International Journal of Communication 17(2023), 4304–4325 1932–8036/20230005 Copyright © 2023 (Sean Phelan and Pieter Maeseele). Licensed under the Creative Commons Attribution Non-commercial No Derivatives (by-nc-nd). Available at http://ijoc.org. Critiquing “Mainstream Media” on Twitter: Between Moralized Suspicion and Democratic Possibility SEAN PHELAN1 Massey University, New Zealand University of Antwerp, Belgium PIETER MAESEELE University of Antwerp, Belgium How can we understand the critique of mainstream media (MSM) in a political moment where intense suspicion of media and journalism has been normalized in reactionary discourses? This article addresses this question from a discourse theoretical perspective that is supported by a corpus-assisted interpretivist analysis of how the terms “MSM” and “mainstream media” were articulated in a January 2021 sample of more than 11,000 tweets from different time zones. We begin by clarifying the political stakes of our argument and situating the historical emergence of “mainstream media” as a discursive category. Our Twitter analysis highlights the “logic of equivalence” established between mainstream media and other identities and the normalization of a moralized representation of media as a corrupt ally of government. We conclude by speculating on how we might affirm a radical democratic conception of media critique in a cultural context where anti-MSM rhetoric can float easily between different discourses and ideologies. Keywords: critique, discourse theory, hermeneutics of suspicion, logic of equivalence, mainstream media, moralized rhetoric, reactionary politics Let us start by describing an experience that we think will resonate with readers.2 Until recently, both authors of this article habitually posited MSM (i.e., “mainstream media”) as a commonsense object of Sean Phelan: s.phelan@massey.ac.nz Pieter Maeseele: pieter.maeseele@uantwerpen.be Date Submitted: 2022-04-07 1 This work was supported by a European Union Horizon 2020 Marie Skłodowska-Curie Individual Fellowship (Grant Agreement No. 897768). It is based on a paper that was first presented to the Media and Politics Group of the annual conference of the British Political Studies Association in March 2021. We would like to thank the reviewers for their thoughtful feedback and the editors for their help with the review and production process. 2 For further reflection, see Phelan (2021). International Journal of Communication 17(2023) Critiquing “Mainstream Media” on Twitter 4305 critique in the classroom. We still sometimes do this today. However, we are increasingly wary and self- conscious about doing so because of a heightened anxiety about how that object might be perceived by students. When we think about how the term is articulated today, what is evoked is not necessarily the left- wing critique of capitalist media that nurtured and nurtures our own interests as critical communication and media scholars, but a deeply reactionary “critique” (if that is the right word) of media and journalism—the kind that is normalized in a world where the former President of the United States casts “the media” as “the enemy of the people” (Carlson, Robinson, & Lewis, 2021, p. 737). This article poses what we think is an important political and normative question for communication, media, and journalism scholars: How can we understand the critique of “mainstream media” in a political moment where intense suspicion of media and journalism has been normalized in reactionary discourses? We address this question from an analytical perspective informed by Ernesto Laclau and Chantal Mouffe’s (2001) discourse theory. We support the argument with a corpus-assisted interpretivist analysis of how the terms “mainstream media” and “MSM” were articulated in a January 2021 sample of more than 11,000 tweets from different time zones. The complex, often playful, dynamics of media critique on Twitter are hardly reducible to an examination of explicit MSM talk; indeed, it is conceivable that many people, including participants in different democratically vibrant Twitter counterpublics (Jackson & Foucault Welles, 2015), now avoid using the acronym because of its increasingly pejorative connotations. Nonetheless, our empirical approach offers useful insights into the affective mood (Felski, 2015) of media critique on Twitter as part of our focus on understanding how media are critiqued in a political context that blurs the boundaries between different ideologies and discourses (Krzyżanowski & Ledin, 2017). Our article does not consider different theoretical conceptions of critique or normative accounts of media/press criticism (Wyatt, 2018). Instead, our Twitter analysis operationalizes the term critique in a deliberately open-ended way to signify its everyday association with negative judgements that might seem completely removed from a theoretically sophisticated understanding of critique. However, Rita Felski’s (2015) conception of critique as a “hermeneutics of suspicion” (p. 2; a term originally coined by Paul Ricoeur) offers a useful perspective for considering the affective resonances of practices that are experienced as critique in different social universes, particularly when suspicion of “the media” takes moralizing forms that, to adapt Chantal Mouffe’s argument (2005a, 2005b), offer one symptom of a generalized culture of moralized politics (Phelan, 2022). Similarly, rather than treating Twitter critique of “mainstream media” as a different species of being from academic critique, we preface our Twitter analysis with a brief history of how mainstream media have been discussed in different academic and political contexts. The rest of the article has four main elements. First, we historicize our analysis by highlighting the relatively recent emergence of “mainstream media” as a discursive category, including in academic journals. We discuss how the term has been historically articulated in a commonsense way that has been over- associated with progressive political commitments that can no longer be taken for granted because of the cultural normalization of right-wing media critique. Second, we outline the theoretical and methodological rationale of our Twitter analysis, which puts discourse theoretical concepts into conversation with corpus- assisted methods. We describe how our January 2021 sample collated everyday Twitter talk about 4306 Sean Phelan and Pieter Maeseele International Journal of Communication 17(2023) “mainstream media” and “MSM” from different empirical contexts at a time when references to COVID-19 and Joe Biden’s presidential inauguration were prominent news topics. Third, we begin our Twitter analysis with an overview of how the signifier “mainstream media” and the more commonly used acronym “MSM” are articulated as part of a “logic of equivalence” (Laclau & Mouffe, 2001, p. 130) with other identities, including “government,” “Big Tech,” and the “left.” This is supplemented by an examination of the moralized dimensions of Twitter media critique and a brief analysis of retweets that illustrate sharper ideological differences than the rest of our sample. Fourth, we conclude by clarifying the significance of our argument, particularly from the perspective of how we might affirm a radical democratic conception of media critique in a cultural context where moralized anti-MSM rhetoric can float easily between different discourses and ideologies. Contextualizing the Argument: A Brief History of “Mainstream Media” Talk The illustrations below are necessarily selective and impressionistic. However, they underscore how, much like the general category of “the media” (Guillory, 2010; Shechtman, 2021), “mainstream media” is a recent linguistic invention. A proper genealogy of the concept might trace its origins avant la lettre, particularly given its affinities with the term “mass media.” In Schectman’s (2021) account, the latter term “gained currency”—alongside the terms “mass culture” and “mass society”—“in the growing transatlantic field of sociology between the world wars” (p. 649). The Oxford English dictionary records the first not-yet compounded use of the noun “main stream” to 1585 (“Mainstream,” 2000). It records the first use of “mainstreams” as a pluralized noun to 1938, to signify “the prevailing trend of opinion, fashion, society,” and the first use of “mainstream” as an adjective to describe an “established tradition” or “field of activity” (“mainstream fiction”) to 1953 (“Mainstream,” 2000). The dictionary suggests more expansive grammatical forms of the word “mainstream” from the 1970s onward, including the use of the verb “mainstreaming” to describe an educational discourse that integrates children with disabilities into classes “for those without special needs” (“Mainstream,” 2000). Accordingly, we can surmise that as the category of “the media” started to be cited more as the name for a distinct social universe from the 1960s and 1970s onward, “mainstream media” started to be used as a commonsense discursive category around the same time. Launched in 1951, the Journal of Communication is the flagship journal of the International Communication Association. Based on a February 2022 search of its historical archive, the term “mainstream media” first appeared in a 1982 review of the edited volume Small Voices and Great Trumpets: Minorities and the Media. The reviewer, Sue Curry Jansen (1982), notes how the book challenges the mythology that the job of journalists is to “afflict the comfortable and comfort the afflicted” (p. 227). The book is described as bringing attention to “the class, race, and sex bias of mainstream media,” a “bias that skews . . . the organizational arrangements of the U.S. culture industry” (Jansen, 1982, p. 227). The use of the term is nondescript, suggesting a taken-for-granted post-1960s countercultural context (Gitlin, 1992) where Jansen (1982) could quite reasonably assume a readership already familiar with the term. “Mainstream” signifies particular kinds of media institutions, aligned with the Frankfurt School connotations of “the U.S. culture industry” (Jansen, 1982, p. 227). International Journal of Communication 17(2023) Critiquing “Mainstream Media” on Twitter 4307 The term appeared in only seven archived files in the 1980s, mainly book reviews and notices. In contrast, it has featured (as of February 2022) in 154 files since 1990, indicating that its routinized scholarly use is a comparatively recent development. The term was not cited in a full article until 1988, in a piece (Heath, 1988) about private sector participation in Kenyan public service broadcasting. It was again invoked in a taken-for-granted way, but this time to reference a singular nominalized entity, “the mainstream media” (Heath, 1998, p. 105). As if to underline the term’s banality, Heath (1998) notes how “there is nothing exceptional” about the situation in which Kenya’s public broadcaster operates: “The mainstream media everywhere serve the interests of the powerful” (p. 105). A search for the term in Journalism and Mass Communication Quarterly (which describes itself as the oldest refereed journal in mass communication) produced similar results. The term only appeared in seven texts in the 1980s, again mainly in book reviews, compared (as of February 2022) to 225 texts across the full archive. It was first cited in a 1980 article containing short bibliographical entries for recent publications, with one described as giving “an overview of mainstream media's neglect and stereotyping of [American] Indians” (McKerns & Delahaye, 1980, p. 379). It did not feature in a full article until 1985, in an article written by Dates and Gandy (1985). The term was again linked to the topic of race and what the authors described as the ideological underpinnings of “the mainstream white press” (Dates & Gandy, 1985, p. 595; emphasis added). The articulation of the term in formal academic discourse is textured by its own particular conventions. To illuminate the term’s historical emergence in a wider public sphere, we also searched for its use in the Hansard record of the U.K. (United Kingdom) parliament. It was first cited in May 1993 by Labour MP (Member of Parliament) Dianne Abbott in a House of Commons debate about “racial violence” (Abbott, 1993), reinforcing the link with critiques of media and race. The debate happened after the killing of Stephen Lawrence, an event that went on to become the subject of a high-profile public inquiry that criticized the London Metropolitan Police Service as institutionally racist. Abbott suggested that “the most disturbing and hurtful aspect of the killing . . . for the black community was the way in which initially it was almost entirely ignored by the mainstream media and the government” (Abbott, 1993, para. 19). The term was not used again in Parliament until December 1995 and again by Abbott to make a point about media and race, as part of a debate about the voluntary resettlement of people to the Caribbean. Abbott interjected with a question to the debate’s sponsor, who had already criticized “the media” for failing to adequately cover the recent death of a young Black man, Wayne Douglas, in police custody, an event that sparked the 1995 Brixton riots. Abbott censured “mainstream media” for criticizing the Caribbean community media that did cover the story—as if there was something inflammatory about “ethnic press . . . printing the facts when the mainstream press would not” (Abbott, 1995, para. 30). Abbott remains a prominent MP on the left of the U.K. Labour party, and her taken-for-granted use of the term reinforces the left-wing connotations of “mainstream media” talk—namely, its strong historical association with left critiques of the functional place of media institutions in a wider infrastructure of power, 4308 Sean Phelan and Pieter Maeseele International Journal of Communication 17(2023) as popularized by authors like Edward Herman and Noam Chomsky (2002).3 However, the ideological connotations of “mainstream media” talk have blurred over time, as the use of the term itself has been mainstreamed as part of media critiques articulated by different political constituencies. Further analysis of the Hansard archive shows how the term has been cited (as of February 2022) in 128 different contributions to U.K. parliamentary debates. One hundred and eighteen of these have been made since 2010, and the most annual citations (24) were recorded in 2021. The 2021 citations show how the term is now used across party-political lines, with most (10 uses) attributed to Tory MPs. Some Tories used the term in a neutral way. Others criticized how “the mainstream media” was used as a “derogatory term” online (Collins, 2021, para. 32). While others again reproduced those same derogatory associations, in a culture war idiom that rails on behalf of “constituents who are sick and tired of the stale, politically correct and ideologically biased output of much of the mainstream media” (Benton, 2021, para. 72). The latter example points to a rather different political genealogy of media critique that overlaps with the first citation of the term in the U.S. Congress Record in January 1984. The term was not invoked directly in debate. Rather, it was cited in a 1982 Los Angeles Times feature article that Republican Senator Jesse Helms tabled in support of an amendment to legislation for “racketeering” in pornography. The article discussed the release of the 1972 “porno movie” Deep Throat (Peraino & Gerard, 1972), noting how “mainstream media attention conferred a measure of legitimacy” on the film (130 Cong. Rec. 848, 1984). Helms’ citation of the article captures a party-political class that was becoming more alert to the politics of media representations and the perceived biases of “liberal media” (Lane, 2019). It evokes a distinctly right-wing media critique that can be traced to the hostility of the Nixon administration to the press in the 1970s (Cimaglio, 2016), and earlier again to the role of politicians like Helms in fomenting hostility to journalists for their sympathetic coverage of the civil rights movement (Greenberg, 2008). From the perspective of the politics of the present, we might describe it as emblematic of an American-style media critique that is increasingly globalized as part of the internationalization of right-wing “culture war” discourses, conflicts, and stereotypes (Hlavajova, 2020; Hunter, 1991; Phelan, 2022). What might be taken from this very fragmentary history that speaks to this article’s empirical focus on Twitter? We think the diverse examples capture a point made by Roman Hájek and Nico Carpentier (2015) about the taken for grantedness of “mainstream media” talk ever since the emergence of “mainstream media” as an object to be talked about. Citing Bruno Latour, they describe “mainstream media” as a “black box concept” (p. 369): the signifier of a complex set of mechanisms that is both obvious and epistemologically elusive at the same time. It assumes the guise of a box “about which we do not need to know nothing but their inputs and outputs” (Hájek & Carpentier, 2015, p. 369): ambiguous qualities that are exemplified by the term’s grammatical capacity to simultaneously represent both a single totality (“the mainstream media”) and a plurality of media outlets. Hájek and Carpentier (2015) make the argument as part of their critique of how “mainstream media” has been undertheorized in the “alternative media” literature. However, we think it can be extended, a fortiori, to an analysis of how “mainstream media” is/are 3 Interestingly, “mainstream media” was cited only once in the original 1988 edition of Herman and Chomsky’s (2002) book Manufacturing Consent: The Political Economy of the Mass Media but featured 34 times in a new introduction to the 2002 revised edition. International Journal of Communication 17(2023) Critiquing “Mainstream Media” on Twitter 4309 talked about on Twitter, a platform where the taken-for-granted aspects of communication are structured in part by its textual limitations (Zappavigna, 2012). Put simply, if even academics habitually talk about mainstream media by presupposing what exactly is being signified, we should expect these tendencies to be amplified on a platform where people habitually use acronyms like “MSM” to signify entities that are assumed to be already known to readers. Hájek and Carpentier (2015) emphasize the historical importance of the category of “mainstream media” to the identity of “alternative media” scholars and activists as the signifiers of a “constitutive outside” (p. 369) to a left-wing and anticapitalist credo. We see this today in how “alternative left-wing media” continue to define themselves in explicit opposition to “mainstream media” (Cushion, 2020, p. 153). However, the historical connotations of the mainstream/alternative media dichotomy have been destabilized by the growth of a far-right media infrastructure, as one manifestation of an ideological shift where the right has more actively assumed the role of populist opposition to an elite establishment (Finlayson, 2022; Peck, 2019). Right-wing media outlets and practitioners are perhaps now just as likely to claim the identity of “alternative media” critics of the putatively “left-wing” values of “mainstream media” (Figenschou & Ihlebæk, 2019; Holt, 2018; Lane, 2019; Nadler & Bauer, 2019; Nadler, Taussig, Natacha, & Wenzel, 2021; Peck, 2019; Roberts & Wahl-Jorgensen, 2020). These developments are most easily linked to the emergence of internationally recognized U.S. media outlets, such as Fox News and Breitbart.4 However, they also capture charges that are transnational in scope, evident in countries as different as Norway and India (Chadha & Bhat, 2022; Ihlebæk et al, 2022). They attest to a normalization of far-right discourses and affective sensibilities (Brown & Mondon, 2021; Knops & De Cleen, 2019; Krzyżanowski, 2020), where antagonism to “MSM” becomes interchangeable with antagonism to other political targets. The literature documenting the rise of right-wing media critiques (whether explicitly directed against an “MSM” or “liberal media” antagonist) therefore forms an important backdrop to our analysis. Our focus on Twitter heeds Broersma’s (2018) suggestion that we should situate the meanings attributed to media and journalism within the culture of the digital and “analyze different media objects and outlets in relation to one another” (p. 518). The rhetorical practice of denouncing “MSM” on Twitter is not without irony; if we were to ask what the black box category of “mainstream media” signifies today, it is not difficult to imagine an answer that would place Twitter (and other “Big Tech” platforms) at the heart of the definition. Nonetheless, as our Twitter analysis suggests, the supposition that “mainstream media” signifies some commonsense entity that is distinct from the thing called “social media” now seems to be curiously naturalized. We only have to point to the extraordinarily anachronistic hubris of the platform’s new multibillionaire owner, Elon Musk, when he declared his intention to save “citizen journalism” from the oligopolistic “elites” of “mainstream media” (Rutz, 2022). 4 It does not seem wildly speculative to trace these rhetorical shifts to the popularization of a meme culture on platforms like Twitter and Tumblr (Donovan, Dreyfuss, & Friedberg, 2022). In tandem, we might point to the growing cultural power of far-right “ideological entrepreneurs” (Finlayson, 2022) like Steve Bannon and Andrew Breitbart who took the antisystem digital politics of Occupy as a “blueprint” (Donovan et al., 2022, p. 45) for their own brand of antiestablishment politics. 4310 Sean Phelan and Pieter Maeseele International Journal of Communication 17(2023) Twitter Analysis: Theory and Method The previous section anticipated our Twitter analysis by stressing the ideologically diverse underpinnings of media critique. This section describes the theoretical and methodological rationale of our approach, which is based on a comparatively novel (for exceptions, see Brown & Mondon, 2021; Dehghan, Bruns, Mitchell, & Moon, 2020; Jacobs, 2020) combination of insights from discourse theory and corpus linguistics. The theoretical and disciplinary orientations of these two perspectives are very different, yet both stress how meaning is socially constructed through a patterned combination of signifying elements. Discourse theorists grasp the process of meaning-making via the concept of articulation (Glynos & Howarth, 2007). Researchers examine how signifiers are articulated to create discursive meaning at the macro level of an overall social formation or group identity, whether by a positively affirmed or negatively Othered identity. Conversely, corpus linguists (McEnery, McGlashan, & Love, 2015; Subtirelu & Baker, 2017) use the concept of collocation to highlight how the meaning of a word is shaped by the words it is associated with. Researchers examine how meaning is constructed at the micro level of the sentence or, even narrower again, at the default computational setting of three words to the right and three words to the left of the keyword. Formulations as mundane as the patterned use of the conjunction “and” to link “mainstream media” to other institutions become a potential topic of analytical interest. Our engagement with discourse theory and corpus linguistics here is necessarily brief, and highlights the affinities between the concept of the logic of equivalence and the concept of collocation. Laclau and Mouffe (2001) argue that any discourse is constructed through the interplay of a logic of equivalence and a logic of difference. A logic of equivalence is a logic of “simplification” (Laclau & Mouffe, 2001, p. 130) that creates a chain-like relationship of discursive sameness between different signifiers, including in forms where patterned associations between signifiers become exemplars of some Other (Salter, 2016; Tuters & Hagen, 2020). For example, let’s imagine a left critique of “mainstream media” that, like much of the literature on alternative media, is habitually linked to a critique of “capitalism.” The two signifiers become, in a sense, equivalent so that critiques of the former are routinely heard as critiques of the latter. A logic of difference, in contrast, highlights how there is nothing essential about the routinized pairing of the two signifiers, as illustrated by right-wing critiques of “mainstream media” that say next to nothing about capitalism. The key point is that examining how the signifier “mainstream media” is habitually collocated with other signifiers offers a useful way of analyzing the ideological valences of media critique today. We might conceptualize it as an exemplary “floating signifier” (Laclau & Mouffe, 2001, p. 113) because of its capacity to move easily between different discourses and do useful rhetorical work for different political and ideological constituencies. The combination of discourse theory and corpus-assisted analysis offers one example of a development in critical discourse studies (Subtirelu & Baker, 2017) that blurs conventional distinctions between “quantitative” and “qualitative” methods. The Twitter corpora described in the next section are comparatively small, at least when compared to a trend in corpus linguists where “corpora are getting ever larger, with ‘mega-corpora’ . . . having hundreds of millions of words” (Koester, 2010, p. 66). Our approach instead sides with the countertrend identified by Koester (2010): small corpus analysis. She argues that the International Journal of Communication 17(2023) Critiquing “Mainstream Media” on Twitter 4311 analysis of a specific topic in a corpus as small as 25,000 words has the comparative advantage of illuminating how language is used in contextual settings. It tempers the risk of producing computational- driven quantitative data that are disconnected from the discursive “contexts in which they make sense” and preoccupied with describing “patterns of [narrowly linguistic] usage and incidence” (Freeden, 2021). Compared with conventional corpus linguistics, our analysis does not try to follow specific subdisciplinary conventions but rather exploits the affordances of computational tools in the kind of “anarchist” fashion commended by Lindgren (2020). For these reasons, we think our method is best described as corpus-assisted interpretivist analysis, where the interpretative resources of discourse theory provide a framework for examining how mainstream media are critiqued in our Twitter corpora. Our empirical analysis should not be conflated with a strict inductive approach, where our knowledge about media critique emerges “ground up” from the data. Furthermore, we do not follow a theoretically deductive positivist approach, where the empirical material becomes a ground for testing the authority of some falsifiable hypotheses. Instead, we follow the retroductive research strategy commended by Glynos and Howarth (2007) for discourse theoretical analysis. The relationship between the theoretical and empirical dimensions of our analysis is conceptualized as a process of cyclical reasoning. We might therefore think of the article as an empirically supported theoretical discussion of the problem of ideologically nebulous media critique, rather than an empiricist endeavor to see if the problem “exists” in the first place. Twitter Sample, Corpora, and Empirical Context Our analysis is based on a sample of Twitter talk that cites the terms “mainstream media” and “MSM” over a 7-day period from January 20 to January 26, 2021.5 The sample was constructed through searches of Twitter’s public API (Application Programming Interface) using the Twitter Archiver app. We sampled six different 4-hour block periods on six different days to increase the chances of picking up original (i.e., not retweeted) tweets from different time zones. The Twitter archiver has some technical limitations as a data-gathering tool, but our sampling rationale ensured some geographical diversity, albeit within the limitations of an English-language data set.6 Our core sample was organized as two data sets of tweets— one for “mainstream media” and one for “MSM”—that filtered out retweets. The MSM data set was three times larger, containing 279,173 words7 and 8,830 tweets. The “mainstream media” sample contained 90,024 words and 2,779 tweets. The top retweets for both terms were saved separately. From one 5 The study’s datasets have been published as anonymized Tweet identification (ID) numbers at the research repository of the University of Antwerp (Phelan & Maeseele, 2023). 6 Our sample needed modification in some cases because Twitter Archiver stopped saving tweets at a certain time. For example, the “MSM” data for day 1 ended at 3.26 a.m., so day 1 begins on 23.27 on January 19 and ends on 3.26 on January 20. The final 4-hour sample is taken from January 26. Twitter Archiver missed data for the relevant 4-hour period on January 25, so we sampled the next day instead (though our retweets sample, which was constructed as a separate file, includes tweets—with a minimum of 50 retweets—for all days from January 20 to 26). These differences are the result of automated processes rather than conscious intent, so the impact on our substantive analysis is likely trivial. For each day, the search recorded tweets within the assigned 4-hour timeframe. 7 The word count figure was computationally calculated by Sketch Engine. 4312 Sean Phelan and Pieter Maeseele International Journal of Communication 17(2023) perspective, “MSM” and “mainstream media” might be seen as the same signifier, since, as used here, they both signify the same nominal entity. However, treating them as the same would miss how acronyms (Billig, 2013) like “MSM” can function as an insider code that conceivably might mean nothing at all to someone unfamiliar with the shorthand conventions of digital culture. Moreover, when transposed to another social context, MSM could signify something entirely different, as it is also a shorthand for “men who have sex with men” in public health discourses. The sample has a U.S.-centric bias, given that Joe Biden’s presidential inauguration on January 20 was the highest profile news event during the timeframe and a focus of Twitter commentary beyond the United States. In addition, many tweets referenced the COVID-19 pandemic, as well as eclectic events and topics from different national contexts. The COVID focus is especially resonant, given the affinities between COVID scepticism and the disparagement of mainstream media in reactionary cultures (Callison & Slobodian, 2021; Topinka, 2022). We chose the timeframe partly because it suited the trajectory of the research project. However, we think the core mix of U.S. cultural politics (in a context where Trump had been banned from Twitter weeks earlier) and COVID-themed content from different national contexts makes for a suitable sample for examining the dynamics of Twitter media critique. The software we used was the Web-based Sketch Engine app, which has some distinct customized features compared with other corpus analysis tools (Kilgarriff et al., 2014). Sketch Engine’s “Word Sketch” function is designed around built-in algorithms that map how particular lexical items are grammatically collocated. It automatically produces an abundance of data, exemplifying some of the general risks of using corpus software to produce quantitatively driven data in a mindless way (Freeden, 2021). For example, the app generates results for—among other options—adjective features of MSM, verbs with MSM as a subject, and nouns modified by MSM. To mitigate these risks and avoid cherry-picking data, we focused on the Word Sketch results for “and/or” conjunctions for both keywords. This meant identifying the other nominalized entities that were collocated (either three words to the left or right) using either conjunction8 for both search terms. We selected this option because of its alignment with our interest in examining how chains of equivalence are constructed between “mainstream media/MSM” and other institutions, such as “government.” We did not conduct a proper bot analysis of the data set, but we did check for obvious signs of BOT-driven distortions, such as cases where the exact same tweet was authored by different accounts within the sampled timeframe. Overview of How “MSM” and “Mainstream Media” Were Used as Part of “And/Or” Collocations The top 10 results for “and/or” collocations for “MSM”9 and “mainstream media” are presented in the three tables below. They incorporate lexical variants of the same lemma (for example, “government,” 8 The algorithm for “and/or” conjunctions also records collocations connected through commas. Thanks to Vlasta Ohlídalová at Sketch Engine for clarifying this and other points. 9 The Word Sketch algorithms recognized “MSM” (7,019 citations) and “msm” (1,058) as separate terms, which are aggregated in Table 1. International Journal of Communication 17(2023) Critiquing “Mainstream Media” on Twitter 4313 “govt,” and “gov”) in recognition of how regularly words are abbreviated on Twitter. The “Other” results for each table are grouped together as a single (and noticeably high) figure underneath the top 10. Both authors checked the table data to avoid inconsistencies. Table 1. “And/Or” collocations for “MSM/msm.” Category Total Percent. 1. Government (includes government/Gov/govt/Govt/gov/gvmnt/Gov’t/govt) 75 7.20% 2. Democrats (includes dem/Dems/democrat/Democrat/Dem/DNC)10 57 5.47% 3. Tech/tech 45 4.32% 4. Politician/politics 36 3.45% 5. Medium/media 35 3.36% 6. BBC (includes @BBC/BBBC/bbc) 33 3.17% 7. Left (includes left/Leftwaffe/Lefty/Leftist) 23 2.21% 8. CNN (includes cnn/@CNN) 16 1.54% 9. Hollywood 16 1.54% 10. Twitter 13 1.25% Other 693 66.49% Total 1,042 100,00% The Table 1 results are unsurprising given the sample timeframe and previous research. The highest result of 75 collocations (7.20%) for “government” illustrates a strong ideological association between MSM and government. Similar associations are suggested by the 36 results for “politician.” These equivalences are given a specific political referent in the results for “Democrats” (57) and “Left” (23). Other top 10 results align MSM with heterogeneous Anglo-American identities: Hollywood, CNN, Twitter, and BBC. Of the 45 collocations for “Tech,” all bar 3 refer to “Big Tech.” While 28 of the 35 collocations for “media” construct a category distinction between MSM and “social media,” suggesting an enduring tendency for MSM to primarily signify an older, journalism-centric, media infrastructure. Aggregated corpus data sets do not typically count collocations with a single entry (we did so here because of our small corpora), so this partly explains the high figure for “Other” results in Tables 1–3. The examples correlating with this category point to the protean character of MSM critique and the capacity of anti-MSM rhetoric to be ideologically paired with heterogenous identities (“scientists,” “parliament,” “war mongers,” “universities,” etc.). The results for “and/or” collocations of “mainstream media”11 in our second corpus are presented in Tables 2 and 3. Two tables were used because the software automatically created separate results for both words. This means that the results record a comparatively rare number of cases where “mainstream” and “media” are linked together by a third word, as in the phrase “mainstream Western 10 The signifier “democrat” can clearly be used differently, but these results all seemed to reference the U.S. Democratic party. 11 Because Sketch Engine prioritizes singular dictionary forms, the “mainstream media” search was conducted as a search for “mainstream medium.” 4314 Sean Phelan and Pieter Maeseele International Journal of Communication 17(2023) media” (personal communication, January 2021). The simple disparity in the size of both corpora and the comparatively lower figures in Tables 2 and 3 illustrate the extent to which the acronym is normalized on Twitter. The figures in Tables 2 and 3 count all collocations but only list individual entries that achieved a minimum frequency level of 3. Table 2. “And/Or” Collocations for “Mainstream.” Category Total Percent. 1. Social 14 20.29% 2. Western 4 5.80% 3. Liberal 4 5.80% 4. Other 4 5.80% 5. American 3 4.35% 6. Corporate 3 4.35% Other 37 53.61% Total 69 100,00% Table 2 records extended noun phrases, which identify either specific kinds of mainstream media (for example, “mainstream liberal media”) or references “other mainstream media” (personal communication, January 2021). Of the 14 collocations for the top result “social,” all bar 1 use the phrase “mainstream social media” (personal communication, January 2021). The implied alternative is usually positioned as a right-wing domain: what one tweet depicts as a refuge for Trump supporters “exiled from mainstream social media” (personal communication, January 2021). Table 3 reinforces many of the associations identified in Table 1, aside from the absence of any collocations for “government” (which records only two collocates) in the top 10. The top results for “tech” collocations are all citations of the phrase “Big Tech.” Table 3. “And/Or” Collocations for “Media.” Category Total Percent. 1. Tech/tech 15 9.15% 2. Democrats (includes Dems/DNC/democrat) 12 7.32% 3. Politician/politics 8 4.88% 4. CNN 7 4.27% 5. Hollywood 5 3.05% 6. Celebrity 3 1.83% 7. Industry 3 1.83% Other 111 67.67% Total 164 100% There is clearly only so much that can be inferred from citing such basic quantitative data, particularly given the low figures. However, further analysis produced some ideologically illuminating insights. For example, of the 75 Table 1 collocates for “government,” most link the terms using the International Journal of Communication 17(2023) Critiquing “Mainstream Media” on Twitter 4315 conjunction “and.” This is hardly surprising since our analysis specifically highlights “and/or” collocations, and the broad grammatical pattern is replicated across Table 1. Nonetheless, they point to the normalization of an everyday Twitter rhetoric where two complex sociological entities are depicted as operating according to a single “agenda” or “narrative.” Even mundane (and comparatively neutral) formulations like the government and MSM “tell us,” “prefer,” or “want” give the impression of two psychologized, anthropomorphized forces acting in concert. This is not a classic liberal fourth estate imaginary that presupposes the default separation of media and government. The comparatively high number of Table 1 collocations for Democrats and MSM reinforces the perception of two ideologically harmonious forces acting against the interests of Trump and Republicans. One tweet accuses Anthony Fauci, the World Health Organization, Democrats, and MSM of colluding to get “Trump out.” This pattern of extending the chain of equivalence to other antagonists is clearly observable in the sentence-level Table 1 results for MSM and Democrat collocates. We find direct equivalences to “Hollywood,” “China,” the “far left,” “Anti-Trumpers,” “FBI,” “leftist democrats,” “Bill Barr,” “political elites,” “Biden,” and even one sweeping equivalence to “The Repubs.” Corpus absences are also illuminating in how they attest to the normalization of right-wing media critique, even after Twitter suspended the accounts of Trump and some of his most extreme supporters. Word Sketch produces only two MSN collocations for variants of (neo)conservative and a single collocation for “right.” Derivates of the signifier “Trump” get five collocations, “Republican” receives 4, while the signifier “centrist” scores 1. The specific results attest to the reactionary valences of anti-MSM rhetoric, a universe where the “Democrats,” “MSM,” and “the left” are depicted as perfectly interchangeable entities. These banal associations displace media critiques that highlight the role played by capitalist MSM (Mondon & Winter, 2020; Pickard, 2016) in amplifying Trump’s demagoguery. MSM Critique and Moralized Suspicion This section presents a closer qualitative analysis of how the relationship between MSM and government is represented suspiciously, often in (hyper)moralized rhetoric that depicts both entities as indistinguishable enemies. Our understanding of moralized rhetoric is (as we clarify in the conclusion) informed by Chantal Mouffe’s (2005a, 2005b) critique of the moralization of politics. Mouffe (2005a, 2005b) made this argument part of a critique of neoliberalism, but we think it can be adapted to capture some of the general features of mediated politics today (Phelan, 2022). Most of this section’s illustrations are taken from the Table 1 data set. In each case, we used tweet ID numbers to check whether the original tweet was still traceable on Twitter before quoting fragments directly.12 This gave us more contextual information about the specific tweet and Twitter account (even if bios might not be trustworthy) and allowed us to better gauge how particular equivalences were being constructed. The University of Antwerp ethics committee that approved the research allowed us to directly quote tweets, but we committed to doing so in a generally anonymized and fragmentary way that minimized the risk of identifying authors (hence, most tweets are referenced as “personal communication”), except in the case of Twitter accounts with a significant public profile. 12 We still refer generically to some deleted tweets, but without quoting them directly. 4316 Sean Phelan and Pieter Maeseele International Journal of Communication 17(2023) Let us first elaborate on how MSM and government (whatever the precise government) are represented as having a perfect alignment of interests and motivations. Consider how the following examples capture a general COVID-mediated conspiratorial mood across countries: MSM and government have “turned” most people “woke,” says someone in a reply to a U.K. TalkRadio tweet (personal communication, January 2021); the questions posed by one user to a U.K. Government Twitter account are described by another as not fitting with the COVID narrative “peddled” by “GOV [government] and MSM” (personal communication, January 2021); the “Government and MSM [are] forcing” COVID “down our neck,” says someone with a bio located in Scotland (personal communication, January 2021); “YOU” have been “Groomed and Abused” by “MSM and govt,” says one person in response to another suggesting they have been duped by COVID-sceptic propaganda (personal communication, January 2021). A linked theme highlights the failure of MSM and government (again acting in concert) to publicize topics that some Twitter users regard as publicity-worthy: “Why is government and MSM not promoting this?” says one person in response to suggestions that taking Vitamin D prevents COVID (personal communication, January 2021). Another person with a bio location in Dublin laments how anyone with an “opinion” that does not side with “the government or MSM” will be labeled as an “anti vaxer” or “covid denier” (personal communication, January 2021). While the dominant impression in our sample suggests a harmonious alliance between MSM and the government, the normalization of this assumption is sometimes the target of wry commentary. This was captured in a since deleted tweet that noted the normalization of a rhetoric that treated anything said by MSM and the government as automatically untrue. However, reflective meta-commentary of this kind is rare, compared with a habitual representation of MSM as liars and propogandists: We have had 12 months of “fraud, lies and deceit” from “Government and MSM,” says an account with a bio inscription denouncing “the brainwashing” (personal communication, January 2021); “MSM and govt lies . . . again and again,” said one user in response to a tweet about COVID funeral restrictions in Canada, from an account with anticommunist imagery (personal communication, January 2021); down with “government and MSM for hijacking democracy,” says one user in response to an opinion piece about Australian government policy (personal communication, January 2021). One metaphor that became a staple of antilockdown rhetoric is the image of the mass of people as “Sheeple,” because of their pitiable tendency to uncritically believe whatever is said by MSM and government. This theme is also clearly evident: “Why” believe everything “MSM and Government are spouting?” asks an account with an antilockdown bio (personal communication, January 2021); it's “sad?” to see what MSM and the government have done to people’s “mindsets,” says one person in what is framed as genuine sympathy for another person’s tweet (personal communication, January 2021). The commonsense authority of this rhetoric is suggested by another tweeter who anticipates the charge they have been “brainwashed,” as part of a thread pleading for people to stay at home and wear masks: Some might think I have been “brainwashed” by MSM and govt, “but that not’s true” (personal communication, January 2021). International Journal of Communication 17(2023) Critiquing “Mainstream Media” on Twitter 4317 For some wanting to protect themselves from brainwashing forces, emancipation only becomes possible through the politics of withdrawing from mainstream governmental and media infrastructure. We might describe this as a subjectivity aligned with Albert Hirschman’s (1970) “exit” option for those who feel they can no longer find agency or voice in an existing structure. Some justify their stance by appealing to the notion of “critical thinking,” as in an account with an anti-left bio, which suggests that the lockdown will only continue as long as people let MSM and government “tell them” how to live their lives. Others affirm the authenticity of their own political “resistance” against what they see as the docile conformism of leftists, as in a tweet that mocked a post from the British Socialist Workers Party. It is worth citing additional corpus evidence that shows how the theme of the MSM exit is normalized. Word Sketch automatically generates a typicality score for different grammatical collocations that indicates how “strong” the collocation is. The higher the score, the stronger the collocation–meaning a high level of co-occurrence between two words in a sample. For verbs with (capitalized) “MSM” as an object, the highest typicality score (10.9) is for the verb “watch.” The two terms are collocated 86 times, but nearly always to articulate disidentification with MSM. For example, many users describe how they “do not” watch MSM or implore others to “stop watching.” This position is justified on various grounds, bringing together themes of ideological antagonism, self-care, and truth-seeking that have been illuminated in analyses of far-right politics (Finlayson, 2022). Consider the following examples: “I” have stopped watching MSM “because of liberal infection” (personal communication, January 2021); do not watch MSM if you “want your life back” (personal communication, January 2021); stop watching and “do your own research” (personal communication, January 2021); you must not care “much for facts” if you watch MSM (personal communication, January 2021); stop watching MSM and “grow some brain cells” (personal communication, January 2021); “I stopped watching” MSM five years ago (personal communication, January 2021); “Stop watching MSM and “search” for “the truth” (personal communication, January 2021); “You’ve lost the plot anyhow” if you watch MSM (personal communication, January 2021). Taken together, our examples point to the normalization of a rather dystopian representation of MSM and its relationship with the government. Many communicate an affective certainty and knowingness— as if any disposition other than blanket suspicion (Gilroy-Ware, 2020) suggests a kind of hopeless naivety and stupidity. Moralizing and reactionary rhetoric are normalized, even when it is not always clear how tweets are framed politically. MSM is constructed as one of a gallery of interchangeable antagonists, acting in concerted opposition to the desired subject position. Retweets and Ideological Differentiation We want to conclude our empirical analysis by summarizing the top 10 retweets because they illustrate clearer (however limited) political differences from the more ideologically nebulous rhetoric of the main sample. The practice of retweeting is central to the gamified design of Twitter, in that the ability to harvest retweets, likes, and followers is indicative of an account’s reach and popularity (Nguyen, 2021). Retweets offer markers of ideological resonance (McEnery et al., 2015), though not always in ways that are flattering to the original tweeter. Many of the tweets cited earlier come from accounts with comparatively low numbers of followers. Unbeknown to us, some are conceivably bot accounts, distributing cookie-cutter 4318 Sean Phelan and Pieter Maeseele International Journal of Communication 17(2023) condemnations of MSM to random targets. Conversely, the top retweets come from accounts with higher numbers of followers, meaning that they circulate more widely on the platform. The most retweeted tweet in the MSM sample (320 retweets when first downloaded) discussed Biden’s “cognitive decline,” mocking “MSM” for not reporting this during the election. Anti-Biden commentary featured in four more retweets, including those ranked second, third, and fourth. Of the remaining top 10, three were included because the acronym MSM was featured in the name of an account that engaged in a comparatively temperate critique of BBC Scotland. Two others offered conventional fourth-estate-style critiques of the capacity of MSM to hold the U.K. government to account. The non-Biden tweets were, in one sense, entirely prosaic, yet nonetheless departed from the dominant emphases identified earlier. Of the top 10 “mainstream media” retweets, five affirm support for Trump in opposition to “mainstream media.” However, the others offer the clearest articulation of progressive media critique in our sample. Three reference American politics, but from a perspective that critiques mainstream media for publicizing every Trump-related PR gimmick, symptomatic of a far-right “bias” that indulges “Trump’s base.” The most retweeted tweet (462 retweets when downloaded, but now more than 2,000) is located in India, another country that has seen the rise of far-right media fixated on the perceived “liberal bias” of mainstream media (Chadha & Bhat, 2022). It was posted by Mohammed Zubair (2021), whose bio describes him as an analyst of misinformation/disinformation across India. The tweet references the Indian farmers’ tractor protests that took place in Delhi on January 26, 2021, highlighting the complicity of the mainstream media with the government’s strategy of ignoring the protesters. It articulates a style of media critique that might be commended in any conventional media literacy course: “Mainstream media will try and avoid covering tractor protest by farmers. Let's amplify by RTing people covering the Tractor parade/March/protest. But make sure it's authentic :) #tractorParade #HistoricTractorMarch #TractorRally” (Zubair, 2021). Conclusion: Locating Media Critique in a Time of Reactionary Politics This article examined the dynamics of media critique by analyzing how the signifiers “mainstream media” and “MSM” are articulated on Twitter. We prefaced our Twitter analysis by mapping the emergence of “mainstream media” talk, emphasizing how the term’s historical embeddedness in left-wing media critique has been complicated by its articulation in far-right media critique. Our analysis illustrates how the term can move easily between discourses and contexts, often in a (hyper)moralized register that is deeply suspicious of media, yet vague in its precise ideological and political moorings. We end by reflecting on the political and normative question that has been animating our argument: How might we affirm a (radical) democratic conception of media critique in a context where reactionary critiques have been normalized alongside an equally normalized culture of progressive and left-wing media critique? Our answer takes inspiration from Mouffe’s (2005a, 2005b) argument about the moralization of politics, which we propose can be productively adapted as an argument about the moralization of critique. Mouffe (2005a, 2005b) critiques neoliberalized political discourses that reduce politics to a moral battle between right and wrong as a counterpoint to her own advocacy of radical democratic politics (Phelan, 2022). The moralization of politics normalizes a friend/enemy image of the political. The Other is depicted as irrational, extreme, or evil, in opposition to the assumed benevolence of the self-avowed identity. The International Journal of Communication 17(2023) Critiquing “Mainstream Media” on Twitter 4319 argument can only be signposted here, but to speak of a moralization of critique is simply to observe how the object of critique examined in this article, the MSM Other, is represented in a similar moralizing fashion. These tendencies are exemplified by our analysis of how the relationship between MSM and government is represented on Twitter. In its most sweeping form, the habitual representation of both identities as a single force becomes symptomatic of a moralized judgmental mood that turns the general workings of power into the workings of an undifferentiated blob: a nebulous Other (Tuters & Hagen, 2020) that annihilates the differences between one thing and another. These discourses reflect, in one respect, the normalization of a reactionary hermeneutics of suspicion: the kind of moralized (anti)politics often pejoratively named “populist,” which was exemplified by the antilockdown movements that emerged during “Coronapolitics” (Callison & Slobodian, 2021). Yet, that analysis is too one-dimensional because, as we know, people sometimes have very good reasons for suspecting how governmental and media actions, representations, and interests converge, and sometimes that suspicion assumes a clear moral(izing) register. To deny that truism would mean renouncing heterogenous traditions of left-media critique—as if left-wing publics and academics should somehow stop critiquing media now that the far right has tried to hegemonize the terrain as their own. It would also mean the eternalization of a postpolitical imaginary, where the defense of (actually existing) “mainstream media” against authoritarian forces becomes the only conceivable democratic move, a tendency evident in the depoliticizing tendencies of the “disinformation” literature (see Farkas & Schou, 2019). We would like to conclude, then, by gesturing toward the notion of radical democratic media critique as an alternative imaginary for thinking about how we might affirm the notion of media critique in a time of emboldened reactionary politics. Developing the capacity to critically assess how the world is represented in different media spaces and forms is an important source of democratic agency. Yet, when we do this today, we need to be more contextually alert (than we have been in the past) to the ideologically confusing valences of media critique, especially in a cultural atmosphere where far-right actors, centrists, and some nominal leftists like to blur assumed boundaries between different discourses and ideologies (Krzyżanowski & Ledin, 2017; Mondon & Winter, 2020). Abstractions like “the media,” “mainstream media,” and “social media” should be legitimate targets of critique for different ideological constituencies, even when the practice of abstracting always risks—as one of our reviewers put it—“flattening out important differences and nuances” that treat “all media as a monolithic entity” (personal communication, July 2022). Our argument highlights how the most vivid illustrations of that “flattening out” process are now textured by moralizing idioms—a suspicious mood that constructs “mainstream media” as an object(s) of hatred and disparagement. Maybe we might therefore think of the speculative notion of radical democratic media critique as a foil to these Manichean impulses—a mode of critique that affirms the politics of media, meaning their contingency and potential to be articulated more democratically (Chang & Glynos, 2011), rather than the repressive image of politics naturalized in reactionary discourses (Phelan, 2022). It would cultivate a media critique that might be called agonistic (Maeseele & Raeijmaekers, 2020; Mouffe, 2005a): a perspective that would encourage us to think of mainstream media as the signifier of a pluralistic universe that mediates the activities of many legitimate democratic adversaries, even when it routinely fails to be sufficiently pluralistic or democratic. At the very least, it would nurture a civic capacity to see something more sociologically complex and politically porous than a one-dimensional target of enmity and denunciations. 4320 Sean Phelan and Pieter Maeseele International Journal of Communication 17(2023) References 130 Cong. Rec. 848. (1984). Abbott, D. (1993, May 21). Racial violence. Hansard, 225. Retrieved from https://hansard.parliament.uk/Commons/1993-05-21/debates/5433da52-1580-4564-91fc- f5a5db99d919/RacialViolence?highlight=%22mainstream%20media%22#contribution-8350efdb- 7075-4804-ac26-10fbcb05ecb0 Abbott, D. (1995, December 19). Caribbean (Voluntary resettlement). Hansard, 268. Retrieved from https://hansard.parliament.uk/Commons/1995-12-19/debates/df33c33f-1f2d-4bf7-96f0- c816a4bc3b72/Caribbean(VoluntaryResettlement)?highlight=%22mainstream%20media%22#co ntribution-19c1361e-76b2-455c-8f36-ea4cec42531d Benton, S. (2021, December 2). Business of the house. Hansard, 704. Retrieved from https://hansard.parliament.uk/Commons/2021-12-02/debates/A4715ABE-60CD-4104-B1AF- 765177BD3452/BusinessOfTheHouse?highlight=%22mainstream%20media%22%20%22sick%2 0and%20tired%20of%20the%20stale%22#contribution-DAA8751B-0068-44E7-A714- 9C1F936F49D0 Billig, M. (2013). Learn to write badly: How to succeed in the social sciences. Cambridge, UK: Cambridge University Press. Broersma, M. (2018). Epilogue: Situating journalism in the digital: A plea for studying news flows, users, and materiality. In S. A. Eldridge II & B. Frankin (Eds.), The Routledge handbook of developments in digital journalism studies (pp. 515–526). London, UK: Routledge. Brown, K., & Mondon, A. (2021). Populism, the media, and the mainstreaming of the far right: The Guardian’s coverage of populism as a case study. Politics, 41(3), 279–295. doi:10.1177/0263395720955036 Callison, W., & Slobodian, Q. (2021, December 1). Coronapolitics from the Reichstag to the Capitol. Boston Review. Retrieved from https://bostonreview.net/articles/quinn-slobodian-toxic-politics- coronakspeticism/ Carlson, M., Robinson, S., & Lewis, S. C. (2021). Digital press criticism: The symbolic dimensions of Donald Trump’s assault on U.S. journalists as the “enemy of the people.” Digital Journalism, 9(6), 737–754. doi:10.1080/21670811.2020.1836981 Chadha, K., & Bhat, P. (2022). Alternative news media and critique of mainstream journalism in India: The case of OpIndia. Digital Journalism, 10(8), 1283–1301. doi:10.1080/21670811.2022.2118143 International Journal of Communication 17(2023) Critiquing “Mainstream Media” on Twitter 4321 Chang, W., & Glynos, J. (2011). Ideology and politics in the popular press: The case of the 2009 UK MPs’ expenses scandal. In L. Dahlberg & S. Phelan (Eds.), Discourse theory and critical media politics (pp. 106–127). Basingstoke, UK: Palgrave Macmillan. Cimaglio, C. (2016). “A tiny and closed fraternity of privileged men”: The Nixon-Agnew anti-media campaign and the liberal roots of the U.S. conservative “liberal media” critique. International Journal of Communication, 10, 1–19. Retrieved from https://ijoc.org/index.php/ijoc/article/view/2633/1519 Collins, D. (2021, May 27). World press freedom day. Hansard, 696. Retrieved from https://hansard.parliament.uk/Commons/2021-05-27/debates/3D70B218-15CB-4F43-AA2B- 7AF6E56B31ED/WorldPressFreedomDay?highlight=%22mainstream%20media%22%20derogator y%20term#contribution-5DB676F4-70F6-4856-BA14-E536D97F7F63 Cushion, S. (2020). Six ways alt-left media legitimatize their criticism of mainstream media: An analysis of The Canary and Evolve Politics (2015–19). Journal of Alternative & Community Media, 5(2), 153–171. doi:10.1386/joacm_00081_1 Dates, J. L., & Gandy Jr., O. H. (1985). How ideological constraints affected coverage of the Jesse Jackson campaign. Journalism Quarterly, 62(3), 595–625. doi:10.1177/107769908506200320 Dehghan, E., Bruns, A., Mitchell, P., & Moon, B. (2020). Discourse-analytical studies on social media platforms: A data-driven mixed-methods approach. In R. A. Lind (Ed.), Produsing theory in a digital world 3.0: The intersection of audiences and production in contemporary theory (vol. 3, pp. 159–177). New York, NY: Peter Lang. Donovan, J., Dreyfuss, E., & Friedberg, B. (2022). Meme wars: The untold story of the online battles upending democracy in America. New York, NY: Bloomsbury Publishing. Farkas, J., & Schou, J. (2019). Post-truth, fake news and democracy: Mapping the politics of falsehood. London, UK: Routledge. Figenschou, T. U., & Ihlebæk, K. A. (2019). Challenging journalistic authority. Journalism Studies, 20(9), 1221–1237. doi:10.1080/1461670X.2018.1500868 Finlayson, A. (2022). YouTube and political ideologies: Technology, populism and rhetorical form. Political Studies, 70(1), 62–80. doi:10.1177/0032321720934630 Freeden, M. (2021). Discourse, concepts, ideologies: Pausing for thought. Journal of Language and Politics, 20(1), 47–61. doi:10.1075/jlp.20051.fre Gilroy-Ware, M. (2020). After the fact?: The truth about fake news. London, UK: Repeater. 4322 Sean Phelan and Pieter Maeseele International Journal of Communication 17(2023) Gitlin. (1992). Whole world is watching: Mass media in the making and unmaking of the new left. Oakland: University of California Press. Glynos, J., & Howarth, D. (2007). Logics of critical explanation in social and political theory. London, UK: Routledge. Greenberg, D. (2008). The idea of “the liberal media” and its roots in the civil rights movement. The Sixties, 1(2), 167–186. doi:10.1080/17541320802457111 Guillory, J. (2010). Genesis of the media concept. Critical Inquiry, 36(2), 321–362. doi:10.1086/648528 Hájek, R., & Carpentier, N. (2015). Alternative mainstream media in the Czech Republic: Beyond the dichotomy of alterative and mainstream media. Continuum, 29(3), 365–382. doi:10.1080/10304312.2014.986061 Heath, C. W. (1988). Government—media—audiences: Private sector participation in public service broadcasting: The case of Kenya. Journal of Communication, 38(3), 96–107. doi:10.1111/j.1460- 2466.1988.tb02062.x Herman, E., & Chomsky, N. (2002). Manufacturing consent: The political economy of the mass media (2nd ed.). New York, NY: Pantheon. Hirschman, A. O. (1970). Exit, voice, and loyalty: Responses to decline in firms, organizations, and states. Cambridge, MA: Harvard University Press. Hlavajova, M. (2020). Foreword. In M. Hlavajova & S. Lütticken (Eds.), Deserting from the culture wars (pp. 9–21). Cambridge, MA: MIT Press. Holt, K. (2018). Alternative media and the notion of anti-systemness: Towards an analytical framework. Media and Communication, 6(4), 49–57. doi:10.17645/mac.v6i4.1467 Hunter, J. D. (1991). Culture wars: The struggle to control the family, art, education, law, and politics in America (Rep. ed.). New York, NY: Basic Books. Ihlebæk, K. A., Figenschou, T. U., Eldridge, S. A., Frischlich, L., Cushion, S., & Holt, K. (2022). Understanding alternative news media and its contribution to diversity. Digital Journalism, 10(8), 1267–1282. doi:10.1080/21670811.2022.2134165 Jackson, S. J., & Foucault Welles, B. (2015). Hijacking #MYNYPD: Social media dissent and networked counterpublics. Journal of Communication, 65(6), 932–952. doi:10.1111/jcom.12185 International Journal of Communication 17(2023) Critiquing “Mainstream Media” on Twitter 4323 Jacobs, T. (2020). The post-Marxist political economy of EU trade. A discourse-theoretical analysis of the construction of political agency in the European Parliament (SSRN Scholarly Paper ID 3620061). Social Science Research Network. Retrieved from https://doi.org/10.2139/ssrn.3620061 Jansen, S. C. (1982). Bias of simple arithmetic [Review of Small voices and great trumpets: Minorities and the media by B. Rubin]. Journal of Communication, 32(1), 227. doi:10.1111/j.1460- 2466.1982.tb00490.x Kilgarriff, A., Baisa, V., Jakubíček, M., Kovář, V., Michelfeit, V., Rychlý, P., & Suchomel, V. (2014). The Sketch Engine: Ten years on. Lexicography, 1(1), 7–36. doi:10.1007/s40607-014-0009-9 Knops, L., & De Cleen, B. (2019). The radical right versus the media: From media critique to claims of (mis)representation. Politics and Governance, 7(3), 165–178. doi:10.17645/pag.v7i3.2131 Koester, A. (2010). Building small specialised corpora. In A. O’Keefe & M. McCarthy (Eds.), The Routledge handbook of corpus linguistics (pp. 66–79). London, UK: Routledge. Krzyżanowski, M. (2020). Normalization and the discursive construction of “new” norms and “ new ” normality: Discourse in the paradoxes of populism and neoliberalism. Social Semiotics, 30(4), 431–448. doi:10.1080/10350330.2020.1766193 Krzyżanowski, M., & Ledin, P. (2017). Uncivility on the web: Populism in/and the borderline discourses of exclusion. Journal of Language and Politics, 16(4), 566–581. doi:10.1075/jlp.17028.krz Laclau, E., & Mouffe, C. (2001). Hegemony and socialist strategy: Towards a radical democratic politics (2nd ed.). London, UK: Verso. Lane, J. B. (2019). Cultivating distrust of the mainstream media: Propagandists for a liberal machine and the American establishment. In A. Nadler & A. J. Bauer (Eds.), News on the right (pp. 157–173). Oxford, UK: Oxford University Press. Lindgren, S. (2020). Data theory: Interpretive sociology and computational methods. Cambridge, UK: Polity. Maeseele, P., & Raeijmaekers, D. (2020). Nothing on the news but the establishment blues? Toward a framework of depoliticization and agonistic media pluralism. Journalism, 21(11), 1593–1610. doi:10.1177/1464884917739476 Mainstream. (2000). In Oxford English dictionary (3rd ed.). McEnery, T., McGlashan, M., & Love, R. (2015). Press and social media reaction to ideologically inspired murder: The case of Lee Rigby. Discourse & Communication, 9(2), 237–259. doi:10.1177/1750481314568545 4324 Sean Phelan and Pieter Maeseele International Journal of Communication 17(2023) McKerns, J. P., & Delahaye, A. N. (1980). Articles on mass communication in U.S and foreign journals: A selected biography. Journalism & Mass Communication Quarterly, 57(1), 184–196. doi:10.1177/107769908005700130 Mondon, A., & Winter, A. (2020). Reactionary democracy: How racism and the populist far right became mainstream. London, UK: Verso. Mouffe, C. (2005a). On the political. London, UK: Routledge. Mouffe, C. (2005b). The “end of politics” and the challenge of right-wing populism. In F. Panizza (Ed.), Populism and the mirror of democracy (pp. 50–71). London, UK: Verso. Nadler, A., & Bauer, A. J. (Eds.). (2019). News on the right: Studying conservative news cultures. Oxford, UK: Oxford University Press. Nadler, A., Taussig, D., Natacha, Y., & Wenzel, A. (2021). Unmasking polarization: How conservatives make sense of Covid19 coverage (TOW Report). Tow Centre for Digital Journalism. Retrieved from https://www.cjr.org/tow_center_reports/polarization-covid-conservative.php/ Nguyen, C. T. (2021). How twitter gamifies communication. In J. Lackey (Ed.), Applied epistemology (pp. 410–436). Oxford, UK: Oxford University Press. Peck, R. (2019). Fox populism: Branding conservatism as working class. Cambridge, UK: Cambridge University Press. Peraino, L. (Producer), & Gerard, J. (Director). (1972). Deep throat [Motion picture]. New York, NY: Bryanston Distributing Company. Phelan, S. (2021, March 2). Media critique in a very online world. OpenDemocracy. Retrieved from https://www.opendemocracy.net/en/can-europe-make-it/media-critique-very-online-world/ Phelan, S. (2022). Friends, enemies, and agonists: Politics, morality and media in the COVID-19 conjuncture. Discourse & Society, 33(6), 744–757. doi:10.1177/09579265221095408 Phelan, S., & Maeseele, P. (2023). Tweet ID dataset for Phelan, S. & Maeseele, P. (2023). Critiquing “mainstream media” on Twitter: Between moralized suspicion and democratic possibility. International Journal of Communication [Data set]. University of Antwerp. Retrieved from https://repository.uantwerpen.be/docstore/d:irua:17449 Pickard, V. (2016). Media failures in the age of Trump. The Political Economy of Communication, 4(2), 118–122. Retrieved from http://polecom.org/index.php/polecom/article/view/74/264 International Journal of Communication 17(2023) Critiquing “Mainstream Media” on Twitter 4325 Roberts, J., & Wahl-Jorgensen, K. (2020). Breitbart’s attacks on mainstream media: Victories, victimhood, and vilification. In M. Boler & E. Davis (Eds.), Affective politics of digital media (pp. 170–185). London, UK: Routledge. Rutz, D. (2022, November 11). Journalists erupt on Elon Musk for saying “media elite” preventing Twitter from “elevating citizen journalism.” Fox News. Retrieved from https://www.foxnews.com/media/journalists-erupt-elon-musk-saying-media-elite-preventing- twitter-elevating-citizen-journalism Salter, L. (2016). Populism as a fantasmatic rupture in the post-political order: Integrating Laclau with Glynos and Stavrakakis. Kōtuitui: New Zealand Journal of Social Sciences Online, 11(2), 116– 132. doi:10.1080/1177083X.2015.1132749 Shechtman, A. (2021). Command of media’s metaphors. Critical Inquiry, 47(4), 644–674. doi:10.1086/714512 Subtirelu, N. C., & Baker, P. (2017). Corpus-based approaches. In J. Flowerdew & J. E. Richardson (Eds.), The Routledge handbook of critical discourse studies (pp. 106–119). London, UK: Routledge. Topinka, R. (2022). “Post-pandemic” politics & reactionary digital cultures: How Covid scepticism fuels extremism online. Birkbeck Centre for Media and Democracy. Retrieved from https://reactionarypandemicpolitics.wordpress.com/ Tuters, M., & Hagen, S. (2020). (((They))) rule: Memetic antagonism and nebulous othering on 4chan. New Media & Society, 22(12), 2218–2237. doi:10.1177/1461444819888746 Wyatt, W. N. (2018). Press criticism. In T. P. Vos & F. Hanusch (Eds.), The international encyclopedia of journalism studies (pp. 1–9). Hoboken, NJ: Wiley Blackwell. doi:10.1002/9781118841570.iejs0104 Zappavigna, M. (2012). Discourse of Twitter and social media. New York, NY: Bloomsbury Academic. Zubair, M. [zoo_bear]. (2021, January 26). Mainstream media will try and avoid covering tractor protest by farmers. Let's amplify by RTing people covering the Tractor parade/March/protest. But make sure it's authentic :) #tractorParade #HistoricTractorMarch #TractorRally [Tweet]. Twitter. Retrieved from https://twitter.com/zoo_bear/status/1353937776522129409