A Social Identity Approach to How Elite Outgroups Are Invoked by Politicians and the Media in Nativist Populism

0162-895X © 2021 The Authors. Political Psychology published by Wiley Periodicals LLC on behalf of International Society of Political Psychology. Published by Wiley Periodicals, Inc., 350 Main Street, Malden, MA 02148, USA, 9600 Garsington Road, Oxford, OX4 2DQ, and PO Box 378 Carlton South, 3053 Victoria, Australia This is an open access article under the terms of the Creat ive Commo ns Attri bution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. A Social Identity Approach to How Elite Outgroups Are Invoked by Politicians and the Media in Nativist Populism

. The party family of nativist populists is therefore similar to those referred to as "populist radical right" or "right-wing populist," which are primarily categorized by their ethnic conceptions of national publics (Rydgren, 2017). However, we use the term "nativist populism" (NP) as it makes clear the two cleavages used for categorization (Gagnon et al., 2018), and that only the sociocultural dimension has been used (Bergmann, 2020). This provides a more transparent label and also has the benefit of being able to encompass populist parties combining sociocultural nativism with socioeconomic socialism (e.g., Five Star Movement) which are difficult for traditionally understood right/left labels.
Within NP rhetoric, we analyze the strategic use of social identities because their importance for people's self-definitions (Tajfel & Turner, 1979), and inherent contingency and adaptability of meaning (Reicher, 2004), mean they are a key site of discursive work in any political project (Reicher & Hopkins, 2001). Such analyses are essential due to the fact that adopting negative attitudes towards outgroups requires an ability to argue one's reasons effectively (Wood & Finlay, 2008), and so, the more we understand about how such attitudes are articulated, the more we can improve countermeasures to promote positive intergroup relations in society. We will first position our article within specific research strands of the social identity perspective, which focus on processes of mobilization and change, then discuss the importance of understanding the use of elite outgroups in NP and outline our research aims.

Mobilizations of Social Identity
Our approach to analyzing NP rhetoric uses perspectives which consider the mobilization of social identities for contestation and change (Hopkins & Kahani-Hopkins, 2004;Reicher & Hopkins, 2001) and the identity entrepreneurship of leaders Steffens et al., 2014). These developments understand social identities as being constructed, and used performatively, in the service of strategic political goals (Elcheroth & Reicher, 2017). The social categories and contexts which underpin any political action are therefore not taken as a priori understandings, but rather, as active processes shaped by social actors who seek to alter shared understandings of social reality (Mols, 2012;Reicher, 2004) and achieve a different set of social relations in the future (Reicher & Hopkins, 2001).
Relatedly, the social identity of leadership argues that effective leadership occurs when leaders and followers see themselves as sharing a social identity (Steffens et al., 2014). However, the perception of a shared social identity between leaders and followers can be the outcome of group leaders acting as "entrepreneurs of identity" in which they construe the nature of intergroup relationships in line with their goals . For example, rhetorically aligning identity content to different political projects was investigated with the emergent national identity of "Northern Irish" (McNicholl et al., 2019). Different political parties could depict its "true" meaning in ways which invoked a shared identity congruent with their ideological aims. It could be construed by a centrist, antisectarian party as a superordinate identity that represented unification, and therefore, that the public supported moving away from contentious Irish/British demarcations. In contrast, a hard-line unionist party could claim it represented a subordinate, regional identity under Britishness, and therefore, demonstrated public support for their unionist political project (McNicholl et al., 2019). Shared social identity is the basis for influence as it anchors perceptions of a given social context from which leaders can persuade followers to enact the group goals they put forth (Steffens et al., 2014).
Another key factor in the success of any leader's group mobilization is their perceived prototypicality: the extent to which followers perceive a leader to be representative of the ingroup (Steffens et al., 2014). Importantly, this representativeness needs to be understood in relation to the relevant dimensions of value within a particular social context (Reicher et al., 2005). However, because any social context is a combination of both objective factors and subjective perceptions, it is open to rhetorical influence (Mols, 2012). Judgments of prototypicality are therefore not just a function of how attributes match up to the objective features of a social context; a leader can also influence contextual understandings so that the relevant dimensions of value are those which are contained within the bounds of their project (Reicher et al., 2005).
It is important to note that, while the agency in meaning making by leaders is emphasized, a social identity approach to leadership does not assume simplistic, top-down processes of acquiescence by potential followers. It is acknowledged that people are generally attentive to, and skilled at reading, the credibility of social identity claims (Abell et al., 2007). This attentiveness is equally applied towards those vying for leadership, including NPs. For example, experimental evidence suggests attempts by mainstream politicians to emulate populist rhetoric produces a backlash, resulting in participants perceiving a reduction in their legitimacy (Bos et al., 2013). This highlights the need for analyses which engage with the social meanings, boundaries, and contexts that NPs seek to mobilize, so as to better elucidate the group understandings between leaders and followers.

Elite Outgroups as Supply-Side Tools
Despite an initial tendency towards demand-side research in populism (Rydgren, 2007) recent years have seen more supply-side investigations which focus on the attitudes and/or grievances populists seek to cultivate within voting publics, particularly those involving intergroup relationships (Mols & Jetten, 2020). Social identity analyses have contributed to these developments by demonstrating the rhetorical functions elite outgroups have within cultivating anti-immigrant attitudes. This work includes analyses of modern fascist parties who also make anti-immigration a mainstay of their political projects, although from a more explicit biological racism than NP parties (Wood & Finlay, 2008).
For example, fascist party leaders have been found to use elite outgroups to distance themselves from accusations of fascism by constructing an elite conspiracy which is "anti-White" and directs immigration and multiculturalism for its own benefit at the expense of the public (Johnson & Goodman, 2013;Wood & Finlay, 2008). Similarly, NP party leaders depict covert, mutually beneficial alliances of elite and immigrant outgroups to construe "ordinary" public ingroup members as the true victims and justify their harsher immigration policies (Mols & Jetten, 2016). Elite outgroups can also serve to defend one's position by depicting criticism of controversial anti-immigration campaigns as reflecting elite interests rather than any true racial prejudice by the NP party (Durrheim et al., 2018). Finally, NP leaders can construe their anti-Islam positions as ingroup prototypical by depicting an elite who exists above the populace, in contrast with the NP politician who shares the public's supposed concerns (Rooyackers & Verkuyten, 2012).
We seek to extend this line of research by addressing gaps in understanding the strategic function of elite outgroups in NP rhetoric. This is because previous work has focused primarily on the cultivation of prejudice, that is, the direct articulation of elite outgroups with anti-immigrant agendas (e.g., Durrheim et al., 2018;Mols & Jetten, 2016). However, demand-side research has found up to a third of European NP voters do not hold anti-immigrant attitudes (Stockemer et al., 2021) and that some voters can support NP parties despite holding culturally liberal immigration attitudes (Van Hauwaert & Van Kessel, 2018). This indicates potential uses of elite outgroups not directly linked to cultivating prejudice and which aim towards mobilizing such voters. In addition, questions remain over certain tensions within NP antielite rhetoric. NP leaders claim to be representatives of "the people" versus "the elite," yet most could be understood as elite, in terms of wealth and/or political influence, to those they decry. Communication scholars have also suggested a paradoxical relationship for NP party leaders and the media; they require mainstream publicity yet receive it in a mostly negative way . It is likely that each tension requires subtle interpretations of inter-and intragroup relationships for NPs to legitimize their positions.

Media as Identity Entrepreneurs
A final consideration is that NP politicians may not be the only mobilizers of social identities. An analysis of 10 European media systems found that while news organizations were largely restrictive in platforming populists, many used their own antiestablishment and anti-immigration rhetoric (Wettstein et al., 2018). The media production of NP content is an important supply-side factor as evidence suggests populist media that blames elites can affect attitudes by increasing political cynicism (Sheets et al., 2016) and increase the voting intentions for populist parties (Hameleers et al., 2018). However, the media-as-identity-entrepreneur is a topic which has been somewhat neglected within social identity research. At times they are considered within the wider national rhetorical context (e.g., Portice & Reicher, 2018), but their processes as identity entrepreneurs are rarely given equal focus to that of political leaders. Of course, each entity has different goals. News organizations do not seek the leadership of political office. However, like political leaders, they do seek to influence the sociopolitical understandings of the national public, particularly through the use of opinion and editorial articles (Coppock et al., 2018). In addition, ingroup prototypicality and social influence are not dependent on being perceived as a group leader (Turner et al., 1987). Any ingroup member can be seen as more or less prototypical and influential in a given context (Reicher et al., 2005). There is therefore potential for considering media organizations as social identity entrepreneurs in their own right.

The Present Study
Using a qualitative analysis of nativist populist rhetoric across three country cases, this study contributes towards supply-side research in two ways. First, we analyze shared processes of how elite outgroups are used for mobilization in addition to the cultivation of anti-immigrant prejudice. Second, we extend the focus beyond NP political leaders to consider news organizations as mobilizers of social identity.
Our interest in shared strategies is due to the concept that the emergence of the NP party family in Europe was aided by parties observing and adapting successes from one national context to another, which resulted in the propagation of potent, cross-national ideological master frames (Rydgren, 2005). This also suggests the possibility for common processes of dissemination and legitimization. This could help identify influential mobilizations through assuming that shared processes across contexts indicate that NPs themselves, either implicitly or explicitly, consider them key for social influence. Other social identity analyses have identified shared strategies across national contexts (Mols & Jetten, 2014, and such work provides important comparative perspectives to past tendencies for single-case studies in the populism literature .

Source Selection
As our concern was for the effects on liberal democracies, we sought democratic country cases with significant NP successes. Due to researcher language limitations and issues with alteration of meaning from translated materials (Liddicoat, 2009), we were restricted to English-speaking countries. We selected the United Kingdom, United States, and Australia. The selection criteria for the political parties related to the definition of nativist populism; parties needed to have deployed antielite and anti-immigration rhetoric. The parties were, respectively, the United Kingdom Independence Party (UKIP) under Nigel Farage, the Republican Party under Donald Trump, and Pauline Hanson's One Nation Party (One Nation). All three leaders used rhetoric which was anti-immigrant and referenced public discontent with democratic representation due to political and/or financial elites. The year selected was 2016 as each country had a national campaign with a NP success; the U.S. Presidential election, the United Kingdom's referendum on European Union (E.U.) membership, and the Australian Senate elections. In relation to the latter, while One Nation's success was a comeback for the party and its leader, the Australian case was not as significant a success as the other two. We chose to include it to reduce the limitations which would result from focusing only on the United Kingdom and the United States in this year. This is due to the multiple connections between the Brexit and Trump campaigns, both from the Cambridge Analytica scandal (Cadwalladr, 2018) and the fact that Farage campaigned for Trump in the United States (Wilson, 2017). Including Australia allowed shared mobilizations to be assessed outside the bounds of these connections.
The Nexis UK database was used to review newspaper op-eds and their use of NP rhetoric. Tabloids were given focus as they tend to take stronger anti-immigration positions and have higher levels of antielitism (Wettstein et al., 2018). The aim was to have the highest circulation possible to maximize influence, while balancing against the effects of ownership structures (Day & Golan, 2005). That is, the intention was that findings would point towards shared strategies of NP media, rather than shared strategies of a media owner over their subsidiaries. For the United Kingdom, the Daily Mail was selected. During the time period it was the second largest national tabloid (Ofcom, 2017) and was highly critical of political elites and immigration. The Daily Telegraph was selected for Australia. It had the largest tabloid audience during the time period (EMMA, 2016) and focused on anti-immigration and animosity towards cultural/political elites. The U.S. context proved more difficult as tabloid newspapers are less common. USA Today was the only nationally circulated print tabloid. However, it was neither particularly antielite nor anti-immigrant, and the major national broadsheets took a negative stance toward the Trump campaign's rhetoric on both (Shorenstein Centre, 2016). The Washington Times was selected due to its strong criticism of immigration and elites. While this tabloid only had a print circulation of 50,000, it also had an online version with 7 million monthly readers (Pew Research Center, 2015), which was the version used.
The materials selected were political speeches and newspaper op-eds. Further details on data collection are contained in the online Supporting Information, along with an OSF link to the full data set. The total number of speeches were Hanson (2), Farage (5), and Trump (5). The op-eds were Daily Mail (20), Daily Telegraph (21), and Washington Times (24). The total rounded word counts were 42,000 for the speeches (111 pages) and 52,000 for the articles (138 pages).

Analytic Process
We adopted a constructionist epistemology for the research aims as it assumes language is used to construct, rather than reflect, reality (Willig, 2017). Thematic analysis was the method selected due to its compatibility with using established theoretical frameworks and its flexibility to search for patterns of meaning across the data set. We followed guidelines in Braun and Clarke (2006) and Terry et al. (2017). Please see the online Supporting Information for more detail. Two research questions guided the analysis: (1) How are elite outgroups used within NP rhetoric? (2) Are there a set of common strategies in how elite outgroups are used within NP rhetoric?

Analysis
We first considered what social categories were invoked by the sources. Common codes for all country cases were: "elite," "public," "media," "nation," while country-specific codes included, "Remain," "Brussels" (United Kingdom), "special interests," "Washington establishment" (United Sttes), and "Labor elites," "suburban Australians" (Australia). We next considered who the ingroup was when contrasted with elite outgroups. This was most commonly a generic national public (e.g., "the American people," "Australians," "the British public"), which also invoked notions of work and ordinariness (e.g., "ordinary people," "ordinary working people"). Elite outgroups were therefore understood as any entity that possessed significant power (political/social/financial) and, further, used this power self-servingly to increase personal wealth and/or status at the expense of those without power.
We then moved to consider how these social identities were used for influence. We sought to understand both the form (the particular rhetorical intergroup constructions) and the function (what social identity mechanisms the forms attempted their influence through). A number of strategies were judged to hold across the country cases, and due to the interest in patterns of shared meaning across the data and space limitations of the article, only these aspects will be discussed in detail. We do not wish to give a false impression, however; there were as many differences as similarities. Some examples were, in Australia, Hanson depicted the nation being sold off. This was through construing the sale of farmland to international investors as a "foreign takeover" of public land and assets. In the United Kingdom, both sources depicted E.U. membership as national subservience to authoritarian control. The E.U. was construed as being able to "impose its will" on the United Kingdom without democratic accountability. In the United States, Trump depicted the nation as having been "cheated" by foreign governments due to unfair trade deals such as NAFTA or through paying the majority of costs towards military alliances such as NATO.
We constructed three themes across the data set: "Colluding Elites," "A Shared Fate," and "Consensus Supressed." To evidence the claims of shared social identity processes, a number of extracts have been provided. However, we would encourage the reader to refer to the full data set to further assess validity. For each extract, the source name, date, and country are given.

Colluding Elites
Self-serving interest provided an explanation for why elite outgroups acted the way they did, but this was only a partial story across the data. The first theme related to how they acted. It existed in two main forms; first were depictions of working in concealed coordination: The Liberal and Labor parties pretend they are in opposition, but we are starting to see that they are one and the same. Good governments are made by only good opposition and in Queensland Australia we have neither. (Hanson,04.07.15,Australia) The central base of world political power is right here in America, and it is our corrupt political establishment that is the greatest power behind the efforts at radical globalization and the disenfranchisement of working people. (Trump,10.13.16,United States) Here Hanson construes the two main Australian parties as part of the same political elite who hide this through "pretend" opposition. Trump similarly constructs the U.S. political system as a singular, collaborative "corrupt political establishment," but he takes it further onto a global scale as "the central base of world political power." Thus, each characterizes the political system as one beset by systematic, pervasive collusion. The other form of the theme implied a desire to mislead the public through depicting ulterior motives for actions and/or utterances. This was commonly applied to mainstream political parties and news organizations but interestingly, in relation to the latter, this could come from the newspaper sources themselves: With an election that threatens the status quo for the establishment, the curtain has finally been completely pulled back, leaving no room for imagination about the condition and intention of establishment journalism. Case in point: an editorial from The New York Times, remarkable not just in the absence of any sort of effort to appear sober in its assessment of Donald Trump, the Republican presidential nominee, but a shocking … screed targeting his supporters. (Washington Times,08.24.16,United States) The Washington Times construes Trump's campaign as a threat to a singular "establishment." It uses a common mythos of opaque control ("curtain pulled back") which has been revealed to show the true aims of other media organizations as being an informational arm of this elite monolith ("establishment journalism"). The implications are that because such media organizations are part of the same "establishment," they work to protect the same interests and will deceive the public towards those ends.
This was the most prevalent theme in the data. It could at times be invoked through a single sentence or phrase (e.g., "media-savvy propagandists," "the Clinton machine"), which appeared to serve as repeated flags to the type of communication that elites engaged in towards the public. Before moving to the potential social identity functions, a key point to note is that, although commonly applied to political and media organizations, a variety of national and international organizations within the public and private sector could be depicted as working in secret coordination and/or with ulterior motives. Each aspect is italicized for clarity: Trying to engineer cultural change is all it has left, a strategy exercised through influence in universities, schools, government nanny-state agencies and media outlets. …. When the elites talk about "the need for widespread cultural change," what they are actually doing is trying to transform human nature-a dangerously authoritarian process. The Daily Telegraph depicts universities and schools working alongside government and media organizations towards a singular "authoritarian process." Hanson depicts a combination of intra and international collusion, where a previous administration worked with the United Nations to remove people's freedom and, subsequently, that "the media and politicians" have continued to conceal this from the public. These extracts demonstrate that collusion enables a greater expansion of the boundaries for who is categorized as belonging to elite outgroups. This feature of outgroup boundary expansion meant that the theme was a flexible tool which could be used to defend political arguments: We have the unholy modern alliance of big corporate business and big politics and overpaid experts who work for the IMF and OECD and the IFS. A group of people … in conjunction with our Prime Minister and our Chancellor, telling us that if we left the European Union we would be worse off. (Farage, 06.06.16, United Kingdom) In a brilliantly organized campaign, big business leaders, universities, "elder statesmen" and the Civil Service … have been corralled into making the case that it would be a catastrophe if Britain left the EU. (Daily Mail,02.04.16,United Kingdom) Farage constructs ideas contrary to his as an "unholy modern alliance" of corporate, political, and institutional elites who work "in conjunction" with the government. The Daily Mail similarly construes a campaign which has been "corralled" (implying organization by some external agency) into speaking against leaving the European Union. Therefore, rather than independent assessments of the same issue leading to some convergence, elite collusion is constructed as the reason the organizations have taken positions contrary to each source.
As stated previously, each national context had specific categories invoked as elite outgroups which were fairly irrelevant cross-nationally. However, abstracting the categories into a common understanding of "the elite," and taking all the extracts together, we argue that a similar social context was being constructed and continually reinforced in all three countries. The use of "colluding elites" implied that the public ingroup existed in a suspicious social context where most information was not produced honestly, but rather, in covert coordination to deceive and manipulate the populace. As can be seen, not only could these invocations be used separately from nativist agendas, but even when linked in a similar manner to previous findings (e.g., Mols & Jetten, 2016), there were additional social identity functions to consider in relation to this cultivated social context and its bearing on intragroup relationships. These narratives appeared to function to delegitimize or dismiss large amounts of a society's information. That is, construing a range of organizations as working together for the same elite interests appeared to undermine social trust in those institutions and the information they produced. The use of this theme then can be understood as vying for control of the ingroup informational influence-who the public should categorize as other ingroup members and with whom they should therefore seek to validate their beliefs (Turner et al., 1987). This aspect was particularly clear in a single instance by one of the tabloids. While the theme was almost exclusively applied to large social institutions and/or politicians, it could also be used with other "ordinary" ingroup members: Make no mistake, these supposed anti-Trump riots are not organic nor are they natural; they are the result of leftist organizing using paid stooges … So let's have a few facts about who has been organizing this from the beginning. In March, Kelly Riddell of this newspaper reported the leftist agitating group MoveOn.org, which has endorsed Bernie Sanders and is funded by Hillary supporter George Soros, sent out a fundraising email after liberal agitators managed to shut down a Trump rally in Chicago. (Washington Times, 06.06.16, United States) The "anti-Trump riots" are not concerned citizens who are exercising their right to protest and express their political concerns; they are "liberal agitators" under the direction of political ("Sanders," "Clinton") and financial ("George Soros") elites. The Washington Times seeks to recategorize protesting public ingroup members as representatives of the outgroup elite to reduce, or eliminate, the ingroup informational influence that would otherwise exist.
This theme then speaks to how NPs work towards a degradation of pluralistic values within liberal democracies in general. Any social entity which criticizes NP positions, whether mainstream political parties, large social institutions, or even other "ordinary" ingroup members, can be construed as part of elite outgroups. Further, that these constructions were deployed by both the politician and media source in each country is important in understanding these as attempts at creating a more generalized shared understanding of the social context. This can also provide a more nuanced approach to understanding statements like "alternative facts" 1 for those who share the NP social perspective. Here, "facts" have been produced in a manipulated informational environment that serves to protect elite power; thus, they are categorized as belonging to an outgroup elite and erroneous by virtue of that construal. In response, "alternative facts" are needed to fight back against this social context. They are facts which are an alternative to elite outgroup power and a more legitimate reflection of the interests of the public ingroup. Finally, there were elements in the extracts above which also appeared to position the source in relation to the elite informational collusion. This is an issue to which the next theme explicitly turns.

A Shared Fate: Under Attack and Equally Susceptible
If the social context is construed as pervaded by elite outgroups colluding and communicating with ulterior motives, then this can become a resource for categorizing oneself. Despite their wealth or social influence, NPs could use their constructions of elite informational manipulation to position themselves as sharing a fate with the public, and therefore, as nonelite. We separated these depictions into two subthemes: "Under Attack" and "Equally Susceptible."

Under Attack
The first subtheme was used to construe negative coverage towards the politician or newspaper as attacks produced by the biased information system: The Clinton machine is at the center of this power structure. … Hillary Clinton meets in secret with international banks to plot the destruction of U.S. sovereignty in order to enrich these global financial powers. … With their control over our government at stake, with trillions of dollars on the line, the Clinton machine is determined to achieve the destruction of our campaign. (Trump, 10.13.16, United States) Trump cannot be part of "this power structure" because it aims for the "destruction" of the Trump campaign. Being under attack from the information system which seeks to protect elite power implies that he is not "one of them," distancing him from the elite outgroups. Additionally, a subtle deixis is used to position Trump in a shared fate with the public; the use of "our government" and "our campaign" positions his presidential campaign and the national public together against the elite attacks.
Like the elite collusion narratives, this rhetorical work to depict a shared fate with the public was at times separate to the cultivation of prejudice; however, it could also be used towards those ends: The media and some areas and others have targeted this as a racist rally before any Australian has had the opportunity to voice their concerns. For too long we have suffered trial by media and those with hidden agendas. Let my fellows Australians judge me by what I say but don't deny me the right to have my say. (Hanson,04.07.15,Australia) Hanson constructs collusion and control by contrasting a public who wish simply to "voice their concerns" against a biased information system with "hidden agendas." Hanson blurs the distinction between herself as a politician and the nonpolitician public through use of the phrase "my fellow Australians" and the group plural pronoun ("we have suffered"). Similar to Trump, this deixis positions herself with the public as sharing the suffering from the biased attacks. Although here elite attacks are used to depict Hanson's, and her supporters', anti-immigration positions as reasonable in contrast to the irrational bias of the elite information system ("targeted this as a racist rally"; "trail by media"). These two extracts demonstrate that while construing a shared fate through elite attacks can be used for legitimizing prejudice, it is not interdependent with this agenda (cf. Wood & Finlay, 2008).

Equally Susceptible
Being under attack from elites was not the only way NPs attempted to craft a shared fate with the public. The second subtheme relates to a politician or newspaper depicting itself as equally vulnerable to the elite manipulations: Now we saw yesterday that the PM, the government, are about to send a booklet to 27 million homes. Did you see that? They're going to spend 10 million pounds of our money telling us what we should think and how we should vote, and frankly my view on that is what right does he have to spend our money telling us what we should think. (Farage,04.10.16,United Kingdom) Farage's initial reference to "our money" seems to be in the national sense of being taxpayers; however, this is then blurred through the multiple uses of "us" and "we." This self-public group is then construed as at the mercy of a domineering government that is "telling us what to think" and "how we should vote," which seem to implicitly position Farage and the public as equally susceptible to the government manipulations. Overall, the use of "a shared fate" was more prevalent for the politicians across all three country cases, but it was not confined to this source type:

Under Attack
But it's already too late. The Democrats and their allies simply don't understand trying to beat us into submission reaffirms our determination to end this charade. (Washington Times, 06.06.16, United States)

Equally Susceptible
This is the stark choice facing us in the coming referendum. For months, we have been bombarded with propaganda from one side: the side that speaks for Brussels. (Daily Mail,02.04.16,United Kingdom) The function of this theme potentially relates to how prototypicality, and the influence it confers, result from the capacity to embody what a particular group means to its members. As stated in the introduction, prototypicality judgments stem from contextually relevant dimensions of value (Reicher et al., 2005). In relation to the context of biased elite information, the NPs depicted themselves and the public as under attack, or equally susceptible, to be seen as more ingroup prototypical on those dimensions. The use of these dimensions also had the benefit of deemphasizing the relevance of a politician's or newspaper's own wealth/influence, which could have potentially questioned the credibility of their antielite rhetoric. This could be seen most clearly for Trump who was unique among the sources in having to work explicitly to deemphasize his elite attributes: In my former life, I was an insider as much as anybody else. And I knew what it's like, and I still know what it's like to be an insider. It's not bad. It's not bad. Now I'm being punished for leaving the special club and revealing to you the terrible things that are going on with our country. Because I used to be part of the club, I'm the only one that can fix it. (Trump,10.13.16,United States) Trump has to explicitly acknowledge his elite attributes to enable him to construe alternative dimensions as more relevant for prototypicality. Being a billionaire business owner is diminished as a "club" which is "not bad," and what is emphasized instead is the attacks the elite now "punish" him with. However, this specific prototypicality management results in an inconsistency when contrasted with Trump's other depictions of a shared fate; here he is the "only one" who can solve the problems of the public, who are implicitly positioned as lacking any shared collective agency.
Finally, both of these types of shared-fate narratives could then be used to argue for collective mobilization against the elites: And I thought the sight of a multi-millionaire former Rockstar shouting abuse, making a variety of hand gestures … and not directed just at me, that would have been ok, but directed at our fishermen, that image for me says it all. That actually it's the vested interests it's the rich it's big business. It's those who were doing nicely thank you against pretty much everybody else. We can do better than this. Tomorrow we can vote for real change. (Farage,06.22.16,United Kingdom) For a subordinate ingroup to challenge that of a dominant outgroup, two status perceptions are necessary: illegitimacy and insecurity (Tajfel & Turner, 1979). Depictions of elite corruption provide status illegitimacy, but elite attacks provide the insecurity. The fact that elite outgroups must resort to open and sustained attacks imply that the NPs represent a genuine threat to the status security of the dominant elite outgroups, and, further, that the subordinate public ingroup has a chance to challenge this power through the campaigns advocated by the politicians and the media.

Consensus Suppressed
While we have primarily focused on the use of elite collusion narratives in addition to antiimmigration agendas, the final theme captures how the context of biased elite information could be used directly towards such aims. This was by contesting the normative understanding of the ingroup identity through constructing beliefs about what others believe. Previous work has demonstrated how NPs can invoke a vague public consensus to distance oneself from the anti-immigration claims being made and to legitimize these attitudes as socially normative (e.g., Mols & Jetten, 2014;Rooyackers, & Verkuyten, 2012).
Within our data, we found similar consensus constructions, but these could also be extended by first claiming that it was only elites who wanted the public identity to be defined in terms of support for immigration and/or minority rights. There was not a single instance of any source construing minority rights as a valid concern for, or lived experience of, "ordinary working" individuals. From this, a counter-consensus, one which was suppressed, could be constructed to further contest representations of "the" public attitude: For decades this elite has stifled debate on mass migration, though it emerges again and again among the public's gravest concerns. The gag must come off. (Daily Mail,05.28.16,United Kingdom) A majority of Americans agree with Mr. Trump's immigration proposals … Despite what the left and their water-carriers like the New York Times would like Americans to believe, their fellow citizens, especially immigrants, also want their families to have a future better than their own. (Washington Times, 08.24.16, United States) Real people are fed up with a sinister new political correctness that has become a form of totalitarianism that threatens people's livelihoods, reputations and freedoms … The vicious condescension of the leftist ruling classes might have succeeded in silencing people who disagree with their cultural vandalism, but it hasn't changed their minds. (Daily Telegraph,03.02.16,Australia) Each extract constructs a level of attitudes "out there" that is actually higher than commonly believed. This is then built on to argue that elite outgroups have suppressed this consensus from being widespread public knowledge. The summative idea is "more people think like this than you are being led to believe." Thus, constructions of elite outgroups who suppress dissent to hegemonic narratives allow claims of a resistant public identity which holds opposing attitudes and "truly" represents it.
These contestations may be useful due to two potential functions based on intragroup processes. For individuals who are sympathetic but undecided, constructions of societal consensus provide a reference of ingroup beliefs to influence their deliberations (Turner et al., 1987). That is, contesting the depictions of NP positions as controversial or fringe suggest that readers/listeners should actually give serious consideration to them. Second, for current supporters, these constructions validate any controversial attitudes they hold. Categorizing others as similar to self can increase expectations of support and motivation to act (Stott & Drury, 2000), and so perceiving one's attitudes as widespread can reduce the fear of social sanction in continuing to hold, or to further express, one's views.

Discussion
We presented three themes demonstrating the use of elite outgroups in NP rhetoric. First, depictions of elite collusion created a social context of mistrust to vie for control of the ingroup informational influence. Second, ingroup prototypicality was managed in relation to this context through dimensions relevant to it (sharing a fate from elite attacks and/or being equally susceptible to elite manipulations). Finally, ingroup norms were contested by claiming positive attitudes towards minority groups were representative of "the elite" and that the "true" public attitudes were suppressed.
Overall, the findings from our comparative analysis lend support to the concept that not only have NP parties adopted a potent, cross-national master frame of antiestablishment populism and anti-immigrant nativism (Rydgren, 2005), but they have adopted similar strategies of mobilization to cultivate this ideological frame within voting publics (Mols & Jetten, 2014. Crucially, however, our study extends previous research in three important ways. First, social identity analyses have demonstrated the use of elite collusion narratives for cultivating anti-immigrant prejudice (Johnson & Goodman, 2013;Mols & Jetten, 2016). We establish functions in addition to this goal. The fact that these narratives occurred across the source types in each country shows they can also have a more generalized function. This is to create a social context of suspicion in intragroup informational exchange, which enables NPs to vie for control of the ingroup informational influence. This is a flexible tool which can expand the outgroup boundaries and defend one's position on any given issue. This therefore contributes towards understanding the mobilization of support from those NP voters who do not hold anti-immigration attitudes (Stockemer et al., 2021) because elite collusion narratives can be used to justify a range of socioeconomic and sociocultural policies. However, they also have important implications for all supporters of NP parties as they relate to the degradation of pluralist values within democratic societies. The expectations to agree with those categorized as ingroup others is a key aspect of the factual foundation for democratic functioning (Turner et al., 1987). It is an important part of the process of how ingroup criticism can produce ingroup change (Haslam et al., 2012). If NPs sow doubt that national entities (e.g., news media, public service institutions, other citizens) can be expected to be sources of mutual influence on a variety of issues, then this reduces the range of criticism and accountability and undermines the pluralism of voices on which liberal democracies depend.
Second, previous work has shown how NPs construe ingroup prototypicality by embedding the self within the public in relation to holding anti-immigration attitudes (Durrheim et al., 2018;Rooyackers & Verkuyten, 2012). We develop an understanding about how NPs can embed themselves within the public by showing that cultivating a social context of biased elite information enables NPs to position themselves and the public as suffering together from its attacks and/or manipulations. The emphasis on these dimensions also reduces the relevance of their other potentially elite attributes and helps explain why they do not pose a threat to ingroup representativeness.
The dimensions of "under attack" and/or "equally susceptible" can therefore be understood as comparative fit based on a common fate. In certain contexts, common fate perceptions can supersede objective characteristics in categorizing individuals as a single social entity (Condor, 2006).
Third, we extend how public-consensus depictions can be used through their articulation with collusion narratives. Previous analyses summarized the invocation of ingroup consensus for antiimmigrant agendas as "daring to say what most people think" (Mols & Jetten, 2014, p. 84): our extension to this is "but are being prevented from saying." That is, NPs can enhance public-consensus constructions, with narratives of elite suppression, to further cultivate anti-immigrant attitudes as normative for the ingroup.
In relation to populism by the media, we contribute to the supply-side literature by demonstrating that news organizations can be treated as social identity entrepreneurs (Reicher et al., 2005) in their own right. However, as previously noted, there are important goal distinctions between the source types, and so the motivations for why news organizations mobilized social identities in similar ways are not clear. Content analyses have found higher levels of NP rhetoric for tabloid newspapers versus broadsheets and suggested commercial motivations for the differences (e.g., Wettstein et al., 2018) but neglected to address why populist rhetoric would contribute to commercial success. Based on the social identity of leadership framework, and our findings, we suggest a theoretical development. To create the perception that other news organizations are organs of elite outgroups and that their news is produced, not for informing publics, but for manipulating them, is to position oneself as a truer source of information. The more successful such constructions are, the larger a readership one could gain. Relatedly, the more a newspaper can increase its ingroup prototypicality, the more it may come to be seen as an exclusive source of "our news" and increase the loyalty of its readership. Each of these social identity processes, if successful, would lead to a larger and more dedicated readership, from which a news organization would ultimately profit.
The analysis also provides some clarity around the paradoxical relationship of NP party leaders requiring mainstream media publicity yet receiving mostly negative coverage from it . Mainstream media platforms, and any negative coverage received, can both be beneficial because negative coverage serves the constructions of being under attack from an elite information system. Negative mainstream coverage is therefore only a paradox when the constructions of the social context, and the ingroup prototypically dimensions which result from it, are missing. Further, while a number of national media outlets may provide critical coverage of their NP parties, the analysis demonstrated that not all do. Some of the extracts showed there could even be a close mirroring of arguments within each country. A potential explanation for this is, due to ideological similarities, the sources were likely inclined to read/listen to each other's public arguments. This points towards an intranational extension to Rydgren's (2005) propagation concept by suggesting it is not just NP politicians/parties that learn from one another. In so far as media organizations seek to produce NP content, they can observe the political parties in their national context and adopt and adapt arguments to suit their own ends, and vice versa.
Finally, recent work has suggested links between conspiracy theories and populism which can be used to shift blame onto elites (Hameleers, 2021). Although there are similarities with the analysis presented here, we argue the findings are distinct in two important ways. First, we analyze social identity functions beyond the cultivation of blame towards elites. Second, we take a different analytic orientation to the rhetorical use of elite outgroups. We have chosen to label the NP claims as narratives of "elite collusion," rather than the more normatively laden "conspiracy theories." This is in line with a social identity perspective which seeks to analyze the social realities of others in a way that elucidates the reasons they may have for seeing the world in the way that they do (Haslam et al., 2012). This is not to validate the NP collusion narratives but to highlight how implicit judgments of irrational political claims effect the analytic orientation taken towards leaders, and more importantly, their followers (Durrheim et al., 2018). More nuanced labeling enables greater differentiation between claims which are conspiracy theories (implausible, no evidence to support), collusion narratives (plausible, no evidence to support), and conspiracy facts (plausible, evidence to support). To group all three of these distinct forms together would be to miss the rational reasons some voters may have for engaging with the second and third forms, by virtue of the outlandishness of the first. That is, all the NP collusion claims are naïve theories in the sense that they are vague and lack any evidence, but many of them are mundane because they are plausible in a way that, for example, fascist conspiracies of a global Jewish cabal (Finlay, 2007) or that all Muslims in U.K. society collude to pretend to be ordinary citizens (Wood & Finlay, 2008) are not. To categorize all collusion claims in antielite discourse as "conspiracy theories" is to invoke notions of the impossible and sensational and to obscure consideration of the plausible and mundane. This is similar to arguments made in the study of nationalism-that focusing only on the extreme obscures consideration of the banal (Billig, 1995). In some ways, our notion of colluding elites could be thought of in these terms; they are banal conspiracies. But it is in their banality that space opens up for a considered analysis of their social identity functions and how they may reasonably serve as shared social realities between leaders and followers within NP.

Limitations
The requirement of a restricted data corpus for a manual qualitative analysis meant only comparisons between sources using nativist populist rhetoric were made. It is not clear if the findings are specific to this form as we did not investigate other populist types which are not nativist, but which also invoke elite outgroups (e.g., Podemos, Syriza). The qualitative methodology also meant representativeness could not be assessed and the findings may apply only to our sources for the chosen timeframe.
The findings for the news organizations were likely increased by the focus on op-ed articles, which are specifically written for influence (Coppock et al., 2018). Such explicit social identity mobilizations may not be found within more standard news reporting, or beyond tabloid newspapers. Further, we treated the politicians' speeches as pieces of text. The performative and/or semiotic styles were not considered within the analysis (e.g., Moffitt & Tormey, 2014), and these could have highlighted more differences between the source types.

Future Research
The analysis contributes towards a growing understanding of NP mobilization logic and the resource pool that can be drawn on for social influence. Future research could explore to what extent the rhetorical strategies identified here extend to other country contexts. These could be operationalized for coding schemes used in content analyses in a similar way to those of "corrupt elites." Additionally, the study raises questions around why the NPs used the rhetorical strategies identified. This potentially speaks to a prevalent and pressing matter for liberal democracies, increasing public mistrust for mainstream political parties and governments (Ziller & Schübel, 2015). Large-scale European surveys have found lower levels of political trust are partially explained by higher levels of democratic ideals, due to attentive and critical citizens perceiving those ideals to be frustrated (Hooghe et al., 2017). Therefore, claims of an alienated citizenry represented by an "establishment class" may be more credible to cultivate where a neoliberal economic consensus of mainstream parties has resulted in the removal of state protections for citizens against free-market forces (Mastropaolo, 2008). Such factors potentially contribute to rising populist support in countries with high GDP but also rising inequality (Jay et al., 2019). Assessing correlative relationships between widening inequality, political mistrust, and operationalizations of the social identity mobilizations outlined here could offer additional insights.