Framing the U.S. news media as a threat: President Donald Trump’s securitising move

While securitisation theory sees that the news media can be both a securitising actor and audience, the existing literature has not explored cases where the media are framed as the threat. This article illustrates how former President Donald Trump framed the U.S. news media as a threat, and argues that this constitutes a securitising move. Discourse analysis informed by framing theory is applied to a sample of tweets posted by Trump. These examples illustrate how Trump has continuously framed the media as a threat, highlighting the referent object as being a threat to the truth, the people and the country, thereby undertaking a securitising move.


Introduction
Former U.S. President Donald Trump consistently utters falsehoods and regularly challenges the legitimacy of the news media in a way that is 'unprecedented in the modern era' (Guess, Nyhan and Reifler, 2017, p. 11) in a democratic state. Trump has consistently responded to criticism and unfavourable reporting of him as 'fake news' (Kellner, 2019, p. 55). Additionally, Trump dubbed the majority of the U.S. news media as 'the fake news media', labelling them as untrustworthy and as the 'enemy of the American people' among other things. In 2019, Reporters Without Borders' (RSF) World Freedom Index classified the media climate for reporters in the United States as 'problematic' (Reporters without Borders, 2019a), citing Trump's rhetoric as part of the cause. Through his rhetoric, Trump capitalised on a longheld perception of a liberal media bias in U.S. news amongst Republican party members and supporters (Guess, Nyhan and Reifler, 2018, p. 4). Moreover, Trump capitalised on -and has exacerbated -the regime of post-truth and the considerable media distrust in the United States. Considering the above, can Trump's language be seen as securitising?
The Copenhagen School (CS) of securitisation argues that a securitising actor can make a 'securitising move' (Buzan et al., 1998, p. 25) by presenting an issue as an existential security threat in its rhetoric to an audience. Security actors were initially perceived to be elite actors, such as leaders or persons of authority in society (Waever, 1995, p. 54); however, second-generation securitisation scholars have expanded upon the initial conceptualisation regarding who may constitute a securitising actor. For example, the media is a securitising actor (Vultee, 2010a;Mortensgaard, 2018;Watson, 2009) when it makes its own securitising framing choices on matters (Mortensgaard, 2018, p. 10). It is important to understand that the media can operate as a securitising actor because the more 'authority or credibility a medium has, the more likely it is to become a successful securitising actor' (Hass, 2009, p. 84). Thus, the media is 'an authority in itself' (Mortensgaard, 2018, p. 10). Ideally, the media are also able to work as a 'watch dog' (Hass, 2009, p. 84) for abuses of power conducted through the guise of extraordinary measures enabled by securitisation (Hass, 2009, p. 84). This latter point highlights the media's importance as the so-called fourth estate, with its implicit ability to check power (Mortensgaard, 2018, p. 10) -an ideal role that is tied to a well-functioning democracy (Hass, 2009, p. 78). Further, the understanding of audience has likewise been expanded from a narrow definition of the citizens of a state, to encompass other (Vuori, 2008, p. 72;Sperling and Webber, 2016, p. 21;Salter, 2008, p. 334), often multiple audiences (Roe, 2008, p. 624;Salter, 2008, p. 334;Côté, 2016, p. 547). For example, the media have been assigned the role of audience within securitisation theory (Vultee, 2010a;Mortensgaard, 2018). The media act as audience, for example, when they further a government's securitising move through the adoption of a governmental framing (Mortensgaard, 2018, p. 10). Despite, however, a widening in the understanding that the media can speak security and be a securitising actor, as well as the audience, securitisation theory has not considered whether the media can be designated as the threat -a notable gap in the securitisation literature.
As such, the initial research question that emerged was: can the media be designated as the threat in accordance with securitisation theory? Moreover, in light of Trump's discourse concerning the news media, this article became guided by the following question: can Trump's discourse concerning the news media be considered a securitising move? A democracy relies on a 'free and independent media sector that can keep the population informed and hold leaders to account' (Repucci, 2019). As Repucci (2019) argues, this is 'as crucial for a strong and sustainable democracy as free and fair elections'. Trusted news media are vital and, when they are absent, 'citizens cannot make informed decisions about how they are ruled, and abuse of power, which is all but inevitable in any society, cannot be exposed and corrected' (Repucci, 2019, p. 5). If the news media become a threat and are not seen as a reliable source on which the public can depend, they are unable to serve their vital democratic role. Additionally, Huysmans (2014) argues that securitisation poses a challenge to democracy, as '[e]nacting relations to others and one's environment as always dangerous, fearful and inimical translates into a politics that limits and hollows out democratic organisations based on principles of freedom and justice' (p. 3). As such, a securitising move against the media is worth studying, as both the action of securitising, and the possible securitising of the news media in particular, pose grave challenges for -and undermine the legitimacy of -liberal democracy and its institutions.

Securitisation and framing theory
This article illustrates former President Donald Trump's framing of the U.S. news media as a threat and as a securitising move. Thus, this literature review delves into securitisation theory, framing theory, the literature on the media in relation to securitisation and framing, and the literature on President Trump in relation to securitisation. This review informs the investigation and the analysis. Importantly, it highlights a clear gap in the existing literature, which outlines how the news media can be the audience of securitisation, speak security or be a securitising actor. It has not explored, however, the case where the news media are framed as the threat, as that which is being securitised, and it is this gap that this article seeks to fill.

Securitisation theory
Securitisation theory was introduced by the CS (Buzan et al., 1998) and has been re-evaluated by a second generation of securitisation scholars who have critiqued, modified and expanded upon the original theory. This article will mainly apply second-generation theory; however, it will begin by reviewing CS theory to provide a comprehensive outline of its original tenets and a foundation for the later conceptualisations.
The CS asserts that securitisation is a form of extreme politicisation in which an 'issue is presented as an existential threat, requiring emergency measures and justifying actions outside the normal bounds of political procedure' (Buzan et al., 1998, pp. 23-24) in order to deal with it. In so doing, the issue becomes prioritised over other issues (Buzan et al., 1998, p. 24). Further, the CS argues that security can occur in the environmental, economic, societal and political sectors (pp. 22-23), thereby extricating the understanding of security from the mere military sector held by traditional security studies (pp. 2-3). As Mortensgaard (2018) recounts, the CS infers that 'most issues can be securitised, that is, come to be perceived as an existential threat to someone/something' (italics in original, p. 5).
In understanding the process of securitisation as conceptualised by the CS, it is necessary to understand the different elements that make up securitisation. In order to initiate a securitisation process, a securitising actor makes a securitising move. The CS defines the securitising actor as a person in a position of authority, such as belonging to a political elite (Buzan et al., 1998, p. 33). The securitising actor attempts a securitising move with the goal of convincing an audience that something, a referent object, is existentially threatened and requires protection and emergency measures to ensure its survival (Buzan et al., 1998, p. 25). Buzan et al. (1998) argue that if the audience does not accept the actor's securitising move, then the move fails (p. 25). If the audience, however, accepts the securitising move, the move is successful, and the securitising actor has the permission to deal with the issue and to forego 'procedures or rules he or she would otherwise be bound by' (Buzan et al., 1998, p. 25). A sign of securitisation, for example, can be 'placing limitations on otherwise inviolable rights' (Watson, 2009, p. 24). The CS highlights that 'security should be seen as negative, as a failure to deal with issues as normal politics' (Buzan et al., 1998, p. 29).
The CS sees a securitising move as an illocutionary speech act and argues that 'the utterance itself is the act' (Waever, 1995, p. 55). The speech act logic makes it possible for the CS to rationalise that the 'utterance of "security" is more than just saying or describing something' (Stritzel, 2014, p. 21), and allows the CS not only to understand security and threat construction through 'the traditional threat-reality nexus' (Stritzel, 2014, p. 22) but to study how speech can construct threats. The CS further argues that those in power may use securitisation 'to gain control over' (Waever, 1995, p. 54) that which is securitised, and that 'something is a security problem when the elites declare it to be so' (p. 54). After a securitising move is made, in order for it to be successful, the audience must accept the designation of the threat. As such, securitisation is also an intersubjective process according to the CS (Buzan et al., 1998, p. 31).

Second-generation debates and developments
One of the main critiques levelled at the CS's conceptualisation of securitisation is that the school relies on the illocutionary act while it simultaneously argues that securitisation is an 'intersubjective process' (Côté, 2016, p. 542). The CS's dual explanation is viewed as a contradiction. Waever argues that three illocutionary felicity conditions are necessary for securitisation to be successful, albeit they do not guarantee success: '(1) the grammar or plot of security, (2) the social capital of the enunciator, and (3) conditions related to the threat' (cited in Vuori, 2008, p. 70). Balzacq (2005) instead argues for a 'pragmatic act ' (p. 178) or, as put by Vuori (2008), a fourth felicity condition: 'conditions related to the audience of securitization' (p. 70). Instead of focusing on the illocutionary speech act, Balzacq (2005) argues for a 'pragmatic act' (p. 178), which allows for the illocutionary force of the speech act while taking into consideration the perlocutionary effects in the audience. This way, the theory becomes 'audience-centered' (p. 171) because the audience is, through perlocutionary effects, granted a larger role in determining the success of the securitisation process and is not merely an actor that either accepts or rejects a securitising move.
Following from this, Balzacq (2005) argues that the audience, the context and securitising agent are 'congruent' (p. 174) factors, which condition the 'effectuation of securitization ' (p. 192). The audience, according to Balzacq, has three components: the 'audience's frame of reference', 'its readiness to be convinced, which depends on whether it perceives the securitising actor as knowing the issue and as trustworthy', and 'its ability to grant or deny a formal mandate to public officials ' (p. 192). Balzacq also outlines contextual factors that impact the receptiveness of the audience to arguments made by the securitising actor, among them the arguments' relevance within the zeitgeist (p. 192). The '[z]eitgeist can be rooted in collective memory, it is mostly constituted by the predominant social views, trends, ideological and political attitudes that pervade the context in which participants are nested' (Balzacq, 2005, p. 186). Finally, the securitising agent component includes 'the capacity of the securitising actor to use appropriate words and cogent frames of reference in a given context' (Balzacq, 2005, p. 192) to resonate with the audience (p. 179) and secure its support (p. 192). Here, for example, the actor may invoke 'metaphors, emotions, stereotypes, gestures, silence, and even lies' (p. 172). Securitisation is, thus, dependent on all of these factors, and focusing merely on speech acts or one factor alone will not capture accurately the process of securitisation. Thus, Vuori argues, '[b]oth linguistic and social felicity is necessary to achieve successful securitization' (p. 94), but success is not guaranteed. According to Balzacq, studying the congruence among these factors allows us to determine what is 'driving and/or constraining securitization ' (p. 192).
Several other scholars critique the CS's illocutionary speech act reliance. Stritzel (2007) argues that we more often experience 'a process of articulations' (italics in original, p. 377) rather than one speech act and, therefore, seeing the speech act as an illocutionary act will 'rarely explain the entire social process' (p. 377) that leads to securitisation. In a similar vein, McDonald (2008) argues that the CS sees 'the moment of intervention only' (p. 564) and fails to consider that security can be 'constructed over time through a range of incremental processes and representations', as well as 'why particular representations resonate with relevant constituencies'. Additionally, Oren and Solomon (2015) argue that '[t]he more constantly and frequently a securitising phrase is being repeated, that is, the more condensed the "historicity" of the phrase becomes, the more likely is the phrase to acquire an illocutionary force, to construct the security threat it ostensibly describes' (p. 320). Thus, Oren and Solomon (2015) draw on Butler's view of 'condensed historicity' (p. 319) arguing that past, present and future invocations of a phrase blend together to represent a threat (p. 324). The idea of 'condensed historicity' (p. 320), they argue, 'makes it possible to preserve an understanding of the act as illocutionary even as, or indeed precisely because, we incorporate temporality into our theoretical account' (p. 320). Moreover, the authors argue that 'simplified phrases' that are repeated by securitising actors are worth analysing (p. 320). When an actor continuously repeats ambiguous phrases, they may become repeated by an/ several audience(s); thus, the audience joins in on a 'ritualised chanting of the securitising phrase' (Oren and Solomon 2015, p. 313). Thereby, the audience is active (Oren and Solomon, 2015, p. 313) and contributes to the securitisation process regardless of whether its repetition of the phrase is in support or in opposition (p. 325). Active audience participation in chanting, thus, is a way of determining audience acceptance (p. 313) and successful securitisation (Oren and Solomon, 2015, p. 322). As such, upon reviewing the debate, it is clear that the CS's focus on the illocutionary act is insufficient when studying securitisation. Instead, it seems necessary, in order to analyse securitisation accurately, to evaluate it as a process with several contributing factors.
Another issue with the CS's original framework is its failure to define the audience in securitisation expansively (Balzacq, 2005, p. 173). Within second-generation scholarship, the audience has gone from being an underspecified homogeneous block, to several different types of audience, to multiple audiences (Côté, 2016, p. 547). For example, the audience can be 'fundamentalists' (Vuori, 2008, p. 72), 'the power elite', member states in an international organisation such as NATO (Sperling and Webber, 2016, p. 21), a single person such as a minister, or a government (Salter, 2008, p. 334) to name but a few. The media have also been designated the role of audience in securitisation (Vultee, 2010a;Mortensgaard, 2018). The media act as audience, for example, when they further a government's securitising move through the adoption of its framing, as this indicates media acceptance of that frame (Mortensgaard, 2018, p. 10). Essentially, as Vuori (2008) argues, 'audiences are dependent on the socio-historical situation: who has to be convinced of the necessity of security action changes with the cultural and political systems in which the securitization is taking place' (p. 72). Notably, Vultee (2010b) underlines that one can differentiate between audiences by arguing that securitisation 'only requires the consent of enough of the audience' (p. 84) and not the entire audience. O'Reilly (2008) conceptualises this as a 'critical mass': securitisation is successful when it obtains the consent of 'enough' (p. 67) of the 'majority of the target group'. Moreover, multiple audiences can exist and simultaneously require persuasion on an issue, such as the public, a political party and parliament (Roe, 2008, p. 624). Côté (2016) argues that 'multiple audiences may exist within a single securitization process, and that audiences often possess differential powers and influence, leading to differing effects on securitization outcomes' (p. 547).
As noted above, the CS frames securitisation as increasing the ability to enact emergency measures and exceptionalist policies. The Paris School (PS), however, places more focus on the 'reproduction of unease and insecurity in everyday life' (Nyman, 2018, p. 107) and while security may involve speech acts, it is not limited to them. Huysmans (2011) highlights that, apart from the speech act, 'little security nothings' (p. 371) can create insecurities, such as 'non-exceptional routine procedures that together add up to securitization, but who inadvertently reintroduce exceptional acts' (Floyd, 2016, p. 683). Huysmans (2014) also distinguishes between exceptional securitising and 'diffuse securitising' (p. 10), the latter of which involves 'dispersing and non-intensifying modes' of securitisation. Bigo (2002) further asserts that issues can become securitised as a 'result of the creation of a continuum of threats and general unease in which many different actors exchange their fears and beliefs' (p. 63). Furthermore, Côté (2016) notes that 'acceptance of a securitising move is, itself, a security action' (p. 544), and the requirement of policy implementation for successful securitisation is too high of a 'threshold' (p. 544). Additionally, exceptional measures are not always how securitising actors respond when dealing with a security threat (Floyd, 2016, p. 678), and a threat does not necessarily need to be existential for it to be securitised (p. 691). It is important to note that since the CS's conceptualisation, second-generation scholars have continued to explore and expand on what may be designated as a security threat in securitisation. For example, non-state actors (Stritzel and Chang, 2015), religion (Cesari, 2009) and (im)migrants (Huysmans, 2000;Buonfino, 2004;Ibrahim, 2005;Ceyhan and Tsoukala, 2002) to name but a few, have been framed as threats. Securitisation scholarship, however, has not explored the case where the media are designated as the threat.
Another significant insight within second-generation scholarship is Huysmans ' (2014) argument that securitisation poses a challenge to democracy, and its institutions and principles (Huysmans, 2014, p. 3). While Huysmans does not particularly discuss the securitisation of the media, this conceptualisation is important to the later investigation, as a free news media is often viewed as 'the fourth estate' (Kellner, 2018, p. 97) in democracies, and because 'Donald Trump's unceasing denunciation of the news media is without historical precedent' (Hyvönen, 2018, p. 126). Huysmans (2014) further underlines that the practice of security creates 'enemies and fear' and is not a practice of responding to them (p. 3). Moreover, Huysmans (2014) draws on Neumann, and notes that 'democracy and politics of fear do not go well together because the latter invited secrecy or emergency laws while democracy demands transparency and due process' (p. 3). As such, 'we should ask if insecurity is becoming the energetic principle of politics. If so, for anyone invested in democratic politics alarm bells should go off' (Huysmans, 2014, p. 5). Huysmans clearly lays out that liberal democracy and securitisation are, in essence, incompatible concepts. Cacciatore et al. (2016, p. 9) argue that framing as a theory has its origins in the fields of psychology (see Kahneman and Tversky, 1985) and sociology (see Goffman, 1974;Gitlin, 1980;Gamson and Modigliani, 1987). Within psychology, framing studies 'how a piece of information is presented to audiences' (Cacciatore, 2016, p. 10), while sociology sees 'framing as a means of understanding how people construct meaning and make sense of the everyday world' (p. 10). Due to the different understandings of framing, however, there is 'ambiguity around the concept' (Cacciatore et al., 2016, p. 8). Pan and Kosicki (1993, p. 57) argue that the different understandings of framing underline that frames can be, as Kinder and Sanders (1990) originally put it, 'internal structures of the mind' (Pan and Kosicki, 1993, p. 74), as well as 'devices embedded in political discourse' (p. 74). In a related understanding, Chong and Druckman (2007a) differentiate between 'a frame in communication' (p. 100), which relates to the words or phrases that a speaker uses, and a 'frame in thought' (p. 101), namely how a member of the audience understands a situation (p. 101). This section reviews scholarship on political and media framing, the function of schemas in framing, framing using repetition and hyperbole, as well as the link between securitisation and framing, due to the literature's applicability to the case studied. This overview allows for a greater methodological sensitivity to the data in the analysis.

Framing theory
Politicians frame an issue in discourse to 'define what a given issue is about' (Lakoff and Wehling, 2016, p. 87), decide what is up for debate (p. 87), choose how the issue should be thought about, and may propose solutions (Chong and Druckman, 2007a, p. 136). Politicians decide what words make up the debate, thereby framing an issue from the outset. In a similar understanding, Entman (1993) argues that 'to frame is to select some aspects of a perceived reality and make them more salient in a communicating text, in such a way as to promote a particular problem definition, causal interpretation, moral evaluation and/or treatment recommendation for the item described' (p. 52). 'Elite frames', those put forth by a person of authority, can effectively target 'the partisan and ideological leanings of the audience' (Chong and Druckman, 2007b, p. 112). Politicians may highlight certain features of a policy which will resonate with people by linking them to certain values or opinions to gain support for a policy (Chong and Druckman, 2007a, p. 106). Compelling frames 'can be built around exaggerations and outright lies playing on the fears and prejudices of the public. Strong frames […] often rest on […] links to partisanship and ideology' (Chong and Druckman, 2007b, p. 111). Frames presented by those seen to be 'credible sources' (Chong and Druckman, 2007b, p. 112) and those that 'invoke longstanding cultural values' (p. 112) have a greater prospect for changing opinions. According to Sniderman and Bullock (2004, p. 347), an audience exposed to different framings will choose the framing that resonates with their values, and are less receptive to frames that contradict these. Those that are more informed about an issue, however, are less likely to be influenced by framing (Chong and Druckman, 2007b, p. 121). Political framing is often viewed as negative, and as allowing 'political elites [to] manipulate popular preferences to serve their own interests' (Chong and Druckman, 2007b, p. 120). As such, like in securitisation theory, a person with authority can be persuasive as a result of their position, and is especially so when they tap into the ideological leanings and concerns of the audience.
The news media is another often-cited actor which engages in framing. Framing is a way of 'constructing and processing news discourse' (Pan and Kosicki, 1993, p. 57). The media thus make decisions regarding what to highlight, stress or omit (Pan and Kosicki, 1993, p. 57), and can 'assign blame, encourage empathy or simply ignore certain aspects of an event' (Mortensgaard, 2018, p. 10). As a result, media frames shape how the public understands issues (Steensland, 2008, p. 4). Moreover, according to Entman (2004), 'the words and images that make up the frame can be distinguished from the rest of the news by their capacity to stimulate support or opposition to sides in a political conflict. We can measure this capacity by cultural resonance and magnitude' (italics in original, p. 6). Culturally resonant terms, such as those that are most 'noticeable, understandable, memorable and emotionally charged' (italics in original, p. 6) are most influential. Magnitude refers to repetition, which when combined with resonance, increases the prospect that the frame will arouse 'similar thoughts and feelings in large portions of the audience' (p. 6). While Entman refers to news framing in this instance, these concepts can be extended and applied to politicians, thereby making them relevant to this study.
Furthermore, according to Chong and Druckman (2007b), the impact and strength of a frame is dependent upon various factors such as 'the strength and repetition of the frame, the competitive environment, and individual motivations' (p. 111). Words evoke frames, and when these are repeated, they 'become common sense' (Lakoff and Wehling, 2016, p. 84) and 'right and real ' (p. 84). This highlights how repetition can be a significant mechanism that can be utilised by a securitising actor, which, as already noted, is what Oren and Solomon (2015) contend. Further, Burgers et al. (2016) argue that the use of hyperbole is an influential persuasion device and a form of figurative framing (p. 412). Hyperboles can be used 'to exaggerate the threat' (Burgers et al., 2016, p. 415) and can thus 'be a powerful rhetorical tool in persuading the public of the existence, importance and imminence of a certain threat' (p. 415). Moreover, when 'repeated often, the exaggerated topic gets a place in the public debate that is different than when the topic is described without hyperbole' (Burgers et al., 2016, p. 415).
Further, Entman's (2004) conceptualisation of schemas provides further insight into frame evocation. According to Entman (2004), schemas function as 'interpretive processes that occur in the human mind' (p. 7). Mortensgaard (2018) argues that a reader has schemas relating to specific terms, which when employed, 'bring to mind people, concepts and valuejudgements' (p. 9). These schemas exist in 'knowledge networks' (Entman, 2004, p. 7) linked to emotions and feelings. They are 'stored in long-term memory' (Entman, 2004, p. 7) and certain terms will bring to mind these emotions and feelings. As such, frames developed at an early stage become dominant (Mortensgaard, 2018, p. 9), and 'future mentions' of an issue activate the 'preestablished schema' (p. 9). Thus, an audience's '[f]irst impressions may be difficult to dislodge' (Entman, 2004, p. 7), while contesting frames are more likely to fail (p. 9). In a similar vein, Lakoff and Wehling (2016, p. 84) argue that individuals will ignore new facts that do not correspond with the frames they already have about an issue (Lakoff and Wehling, 2016, p. 84). Additionally, an audience's behaviour and responses to issues are dependent on their schemas and frames (Mortensgaard, 2018, p. 9), which highlights the potential power of framing actors.

Framing and securitisation
Framing theory has been used in conjunction with securitisation theory by scholars such as Mortensgaard (2018), Watson (2012) and Vultee (2010b). As seen from the above, framing theory provides a way to analyse discourse. It is also linked to securitisation, highlighting its significance to this article's aims. This assertion is supported by Watson, who argues that 'securitisation should be viewed as a subfield of framing' (Watson, 2012, p. 279). Similarly, Vultee (2010b) argues that securitisation is a 'form of framing that highlights the existential threat of an issue' (p. 79). Frames define a problem, underline how it can be dealt with and decide 'which actors are protagonists and antagonists' (ibid.).

News media as the threat
Having reviewed the above, it is clear that the media engage in framing but what is their role in securitisation? The media have been attributed various roles in securitisation by different scholars. The CS does not see the media as a securitising actor and would rather label it as a functional one because the media 'popularize the security discourse' (Karyotis and Patrikios, 2010, p. 44). By this logic, the media are not in a position of authority in the same way as security actors are but can have a meaningful influence on security decisions. The media are, thus, a 'mouthpiece for governing elites' (Watson, 2009, p. 21). O'Reilly (2008) builds on the CS's view and conceptualises the media as a 'major functional actor' (emphasis added, p. 66) that can play a pivotal role in securitisation. Further, Vultee (2010a) examines the role of the media in speaking security through framing, and argues that the media play an important role in 'amplifying or moderating a security frame' (p. 34), and that media frames are 'the lens through which the public sees an issue' (p. 33). The scholar also argues that 'in addition to (or instead of) being an independent check on centers of authority, journalism is itself a center of authority' (Vultee, 2010b, p. 82) and can decide who speaks security (p. 83).
Securitization, in short, is conceptualized as a news frame that cues several results. When the right actor invokes the right threats under the right conditions to the right audience, the results should reflect a greater willingness to place authority, as well as civil liberties, in the hands of the government for the duration (Vultee, 2010b, p. 84).
Still, however, 'not all media securitize all things equally' (Vultee, 2010b, p. 83). They can adopt different frames, and different news media can have different impacts (Mortensgaard, 2018, p. 11). Further, drawing on Vultee, Mortensgaard (2018) argues that the media function dually as both securitising audience and actor (p. 10). The former role materialises when the media further a government's securitising move by adopting the government's frame, and the latter occurs when the media make their own securitising framing choices. As Mortensgaard (2018) states, 'the media does not necessarily accept and replicate a securitizing move uncritically' (p. 10). The media may also, ideally, function as a check on abuses of power conducted as a result of emergency measures legitimised through securitisation (Hass, 2009, p. 78). Further, Watson (2009) argues that the media can take on different functions within a securitisation, arguing that the media 'communicates the securitizing claims of other actors, […] can make securitizing claims of its own and […] can expose securitizing claims to contesting views. In most cases it serves all three purposes' (p. 22). As such, the media have been categorised as a (major) functional actor, securitising actor and audience. The literature, however, has not considered the possibility of the media being designated as the threat in a securitisation process, a notable gap which this article seeks to fill.
Another gap in the literature is revealed upon surveying the literature on President Trump and securitisation. Trump has made several securitising moves on a number of issues. For example, Magcamit (2017) examines Trump's attempts to securitise the U.S. economy (p. 17). Additionally, Floyd (2019) notes President Trump's securitisation of terrorism, which led to several 'security measures' (p. 61), while Shipoli (2018) examines President Trump's securitisation of Islam and argues that Trump's election win was a consequence of the 'unsuccessful desecuritization of Islam in American politics' (p. 96). These articles underline that Trump can be a securitising actor; however, while covering several issues that Trump has securitised, the authors do not consider whether Trump has securitised the news media, despite, in the case of Shipoli (2018), acknowledging that Trump has a hostile and unconventional relationship with the media (pp. 211, 238). This article also seeks to address this issue.

Context
Post-truth and fake news The election of President Donald Trump is an often-cited example used to illustrate that we currently live in 'an age of post-truth politics' (Forstenzer, 2018, p. 5). This assertion is based on the amount of 'falsehoods and outright lies' (p. 5) produced by the former President and his campaign. It was Steve Tesich who in 1992 coined the term 'post-truth' as it is understood today (Legg, 2018, p. 44). Tesich, in an article discussing Watergate, the Iran/Contra scandal and the Persian Gulf War (Tesich, 1992, pp. 12-14), used the term to argue that the U.S. 'had become a society where truth is politically unimportant' (Forstenzer, 2018, p. 5). Tesich (1992) stated: we are rapidly becoming prototypes of a people that totalitarian monsters could only drool about in their dreams. All of the dictators up to now have had to work hard at suppressing the truth. We, by our actions, are saying that this is no longer necessary, that we have acquired a spiritual mechanism that can denude truth of any significance. In a very fundamental way we, as a free people, have freely decided that we want to live in some post-truth world (p. 13). (2017) argues that, according to Tesich, these events traumatised the U.S. population to such a degree that the public abandoned the truth and began to 'collude wearily in its suppression' (p. 9). Following Trump's election and the Brexit referendum (Forstenzer, 2018, p. 5), the concept of post-truth has 'exploded' (Legg, 2018, p. 44) and, in 2016, Oxford Dictionaries nominated the term for "word of the year" defining it as 'relating to or denoting circumstances in which objective facts are less influential in shaping public opinion than appeals to emotion and personal belief' (Oxford Languages, 2016). Today's post-truth has come about as a result of 'societal mega-trends such as a decline in social capital, growing economic inequality, increased polarization, declining trust in science, and an increasingly fractionated media landscape' (Lewandowsky et al., 2017, p. 353).

D'Ancona
Further, according to Harsin (2015), we are currently experiencing a 'regime of posttruth' (p. 1), whereby 'regime' underlines that 'this epistemic turn is newly systematic' (Legg, 2018, p. 44 -italics in original). This regime is, among other things, characterised by a 'deliberate, continued repetition of talking points even when they have been clearly rebutted by easily verifiable facts. Consumers of the talking points are often even aware that these rebuttals exist, but don't seem to care.' (Legg, 2018, p. 44). Trump and his administration were noted for having employed the former behaviour, while his supporters the latter (Legg, 2018, p. 45). Although the perception that politicians can be dishonest is widely held in society (Stolee and Caton, 2018, p. 157), traditionally, when politicians are caught lying, they tend to 'provide complex justifications and near-apologies' (Sismondo, 2017, p. 3). Trump, however, departed from this practice (Sismondo, 2017, p. 3). His 'falsehoods seem to usher in an era of unprecedented political mendacity' (Stolee and Caton, 2018, p. 157). The Trump administration consistently rebuked criticism and unfavourable reporting as 'fake news', which also marked 'the first time that a president has so broadly delegitimized the mainstream media' (Kellner, 2019, p. 98). The Administration also introduced the controversial term 'alternative facts' in response to an accusation of lying from the news media (Kellner, 2018, p. 97). As such, post-truth is the current "zeitgeist". As already noted, Balzacq argues that zeitgeist is a contextual factor which impacts securitisation and, thus, post-truth and the public's apathy concerning the factual truth in this era is taken into account in the following analysis.

Trump and trust
As noted above, Trump's supporters trust him despite his factually incorrect statements. Supporters point to his lack of 'inhibitions' and that 'he speaks his mind', and argue that 'his language is more genuine' (Stolee and Caton, 2018, p. 157). These qualities are valued more than whether his statements are factually correct. Trump, in the eyes of his supporters, does not hide his intentions like other politicians (Stolee and Caton, 2018, p. 157). While several news outlets consistently fact-checked Trump's statements and tweets, pointing out when the President lied, this did little to sway his supporters' trust. Instead, the consistent scrutiny and fact-checking caused his supporters to believe that journalists were 'out to get him' (Rutenberg, 2018), which thus nullified the findings and reinforced the feeling of a 'biased' press (Smith, 2016) -an already prevalent feeling about the media in the U.S. explored further below. This ineffectual fact-checking highlights the cornerstone of the post-truth phenomenon, namely that truth is unimportant (Jandric, 2018, p. 103). Post-truth also encompasses a reduced trust in experts and the 'elite' (D'Ancona, 2017, p. 8) to which, according to his supporters, Trump does not belong. Thus, Trump holds a certain position of authority and has the confidence and trust of his supporters, which highlights the persuasive potential of his words towards the audience to which he speaks. The above are critical assertions to understanding Trump's securitising move towards the news media.

Distrust in the media in the United States
Another element to consider in relation to both context and audience is that of trust in the news media in the U.S. According to Turcotte et al. (2015), trust in the news media has 'reached historic lows' (p. 520). A study conducted by the Poynter Institute found that while the majority of the public support the news media, this support is weak, and there is a 'dramatic polarization in media attitudes' (Guess, Nyhan and Reifler, 2017, p. 1). Comparing Republicans, Trump supporters, Democrats and Trump opponents, the former two have 'vastly more negative views of the press' than the latter two (Guess, Nyhan and Reifler, 2017, p. 12). Trump supporters and Republicans were 'far more likely to endorse extreme claims about media fabrication, to describe journalists as an enemy of the people, and to support restrictions on press freedom' (Guess, Nyhan and Reifler, 2017, p. 1) with approximately half of Republicans supporting the latter (ibid., p. 11). In the updated 2018 report, however, Poynter found that since Trump became president, there has been some increase in trust in the press for both Republicans and Democrats (Guess, Nyhan and Reifler, 2018, p. 7), albeit still highly polarised. Notably, in both the 2017 and 2018 reports, the opinion that 'news organizations keep political leaders from doing things that shouldn't be done' (Guess, Nyhan and Reifler, 2018, p. 2) was upheld by around two-thirds of the public. The report also noted that conservatives have argued about a liberal bias in the news 'for decades' (Guess, Nyhan and Reifler, 2018, p. 4), with fewer than ten percent of Republicans viewing the press as balanced. Thus, while Trump exploited and exacerbated these polarised attitudes, he is not their cause (Kalb, 2018, p. 130). The former president, however, has a 'remarkable ability to read, influence and change public opinion, especially among Republicans' (Kalb, 2018, p. 130). Trump invokes fear 'in concrete and abstract ways' (Ball, 2016) and 'grasps and channels the fear coursing through the electorate'. Guess, Nyhan and Reifler argue that elite messaging influences media trust, especially in regard to bias (2018, p. 4). Thus, Trump was in an influential position of power and seemed to understand the already prevalent attitudes of many within his party and amongst his supporters, on which the former President capitalised.
Notably, the U.S. is also in an 'era of media choice' (Turcotte et al., 2015, p. 521), where people are able to choose outlets that support their 'ideological leanings'. This has led to individuals placing greater trust in the news outlets which align with their ideology, and are distrustful of ones that do not. Media choice exacerbates partisan divides and the production of echo chambers, which is further aggravated by social media. Twenty percent of U.S. adults get their news from social media (Shearer, 2018), with sixty-eight percent stating that they do so 'at least occasionally' (Matsa and Shearer, 2018). These findings underline that there is lack of trust in the media generally, and that there are ideological divides on which a securitising move can capitalise. Moreover, as social media is increasingly becoming a platform for news, the usefulness of conducting a securitising move via social media, as this article argues that Trump does via Twitter, is made clear.

The Age of Twitter
The rise of Twitter has further consequences. According to Ott (2017), we are currently experiencing a 'changing character of public discourse in the Age of Twitter' (p. 59). Twitter, Ott (2017) argues, 'promotes public discourse that is simple, impetuous, and frequently denigrating and dehumanizing' (p. 60). The platform's character limit leads to simplicity because it is not possible to post 'detailed and sophisticated messages' (Ott, 2017, p. 60). This hinders real discussion and the ability to 'think about issues and events in more complex ways' (Ott, 2017, p. 61). It is also impulsive and does not command much reflection about the effects of a tweet (Ott, 2017, p. 61). Finally, it 'encourage[s] uncivil discourse' (Ott, 2017, p. 62) as a result of its informality, and because it 'depersonalizes interactions' (p. 62). These insights are important to note, as they show that Twitter fosters lack of critical reflection in users/consumers. This highlights the persuasive potential tweets can have and, thus, makes it an ideal tool for a securitising actor. Importantly, tweets that are 'emotionally charged' receive more retweets than those that are not (p. 61), highlighting that the use of emotion on Twitter is also effective for a securitising actor.

Trump and Twitter
Trump's use of Twitter as 'one of his primary means of presidential communication' (Newport, 2018) was unprecedented. Stolee and Caton (2018) assert that we may be experiencing a 'major cultural shift' (p. 149) in presidential talk. It may be the case that there is a 'shift in the presidential rhetorical strategy from an address of a wide consistency (built on coalitions) to a core consistency (built on a base)' (Stolee and Caton, 2018, p. 164). This is because Trump's tweets and rhetoric were directed towards his base rather than meant to resonate with the greater American public (Stolee and Caton, 2018, p. 151). Thus, Trump framed issues in ways that his supporters identified with, sometimes 'at the expense of a wider reach of the electorate' (Stolee and Caton, 2018, p. 150). As such, it seems that Trump's audience, in terms of securitisation theory, is his base.
Trump used Twitter in interesting ways. Trump's tweets gave him an advantage as they 'always seem ahead of everyone else's in framing the message' (Stolee and Caton, 2018, p. 157). Thereby, Trump's framing had the potential to become the dominant frame. Additionally, Lakoff (cited in Ross and Rivers, 2008) argues that Trump used the 'trial balloon' (p. 4) tactic on Twitter; he would post a tweet with a contentious statement in order to test the audience's reaction to it. Trump also used exclamation points and caps lock, which 'reinforce the negative sentiment' (Ott, 2017, p. 64) of the content and 'heighten [its] emotional impact (p. 64). This emotion was then transferred to his followers in a manner known as 'emotional contagion' (Auflick, cited in Ott, 2017, p. 64). Thus, again, emotion becomes a noteworthy tool that will be brought into the analysis later on.
Notably, Trump's usage of Twitter led to his temporary suspension from the platform in January 2021 in the midst of the Capitol riot/insurrection 1 (The Guardian, 2021). Trump was subsequently permanently suspended from the platform 'due to the risk of further incitement of violence' (Twitter Inc., 2021). Trump had argued that the election was 'stolen' (Kahn et al., 2021) and that his loss was due to voter fraud, while he advocated for an overturn of the election. Through tweets (among other things) Trump encouraged his supporters to protest his 2020 election loss with a march on the Capitol, which resulted in a riot/insurrection (The Guardian, 2021).
A problematic media climate Finally, in order to grasp fully the situation of the U.S. media in the current zeitgeist, the findings of the Reporters Without Borders (RSF)'s 2019 World Freedom Index should be reviewed. The report classifies the media climate in the U.S. as 'problematic' (Reporters without Borders, 2019a). The Index has categories ranging from 'good', 'satisfactory', 'problematic', 'difficult', and 'very serious' for journalists. The report cites the reasoning for the 'problematic' classification as 'an increasingly hostile climate that goes beyond Donald Trump's comments' (Reporters without Borders, 2019a). Further, the report argues that the Administration has denied 'access to information and events of public interest' (Reporters without Borders, 2019b) to reporters. The White House revoked CNN reporter Jim Acosta's press pass in November 2018 (Reporters without Borders, 2019b) (Wang and Farhi, 2018), while the Administration first went an unprecedented 42 days without a press briefing (Reporters without Borders, 2019b) followed by 100 days in June (Stelter, 2019b). The report further noted a record number of death threats towards journalists from the public, and noted that several newsrooms had received 'suspicious packages' (Reporters without Borders, 2019b) and 'bomb threats'. Additionally, reporters attending Trump rallies were 'physically accosted' (Reporters without Borders, 2019b) and intimidated by attendees. The Index also cited the 2018 Capital Gazette newsroom shooting, which led to the death of four reporters. Trump continued with his 'anti-press rhetoric' (Reporters without Borders, 2019b) in the direct aftermath of the shooting. While the media climate in the U.S., and these occurrences, are not solely attributed to Trump's rhetoric against the media, Trump's contribution is underlined as a contributing factor by the report (Reporters without Borders, 2019b).
Trump was impeached and charged by the House of Representatives for inciting the 2021 Capitol insurrection (Fandos, 2021), 2 which also saw attacks on the media by parts of the crowd. Hsu and Robertson (2021) describe journalists as the 'secondary target' of the insurrection. The words 'Murder the media' were etched into a Capitol door by insurrectionists, parts of the crowd chanted 'CNN sucks' and there were reports of journalists being surrounded, threatened, physically attacked and forced to take shelter. Associated Press camera equipment was destroyed, and a noose was seen 'fashioned out of a camera cord and hung from a tree' (Hsu and Robertson, 2021). As a precautionary measure, some reporters had opted not to wear their news organisation's insignia, and another reporter wore protective gear traditionally used for reporting in conflict zones, when covering the protest/insurrection (Hsu and Robertson, 2021). These occurrences further highlight the problematic climate for journalists in the U.S., as well as the views Trump's base has towards the media, to which Trump has contributed.

Aims, theory and method
The aim of this article is to analyse President Trump's discourse concerning the U.S. news media, and to illustrate that this discourse frames the news media as a security threat, constituting a securitising move. In order to do so, securitisation theory is employed. More specifically, this article applies second-generation literature as a result of the critiques and expansions of the CS's original theory of securitisation. The qualitative method of discourse analysis is employed by applying framing theory in conjunction with securitisation theory to a sample of rhetoric posted by Trump on Twitter. Framing was chosen for the discourse analysis, as 'frames influence the way in which we view particular situations and issues' (Kuypers, 2009, p. 300) and because securitisation is seen as a type of framing that 'highlights the existential threat of an issue' (Vultee, 2010b, p. 79).
Further, as Twitter becomes an increasingly popular way through which the American public consumes their news, and as it was one of President Trump's 'primary means of presidential communication' (Newport, 2018), its use to study Trump's discourse is deemed appropriate. TheTrumparchive.com, a website that serves as an online archive of the President's tweets, was utilised to conduct word searches and obtain the President's tweets -including ones since deleted from Trump's Twitter profile. It is a source cited by the likes of The BBC, The Guardian, The Washington Post and other news outlets, rendering it a reliable platform. The time span chosen for this study ranged from Trump's inauguration (20 January 2017) to 22 August 2019. This period was chosen to investigate Trump's rhetoric while he presided as sitting president and extended to the latter date to encapsulate a picture of the consistency with which this discourse had been uttered over time.
Word searches were conducted on thetrumparchive.com using the following words, noting the respective number of tweets that they appeared in (up to 22 August 2019): 'enemy of the people' (35 times), 'enemy of the American people' (2), 'fake news' (496), 'fake news media' (159), 'dishonest media' (29), 'dishonest fake news media' (2), 'corrupt media' (19), 'corrupt news' (8), 'corrupt news media' (3) and 'biased media' (6). Further, "enemy of the people" without "media" (12) and "enemy of the people" without "news" (8) were searched; however, all of the tweets that included 'enemy of the people' used either 'news' instead of 'media' or the reverse, and/or included references to individual news outlets. This word search was done in order to make sure the phrase had not been used in other contexts, which could distort frequency findings. Finally, individual outlets were searched independently, which brought to light 'failing @nytimes' (75), 'failing new york times' (31) and 'fake news CNN' (21) to name but a few.
This article has chosen to focus on the phrases 'fake news', 'fake news media' and 'enemy of the people'. These particular samples were chosen in part due to the frequency of their appearance and also due to the publicity they have received in the news media. When researching Trump's relationship with the media, the phrases 'enemy of the people', 'fake news' and 'fake news media' were consistently mentioned, which underlines the rationale behind their choice. Moreover, they clearly highlight an anti-media rhetoric which, for me, rang securitisation "alarm bells". Furthermore, as securitisation should be seen as a 'process of articulations' (Stritzel, 2007, p. 377) and studying a single speech act will seldom explain the process of securitisation, and because security is 'constructed over time' (McDonald, 2008, p. 564), analysing a sample of tweets posted over a longer period of time was deemed appropriate.
By conducting the above, this article seeks to fill the aforementioned gap in the literature. Thus, the discussion adds to securitisation theory -through analysing Trump's anti-media discourse -by identifying that the news media can be framed as a security threat, by illustrating Trump's securitising move against the U.S. news media, and by highlighting the danger such a securitising move has on democracy.

Limitations and disclaimers
One limitation that this article has had to deal with is space. As such, other discursive practices were omitted from the study, which could have contributed to the data. Likewise, this article only uses a sample of Trump's tweets and phrases -obtained during the data collection period. It is also important to note that the terms searched for are targeted towards negative/securitising language. While these tweets are, I believe, representative of Trump's general tone concerning the media, they are only a fraction of the available tweets. Moreover, TheTrumpArchive.com, the source utilised to obtain and search through Trump's tweets, is missing 4,000 of those; however, these are mostly tweets posted prior to 27 January 2017. As noted, this article has chosen to study tweets posted while Trump was president, meaning from his inauguration on 20 January 2017 (Kalb, 2018, p. 1) onwards. 3 Further, it is necessary to underline that this article uses the phrases "U.S. news media", "news media" and "media" interchangeably throughout. All of these phrases always refer to the news media in the United States. Notably, Trump has criticised some news media organisations more than others, namely those which he classifies as 'mainstream' versus conservative outlets. This is taken into consideration in the analysis.

Analysis and illustration
Trump's hostile rhetoric concerning the U.S. news media was present during his 2016 election campaign. During this time, the Republican presidential nominee referred to the media as 'disgusting' and 'dishonest' (Acosta, 2019). Trump's rhetoric extended into his presidency, and he consistently challenged the media's legitimacy (Guess, Nyhan and Reifler, 2017, p. 11). As president, Trump was in a position of authority, since the presidency is 'the apex of political authority in designating threats for the United States' (van Rythoven, 2015, p. 461). Thus, Trump, as an elite securitising actor and in line with the CS's conceptualisation (Buzan et al., 1998, p. 33), has the power and influence to perform a securitising move (Waever, 1995, p. 54). As already examined, Trump also speaks to his base, an audience that is receptive to his arguments (Stolee and Caton, 2018, p. 151). Moreover, as noted earlier, Trump used his Twitter handle @realDonaldTrump as one of his main channels of presidential communication (Newport, 2018). Some of his most used phrases on Twitter, in his "anti-media" tweets, included: 'fake news', 'fake news media' and 'enemy of the people' (see Aims, theory and method). Thus, this section analyses a sample of these tweets, applying both framing and securitisation theory, and draws on the contextual information surveyed in the previous sections to illustrate that Trump's framing of the U.S. news media constitutes a securitising move.

Fake news and the fake news media
Truth and country as referent objects Throughout his presidency, Trump consistently referred to 'fake news' alluding to the U.S. news media and their unfavourable reporting of him (Kellner, 2018, p. 55). The former president tweeted the phrase a total of 496 times according to my research (see Aims, theory and method). The extended version, 'fake news media', was tweeted 159 times. By doing so, Trump framed the news media as untruthful and accused them of producing fake news. This is exemplified in the following tweets: Feb 24, 2017 10:09:18 PM FAKE NEWS media knowingly doesn't tell the truth. A great danger to our country. The failing @nytimes has become a joke. Likewise @CNN. Sad! (The Trump Archive V2) Aug 16, 2018 09:10:17 AM There is nothing that I would want more for our Country than true FREEDOM OF THE PRESS. The fact is that the Press is FREE to write and say anything it wants, but much of what it says is FAKE NEWS, pushing a political agenda or just plain trying to hurt people. HONESTY WINS! (The Trump Archive V2) Jan 7, 2019 08:09:03 AM ...The Fake News will knowingly lie and demean in order make the tremendous success of the Trump Administration, and me, look as bad as possible. They use non-existent sources & write stories that are total fiction. Our Country is doing so well, yet this is a sad day in America! (The Trump Archive V2) As noted earlier, the state has traditionally been viewed as the referent object in security studies (Buzan et al., 1998, p. 37). Since then, several second-generation scholars have challenged this view and expanded what can be considered a referent object (Buzan et al., 1998, pp. 37-39;Watson, 2011, p. 5;Hansen andNissenbaum, 2009, p. 1162). Applying this understanding to the above tweets, it becomes clear that by framing the news media as producing fake news and intentionally not telling the truth, Trump framed the truth as being under threat. Thus, truth is the referent object that must be protected. Simultaneously, in the first tweet, Trump refers to 'our country', the United States as a whole, framing it as also threatened by the extension of the threat to the truth.

The (mainstream) media as a threat
By claiming that the news media threatens both the truth and the U.S., Trump framed the news media as the threat. According to the CS's conceptualisation of securitisation theory, a threat is framed as existential in securitisation (Buzan et al., 1998, p. 23). Whether Trump framed the media as an existential threat, however, can be debated. On the one hand, framing it as a 'danger to our country' could warrant it existential status. One could infer that this means that it threatens the American way of life as we know it. Trump, however, does not always use obvious existential terminology in his framing of the media. He does not always reference the state and, instead, often uses words such as 'corrupt' and 'inaccurate', as exemplified in the following tweet: Apr 9, 2019 07:44:17 AM The Mainstream Media has never been more inaccurate or corrupt than it is today. It only seems to get worse. So much Fake News! (The Trump Archive V2) On the other hand, however, Floyd (2016, p. 677) argues that a threat does not necessarily need to be existential for it to be securitised. One can, therefore, make the case that establishing the media as a threat is enough for it to become a threat in line with second-generation securitisation theory.
It is important to note that Trump has distinguished between the 'fake news media' and other news media. Initially, Trump did not label all media as 'fake news', referring to the 'mainstream media' (see tweet above) as those which produce fake news (Kalb, 2018, p. 3). These news outlets are argued to be those that are willing to criticise him (Kalb, 2018, p. 3). The news outlets to which Trump refers are among others: The New York Times (The Trump Archive V2, Feb 24, 2017), MSNBC and CNN (see below). Trump has mostly refrained from labelling conservative news outlets such as Fox News as fake news. Instead, he gives them praise more regularly, as exemplified in the tweets below: On occasion, however, Trump has accused Fox News of fake news, as seen in the tweet below, where he references Fox News host Bret Baier: Jun 17, 2019 06:49:43 PM .@FoxNews Polls are always bad for me. They were against Crooked Hillary also. Something weird going on at Fox. Our polls show us leading in all 17 Swing States. For the record, I didn't spend 30 hours with @abcnews, but rather a tiny fraction of that. More Fake News @BretBaier (The Trump Archive V2) Stelter (2019a) argues that Trump has a 'back-scratching relationship' with Fox News. The network has been criticised for serving as a platform for the President's 'propaganda' (Mayer, 2019), for killing stories unfavourable to Trump (Mayer, 2019), and for the 'steady flow of personnel from Fox News to the Trump White House' (Hemmer, 2019) among other things. This does not mean that Fox never reports news that are damaging to the President, and the network does criticise him on occasion. When it does so, however, Trump lashes out at Fox as well, and labels them fake news. As previously mentioned, the consistent labelling of "fake news" is generally applied to the news outlets that are considered a part of the "mainstream media", and those that are more inclined to criticise Trump or point out when he lies. It seems, however, that Trump is willing to criticise conservative outlets when they present him in a way that he deems unfavourable. 4 Framing the news media as a threat taps into an already prevalent negative view of the media held by Republicans in the U.S. This negative sentiment, as noted previously, is based on a perception that the news media have a liberal bias (Guess, Nyhan and Reifler, 2018, p. 4). Trump also tapped into this sentiment at a time when there was a historically low level of trust in the media in the U.S. (Guess, Nyhan and Reifler, 2017, p. 12), in the era of post-truth. As such, a negative view of the media was already a prevalent frame within Trump's base and, as already determined, this is the audience to which he spoke on Twitter. Thus, a frame in thought (Chong and Druckman, 2007a, p. 101) within the audience was readily available for Trump to tap into by using a frame in communication to exacerbate the former further. Trump took advantage of prejudices and ideological partisan leanings already held by a certain portion of the audience, a method of framing outlined by Chong and Druckman (2007b, p. 111). Moreover, this also underlines van Rythoven's (2015) argument that 'audiences fear threats because they represent phenomena they have already learned to fear or imaginably foresee fearing ' (p. 467). This points to the fact that Trump's arguments are relevant within the zeitgeist -one of Balzacq's contextual securitising factors. By tapping into prevalent social views and political attitudes within the zeitgeist, a securitising actor can influence the receptiveness of the audience towards their arguments (Balzacq, 2005, p. 192). Balzacq's audience factor, the audience's fitting frame of reference, is thus clearly underlined.
As Trump spoke to his base, who perceive him as a trustworthy actor and an expert on the issues, he fulfilled another audience factor outlined by Balzacq. As such, Trump employed words and phrases that would resonate with his audience and win their support, utilising all of Balzacq's congruent securitising agent components of securitisation. Balzacq's congruent factors that condition the 'effectuation of securitisation ' (p. 192) were thereby prevalent. Likewise, the CS's felicity conditions were also met, as Trump clearly used the grammar of security through words that resonated. He had social capital and authority, and could effectively point to the media as the threat considering the predominant hostile sentiments. As such, Trump's designation of the news media as a threat can be seen as a securitising move.
Trump, the protector of truth Trump also framed himself as the protector of the truth, and as the person that enlightens his audience by giving them the truth that the media does not. He argued that he was able to do so via social media. This notion comes to light in the following tweets: Jun 8, 2019 11:08:26 PM I know it is not at all "Presidential" to hit back at the Corrupt Media, or people who work for the Corrupt Media, when they make false statements about me or the Trump Administration. Problem is, if you don't hit back, people believe the Fake News is true. So we'll hit back! (The Trump Archive V2) Jun 12, 2019 07:46:43 AM The Fake News has never been more dishonest than it is today. Thank goodness we can fight back on Social Media. Their new weapon of choice is Fake Polling, sometimes referred to as Suppression Polls (they suppress the numbers). Had it in 2016, but this is worse… (The Trump Archive V2) Here, the former president underlines that this behaviour may not be 'Presidential', which, evidently, is one of the reasons that Trump's supporters trust him. As noted earlier, Trump is not a career politician. He is not part of the establishment that uses correct facts but has 'entirely suspect' intentions (Stolee and Caton, 2018, p. 157). Rather, in the eyes of his supporters, he is an outsider who speaks his mind (Stolee and Caton, 2018, p. 157). Thereby, Trump's supporters see him as honest and a source for truth, despite the fact that he makes statements that are 'factually incorrect' (Stolee and Caton, 2018, p. 157). As mentioned above, in a post-truth world, factual truth becomes unimportant (Jandric, 2018, p. 103); however, that is not to say that his supporters do not believe there is a truth. Instead, they just do not seem to care when their belief in what is true is disproved (Legg, 2018, p. 44). Thus, Trump protects his "truth" and his supporters' "truth", claiming it to be the factual truth, and attacks the media when they report the actual truth, which includes news that are damaging to Trump. Additionally, when the media criticise Trump and his factual inaccuracies, this further solidifies Trump supporters' trust in him and distrust in the media (Rutenberg, 2018). Thus, objective factual truths, the essential political and democratic good, are unimportant to his supporters. Trump argues that if he does not 'hit back' at 'fake news' -meaning if he does not call out the media's behaviour -then the people will suffer because they will believe falsehoods. By doing so, he frames himself as using his office to help the people. Additionally, Trump highlights that if the people want the truth, they should turn to him and social media, the new arena for that truth. As such, Trump frames himself as the uninhibited, unpresidential outsider and populist truth-teller, a frame already adopted by his supporters. By framing himself in this light, Trump was able to gain further legitimacy among his supporters as a securitising actor.

Trump's use of repetition
Trump also employed the framing tactic of repetition. As shown in my research (see Aims, theory and method), Trump tweeted the phrases 'fake news' and 'fake news media' 496 and 159 times respectively, during the period studied. Moreover, Trump also had a tendency to tweet several times in one day, using the same terms, as seen in the employment of the phrase 'Fake News' in the tweets below: Aug 15, 2019 11:30:03 AM Wow! The Deputy Editor of the Failing New York Times was just demoted. Should have been Fired! Totally biased and inaccurate reporting. The paper is a Fraud, Zero Credibility. Fake News takes another hit, but this time a big one! (The Trump Archive V2) Aug 15, 2019 11:52:53 AM The Fake News Media is doing everything they can to crash the economy because they think that will be bad for me and my re-election. The problem they have is that the economy is way too strong and we will soon be winning big on Trade, and everyone knows that, including China! (The Trump Archive V2) Aug 15, 2019 03:18:25 PM Walmart, a great indicator as to how the U.S. is doing, just released outstanding numbers. Our Country, unlike others, is doing great! Don't let the Fake News convince you otherwise. (The Trump Archive V2) As Lakoff and Wehling (2016) argue, repetition of words and phrases make a frame 'common sense' (p. 84) or 'right and real' to an audience, while contradictory facts are discounted. Additionally, as a result of repetition the phrase becomes memorable. As Entman (2004) states, repetition combined with cultural resonance, such as emotion, has the ability to create similar 'thoughts and feelings' (p. 6) in an audience. Moreover, short phrases, such as the simple "fake news", become readily memorable with repetition. As already noted, the regime of post-truth is characterised by actors using 'deliberate, continued repetition of talking points even when they have been clearly rebutted by easily verifiable facts' (Legg, 2018, p. 45). The former president employed this tactic. In this post-truth regime, the audience is oftentimes aware of such refutations, but tends to discount them in favour of appeals to emotion and feelings (Legg, 2018, p. 45). As such, Trump's repetition in his tweets also made his audience more likely to accept his securitising language. Moreover, as noted earlier, Oren and Solomon (2015) argue that repetition of ambiguous phrases is one way to securitise an issue (p. 320). The authors note that '[t]he more constantly and frequently a securitising phrase is being repeated, that is, the more condensed the "historicity" of the phrase becomes, the more likely is the phrase to acquire an illocutionary force, to construct the security threat it ostensibly describes' (p. 320). Successful securitisation can be possible when the audience partakes and contributes to the repetition of the phrase (Oren and Solomon, 2015, p. 322). What is apparent from these tweets is that Trump continuously repeated 'fake news' / 'fake news media' -both ambiguous phrases -with the aim of framing the news as untrustworthy and decisively dangerous. Thus, this repetition created the opportunity for Trump's audience (including those who oppose him) to participate in the chant by echoing the phrases, thereby aiding Trump's initiated securitisation process.

Framing using emotion
As noted, Trump also appealed greatly to emotion in his framing of the media, as exemplified in the following tweet: As discussed above, the use of exclamation points and caps lock heightens the emotional effect and negative feeling of a tweet, and this is transferred to the audience (Ott, 2017, p. 64). Tweets that are laden with emotion receive a higher number of retweets than those that are not, which heightens the impact and circulation of the message. Moreover, the function of retweeting shows that the portion of the audience that retweets plays a role in the spread of Trump's message, underlining the active role of the audience. Further, according to Balzacq's pragmatic act of securitisation, securitising discourse works at 'a level of persuasion ' (2015, p. 172), where emotion is an artefact that can be employed. This is an effective instrument in the zeitgeist in which these tweets are posted because, in a post-truth era, appeals to emotions and opinions are more important for an audience than appeals to facts (Oxford Languages, 2016;Jandric, 2018, p. 106).

Emergency measures
Securitisation, according to the CS, stresses that a speech act designates an existential threat which necessitates 'emergency action or special measures and the acceptance of that designation by a significant audience' (Buzan et al., 1998, p. 27). The CS has generally described policy measures as emergency action; however, Côté argues that this may be too high of a 'threshold ' (2016, p. 544). Secondary scholars and the PS have underlined that securitisation need not focus on exceptionalist policies but rather on the 'reproduction of unease and insecurity in everyday life' (Nyman, 2018, p. 107). Considering these arguments, the following tweets are of significance: Jan 29, 2017 08:00:32 AM Somebody with aptitude and conviction should buy the FAKE NEWS and failing @nytimes and either run it correctly or let it fold with dignity! ( As already noted, Lakoff (cited in Ross and Rivers, 2018) argues that Trump used the 'trial balloon' (p. 4) tactic, in which he would post a tweet with a contentious statement in order to test the audience's reaction to it. The aim was to aid the President's understanding of how to proceed. This is especially interesting when considering the above tweets. The first can be interpreted as arguing that The New York Times should be 'bought' or 'fold' before its reputation worsens. The next two tweets offer more controversial statements, such as the questions 'at what point is it appropriate to challenge their License?' and 'Take away credentials?'. Such measures would be seen as extreme, and if put into effect they would challenge the right to free speech (Shepardson, 2017). '[P]lacing limitations on otherwise inviolable rights' (Watson, 2009, p. 24), however, is a sign of securitisation. As such, in line with securitisation theory, it may be understood that, in a sense, Trump is 'trial balloon'ing emergency measures by testing audience acceptance. Whether or not one interprets emergency measures as necessitating policy changes, Trump is known to invoke fear 'in concrete and abstract ways' (Ball, 2016) and 'grasp[ing] and channel[ling] the fear coursing through the electorate', as exemplified by the studied tweets. Moreover, these tweets underline the consistent enunciation of a security threat and the production of unease. As such, Trump clearly used securitising language in reference to the U.S. media.

Enemy of the people
On 17 February 2017, Trump tweeted: The FAKE NEWS media (failing @nytimes, @NBCNews, @ABC, @ CBS, @CNN) is not my enemy, it is the enemy of the American people (Kalb, 2018, p. 1) This was the first instance of the news media, and specific news outlets, being called 'the enemy of the American people' by the recently inaugurated president (Kalb, 2018, p. 1). According to the available data, Trump tweeted 'the enemy of the people' and 'enemy of the American people' in reference to the news media 35 and two times respectively (see Aims, theory and method). The following is a sample of some of these tweets: The media as the enemy By using the word "enemy", Trump unmistakably labelled the U.S. news media as the opposition. The word "enemy" elicits a schema of danger and of the other, an antagonist and a threat that must be protected against. In order to solidify this frame of the media as the enemy of the people, Trump stated that 'they can also cause War!', which likewise elicits a clear threat frame. The term "war" can allude to several other schemas related to threat, such as destruction, death and fear. Further, stating that 'they are very dangerous & sick' is another obvious threat image. If something is 'very dangerous' or 'sick', it is decisively threatening or repulsive. Moreover, if it is 'corrupt' and 'deranged', it is immoral and unreliable. These are clear negative connotations that aid the framing of the media as a threat. Here, it seems that the media is also plainly framed as an existential threat, because schemas relating to 'enemy' in conjunction with 'America' may also activate the perception of a threat to the American way of life. This alludes to historical schemas related to previous enemies and wars, such as the language utilised in the run-up to the war in Iraq (Reese and Lewis, 2009, p. 779). If the media are 'the enemy', which can 'cause War!' and are 'very dangerous & sick', then the media are undoubtedly a considerable threat, not only to the truth but to the security of the state itself.
'The people' as referent object Trump positioned the word 'enemy' in contrast to 'the people' and 'the American people', thereby framing the people as being threatened. If the media continue to produce fake news, then the U.S. may face severe threats, which the people must be protected from. In these tweets, Trump seems to refer to the entire U.S. public by using the phrase 'the people' and 'the American people'. As noted, Trump speaks decisively to his base, framing his messages with this audience in mind. Thus, the framing of 'the people' will likely create a feeling of collectivity for Trump's supporters and has a large chance of resonating with them because of the fulfilment of Balzacq's contextual, audience and securitising actor congruent factors, as well as Chong and Druckman's arguments concerning framing outlined above. The audience, however, may not accept this framing if they already have a very different and established schema concerning the phrase.
The media as referent object: established schema The phrase 'Enemy of the People' as a whole may elicit very different reactions and feelings for those who have previously established a negative schema related to it in their long-term memory and knowledge networks. The phrase has a long history and has been associated with the likes of Mao, Hitler, Lenin, Stalin, Khrushchev and Nixon (Kalb, 2018, pp. 22-31). Kalb contends that the phrase has always consistently contained a similar understanding: '[i]n the land of the dictator, whether in Russia or China, the critic, the reporter, was considered the "enemy of the people" ' (2018, p. 25). During Khrushchev's rule, the leader abandoned the term used by his predecessor Stalin, seeing that its usage was employed in order to eliminate any form of political disagreement. Speaking at the Twentieth Congress of the Soviet Communist Party in 1956 (Kalb, 2018, p. 26), Khrushchev stated that '[t]he formula "enemy of the people" was specifically introduced for the purpose of physically annihilating individuals who disagreed with Stalin, even communists thought to be in good standing ' (Khrushchev, cited in Kalb, 2018, p. 28). In the United States, historically, the term is linked to President Nixon's 'enemies list' (Kalb, 2018, p. 31), which 'was his way of labelling journalists -those who pointed out the fallacies and fantasies about his policies -as "enemies of the people"' (p. 31). As such, the term may activate already established negative schemas of authoritarianism, oppression, distaste and fear of the actor that employs the phrase. Therefore, individuals who have this negative schema stored in their long-term memory are less likely to accept Trump's framing. This is because, as already noted, 'first impressions are difficult to dislodge' (Entman, 2004, p. 7), and once a schema is established 'all succeeding information about any of these ideas has the potential to bring to mind (online, into working memory) associated feelings and concepts from the knowledge network' (p. 7). As noted earlier, Trump's framing sometimes came 'at the expense of the wider electorate' (Stolee and Caton, 2018, p. 150). Trump, however, may not have needed to convince the entire U.S. public in order to securitise the media, because as Vultee (2010b) argues, one only needs 'the consent of enough of the audience', or O'Reilly's 'critical mass ' (2008, p. 67). Moreover, as Oren and Solomon (2015) contend, audience acceptance does not have to be measured by level of persuasion but rather in having enough of the audience repeating a securitising phrase (in agreement or opposition) (pp. 313, 321).

Framing using hyperbole
As exemplified by tweets already mentioned as well as those that follow, Trump is no stranger to hyperbole: Hyperboles can be utilised to exaggerate a threat and can effectively persuade an audience of the reality and 'imminence' of that threat (Burgers et al., 2016, p. 415). As we know from reviewing the literature, when a securitising actor wishes to persuade an audience of a threat, it is important to highlight its imminence. Additionally, the securitisation of an issue is a 'political choice' (Nyman, 2013, p. 53) and whether or not a security threat actually exists is irrelevant, underlining the need for hyperbole. Hyperbole can also increase the 'emotional attitude attached to a subject' (Burgers et al., 2016, p. 415) and, thus, the 'message['s] persuasiveness'. Repetition of hyperbole also effectively helps put the issue 'in a place in the public debate' (p. 415). Trump uttered the phrase 'enemy of the people' 35 times, which underlines his effective use of the framing mechanisms of hyperbole and repetition, in aid of his securitisation of the media.
Trump's securitising move Considering the above, Trump's securitising move comes to light. Trump employed securitising language in his tweets by identifying the media as a threat. He additionally presented the referent object to this threat by highlighting a threat to truth, the people and the country. Further, Trump directed his rhetoric (tweets) towards a specific audience, his supporters. To these supporters, Trump is a person of authority. He has credibility and knows the issues. Trump appealed to this audience by using framing in line with their ideological views and views of him. He framed his discourse in ways which appealed to the context in which his securitising language took place: a regime of post-truth, social media and distrust in the "mainstream" news media in the country. By framing his discourse in line with the context, Trump attempted to persuade his audience. He employed several framing tactics, such as the use of emotion, repetition and hyperbole, and may also have trial-ballooned his audience's acceptance into action. By doing so, he further used securitising language, framing the media as a threat that needed to be addressed. Trump's securitisation move is, thus, clear. The former president decisively framed the media as a threat through his many tweets.
Having considered the above, a note on the state of insecurity in the U.S. in relation to Trump's rhetoric requires consideration. As noted in the literature review, Huysmans (2014) argues that securitisation poses a challenge to democracy, as it 'hollows out democratic organisations based on principles of freedom and justice' (p. 3). As the media is known as the 'fourth estate' (Kellner, 2018, p. 97) in democracies, its securitisation further underlines the extent to which securitisation can challenge a democracy. Trump is said to have generally engaged in a 'politics of insecurity' (Rojecki, 2016, p. 76). As Huysmans states, we need to consider when 'insecurity is becoming the energetic principle of politics. If so, for anyone invested in democratic politics alarm bells should go off' (p. 5). By consistently tweeting and framing the media as a threat, Trump regularly spread insecurities not only among his supporters but also within the news media. As noted in the context section, the climate for journalists in the U.S. has become 'problematic' (Reporters without Borders, 2019a), in large part due to Trump's securitising rhetoric, which has clearly spread insecurity not only among his supporters in relation to the media, but also among those which he labels as a threat. This further highlights the impact of Trump's use of securitising language.
Finally, while this article's aim was not to prove whether Trump was successful in his securitisation of the U.S. news media, recent developments may suggest that this is the case. The protests/riot and storming of the Capitol demonstrate clear lack of belief in the legitimacy of the 2020 election results reported by the media, and highlight the trust placed in Trump over the media by a large portion of his audience -his loyal base. Additionally, the physical, verbal and symbolic attacks towards members of the news media further underline Trump's base supporters' activated hostility and belief in Trump's securitising language. While Trump lost the 2020 election, his securitising move has had, at the very least, a real impact on his supporters and their perception of the truth and trust in the media, and contributed to a poorer climate for U.S. news media.

Conclusion
This article argues that Trump conducted a securitising move against the U.S. news media. This finding has become apparent through the use of discourse analysis informed by the application of framing and securitisation theory to a sample of Trump's tweets. Trump framed the media as a threat in his tweets by using the phrases 'fake news', 'fake news media' and 'enemy of the people'. These phrases were supported by other words with negative and threatening connotations in order to solidify this frame. Furthermore, Trump underlined that the media are a threat to the truth, the American people and, thus, the country, thereby designating the referent object. Trump's audience is his base, an audience who sees him as a trustworthy speaker imbued with authority. The former president used several framing tactics in order to frame the media as a security threat, such as appealing to emotion, and using repetition and hyperbole, which strengthened his securitising language and increased the likelihood of audience persuasion. Trump's framing seems constructed to resonate with his audience, as he appealed to their prejudices, drawing on already established schemas and views which he then exacerbated. He effectively tapped into the contextual situation in the U.S., of a polarised political and media environment, notions of liberal bias to what he designated as the 'mainstream media', high levels of distrust in the media, a post-truth zeitgeist and the increasing use of social media as a source for news. Thus, over time, Trump produced securitising language in all of his "anti-media" tweets. He framed the media as a threat to truth and country, thereby performing a securitising move.
This article adds to securitisation theory by filling two gaps in the securitisation literature. The existing securitisation scholarship has not explored the conceptualisation of the media as a threat, while scholars have not explored Trump's securitisation of the media. More importantly, however, an investigation into the potential securitisation of the media in a democratic state is worth studying because the designation of a vital cornerstone for democracy as a threat has consequences for the functioning of that democracy. The news media perform vital functions in democracies, such as informing citizens of necessary facts, informing political discussion and debate, exposing wrongdoings and abuses of power, and holding leaders to account. When the media are unable to perform these functions because they are distrusted, stripped of legitimacy and seen as threatening, the public suffers from lack of objective information, informed political debate and even an understanding of objective reality. As noted, the climate for reporters in the U.S. has worsened in part due to Trump's rhetoric (Reporters without Borders, 2019b), and journalists have been threatened, intimidated, physically accosted and shot (Reporters without Borders, 2019b). Trump consistently spread a message of insecurity and, as Huysmans (2014) contends, politics of insecurity and securitisation itself threaten democracy and its organisations (pp. i-5).
There are several ways future research could build on and further the study of Trump's securitisation of the media. Future research could incorporate other methods of discourse than Twitter, such as public statements, comments, and discourse uttered at campaign rallies and interviews. The securitisation of the media could also be studied in relation to other actors. Further, potential future studies may wish to explore how to desecuritise the media, possible counter-securitisations by the news media, as well as study potential contestations by the news media in response to the securitising move outlined above. The Washington Post's slogan 'Democracy dies in darkness' (Fahri, 2017) that emerged in February 2017 was mistakenly understood to be a rebuke to Trump's hostile rhetoric towards the media. This turned out not to be the case. Securitisation theory, however, would benefit from a study of occurrences of actual contestation.
Notes 1 There has been disagreement regarding which words to use to describe the events (see for example Bauder, 2021). 2 Trump was impeached by the House of Representatives but acquitted by the Senate (Cowan, Morgan and Brice, 2021;U.S. Congress, 2021). 3 Some tweets between 20 and 27 January 2017 are not available, which is a limitation when it comes to tallying the frequency with which terms and phrases have been used in tweets.
One of the tweets posted by the President on 24 January 2017, which has since been deleted by him, was retrieved from Kalb (2018, p. 1). The remaining tweets in this article were obtained from www.thetrumparchive.com. 4 Trump's relationship with Fox News seems to have deteriorated to some extent since he lost the 2020 election (Wilson, 2020). Trump has more recently advocated for his supporters to tune instead into the 'far-right media outlet' (Wilson, 2020) One America News Network (OANN). Reportedly, Trump feels betrayed by Fox News for how it covered the election results (Wilson, 2020), for projecting Biden's win in Arizona, and for calling Biden 'president-elect' among other things (Sumlin, 2020).