Leading up to the Brexit referendum in 2016, one of the hallmarks of the Vote Leave-campaign was the “battle bus”, a red double-decker featuring the slogan: “We send the EU £350 million a week, let’s fund our NHS instead – Vote Leave.” Then British Prime Minister Boris Johnson prominently posed in front of the bus which displayed the official logo of the National Health Service (NHS). The figure of £350 million was questioned by many political actors in fact checks and on social media where it was portrayed as being too high or fabricated. In the aftermath of the referendum over whether to stay in the European Union (henceforth EU) or pull out of it, key actors of the Leave campaign such as political strategist Dominic Cummings (Cummings 2017) and – significantly later and indirectly – Boris Johnson backtracked from the claim of the £350 million (Read 2019). Political pundits partly attributed the narrow margin with which the referendum was concluded to the Leave campaign’s proclaimed lies – such as the alleged promise of the large sum of £350 million freed up to finance the NHS.

The claim of the £350 million has been scrutinized in fact checks by the British press (Hutton 2021; Worrall 2016) and by opposing parties such as the Liberal Democrats who declared it “perhaps the most famous of the Brexit lies” (Liberal Democrats 2019). The latter claimed that “Johnson knew it was a lie, but he put it on a bus anyway” (ibid.). Yet the researcher may ask: to what extent did it matter in the Brexit process that statements about fiscal circumstances were truthful or factual? In a blog post on political strategy from January 2017, Cummings states: “Would we have won without £350 m/NHS? All our research and the close result strongly suggests [sic] No” (Cummings 2017). He sees the NHS bus as a crucial ingredient of a “simple and psychologically compelling story” about “the economy and living standards” and as “the most effective argument” with all demographics. He agrees that there was “no definitive figure” (ibid.), suggesting that for Leave voters, the accuracy of the NHS narrative played only a minor role. More importantly, the broader narrative associated with the NHS bus resonated with voters even as it raised scepticism.

This article takes the NHS bus example – an instance of political communication – as a starting point to investigate the role of factuality and truthful representations in populist discourse. It aims to show how approaches from narratology and folklore studies can contribute to digital anthropology. Based on an analysis of social media debates on the NHS bus, I probe whether factuality is a subordinated category in this and other instances of Euroscepticism, and whether, from an analytic perspective, it is more important to examine the contextual appropriateness of narratives to understand their impact in political debate. In line with anthropological research, I define Euroscepticism as the “wide spectrum of sentiments, political strategies, voting behaviours and opinion poll results that mark a degree of opposition to the project of European integration in both Western and Eastern European countries” (Shore 2021; Ilieva & Wilson 2011). I understand appropriateness here as the fit for specific audiences that includes a range of factors involved in judgements about the conventionality and propriety of statements, including primarily normativity, political stance, and social contexts. I argue that non-factual statements are conventional, accepted, and expected in Eurosceptic discourse, and specifically in the Brexit debate. They are practices of conflicting Europeanization that here take the form of EU policies being harshly critiqued and alternative pathways of development being imagined in relation to Europe – ultimately leading to a departure from the hegemonic European project. From an analytic perspective, conditions of appropriateness for Eurosceptic narratives and, more broadly, perceptions of this appropriateness are vital for an understanding of how they are taken up, circulated, and responded to, especially vis-à-vis the role of social media in populist discourse (Engesser et al. 2017). To examine the role of factuality as a possible criterion in Eurosceptic debates, I draw on work from narratology and folklore studies.

Source Data and Methods

My research is based on an analysis of Twitter debates on the “Brexit bus” as one of the cornerstones of public debate on Brexit. My corpus is approximately 200 Twitter posts from 2016 to 2017 and selected tweets from 2019 to 2021 in which the NHS bus features. Using Twitter’s search API and respective keywords, both textually and visually, I probed into common arguments and claims in the data to investigate different context conditions and evaluation criteria. At the centre of my analysis was the narrative attached to the NHS bus and its claim on the relation between the United Kingdom’s (henceforth UK) membership and domestic NHS financing. This data from the platform Twitter was combined with news articles and blog posts both referring to and being referred to in tweets.

This article begins by discussing the relation of appropriateness and factuality in terms of its epistemological assumptions, making an argument for a contextual and layered understanding of appropriateness. I discuss the role of expertise and factuality for Eurosceptic narratives in relation to their truthiness, defined as “the conviction that something is true based on what feels true” (Wiese 2015: 66). I argue that contingency and anticipation are significant in people’s circulating and taking up of truthy narratives. I conclude by arguing that the contextual appropriateness of narratives is a decisive aspect of their acceptance and circulation, crucial for the analysis of populist discourse.

“Truthiness” as Epistemology

The narrative of the NHS bus is a central example of narratives in the context of Brexit (Reid 2019), which culminated in a close referendum and initiated the UK’s exit from the EU. The narrative highlights an increased polarization in everyday communication on politics (Coleman 2022), specifically illustrating how social media contributed to division in Brexit debates (Brändle, Galpin & Trenz 2022). On its surface, Brexit in Great Britain appeared to represent a fierce opposition between Leavers and Remainers, the former represented prominently by the Conservative Party’s Boris Johnson and UKIP’s Nigel Farage, the latter by the Labour Party and the Liberal Democrats. However, Brexit has later been argued to be the result of an amalgam of power-related issues, among them

…anti-European xenophobia, English exceptionalism, a fixation with borders, the triumph of fake news or demagogic populist nationalism, or imperial nostalgia. These were all elements in the Vote Leave narrative, but other factors were also at work, including a backlash against decades of austerity, anger against the government’s austerity policies, disillusionment with Britain’s membership of the EU and a rough cost-benefit calculation of the economic advantages of remaining. (Shore 2021: 3)

The Brexit process has been impacted by issues of relatedness and belonging (Balthazar 2017), political representation and the layout of the political system (Koch 2017), and general domestic policy (Powell 2017). These issues are tied less to specific arguments or factual debates and more to perceptions, positionality, and specific political conjunctures (Clarke & Newman 2017).

For the NHS bus narrative, decentring factuality as a guiding dimension for evaluating narratives means that one must ask whether Leave-voters in the UK believed the statement portrayed on the bus, or whether they deemed believable the narrative connected to it: the narrative that the UK was a net contributor to the EU and that leaving the EU would bring in additional funds which could be spent, among other things, on the NHS. The NHS bus is one case of many in which political processes have centred around prominent claims that do not withstand closer scrutiny in fact checks. However, based on the findings of this paper, for some audiences it seems to have sufficed that they were “truthy”, that is, felt true enough or pointed to a narrative kernel, that is, an abbreviated or condensed expression of an extended narrative (cf. Kitta 2018: 406 on narrative kernel) that feels appropriate. Statements such as the one depicted on the NHS bus do not need to be fully and precisely believed to fulfil their purposes. They must merely be, to some extent, believable and plausible enough to be taken up and influence political discourse. The increase of questionable1 narratives and statements in the decade 2013–2023, especially regarding Euroscepticism, Brexit, US elections, or the Covid-19 pandemic, has led to a profusion of initiatives on social and news media. Partly state-sponsored, partly established by social media platforms following public pressure to stop the circulation of fake news, and partly organized by NGOs, fact-checking initiatives are at the time of writing a staple ingredient of public debates on policy issues. Platforms such as Twitter and Facebook have begun to flag false news stories and to caution users not to rely on untrustworthy sources. Despite such efforts, stories that do not stand up to closer scrutiny are still intensively shared, often on less regulated communication infrastructures as Telegram or Gab, or in more enclosed fora. In many instances, it does not seem to matter whether statements are factually true. At times, it is even openly acknowledged by political or other actors such as Dominic Cummings that statements are in fact false or not even believed in (Polletta & Callahan 2017; Hodges 2019; Kitta 2018). In such instances, fact-checking faces the challenge that the proven non-factuality of statements does not always change political stances or narratives circulated in online popular media: the mere truthiness of statements suffices.

The term truthiness was first widely established by the comedic infotainment format The Colbert Report in 2005 (Hoskinson 2005). It describes a “modality of apprehending truths emotionally or instinctively, without regard to evidence” (Dent 2019: 180). Truthiness is a central quality of the post-truth era and Dent argues that “rather than indicating a time when people no longer assert truths (as the term would suggest), post-truth signals a particular approach to discovering and authorizing them – an epistemology” (ibid.). Truthiness constitutes new qualities of truth claims as it introduces distinct markers of plausibility and felicity (Austin 1962). These markers are based not on facts or verification but on sociocultural, affective, and contextual appropriateness. Instead of focusing on the ability or inability of actors to critically assess the factuality of claims, a perspective on truthiness as an epistemology or epistemic mode posits that actors are not necessarily gullible or naïve in following unchecked or false narratives (cf. Polletta & Callahan 2017: 393). Rather, in taking up, circulating, or modifying truthy or believable narratives, they consider the possibility of them being false or fake, yet dismiss a dichotomic true/false distinction in favour of a close-enough and true-enough interpretation.

Narratives such as that displayed on the NHS bus are not necessarily really believed but taken as typical expressions or ideal types of specific – and more complex – statements. They contain ideas that are believed to be true, appropriate, or right, such as the statement that the UK would gain back financial autonomy if it left the EU. These various epistemic modes of knowledge production or truth-making bring together belief systems, normative frameworks, and political convictions. The gist of the story, that is, the underlying claim about the autonomy of spending funds otherwise allocated to the EU does not change with the factuality of figures. In probing narratives such as the NHS bus narrative, normativity, political stances, and social contexts are central and overshadow questions of factuality. In his work on conspiracy theories in contemporary Turkey, Saglam speaks about “conspirational socialities” (Saglam 2020: 19) to point to the influence of social factors on the acceptance of debunked facts. He argues that images of masculinity and socially accepted perceptions of world orders can give rise to conspirational narratives in that their underlying normative or political convictions are deemed to be desirable while their worth as truth claims is ignored:

Conspiratorial accounts operate independently of truth. Since they are not truth claims, they cannot be simply refuted by revelations of truth either. Rather, […] their circulation may very well work to forge the political subjectivities of the men involved as well as the socialities they strive to fit into. (Saglam 2020: 21)

Focusing on subjective perspectives in sharing narratives, Saglam illustrates how rejecting conspirational narratives can run counter to social desirability. If one follows an understanding of narratives as social activity (Polletta & Callahan 2017) and practice (De Fina & Georgakopoulou 2008), then it is clear that accepting and rejecting narratives have social implications. In cases where narratives feature in processes of identity construction (Bamberg, De Fina & Schiffrin 2007), questioning narratives viewed as typical expressions of a collective political stance can be interpreted as taking a position against the collective, rather than only against the narrative’s factuality. For instance, narratives ascribing “economic downturns” in Turkey may not be presented as domestic problems but as malign plots by outside actors such as “Western powers or Israel” (Saglam 2020: 19). Sharing such narratives serves to shape and strengthen a masculine sociality of being able to identify such plots. Even if such narratives are questioned, they are nonetheless communicatively shared as they meet embedded conditions of appropriateness, echoing shared political convictions. Viewing the sharing of such narratives (and the lack of dissent towards them) only in terms of their truth claims misses their performative dimensions in enacting social desirability. Similar observations have been made in the context of the 2016 US elections: “By sharing, liking, and commenting on outrageous stories – and by determinedly not questioning their factual accuracy – people signalled that they were savvy, scrappy, and clearly on one side of the partisan divide” (Polletta & Callahan 2017: 404).

Binaries of Friends and Bogeys

These social narratives have been discussed in relation to contemporary legends on social media that “are believable, even if they are not believed” (Kitta 2018: 405) and “express fears and anxieties, warn of potentially dangerous situations, and often contain an element of social control” (Kitta 2018: 406; cf. also McNeill 2018). Narratives such as the NHS bus narrative share a crucial feature with contemporary legends in that they can be openly acknowledged as false, but still enter circulation and shape political discourse. As Noyes shows for the French referendum on the European constitution in 2005, this fakeness can take the form of “phantoms” in debates (Noyes 2018). She describes how in France, the “Polish plumber” entered the political arena in the form of a fear that craftsmen from Eastern Europe would be dangerous competition for French workers in a free EU labour market. As Noyes argues, the plumber was a “bogey, a negative phantom openly acknowledged as false but attributed to popular belief” (Noyes 2018: 426). Such bogeys are considered to be plausible and possible without a need to probe into their factuality.

As tokens (or instances) of specific types (i.e. overarching categories, see Wetzel 2018 for the token–type-distinction) of narratives, individual truth conditions can be imposed by generalized or ideological convictions about policies or world orders. In this regard, the narratives of the NHS bus and the Polish plumber are evaluated not only in relation to facts or utterances directly connected to them as specific tokens. They are additionally evaluated in relation to more abstract propositions about types prevalent in political discourse: the narrative of the NHS bus entails statements about state autonomy vis-à-vis transnational governance as a cornerstone of current populist and Eurosceptic discourse; similarly, the narrative of the Polish plumber assesses policies of labour market integration, resonates with discriminatory ethnic stereotypes and invokes notions of Eastern Europe as an “ontological insecurity trope” (Mälksoo 2019). Both narratives (as tokens) include truth claims that can be fact checked. At the same time, they index alternative conditions of appropriateness vis-à-vis normative convictions and invoke truth propositions for overarching assumptions about narratives (or types). Probing into this token–type-relationship shows how the use of bogeys and more generally of truthy narratives simultaneously invokes different sets of epistemic modes. These modes are tied to sociocultural contexts and operate using embedded evaluative criteria; they are not objective.

Discursively, bogeys often stand in for larger narratives reduced to abbreviated symbolic representation (such as the display of a red double-decker) or kernel narratives (such as the indexing of the “NHS bus” or “£350 million”). Given their media prevalence and akin to memetic discourse, bogeys have the potential to affect communicative processes by occupying cognitive slots, that is, by taking up a space in political debate that prevents other issues from being discussed (cf. Stanley 2015; Taylor-Neu 2020). News stories, fact checks, and social media debates dealt with the NHS bus claims long after they had been debunked; the bus and the figure of £350 million became stand-ins for both Leave and Remain positions. In response to fuel shortages and a lack of lorry drivers in the UK in late September 2021, and in the debate on the NHS reform in November 2021, the bus served as a background for critiques of government policies. One example in which the image of a red double-decker was used on Twitter to satirize its previous use in the Brexit campaign reads:

We have no fuel for this bus, but that’s ok cos we don’t have a driver either #ToriesOut #BorisOut #JohnsonOut #BrexitHasFailed #Brexit2

Referring to the shortage of lorry drivers, this and other posts on Twitter capitalize on the capacity of the NHS bus image to allude to criticism of Brexit policies and their economic effects. A second example revolves around debates over an NHS reform and its austerity measures:

BREAKING: “£350 million a week” turned into the government removing the obligation of the NHS to provide everyone with free healthcare. I still can’t believe the bus lied to us [crying emoticon]3

Here, the NHS bus as a stand-in for Brexit policies and promises is alluded to in later debates. The reference works to signal both a stance in political debate (here: opposing Brexit and conservative politics) and to evoke affective dimensions tied to the lies of the NHS slogan.

The inversion illustrates the recursive binary character of the bus narrative which has moved from a true/false distinction to the level of political, normative, and emotional convictions. Rejecting the narrative evaluates it not only as untrue but as fake, fraudulent, and normatively bad with regard to underlying policy decisions and ethical convictions. Accepting it would mean evaluating it as essentially not-factual but truthy, right in principle, and normatively good. This layered appropriateness brings with it communicative pitfalls which can be strategically harnessed. Fact checks can easily be ignored or flagged as irrelevant when normative stances or political convictions support utterances in principle. Focusing on the factuality of numbers or the accuracy of broad statements can be countered by switching to other layers of appropriateness. In the NHS bus example, this switching took various forms on Twitter: emphasizing the principle of the UK’s autonomy vis-à-vis the European Union (prominently by using the hashtag #takebackcontrol); downplaying the importance of the specific figure (“figures and interpretation was stretched […], but sentiment true”4); and contesting that the statement on the bus could linguistically be interpreted as a promise (“an idea not a promise”5). In this context, questions of appropriateness rather than verification become central. For example, in response to a tweet by Labour politician Jeremy Corbyn on the slogan on the bus, a user wrote:

More Labour lies. Please point [to] the promise of £350 million a week for the NHS. It say[s] LET’S now LET’S is an idea not a promise. But you was elected on a Leave manifesto you & all your Lying MPs where [sic!] elected on that promise & you didn’t even try and deliver.6

In this example, the NHS narrative is evaluated vis-à-vis its pragmatic implications. While the factuality of the claim is not discussed, the focus is on how it could have been interpreted. The slogan’s appropriateness is, accordingly, portrayed in terms of its perlocutionary force, not its referential value. Like conspiracy narratives and outrageous stories from the 2016 US elections (Saglam 2020; Poletta & Callahan 2017), signalling political stance or sociality is foregrounded while factuality is neglected.

Appropriateness, Expertise, and Approximation

Truthy narratives partly rely on affirmations of belief (a narrative is accepted as appropriate as it confirms or approximates one’s own belief or political stance). An additional factor is the increased contestation of expertise (Eyal 2019), specifically in digital media (cf. Polletta & Callahan 2017). Clarke and Newman view an “antipathy to ‘expert’ knowledge” (2017: 101) as one of the major tenets of Brexit debates. In this context, narratives are not only considered to be truthy because they entail elements of scepticism towards institutional expertise such as fact-checking initiatives or tech companies flagging fake news. Shore (2021) cites Michael Gove who, in his function as Lord Chancellor in 2016, elaborated on the scepticism of EU institutions: “I think the people of this country have had enough of experts with organisations with acronyms saying that they know what is best and getting it consistently wrong” (Shore 2021: 8–9). On Twitter, this stance towards experts was mirrored in numerous posts that exclaimed “so sick of experts”,7 “[w]e’ve had enough of experts!”8 and “[I] think we’ve jolly well had enough of these so called legal experts (judges) interpreting laws”.9 Such criticism of experts was closely tied to their perceived legitimacy. The phrase “unelected faceless bureaucrats” or variations of it were used to point to the argument that EU institutions and officials did not democratically represent EU citizens – a cornerstone argument of Eurosceptic debates that resonated with a broader populist stance against “the elites”:

#MichaelGove calls for #referendum on #riemannhypothesis. Question can’t be left to unelected “experts” & “elites”. Must #takebackcontrol10

Such arguments about legitimacy were often coupled with claims of conflicts of interests or even corruption, implying biased or skewed expertise as a result, as these two examples illustrate:

I am listening to the experts: just not the ones who appear to be in the pockets of the EU. #TakeBackControl #VoteLeave11

All I keep hearing is #Experts … how fucking wrong they were at the #Elections!!!! #VoteLeave #TakeBackControl #ITVEURef12

This contestation of expertise and its relation to campaign claims such as the NHS bus slogan are further illustrated by posts mocking the Leave campaign’s predictions and its stance on expert opinions:

Remember when the “politicians” told us to stop believing in experts and instead invest faith in what THEY wrote on the side of a bus? #NHS13

We were warning of this all the way through the referendum debate. Experts said Brexit & Brexiteers bad for NHS – but people believed a bus.14

Expertise and its relationship to populist claims constitute a central facet in the debate on the NHS bus narrative, both in statements sceptical of the role of experts in policy processes and in posts defending them. In evaluating developments after Brexit, the cited social media posts on Twitter hinted at a crisis of expertise caused in part by unfounded claims. Referring to the US elections and the impact of fake news, Dominic Boyer posits an “authenticity crisis” grounded in a disconnect between regimes of expertise and their audiences:

The authenticity crisis pivoted on the revelation that the Erfahrungswelten [worlds of experience] of left- and right-liberal expertise no longer resonated with the aspirations, anxieties and miseries of the white non-urban American working class. (Boyer 2019: 84)

The presuppositions and supposedly rational underpinnings of expertise as an established epistemic qualifier in contemporary societies has been challenged in Eurosceptic processes, US elections, and the Covid-19 pandemic. “[A]uthenticated expertise”, that is, “an expertise that is impregnated with a certain authenticity that – in the eyes of everybody involved – enhances the trustworthiness and reliability of disseminated knowledges” (Beck 2015: 18) is met with scepticism. While it would be easy to dismiss such scepticism as naive or nefarious, it should be understood in relation to discourses on the legitimacy of expertise and what can be termed contingent expertise, that is, the multiplicity of expert opinions on a given topic, the anticipatory and prognostic qualities of expertise and its fragility in complex societies and technological settings. Research from the ethnography of policy has illustrated the construction of rationality and the contingency of knowledge production (Carr 2010; Boyer 2009). Additionally, the increased transparency of institutions and the mediatization of institutional processes have made the resulting contingencies part of public debate. Contradictory and contingent expertises abound.

In the NHS bus example, the dispute about the exact costs of EU membership and Brexit is an example of contingent expertise as it hinged on a myriad of factors and was accompanied by insecurities. Contingencies in knowledge production reinforce truthiness as an epistemic mode. Absolute certainty regarding the costs was an epistemological impossibility as figures could only be approximations and future developments could only be anticipated to some extent (cf. Lockie 2017: 3). Statements on the NHS bus were thus accepted or taken up as legitimate approximations rather than factually true.

The shift away from predictions based on authenticated expertise to a debate inherently sceptical of expert opinions leads to a situation in which contingency plays a central role:

The troublesome function of the NHS bus, then, is to neutralise the discussion of the economic impact of Brexit by increasing scepticism of all factual claims, and presenting all predictions as equally bad. It nullifies the issue by reducing the set of commonly accepted, relevant facts and therefore making it harder to come to judgments about the effects of different policy options. (Reid 2019: 628)

Tendencies towards approximations are not limited to populist discourse but are an increasingly important mode of policymaking in different fields. In climate governance, speculative policies about potentially effective pathways of development and regulation shape future developments without being able to provide certainty (DeLeo 2017). Rather, contingencies and risk are priced into such processes. Here, authenticated expertise is not primarily grounded in categories of factuality or accuracy but in probability and prognosis derived from contingent modelling. This involves “anticipatory problems” (DeLeo 2017) which might occur in the future, and which require pre-emptive reactions in policy processes. In the case of Brexit, the anticipatory problem of a downward economic spiral was projected by Remain-proponents while Leave-supporters championed the potentials of bilateral free trade deals with the United States that were not possible when a member of the EU. Likewise, debates on the economics of the debt union, in other words, the policy that meeting the debt obligations of EU member states is guaranteed by the union, produced drastically different expert statements on anticipated outcomes, ranging from shrill warnings of economic instability to prognoses of stable growth enabled by this policy. As institutionally authenticated experts such as policy analysts and scholars voiced their evaluations of policies, these always included anticipatory elements; contingencies were inherently transparent in the process.

Analytically, anticipation and contingency require considering degrees of confidence, the legitimacy of approximations and the handling of uncertainties. These include modes of calculating and mapping contingencies and handling predictions in policy making, which introduce a broad range of analytic questions pertinent to the evaluation of knowledge and expertise as modelled or envisioned: How do actors explicate these contingencies, how transparent are they regarding uncertainties in their prognoses of future developments, to what extent is their anticipation of future developments grounded in observable data? The concept of truthiness as an epistemic mode acknowledges insecurities and approximations attached to narratives. Crucially, the role of contested expertise in the example of the NHS bus illustrates how different criticisms overlap and amplify each other: the democratic legitimation of bureaucrats in supranational constellations (“unelected faceless bureaucrats”), the limits of expertise in predictions (the inherent contingency of prognoses), and political stance as part of expertise (underlying normative convictions of policies) lead to situations in which factuality matters less than the truthiness of narratives.

Stickiness and Modes of Circulation

An analytic interest in truthiness as a reflexive epistemology must account for how truthy narratives become sticky, that is, how they gain traction and momentum and affect political discourse – even when they are rebutted and shown to be factually untrue. Signs become sticky when they are used repeatedly in a certain way (Ahmed 2014: 91). This pertains to the circulation of narratives such as the NHS bus or the Polish plumber, and the sharing and commenting on them on social platforms. Such narratives can be used as “epistemic obstacles” (Stanley 2015) or as “informatic noise” (Taylor-Neu 2020). The stickiness of the NHS narrative fixed the debate on NHS financing to a specific claim and blocked other potential arguments in this context. Cummings, in reflecting about the NHS bus narrative, outlined this openly: “Sometimes we said ‘we send the EU £350 m’ to provoke people into argument” (Cummings 2017). Provoking “people into argument”, as Cummings put it, with such provocative narratives made them sticky, made them occupy a place in discourse and shifted attention from other topics to themselves. The NHS bus was, even in its satirical uptake by Leave-opponents, an example of a successful and sticky narrative which made visible the opposition and reinforced political convictions. Even the kernel narrative of the bus itself as a symbolic representation constituted an epistemic obstacle that motivated actors to address it. Such stickiness is grounded partly in the allusiveness of narratives which enabled audiences to quickly identify with them (Polletta 2006: 88; Polletta & Callahan 2017: 394). In public debate on Brexit, the bus along with the take-back-control-claim constituted a major point of reference for Leave- and Remain-voters alike. It indexed socio-political positions in a debate characterized by a partisan distinction of pro- and contra-EU.

The narrative about the NHS bus was part of a differential circulation that affected audiences in specific ways. The counterfactual claim of £350 million provided a basis for ridicule and accusations of lies on part of opponents to Brexit, connected to affective reactions of indignation about the referendum outcome and claims for rectification. By contrast, Leave-proponents took up the narrative as a story of autonomy and disengagement from what were perceived to be illegitimate EU-policy processes. Some audiences highlighted truthiness and contextual appropriateness as collectively approved genre conventions for narratives such as the NHS bus while other audiences approached them through the lens of verification and fact checking. Seemingly incommensurable criteria and norms of interpretation and interaction were enacted in discourse on the same narrative. Pro-Brexit tweets with reference to the NHS seldom addressed specific monetary figures of support after Brexit but highlighted political convictions such as statements about fiscal sovereignty and a distrust of EU-bureaucracy:

The NHS needs more money. Let’s fund our hospitals – not private jets for EU bureaucrats. #TakeBackControl #VoteLeave15

Yes we need to fund r NHS NOT THE EU!!! #TakeBackControl #SaveOurNHS16

On the other hand, posts critical of Brexit frequently took issue with the figure of £350 million, including alternative and elaborate calculations:

Or, why Brexit will kill the NHS. The phantom £350 m a week simply wouldn’t cover this. #takebackcontrol17

Brexit could cost the UK £1.27 billion who needs the NHS & it’s £350 m a week when you have sovereignty over everything. #takebackcontrol18

This dissimilar uptake of the NHS narrative illustrates a disjuncture in evaluative criteria for political statements. Narratives create different social roles which are specific to social fields and work with different assumptions about their reception. These social roles and the reactions they provoke follow distinct hierarchies, so that some narratives are stickier than others and can overlay other narratives – as in the cases of the NHS bus and the Polish plumber.

The NHS bus is still referenced in criticism of Tory policies today. While positive evaluations of the narrative have decreased, the kernel narrative remains in circulation and along with it, binary oppositions about underlying normative claims, specifically regarding differences in anticipating economic developments.

Condensing and Extending Narratives

The multiplicity of potential contexts for narratives influences their circulation. An innate quality of narratives such as the one of the NHS bus, is that they can be condensed from very complex forms (e.g., extensive stories or policy briefs about how UK spending will change after Brexit) to the level of simple images and references (e.g. a red bus, the phrase “£350 million”) that allude to, but can also be tied to, everyday concerns. For such narratives, discrepant conditions of appropriateness can exist not only in different contexts but also regarding different complexities of narratives. Extended policy statements or press briefings about the economics of Brexit are evaluated by audiences differently than sticky narratives in condensed referential form. The juxtaposition of expertise and faith in a bus slogan does not require an elaboration on the details of the statement nor on its factuality; here, it is – in the context of a Brexit-critical thread – seen as given that the original claim was false. By contrast, policy briefs as well as media reports throughout Europe have delved meticulously into the details of UK contributions to the EU, enacting different context conditions which required evidence, sources, and sound argumentation.

The complexity of these narratives is especially pertinent for social media platforms such as Twitter. The specific qualities of Twitter afford processes of condensing by retweeting, liking, and commenting. In the process, posts pertaining to the NHS bus narrative primarily mentioned the kernel narrative, that is, a very abbreviated reference to underlying claims and convictions, including mentions of a bus, the sum of £350 million, hashtags referring to the NHS, or a combination thereof. Still today, the search combination of “NHS” and “bus” evokes the policy debate of 2016 and the use of the slogans in influencing public opinion. Condensing such narratives erases some features, so that an allusive kernel narrative about the NHS bus works simultaneously to evoke broad political convictions and beliefs about leaving the EU, and to motivate other audiences to contest the factuality of specific figures which are not pertinent to the former audience. This evocation of differential reception and interpretation is one of the reasons why the narrative of the NHS bus is so effective. It evokes different sets of normative evaluations of narratives which connect to conditions of appropriateness specific to individual audiences. As Gal and Irvine suggest, ideological work is required to scale between different levels (Gal & Irvine 2019), for example, between social media debates on sticky narratives and formal political debate. Part of this ideological work is the construction of differential conditions of appropriateness and modes of circulation. Because they are ambiguous in content and social functions, allusions and kernel narratives require references to context to be extended to broader arguments and themes. In the process of condensing narratives, allusions are inserted into discourse which can boil down to the use of specific emojis (a red bus) or condensed phrases (“NHS bus”). This process is afforded by the context of social media platforms. By, for example, limiting the number of characters of tweets or by providing features of instant re-circulation (retweeting), ambiguous kernel narratives are more likely to enter circulation in political discourse than are lengthy evaluations of factuality or detailed discussions of policy. The brevity of posts on Twitter, while in principle extendable by linking to other platforms, favours allusiveness as a mode of approximation. Accordingly, different audiences enact and enforce different conditions of appropriateness (i.e. those of expertise versus those based on everyday affective experiences) for such references.


The impact of social media on debates in the context of Euroscepticism has been highlighted in recent years, tied to discussions on populism and fake news. Here I have combined scholarship from narratology and digital anthropology to investigate the specificities of narratives in social media, their circulation, and centrally the conditions of appropriateness pertinent to the evaluation of Eurosceptic narratives. Focusing on the example of the NHS bus in the context of Brexit, I have analysed Twitter debates on fiscal policies of the UK and issues of autonomy. Such economic themes are prominent in Eurosceptic discourse, often involving poignant figures and calculations to argue against EU policies or against the EU’s political project. I argue that the factuality of claims has not mattered as much as the political stances or normative convictions signalled by sharing, liking, or commenting. This leads to a situation in which different evaluation criteria are used by different audiences to judge appropriateness, favouring narratives that feel true in some sense, that is truthy narratives as approximations over attempts to arrive at absolute conclusions. Statements on the NHS bus were truthy or believable since they felt true to particular audiences and were compatible with existing belief systems and political convictions. In other words, I find that a central problem of fact-checking and verification in digital contexts is that the true/false dichotomy with which they often operate is framed within genre conventions in which categories of believability and appropriateness are paramount and can be diverse.

The success of truthy narratives has also been due to public scepticism over the contingent nature of expert knowledge as embodied by EU bureaucrats. A narrative approach to such debates enables the analysis of contextual features of Eurosceptic narratives, considering their felicity conditions, divergent contexts, and the allusive aspects of condensed or abbreviated narrative references. This approach entails a focus on appropriateness – defined as the fit to specific audiences which includes a range of factors involved in judgements about the conventionality and propriety of statements – as a central concept for the evaluation of narratives to probe into how they are received. Regarding Eurosceptic discourses and specifically in relation to forecasts of economic and political developments, appropriateness is pertinent to understanding the circulation and stickiness of narratives, even if they are not entirely factual or truthy.

Analytically, this makes it necessary to look more closely at how digital contexts of platforms such as Twitter affect the circulation of narratives, and how they enable them to be scaled up. It is crucial to scrutinize the differential production of the criteria of factuality and appropriateness regarding their underlying conditions which are tied to divergent social fields and socialities. As this analysis of Twitter posts on Brexit has shown, this entails analysing how narratives resonate with existing political convictions, normative frameworks, and belief systems, even if they are proven to be non-factual or false as part of fact-checking efforts. This involves looking into the communicative construction of such criteria of factuality and appropriateness and how they are contested. To understand how different audiences receive narratives either as truthy and believable or as fake and false is then less a question of fact-checking and truth-making but rather an issue of the situated production, reception, and circulation of narratives in digital contexts. I have proposed here an approach that considers the layered appropriateness of narratives. These layers are irreducible to binary distinctions but different aspects are given different weight in context. This constitutes a relational approach to narratives, centrally involving the social roles and overarching sociocultural functions of narratives, as well as their relation to expertise, policy programmes, and normative convictions.


  1. I use the term “questionable” here to index that I am not analytically discerning between true and factual or false and non-factual but rather refer to societal debates revolving around truth-qualities of statements. [^]
  2. T1, September 26, 2021. [^]
  3. T2, November 22, 2021 (now deleted). [^]
  4. T3, December 23, 2017. [^]
  5. T4, October 31, 2019. [^]
  6. T4, October 31, 2019. [^]
  7. T5, November 3, 2016. [^]
  8. T6, December 4, 2016. [^]
  9. T7, November 3, 2016. [^]
  10. T8, June 25, 2016. [^]
  11. T9, June 22, 2016. [^]
  12. T10, June 9, 2016. [^]
  13. T12, August 20, 2017. [^]
  14. T13, November 30, 2016. [^]
  15. T14, June 15, 2016. [^]
  16. T15, June 21, 2016. [^]
  17. T16, September 4, 2016 (now deleted). [^]
  18. T17, October 24, 2016. [^]


Ahmed, Sara 2014: The Cultural Politics of Emotion. Edinburgh: Edinburgh University Press.

Austin, J. L. 1962: How to Do Things with Words. Glasgow: Oxford University Press.

Balthazar, Ana Carolina 2017: Made in Britain: Brexit, Teacups, and the Materiality of the Nation. American Ethnologist 44(2): 220‒224. DOI:

Bamberg, Michael G. W., Anna De Fina & Deborah Schiffrin (eds.) 2007: Selves and Identities in Narrative and Discourse. Amsterdam: John Benjamins. DOI:

Beck, Stefan 2015: The Problem of Expertise. From Experience to Skillful Practices to Expertise: Ecological and Pragmatist Perspectives. European Journal of Pragmatism and American Philosophy 7(1): 8–23. DOI:

Boyer, Dominic 2009: Thinking through the Anthropology of Experts. Anthropology in Action 15(2): 38‒46. DOI:

Boyer, Dominic 2019: Our Post-Post-Truth Condition. In: Laboratory: Anthropology of Environment, Human Relations (ed.), After Practice: Thinking through Matter(s) and Meaning Relationally. Berlin: Panama Verlag, 83‒90.

Brändle, Verena K., Charlotte Galpin & Hans-Jörg Trenz 2022: Brexit as “Politics of Division”: Social Media Campaigning after the Referendum. Social Movement Studies 21(1‒2): 234‒253. DOI:

Carr, E. Summerson 2010: Enactments of Expertise. Annual Review of Anthropology 39(1): 17‒32. DOI:

Clarke, John & Janet Newman 2017: “People in this country have had enough of experts”: Brexit and the Paradoxes of Populism. Critical Policy Studies 11(1): 101‒116. DOI:

Coleman, Stephen 2022: How People Talk about Politics: Brexit and Beyond. London: Bloomsbury. DOI:

Cummings, Dominic 2017: On the referendum #21. Dominic Cummings’s Blog (blog), January 9, 2017, (accessed August 16, 2022).

De Fina, Anna & Alexandra Georgakopoulou 2008: Analysing Narratives as Practices. Qualitative Research 8(3): 379‒387. DOI:

DeLeo, Rob A. 2017: Anticipatory Policymaking in Global Venues: Policy Change, Adaptation, and the UNFCCC. Futures 92: 39‒47. DOI:

Dent, Alexander S. 2019: Ludic Authoritarianism. American Anthropologist 121(1): 193‒196. DOI:

Engesser, Sven, Nicole Ernst, Frank Esser & Florin Büchel 2017: Populism and Social Media: How Politicians Spread a Fragmented Ideology. Information, Communication & Society 20(8): 1109‒1126. DOI:

Eyal, Gil 2019: The Crisis of Expertise. Cambridge: Polity Press.

Gal, Susan & Judith T. Irvine 2019: Signs of Difference: Language and Ideology in Social Life. Cambridge: Cambridge University Press. DOI:

Hodges, Adam W. 2019: When Words Trump Politics: Resisting a Hostile Regime of Language. Stanford: Stanford University Press. DOI:

Hoskinson, Jim (dir.). 2005: The Colbert Report. Comedy Central. TV show.

Hutton, Will 2021: The Case for Brexit was Built on Lies. The Observer, June 27, 2021, (last accessed August 16, 2022).

Ilieva, Polya & Thomas M. Wilson 2011: Euroscepticism and Europeanisation at a Margin of Europe. Anthropological Journal of European Cultures 20(2): 87‒113. DOI:

Kitta, Andrea 2018: Alternative Health Websites and Fake News: Taking a Stab at Definition, Genre, and Belief. The Journal of American Folklore 131(522): 405‒412. DOI:

Koch, Insa 2017: What’s in a Vote? Brexit beyond Culture Wars. American Ethnologist 44(2): 225‒230. DOI:

Liberal Democrats 2019: The 12 Lies of Brexit. Liberal Democrats, 2019, (last accessed August 16, 2022). An archived version is available at:

Lockie, Stewart 2017: Post-truth Politics and the Social Sciences. Environmental Sociology 3(1): 1‒5. DOI:

Mälksoo, Maria 2019: The Normative Threat of Subtle Subversion: The Return of “Eastern Europe” as an Ontological Insecurity Trope. Cambridge Review of International Affairs 32(3): 365‒383. DOI:

McNeill, Lynne S. 2018: “My friend posted it and that’s good enough for me!” Source Perception in Online Information Sharing. The Journal of American Folklore 131(522): 493‒499. DOI:

Noyes, Dorothy 2018: Blaming the Polish Plumber, Blaming the French Voter: Bogeys and Attributions of Belief in Liberal Politics. Journal of American Folklore 131(522): 426‒434. DOI:

Polletta, Francesca 2006: It Was Like a Fever: Storytelling in Protest and Politics. Chicago: University of Chicago Press. DOI:

Polletta, Francesca & Jessica Callahan 2017: Deep Stories, Nostalgia Narratives, and Fake News: Storytelling in the Trump Era. American Journal of Cultural Sociology 5(3): 392‒408. DOI:

Powell, Kathy 2017: Brexit Positions: Neoliberalism, Austerity and Immigration ‒ the (Im)possibilities? of Political Revolution. Dialectical Anthropology 41(3): 225‒240. DOI:

Read, Jonathon 2019: Boris Johnson appears to finally admit his “£350m a week” claim was wrong. The New European, September 11, 2019, (accessed August 16, 2022).

Reid, Andrew 2019: Buses and Breaking Point: Freedom of Expression and the “Brexit” Campaign. Ethical Theory and Moral Practice 22(3): 623‒637. DOI:

Saglam, Erol 2020: What to Do with Conspiracy Theories? Insights from Contemporary Turkey. Anthropology Today 36(5): 18‒21. DOI:

Shore, Cris 2021: Britain, Brexit and Euroscepticism: Anthropological Perspectives on Angry Politics, Technopopulism and the UK Referendum. Anthropological Journal of European Cultures 30(2): 1‒22. DOI:

Stanley, Jason 2015: How Propaganda Works. Princeton: Princeton University Press. DOI:

Taylor-Neu, Robyn Holly 2020: Parasites and Post-truth Climate. Journal of Linguistic Anthropology 30(1): 4‒26. DOI:

Wetzel, Linda 2018: Types and Tokens. In: Edward N. Zalta (ed.), The Stanford Encyclopedia of Philosophy. Metaphysics Research Lab: Stanford University. (accessed August 16, 2022).

Wiese, Annjeanette 2015: Telling what Is True: Truthiness and Fictional Truths in Hybrid (Non)Fiction. Prose Studies 37(1): 66‒82. DOI:

Worrall, Patrick 2016: FactCheck: Do we really send £350 m a week to Brussels? Channel 4 News, April 19, 2016, (last accessed August 16, 2022).

Stefan Groth is Privatdozent at the Department of Social Anthropology and Cultural Studies (ISEK) at the University of Zurich and works as a consultant on political communication. He has researched and taught in Göttingen, Bonn, and Zurich. Among his publications are a monograph on multilateral negotiations in the United Nations (Negotiating Tradition, 2012), and a special issue on political narratives (Narrative Culture 6[1], 2019).