative method focused on implementing content analysis and participant observation. Most content analysis was done using posts from online websites 8kun, 4chan, patriots.win, and rumble—all popular fora for QAnon discourse. Notably, 8kun is a central location of QAnon discussion, whereas 4chan, patriots.win, and rumble operate more as generalist forums, disseminating ideas to a broader cultural sphere (on this “normiefication,” see de Zeeuw et al., Reference de Zeeuw, Hagen, Peeters and Jokubauskaite2020).
Each generalist website was surveyed as a snapshot in random time, with all posts or comments viewed and identified as either related or unrelated to QAnon. Therefore, the method was opportunity or convenience sampling based on each website. When examining 4chan, for example, the /b (random) and /pol (political) boards were analyzed on random dates at random times. Each board averages about 15 posts per page, for 10 pages, therefore yielding approximately 300 posts analyzed for QAnon content. Depending on the page and website, QAnon content could be high or low. The /pol board was more likely to yield QAnon content when compared to the /b board, which averaged at an estimated 60% pornography. Rumble was approached similarly, focusing on comments below QAnon conspiracy videos. Even though it is counted as a specialist board, patriots.win was approached with similar methods because its format is also comment based (comparable to Reddit).
For 8kun, a specialist website which is primarily known for hosting QAnon depositories, all Q “drops” and the /qresearch board were analyzed, roughly at 25 posts per page for 25 pages, yielding about 625 posts. These were obviously more directly related to Q discourse, so random systematic sampling was used. By combining both specialist and generalist fora, the breadth of Q discourse dissemination can be observed. Overall, roughly 1,500–2,000 posts or comments over four websites were skimmed for Q content. We then filtered our convenience and systematic sampled data into our coding scheme to draw our qualitative results, which are quoted in full and interpreted theoretically.
Participant observation was done by joining closed online QAnon groups on social media platforms such as Facebook, Parler, and Discord, although these sources were not quoted directly, to avoid potential identification. Many of these groups have been banned repeatedly, especially on Facebook.
Our coding choices focused on keywords that highlighted our two proposed categories: religion (with attendant words like “cult,” etc.) and politics (“populism,” “nativism,” etc.). For religion, we focused on aspects of religious appeals and references, as well as expressions of religious authority and doctrine. This included analyses of the legitimacy of Q or the many “Teachers” and “Bakers”Footnote 6 within the movement, the propensity of the movement to resist disconfirmation (Weiser, Reference Weiser1974; Bader, Reference Bader1999), homophobic language, anti-satanic (Foertsch, Reference Foertsch2022a, Reference Foertsch2022b), and anti-Catholic rhetoric. For politics, we focused on discussions of the state, political authorities, racial and ethnic language (such as anti-Semitic or anti-Black slurs or discussions), sexism, and the construction of populist us/them dichotomies (Schmitt, Reference Schmitt1976; Laclau, Reference Laclau2005; Hofstadter, Reference Hofstadter2008; DeCanio, Reference DeCanio2011; Arato and Cohen, Reference Arato and Cohen2017; van der Linden et al., Reference van der Linden, Panagopoulos, Azevedo and Jost2021; Armaly et al., Reference Armaly, Buckley and Enders2022; Young and Boucher, Reference Young and Boucher2022). Internal validity was tested between the first and second authors.
These findings from content analysis and participant observation were then compared with expectations arising from the politics and religion literature, forming our findings section. This means that our findings are interwoven into a theoretical discussion rather than presented independently, in a formal “Results” section.
Opportunity and systematic content analysis of QAnon sources occurred for a 6-month period between July and December of 2022. Participant observation occurred over 6 months from September 2022 to February 2023.
Results
Quantitative results
Support for QAnon
Initial chi-squared tests revealed statistically significant differences between harboring nativist beliefs, feelings of anger, poor sense of self, and being a Trump supporter in support for the QAnon movement (χ2: 13.51***, 19.50***, 8.68**, 19.47***). This lends initial credence to the variables being useful indicators of the odds of supporting QAnon in our regression models.
Given the relatively small number of respondents who overtly declared support for QAnon and the amount of uncertainty respondents reported, weighted or otherwise, our ability to observe statistical significance was limited (unweighted N = 34, supporters for Trump = 19, Biden = 13, and Other = 2). However, we initially observed that having less than high-school education and living in a small town and/or rural setting were significant predictors of QAnon support (p < 0.05). Hispanic Americans in the sample had significantly greater odds of self-identifying as a QAnon supporter than did their white counterparts (p < 0.05).
Controlling for politics with no religious variable (model 2), we see Hispanic respondents' odds of QAnon support increase (p < 0.01). Having less than a high-school education and being in a suburban or rural community retained significance in predicting increased odds of supporting QAnon; however, no political variables were significant predictors of increased Q support, with Trump supporters only having marginally increased odds of supporting QAnon (p < 0.10).
Controlling for literalism and politics in tandem (model 3), living in a small town becomes a marginally significant predictor of increased QAnon support (p < 0.10). Additionally, significant associations with nativism and the feeling that one can't make ends meet begin to emerge, with each increasing odds of support for QAnon (p < 0.05). While biblical literalism had a higher level of significance (p < 0.01), the feeling that one cannot make ends meet and being a biblical literalist had effect sizes that were substantively close in terms of their impact on increased QAnon support. Trump support was not significant at all after controlling for religion and politics together and only marginally significant in model 2, which is in line with Enders et al.'s (Reference Enders, Uscinski, Klofstad, Wuchty, Seelig, Funchion, Murthi, Premaratne and Stoler2022) findings on partisanship and contradicts van der Linden et al.'s (Reference van der Linden, Panagopoulos, Azevedo and Jost2021) claims of paranoid styles being exclusive to ideological conservatives.
When considering the finding that socioeconomic status (SES) and nativism are major drivers of Q belief, the QAnon that emerges is of a populist movement fueled by anti-globalization and fundamentalist rhetoric; or a status anxiety response borne of perceived struggle and disenfranchisement (Hofstadter, Reference Hofstadter2008; DeCanio, Reference DeCanio2011; Armaly et al., Reference Armaly, Buckley and Enders2022). This contributes to the plausibility of our proposition that Q support is a mixture of political and religious ideotypes. This becomes especially important when considering our next model examining belief in democrats collaborating with global elites who sex traffic children—a core tenet of the QAnon movement.
Belief in democrat collusion with pedophile elites
Initial chi-squared tests revealed statistically significant differences between individuals with higher and lower levels of nativism, those who backed Trump versus other candidates, and between different levels of perceived income regarding the belief that democrats are involved in elite sex trafficking (χ2: 216.58***, 363.93***, and 67.90***). No statistically significant differences were detected in feelings of anger or sense of self.
Model 4 has mixed racial significances, with Hispanic respondents having increased odds of belief in collusion (p < 0.05) and Black respondents having marginally lower odds of belief in collusion (p < 0.10). Having a high-school education or less also increases odds of belief in collusion (p < 0.05). However, the largest odds ratio and most significant variable was biblical literalism (p < 0.001).
Model 5 showed that recent feelings of anger reduced odds of believing in collusion (p < 0.05), with nativism also increasing odds of belief therein (p < 0.05). People who favored candidates other than Biden also had greater odds of belief in Democrat/elite pedophile collusion (p < 0.001). We also observed, however, that Black and Hispanic respondents had increased odds of this belief relative to their white counterparts (p < 0.01). Lower educational attainment was also positively associated with this belief (p < 0.05). Additionally, perceived social disadvantage also increased odds of belief in Democrat/elite pedophile collusion, with both those who made just enough to get by and those who could not make ends meet (p < 0.10, p < 0.05). This falls in line with status anxiety literature (DeCanio, Reference DeCanio2011; Armaly et al., Reference Armaly, Buckley and Enders2022).
In model 6, anger is a marginally significant predictor of decreased support for QAnon (p < 0.10), while educational attainment was no longer a significant predictor in either direction. Black respondents also no longer had significant odds of increased support for QAnon. However, nativism retains significance, as does the feeling that one's income is not sufficient to make ends meet (p < 0.01, p < 0.05). Hispanic respondents also continued to have greater odds in belief of collusion relative to their white counterparts (p < 0.01).
As was the case with the Q model, biblical literalism was also a significant positive predictor of belief in collusion (p < 0.001). While it was also a substantively larger predictor than our other controls, we also noted that biblical literalists were likelier to report a negative perception of their income than non-literalist counterparts (χ2: 16.655*).
As such, while respondents may have been reticent about outright declaring their support for the QAnon movement, the impact of religion (Cohen, Reference Cohen2022; Hughes, Reference Hughes2022) is most strongly observed in those who harbor populist beliefs and/or those who feel they are struggling to get by, even after controlling for staunch religious beliefs. This suggests that religious beliefs are being used as a justification of Q-style anti-globalist belief, which is a status anxiety response to a perceived decline in material conditions (Hofstadter, Reference Hofstadter2008; DeCanio, Reference DeCanio2011; van der Linden et al., Reference van der Linden, Panagopoulos, Azevedo and Jost2021; Armaly et al., Reference Armaly, Buckley and Enders2022) (Tables 2 and 3; Figures 1 and 2).
Reported nativist sentiments and support for QAnon movement.
Perceived income and belief in democrat involvement in elite child sex trafficking.
Support for QAnon
***p < 0.001, **p < 0.01, *p < 0.05, †p < 0.1.
Belief in pedophile elites controlling democratic party
***p < 0.001, **p < 0.01, *p < 0.05, †p < 0.1.
Qualitative results
We can readily discern populist in/out group majoritarian discourse (Schmitt, Reference Schmitt1976; O'Donnell et al., Reference O'Donnell, Schmitter and Whitehead1986; Linz and Stepan, Reference Linz and Stepan1994; Mouffe, Reference Mouffe1999; Laclau, Reference Laclau2005; Linz, Reference Linz2009; Arato and Cohen, Reference Arato and Cohen2017; Revelli, Reference Revelli2017; Joosse and Zelinsky, Reference Joosse and Zelinsky2022; Young and Boucher, Reference Young and Boucher2022) within Q discussions online. These populist boundaries are often stylistically conspiratorial (Armaly et al., Reference Armaly, Buckley and Enders2022; Young and Boucher, Reference Young and Boucher2022; Bromley and Richardson, Reference Bromley, Richardson and Miller2023) and “paranoid” (Hofstadter, Reference Hofstadter2008; DeCanio, Reference DeCanio2011; van der Linden et al., Reference van der Linden, Panagopoulos, Azevedo and Jost2021). Notably, the QAnon-dedicated boards are remarkably tame when discussing outgroups. The following quotes were taken from “anons” posting on a QAnon-specific board, which focuses on “researching” Q dropsFootnote 7 and their associated sources:
Anonymous 10/31/22 (Mon) 19:50:28: pedophilia, along with infanticide, child transgenderism, etc, is part of cabal/deep state religion (baal/moloch worship).
Here, we see expressions of anti-transgenderism and anti-Satanism which in turn are mobilized in the service of an overriding concern about the state. Or consider the efflorescent range of evil described in the quote below:
Anonymous 09/27/22 (Tue) 18:19:33: There are no coincidences why I have all the “declassified” documents and files from QResearch. I spent sleepless and painful nights, researching and archiving all the “declassified” information. The “fake” news media knows I exposed a worldwide, demonic, trafficking/pedophile ring. The demons infiltrated the Vatican, our government, Hollywood and royal family. The “fake” news media knows, I exposed all the names of government swamp, and hollywood elites, in Epstein's flight log. They also know, I exposed Nancy Pelosi, in the underground rooms of Epstein island, caught on the security cameras. The “elites” hide a dirty little secret drug derived from. adrenalized blood. The drug (andrenochrome) exported from China, fetches more cash than the purest heroin. There are no coincidences why I was chosen to expose everything. You are watching the biggest coverup in USA history. The “swamp” got caught and I have everything. The news articles trashing “QAnon” are intentional and disinformation. Rhe news media is attacking a military zintelligence operation. Anyone who has gone down this rabbit hole, could never harm a child. This truly is a “silent” holy war of good vs evil. President Trump (aka Q+) told me to trust the plan and justice was coming MAGA.
Here, demonic elements abound, but in every case, they are infecting and by turns issuing from structural power, whether it be in the form of the Royal Family, the Vatican, China, Hollywood, the U.S. Congress, or the media. In short, it is worldly power that is at the center of concern.
Anonymous 11/01/22 (Tue) 00:10:44: When I was younger I would fantasize about moving to some place better. But then I realize that the Jew has spread the n*gger all over the world. Every land and culture is being raped by the n*gger, so the Jew can usher in the New World Order.Footnote 8
This “anon” thus expresses phobic and racist attitudes toward Jews and blacks because of the supposed political telos behind their actions; namely, the creation of a New World Order.
To summarize, then, some key themes in the texts above are clearly nativist in origin: there is a preoccupation with whiteness, white replacement theory, anti-Semitism, anti-globalism, antiqueer/homosexuality, anti-Satanism, and anti-papism. These are major recurring themes in American populist nativism (Hofstadter, Reference Hofstadter2008; DeCanio, Reference DeCanio2011) which historically fuses political and religious rhetoric (Arato and Cohen, Reference Arato and Cohen2017; Martí, Reference Martí, Grace and Penny2020; Armaly et al., Reference Armaly, Buckley and Enders2022), and QAnon emerges here as something that stands in continuance with this historical tendency. When we move away from QAnon-specific fora and to a more catch-all Trump supporter page like patriots.win, we see a similar focus. To wit:
[On COVID] Anonymous: The answer then and now was always, isolate elderly and immunocompromised and allow the rest to live on. It allows natural/herd immunity to grow while the virus weakens it's mutations. F*cking satanic globalist criminals.Footnote 9
Again, the complaints of the moment (governmental responses to COVID) are given their political valence through an interpretive frame that is first and always preoccupied with a global conspiracy. In the participant interaction below, what is initially and primarily a political debate about President Biden's leadership devolves into violent posturing and the use of a homophobic slur.
[On Biden] Anonymous 1: Eat sh*t and die. You start a world war and I swear on the Bible I'll join any f*cking resistance that involves putting you down like a dog in the god damn street to reclaim our god damn country.
Anonymous 2: Not like a dog, though. Dogs are good. How about a feral hog or something?
Anonymous 3: Put him down like a f*ggot.Footnote 10
On the issue of tech company cooperation with government censorship we have another example:
Anonymous: You forgot the DHS directly provably working to censor for ADL and SPLC Jews, though a suspect mod will delete my words…It started in 2018, prior to later US DHS big funding, with Jewish bot tools auto-commenting on “racist White people” scanned daily on Facebook and Twitter and targeting for relentless disenchantment later…the ADL also targets “anti-semite” online facts on sites they cant delete by using the FBI Directly! Proven. (watch video) FBI director Christopher Wray admitted on a call with Jonathan Greenblatt that the FBI works directly with the ADL for tips and targeting of those they have labeled “anti-semites.” This can be as straight forward as someone criticizing Israel…The FBI does the bidding of the ADL!Footnote 11
This anonymous subject thus claims that the American Department of Homeland Security and Federal Bureau of Investigation both work in collaboration with Jewish individuals in the Anti-Defamation League and Southern Poverty Law Center to label anti-Israeli political dissent as anti-Semitism.
Again, across the above set of content we see an emphasis on homophobia, anti-Semitism, and anti-Satanism among our participants. These phobic responses are typically brought forward when discussing political issues. Additionally, we note that in the more mainstream outlets like patriots.win which are made for a more general audience, participants seemed more likely to recourse to the use of slurs when discussing outgroups. Such discursive practices actively construct an us/them dichotomy (Schmitt, Reference Schmitt1976) in the worldview of the user, creating a symbolic or sacralized eschatological conflict between the “globalists” and the “people.” In general boards like 4chan, which are comparatively more easily accessible to the public, such linguistic signaling quickly becomes vulgar and slur oriented. Interest in elaborating complex conspiratorial “theory” seems to decline in favor of simply constructing stark us/them symbolic boundaries.
Many of the expressions fall clearly in line with a long-standing American preoccupation with conspiratorial anti-Semitism. For example, an anon posted a silhouette of Donald Trump performing fellatio on an orthodox Jewish man, implying the President's demeaning subservience to a shadowy cabal—anti-Semitic preoccupations were common on the site.
This above content was pulled both from the politics-specific board on 4chan dubbed “/pol/” and the random board “/b/.” Board /b/ tends to be dominated by pornographic images, with incendiary political posts like the first quote above interspersed among this material. Overall, /pol/ is far more likely to use shock tactics and contain slurs than patriots.win or QAnon boards on 8kun, which seem to be more directly preoccupied with either proving their nativistic bona fides or reacting to current events. This is likely due to the different mandates for these boards—4chan /pol/ and /b/ are both general anonymous internet boards, whereas patriots.win and 8kun are intended for specific political purposes.
Note in both cases that when religious language and references are used, it is to construct conspiratorial nativist boundaries (Jesus being white, in one example) and legitimize calls for violence (from earlier: “I swear on the Bible I'll join any f*cking resistance that involves putting you down like a dog in the god damn street”) (see Armaly et al., Reference Armaly, Buckley and Enders2022; Setter and Nepstad, Reference Setter and Nepstad2023, for more on political violence). That is, rather than being used as a base epistemological starting point, it is instead used in a post-hoc justificatory manner.
In summary, then, we can see a frequent occurrence of us/them dichotomization (Schmitt, Reference Schmitt1976) and of populistic nativism, both of which have many antecedents in American political culture (such as the Know Nothing party, Ku Klux Klan, and the Tea Party movement) (Hofstadter, Reference Hofstadter2008; DeCanio, Reference DeCanio2011). We can view QAnon, then, as a reactionary populist movement driven by ressentiment (Nietzsche, Reference Nietzsche1998; Rico et al., Reference Rico, Guinjoan and Anduiza2017) in the sense that it seeks to reinstate bygone categorical demarcations that privilege Christian whiteness,Footnote 12 heterosexuality, and patriarchal power. These categorical efforts, in turn, are motivated by a sense of status anxiety (Hofstadter, Reference Hofstadter2008; DeCanio, Reference DeCanio2011) over the emergence of an increasingly global and pluralistic society marked by capital flight. It does this by invoking and reifying traditional (often religious) symbols that are thought to be emblematic of the “people” (O'Donnell et al., Reference O'Donnell, Schmitter and Whitehead1986; Linz and Stepan, Reference Linz and Stepan1994; Mouffe, Reference Mouffe1999; Linz, Reference Linz2009; Arato and Cohen, Reference Arato and Cohen2017). In the long-standing American tradition, such efforts have been alternately expressed through anti-papism, anti-immigration, anti-Satanism, and anti-Semitism.Footnote 13
While there is very little to suggest that QAnon supporters are members of a new religious movement, and while they do not conform to standard criteria for determining religious group membership (“believing,” “belonging,” and “behaving” [Clements, Reference Clements2015]), they do display a significant overlap with historical populism. The QAnon that emerges in this research is a form of reactionary response to structural forces of economic disenfranchisement that occasionally uses religious rhetoric as a means of acquiring legitimacy (Habermas, Reference Habermas1975). Overall, then, QAnon is a political movement with a political telos (Amarasingam and Argentino, Reference Amarasingam and Argentino2020; Beverley, Reference Beverley2020; Moskalenko and McCauley, Reference Moskalenko and McCauley2021; Young and Boucher, Reference Young and Boucher2022; Bromley and Richardson, Reference Bromley, Richardson and Miller2023) rather than an exclusive religious phenomenon (Cohen, Reference Cohen2022; Hughes, Reference Hughes2022). Nevertheless, there is an ongoing blurring effect (Sölle, Reference Sölle1982; Arato and Cohen, Reference Arato and Cohen2017; Margolis, Reference Margolis2018; Whitehead and Perry, Reference Whitehead and Perry2020; Armaly et al., Reference Armaly, Buckley and Enders2022; Foertsch and Pieper, Reference Foertsch and Pieper2023) in which religion and politics are subject to an asymmetric conflation: religion and politics are both in the mix, but what we see is best described as a political cooptation of religion rather than a religious cooptation of politics.
Discussion and conclusion
To restate the proposition initially drawn from the existing literature on QAnon:
P: QAnon can be considered a blend of both religion and politics (Sölle, Reference Sölle1982; Margolis, Reference Margolis2018; Whitehead and Perry, Reference Whitehead and Perry2020; Armaly et al., Reference Armaly, Buckley and Enders2022; Foertsch and Pieper, Reference Foertsch and Pieper2023).
We see from our quantitative analysis that while biblical fundamentalism remained meaningful across our models as a predictor of QAnon support or belief in democrat pedophilia, it was most impactful when individuals were experiencing some form of perceived or actual disenfranchisement. This suggests that religious rhetoric is being used to channel latent class dissatisfaction (or status anxiety) for political aims. Given the results of our quantitative analysis, our proposition which posits QAnon as a blend of religion and politics seems to be accurate.
When we move to our qualitative findings, we find that QAnon conforms to a larger sociohistorical context, exhibiting a family resemblance with past nativist movements in America in its penchant for constructing us/them dichotomies pertaining to who is and is not considered “the people.” The white Christian heterosexual man is considered legitimate, and this expression of value is buttressed by a negative discursive field that employs anti-Semitic, anti-Black, homophobic, antiqueer, anti-satanic, and anti-Catholic rhetoric.
This interpretive frame was exhibited repeatedly on the internet sites we examined which host QAnon discussions. When religious rhetoric was found, it was only as an admixture to what was otherwise political discussion and conclusions. Religion, here, was used to impart legitimacy (Habermas, Reference Habermas1975) to political and populist boundaries by demarcating “us versus them.” Thus, we conclude with our qualitative analysis—again—that our proposition closely describes the nature of QAnon. This finding is largely in line with Armaly et al. (Reference Armaly, Buckley and Enders2022). But our qualitative findings allow for a more nuanced characterization; namely that while QAnon culture involves the conflation of religion and politics, it manifests as an asymmetrical conflation in which religion is being invoked to serve what is otherwise a predominantly political–cultural project. Perhaps the broader populist social movement of which QAnon is a part of will with time go on to produce cults with their own distinct religious doctrines, organizational structures, and channels of authority (Clements, Reference Clements2015). Until that day, however, we as academics must ensure that our categories of analysis responsibly reflect underlying social realities while resisting stereotypical characterizations.
There are some limitations to this work, which future research should seek to amend. Our quantitative analysis was limited due to the small number of observations we had in our sample. Ideally, a survey on QAnon believers would assist in generalizing about the population. To amend this issue, we turned to mixed methods and included a broad qualitative content analysis of QAnon discourse. Qualitative work though, especially random convenience sampling, could also have issues with generalizability. We have attempted to address this issue by including both generalist and specialist websites in our sample, but we are under no pretense that our sample is a systematic representation of the whole QAnon constituency. If possible, future research should attempt to corroborate our findings with more representative samples.
This work suggests several paths for future research. First, a fruitful comparison could be made between the discursive characteristics of Christian nationalist, Christofascist, and QAnon conspiratorial populist rhetoric. This comparison could lend itself to further theorizing about syncretism among reactionary conservative political movements, which Stewart (Reference Stewart2020) dubs “far right civilizationism.” Second, future analyses of QAnon would also greatly benefit from a thicker historical-comparative assessment of relations to status anxiety as it has existed in American populist movements of the past such as McCarthyism, the Know-Nothings, the KKK, and the Tea Party, to name a few (Hofstadter, Reference Hofstadter2008; DeCanio, Reference DeCanio2011). Third and finally, even though much of our analysis focuses on the American context, QAnon-style status anxiety is not a uniquely American phenomenon. Studies of QAnon's dissemination internationally could help expose the rise of anti-globalist politics across the world (Stewart, Reference Stewart2020). In order to understand this important and emerging global political trend, continued academic attention is imperative.
Acknowledgments
The authors thank Paul Froese, George Yancey, Kevin Dougherty, and James T. Richardson for reading drafts, suggesting edits, and making this project possible.
Steven Foertsch is a doctoral candidate in the Department of Sociology and a research assistant at the Institute for Studies of Religion at Baylor University. His research focuses on the study of ideology and the intersection of politics and religion, deviance, social and political philosophy, and emergent religious groups. A recent work of his on politics and religion titled “A Social History of Christofascism” can be found in the Routledge International Handbook of Sociology and Christianity.
Rudra Chakraborty is a doctoral student at Baylor University in the Department of Sociology. His research interests primarily revolve around shifting marriage market dynamics and their impact on mental health, radicalization, and the formation of new social movements (particularly online).
Paul Joosse an associate professor of sociology at the University of Hong Kong. His work on Weber, political sociology, and culture has been published in Social Forces, Sociological Theory, and Theory and Society, among others. He is the recipient of the American Sociological Association's Junior Theorist Award, the International Sociological Association's Junior Theorist Prize, and the ASA's Clifford Geertz Prize for cultural sociology.
Open access
