1. Introduction

The present article was inspired by the EU referendum of 2016 and is framed as a linguist’s response to this historic event and its complex consequences, which we see unfolding three years later. I am encouraged in this endeavour by the fact that the role of language and discourse is regularly recognised as important in analyses of Brexit published by historians, legal experts, political scientists and journalists. Topics dealt with range from the rhetorical strategies employed by ‘Brexiters’ and ‘Remainers’ (e.g. Katz, 2016; Armstrong, 2017; Clarke, Goodwin & Whiteley, 2017; Saunders, 2018) to the details of the legal framework in which English can continue to be used as the de facto working language of the European Union after March 2019, when its native-speaker base and constitutional status in the European Union will be reduced considerably – as the second official language of the Republic of Ireland and, alongside the national language Maltese, as one of the two official languages of Malta.1

Given this widespread interest, an analysis of the language and discourse(s) of Brexit from a linguistic point of view will not only serve the discipline itself, but also hold some interest for neighbouring fields. For it to be comprehensive and systematic, such an analysis will have to meet the following two requirements.

  1. Synchronically, it should go beyond the lexicographical level (i.e. charting the origin and spread of referendum-inspired neologisms such as Brexit, brexity, to Brexit-proof). After all, Brexit-related discourses are characterised not only by the use of Brexit neologisms, but also by the many new ways in which existing words are combined or used. In this sense, phrases such as take back control or take/want one’s country back are as much part of Brexit rhetoric as the word Brexit itself.

  2. Diachronically, a comprehensive Critical Discourse Analysis (henceforth CDA, cf. Fairclough, 2010 [1995]) of Brexit discourse must not confine itself to the present, but needs some historical time-depth (ca. 70 plus years, in the present case), as a historic event such as the June 2016 referendum cannot be interpreted without placing it in the wider historical context.

Methodologically, the present study will rely on the standard mix of quantitative and qualitative text-analytical methods which has been developed in corpus linguistics, the computer-aided analysis of large amounts of digital data. As an interdisciplinary effort, it additionally connects with research in two different but complementary traditions. On the one hand, it builds on work in CDA, and more generally in cultural studies, which typically derives its insights from close readings of key documents, informed by a coherent critical theory of discourse and society. On the other hand, it reaches out to cultural analysis in the ‘distant reading’ (Moretti, 2013) mode pioneered in the Digital Humanities. No corpus-based discourse analysis of any socially or politically controversial topic is complete without the acknowledgement of the fact that in every society there are groups who exercise political power and agency without fully and openly articulating their motives in mainstream public forums. Such absence from public discourse may be voluntary, when people prefer to keep their agendas secret, or involuntary, when people lack the competences necessary for access to the relevant channels of communication or have their access blocked or restricted by others. In the case of the EU referendum, this fact manifested itself in frequent allegations that Brexit discourses showed all sorts of biases: an expert bias – the political, economic and cultural elites against the common sense of the ordinary voter; a class bias – the middle class against the working class; a London bias – the cosmopolitan and diverse capital against the rest of England; and a regional bias within the UK – England and Wales against Scotland and Northern Ireland. To compensate for possible biases of this kind, discourse analysis must be sensitive not only towards expressed content and the linguistic form in which it is expressed, but also towards silences and taboos.

Section 2 below will present an analysis of the immediate linguistic fall-out from the EU referendum, while section 3 will complement this with a longer-term discourse-historical study based on the Hansard parliamentary record from 1803 to 2005 and selected further documents.2 As will be shown, since 1945 Europe has rapidly and massively superseded the Empire and the Commonwealth as the major arena in which the UK has defined its transnational connections. The concluding section will summarise the major findings, pointing out their potential relevance to neighbouring fields, such as history and cultural studies. The analysis will show that at various stages in the developments since 1945 Euroscepticism has meshed both with the rhetoric of the political left and the political right and that, ultimately, the UK’s position vis à vis Europe has been negotiated in the context of massive sociocultural transformations such as the dissolution of the British Empire and the postwar economic boom.

2. The language of Brexit: neologisms and beyond

With regard to the linguistic impact of Brexit, attention was initially focused on vocabulary and terminology (e.g., Katz, 2016; Fontaine, 2017; Lalić-Krstin & Silaški, 2018; Pullum, 2018). More recently, this research has been complemented by publications taking a wider, discourse-analytical approach (e.g. Buckledee, 2018; Koller, Kopf & Miglbauer, 2019; Zappettini & Krzyżanowski, 2019). While some of these studies use linguistic corpora, often customised small ones assembled for the purposes at hand, or at least work with digital textual data, comprehensive and systematic statistical profiling of large masses of text has not generally played a major role so far. The present study will go beyond previous lexicographical and discourse-analytical research on Brexit in two ways.

First, its targeted list of ‘Brexit words’ is made up not only of neologisms, but also of novel uses and combinations of existing words. Secondly, it integrates quantitative methods from corpus linguistics and qualitative methods from discourse analysis, partly inspired by CDA. This mixed-methods approach, exploiting the obvious synergies between the critical ‘close reading’ of individual texts in CDA and the ‘distant reading’ (Moretti, 2013) of large masses of relevant textual data, has already been used successfully in the study of the linguistic fall-out from a number of other social controversies (cf., e.g., Baker, 2006; Gabrielatos, McEnery, Diggle & Baker, 2012; Baker, Gabrielatos, KhosraviNik, Krzyżanowski, McEnery & Wodak, 2008; Germond, McEnery & Marchi, 2016; Baker, Brezina & McEnery, 2017), though not of the Brexit process.

Not unexpectedly, quantitative corpus-linguistic methods of text analysis and Digital Humanities approaches to data visualisation dominate the field where ‘born-digital’ data on Brexit are analysed, as in Simaki et al.’s (2018) work on the Brexit Blog Corpus or Cristianini and his group’s use of Twitter data to assess changes in public sentiment and mood during various stages of the Brexit process (cf. Lansdall-Welfare, Dzogang & Cristianini, 2017).3

The primary databases for the present study are the News on the Web (NoW) corpus, which will be used to explore the immediate linguistic changes caused by Brexit, and the Hansard Corpus, which will serve to explore the longer-term history of Eurosceptic discourses in Britain. NoW (Davies, 2013) is a regularly updated ‘monitor’ corpus covering the years since 2010 and currently comprising ca. 8.1 billion words of running text.

This period spans the crucial years from the run-up to the referendum to the present. Analyses of the data show that Brexit discourse comprises not only the neologisms inspired by the process itself, but also a large number of tropes such as take/want (one’s) country back, take back control, (a truly) global Britain, and more, most of which have developed from discourses of long historical standing. This history will be explored using the Hansard Corpus of British parliamentary debates. The ‘Hansard’ (named after the publisher and entrepreneur Thomas Hansard, who helped start the venture in the early nineteenth century) is the official record of British parliamentary debates and decisions, now available in print and online formats (https://hansard.parliament.uk/). It claims to provide ‘substantially verbatim’ coverage, based on edited versions of the original raw transcripts. Obviously, substantially verbatim coverage may mean different things to political scientists interested in content analysis and linguists studying more subtle kinds of grammatical variability, such as the choice between contracted (don’t they) and non-contracted (do they not) variants of the interrogative. Moreover, Sutherland & Farrell (2017) point out considerable changes in transcribing conventions over time. Mollin (2007) has assessed the potential and limitations of the material from a historical linguist’s point of view. All caveats notwithstanding, though, the Hansard data remain an extremely valuable resource for the purpose at hand – all the more so as there have been no major changes in transcription conventions for the period in focus here (the years since the end of World War II). In particular, the present study benefits from the fact that the Hansard data for the period from 1803 to 2005 were annotated for corpus-linguistic search by Mark Davies as part of the project SAMUELS (‘Semantic Annotation and Mark-Up for Enhancing Lexical Searches’), coordinated at the University of Glasgow (https://www.hansard-corpus.org/, https://www.gla.ac.uk/schools/critical/research/fundedresearchprojects/samuels/).4

The word Brexit, a blend of Britain/British and exit – modelled on an earlier coinage Grexit, which was used informally to refer to a possible exit of Greece from the Eurozone as a result of its sovereign debt crisis – was apparently first used in 2010 (Armstrong, 2017: 1–2; Fontaine, 2017), but remained rare until 2016, when it emerged as the most popular term to refer to the phenomenon in the months before and after the June referendum. This rise to prominence – and permanence – was recognised by the publication of a separate entry for the word in the March 2017 update of the OED Online (OED Online, s.v. Brexit).

In the NoW corpus, the word Brexit is sporadically attested from 2010. The upper half of Figure 1 below shows it expectedly shooting to prominence during the first half of 2016, which is therefore analysed in ‘close-up’, by periods of ten days, in the lower half.

Figure 1
Figure 1

Frequency of Brexit in the NoW corpus, 2010–1018 (top) and 1 January 2016 to 30 June 2016 (bottom).

The crucial measure here is the normalised frequency per million words, which rose from <100 to 440.74 in the 10-day period encompassing the 23 June referendum (see rightmost bar in the lower chart). This topical spike is temporary, but frequency of use has remained high.

‘Wildcard’ searches (for Brexit*, i.e. the root Brexit and any sequence of characters following it) show that the word almost instantly gave rise to numerous derivations, among them the frequent Brexiteer(s), the less common Brexiter(s), fairly common transparent combinations such as Brexit-related, Brexit-induced, Brexit-obsessed, Brexit-battered or Brexit-supporting, but also a long tail comprising dozens of rarer forms such as the verb to Brexit/brexit (i.e. ‘to exit from the EU’, but also found in a number of phrasal combinations such as ‘we’re all brexited out’; see also Lutzky & Kehoe, 2019), Brexit-proof (attested both as an adjective and as a verb), brexity, Brexit-ready, Brexitland, Brexitannia, Brexit-lite, brexitise, Brexitism, etc.

In cases of direct competition among neologisms, it is often the more expressive and emotionally charged terms which win out, lending credence to the fact that the ‘EU question has become more polarized ideologically in Britain than anywhere else in Europe’ (Tombs, 2016). Thus, in the period between January 2010 and December 2018 the term Brexiteer, with its activist connotations (cf. musketeer, volunteer, buccaneer, pioneer) clearly outnumbers the purely descriptive Brexiter: there are 9,103 instances of Brexiteer(s) against 2,547 for Brexiter(s). Remainers, on the other hand, tend to be referred to mostly by this descriptive label (4,483 attestations). Significantly, the expressive alternative in this case is a negatively loaded word, remoaners (with 441 attestations), which creates a terminological asymmetry in the way the two opposing political camps are referred to. Political observers have commented on the persuasive potential of such terminology (Heffer, 2016; Katz, 2016), and constitutional law expert Kenneth Armstrong has characterised the resulting bias as follows:

Over time, the language of Brexit has been adapted and supplemented as a means of characterising responses to the referendum result. When used as a way of describing pro-withdrawal supporters – especially in the form of ‘Brexiteer’ – it conjures up imagery of individuals battling to restore control to British institutions, to be contrasted with the ‘Remoaners’ unwilling and apparently unable to accept the outcome of the referendum (2017: 1–2).

This assessment is echoed in several contributions to Goodbye Europe, an anthology collecting public figures’, writers’ and intellectuals’ responses to the 2016 referendum. Here is an example from the contribution by author Kate Eberlen:

Is it a sense of humour failure to object to the term ‘Remoaner’? It’s a word that comes from our exuberant tradition of tabloid puns. Even if you disagree with the politics, you admire the verbal dexterity. But is it really clever or amusing to belittle the passionately held views of so many people? (2017: 75)

As a final illustration of a direct Brexit-induced neologism, consider the adjective brexity, which has inspired the title of the present article and which denotes a person or mindset strongly favouring and actively encouraging the UK’s exit from the EU. In terms of context-free derivational morphology, the adjective brexity has the broad meaning potential of ‘having to do with Brexit’, which is emotionally neutral. This meaning is occasionally attested in the data, for instance in example (1) below:

  1. Even though the currency and stock markets have been in turmoil since the referendum vote, nothing ‘Brexity’ can happen in Brussels until the resident [sic] of the European Council, Donald Tusk, receives a properly accredited letter from the British prime minister notifying him of Britain’s intention to leave the EU. (NoW, Irish Times, 17 July 2016)

Even most early uses, however, convey the emotionally charged notion of energetic advocacy of Brexit:

  • (2) Boris Johnson is not the only one to have calculated that quitting on the grounds that Mrs May is ‘betraying’ Brexit might well please the Tories’ extremely Brexity membership. (NoW, Guardian, 14 October 2018)

  • (3) None of the most Brexity members have been invited. (NoW, Daily Telegraph, 11 October 2018)

  • (4) The bluster from hardliners initially put the prime minister along the road to a much more Brexity Brexit, but not only has it backfired by painting the UK into a corner, it also smacks of desperation: it is one thing for remainers to be bad losers, but it is quite another for leavers to be bad winners. (NoW, Reaction online newsletter, 11 October 2018)

In these three examples the adjective brexity refers to political opinion. In the following two (5 and 6), on the other hand, its meaning is extended to cover a generally jingoistic mindset, with associated behaviour:

  • (5) Much of Brits’ behaviour abroad is what might be termed Brexity […] I doubt that this will have been helped by Brexit. Our standing on the continent is at an all-time low. Indeed, much of Brits’ behaviour abroad – the sense of entitlement, the determination to behave in ways one never would at home, the lack of respect for cultural or linguistic diversity, the pig-headed belief (even when your head is in a toilet) that you’re somehow superior to other countries – is what might be termed Brexity. (NoW, Guardian, 28 July 2017)

  • (6) Given the Brexity, Trumpity state we find ourselves in, does that mean we are now living in the Pub Landlord’s world now? (NoW, The Herald, 17 December 2017)5

Another area of considerable lexical creativity is the formation of fixed expressions with Brexit as their head element. A search in NoW for the second half of 2016 for adjectives collocating with Brexit6 reveals two conventionalised expressions that occur with frequencies above 300: hard Brexit (1,965) and soft Brexit (488), with a notable frequency bias towards the former. Two years later (second half of 2018) and reflecting the more advanced stage of the negotiations, the three most frequent collocates in fixed expressions are no-deal (4,306), hard (1,763) and soft (505). No-deal seems to have emerged successful from competition with possible alternatives such as disorderly (391), chaotic (136) or cliff-edge (88). None of these alternatives are fully synonymous, and the consensus on no-deal (Brexit) should be seen and analysed as what it is: the result of a struggle in the discursive arena that has gone on largely below the threshold of speakers’ conscious awareness. This struggle has established no-deal, a term which – unlike disorderly or chaotic – has considerable euphemistic potential and may therefore disadvantage people who advocate maintaining close ties with the European Union or even reversing the referendum decision. On the other hand, it would be highly unlikely for chaotic Brexit to emerge as the default term a mere two years after a majority of the electorate voted for Brexit – out of a sense of frustration with the status quo, but also with conviction and hopes for the future.

As emphasised above, neologisms are just the tip of the iceberg, and Brexit discourse relies as much on retooling the existing vocabulary as on the creation of new words. Very often, new combinations of old words show up as brief statistical ‘topicality’ spikes in the NoW data and disappear when the cause which triggered them has been forgotten. A case in point is the snap-election slogan ‘strong and stable [government]’ promoted by Prime Minister Theresa May’s campaign in 2017. In the NoW data, it occurs 80 times in 2015, 292 times in 2016, peaks at 1206 attestations in 2017 and then goes back to 351 in 2018.7

Other Brexit-related tropes, however, seem to have left a more lasting impact. Figure 2 below shows the results for a search in the NoW corpus for ‘TAKE back control’ (i.e. all forms of the verb take followed by the two other words). Visualisation (column 4) is based on the normalised frequencies per million words, as the statistically relevant measure.

Figure 2
Figure 2

TAKE back control in NoW.

This perfectly ordinary phrase – little used up to 2016, when one of its variants, the imperative Vote Leave, take back control, became the slogan of the Leave campaign – shot to prominence with the referendum, but unlike strong and stable, has come to stay. Note that it is precisely this phrase that is singled out for representation in a long litany of changes pervading Brexit Britain in Ali Smith’s topical novel Autumn:

All across the country, the media was insane. All across the country, politicians lied. All across the country, politicians fell apart. All across the country, promises vanished. All across the country, money vanished. All across the country, social media did the job. All across the country, nobody spoke about it. All across the country, nobody spoke about anything else. All across the country, racist bile was general. All across the country, people said it wasn’t that they didn’t like immigrants. All across the country, people said it was about control. (Smith, 2016: 61, emphasis mine)

Slightly different findings are obtained in the search for APPGE country back (i.e. any possessive pronoun + country back), see Figure 3. This collocation, present as a kind of discursive ‘background noise’ even before the heat of the Brexit campaign, shows the expected topicality spike in 2016 and then declines back to the longer-term average.

Figure 3
Figure 3

APPGE country back in NoW.

Further examples could be provided. Global Britain, a collocation which has been analysed from a CDA perspective by Zappettini (2019), increases from 75 instances in 2016 to 528 in 2017 and then declines to 357 in 2018. Backstop, a cricket term with modest potential for metaphorical use in other contexts, varies between frequencies of 122 (in 2010) and 239 (in 2015), but then rapidly rises to 627 (in 2016), 741 (in 2017) and 9,864 in 2018, most of them in the Brexit-related sense of ‘fall back plan to prevent a hard border in Ireland’.

In sum, the veritable explosion of lexical creativity in the wake of the EU referendum should be read as a sign that the responses to the result were not exclusively rooted in rational analyses of the political and economic factors guiding voters’ choices. Rather, they reveal a deeper emotional and cultural stratum, as is recognised by Robert Eaglestone:

Brexit is not only political, economic and administrative: perhaps most significantly it is an event in culture, too. Brexit grew from cultural beliefs, real or imaginary, about Europe and the UK; the arguments before, during and after the referendum were – and are – arguments about culture; its impact on the cultural life of these islands may last for generations. (2018: 1)

It is these cultural roots of a complex and sometimes fraught relationship with Europe that will be explored in the longer historical term in the following section.

3. Europe First? The UK’s geopolitical position as reflected in Hansard

In spite of a temporary setback caused by the American War of Independence, Britain managed to systematically assemble and consolidate from the late 18th century what was to become the largest colonial Empire the world had ever seen. Unsurprisingly, despite geographical proximity, Britain’s European entanglements were not always a priority during this period. The following lexicostatistical profile of the 1803–2005 digital Hansard record is an attempt to document the relative prominence of the Empire (and, subsequently, the Commonwealth) and Europe as political topics in UK parliamentary debates.8 In a thorough and thoughtful assessment of data authenticity and quality, Mollin (2007) has discussed both the high potential and the many limitations of this source for historical linguistic and discourse studies. Fortunately, for content-word searches of the type undertaken here, the potential clearly outweighs the limitations, so that the findings can be expected to be generally robust. For the topic of the (British) Empire, the words Empire and imperial were taken as the central pointers (searches being case-insensitive in both cases). For the topic of the Commonwealth, it was decided to search upper-case and lower-case mentions so as not to miss potentially interesting borderline cases between the common-noun and proper-noun senses. It was somewhat more difficult to cast the ‘wordnet’ for mentions of the topic of Europe. In the end, the results were calculated from the following individual searches (case-insensitive):

  1. the geographical/political terms Europe and European/Europeans

  2. the abbreviations EEC (for ‘European Economic Community’), EC (‘European Community’) and EU (‘European Union’), plus the name of the European currency, the euro, in singular and plural form9

  3. mostly hyphenated words containing the initial combining form Euro- (as in Euro-bonds, Eurobonds, euro-bonds, eurobonds). These words comprise a small number of frequent and well-entrenched items (e.g. Europhile, Europhobe) and a very long tail of infrequent creations which are often emotionally loaded and rhetorically charged, as is shown by the nonce-formation Euro-wanking (in the context of a ‘marvellous Euro-wanking make-work project’).10

In Figures 4 and 5 below all these words are collectively referred to as ‘Euro-words’.11 Frequencies for all search items are listed per decade (or, for the first and last interval, part of a decade): 1803–1809, 1810–1819, 1820–1829, …, 2000–2005. The material analysed amounts to a total of 1.6 billion words. As periods contain different amounts of text, normalised frequencies are calculated per million words, as was done in the case of the NoW corpus. Figure 4 gives frequencies for the entire 202 years of coverage, for a first general orientation.

Figure 4
Figure 4

Empire, Commonwealth and Europe as topics in the Hansard corpus (1803–2005).

Figure 5
Figure 5

Empire, Commonwealth and Europe as topics in the Hansard corpus (1940–1990).

Figure 4 shows two things. First and most importantly, the topic of Europe eclipsed both Empire and Commonwealth dramatically from the 1940s. Today, at frequencies of more than 1000 words per million, ‘Europe’ dominates at levels never reached by the Empire-topic in previous periods. Secondly, Empire itself dominates in the half-century from 1880 to 1930. Figure 5 zooms in on the half-century from 1940 to 1990:

What we see here is the demise of Empire, the rise and fall of the Commonwealth as a topic and the steep increase in Euro-words throughout the 1960s and 1970s. Note in particular that Europe and the Commonwealth were equally strongly represented as topics of parliamentary concern during the 1960s, but Europe clearly dominated from the 1970s onwards. In this light, the result of the 1975 referendum on EU membership, which – with the exception of the Outer Hebrides and the Shetland Islands – showed majorities in favour of membership in the EEC throughout the UK, can be read as a sign of the fact that the electorate had overcome the phantom pains resulting from the dissolution of the British Empire and endorsed a new vision of Britain as an active participant in the European project. By 1975, the Empire and the Commonwealth had ceased to function as effective frames of interpretation in day-to-day politics. By contrast, Britain had shared with Western Europe the generally positive experience of a post-World War II economic, social and cultural transformation. This Europe-friendly constellation, the ‘1975 moment’, has remained a singular event – and was in fact negated by the referendum of 2016.

As the Hansard record and other political documents of the 1960s and 1970s make clear, favourable and hostile attitudes towards European integration cut across the political left-versus-right divide. Out of context, the reader of the resounding statement ‘When the British people speak, everyone, including members of parliament, should tremble before their decision’ might well be forgiven for attributing it to very brexity Conservative MP Jacob Rees-Mogg. In fact, it was uttered by hard-left Eurosceptic Labour MP Anthony Wedgwood Benn in the wake of the 1975 referendum (BBC, 1975). The motivations for Tony Benn’s opposition to the EEC were ideological and constitutional in nature. Ideologically, he feared that membership would slow down and ultimately stall the reorganisation of Britain on socialist lines; constitutionally, he argued that the democratic accountability of European institutions lagged far behind the Westminster parliamentary standard (Benn, 2016 [1975]). Sentimentality, let alone sentimentality about imperial and Commonwealth connections, was not involved.

This had been different a dozen years before, in the case of another Labour Eurosceptic, Hugh Gaitskell, who shortly before his death in 1963 came out firmly against European integration in an impassioned speech (Gaitskell, 1962). To give an impression of the period flair of this speech, I quote the opening lines of the anti-Europe argument:

There are certain ways in which we should not decide this issue. It is not a matter to be settled by attractive pictures of nice old German gentlemen drinking beer on the one hand or, on the other, by race or national hatred stimulated by past experiences. It should not be decided because on the one hand we like Italian girls, or on the other, we think we have been fleeced in Italian hotels. It should not be decided on the basis of whether we think French food is the best in the world, or because, as one of my correspondents put it, she was afraid Europe was out to poison us! (Gaitskell, 1962: n. p.)

World War II still casts a shadow here, with the counterbalancing of the positive stereotype of peaceful German beer drinkers and the ‘race or national hatred stimulated by past experiences’. The paternalistic condescension in the reference to ‘Italian girls’ may offend the feminist sensibilities of some present-day readers. The most obvious feature of the passage, however, is the open and direct way in which national stereotypes about the various European countries are evoked without critical distance. Gaitskell’s substantive arguments revolve around the two topics of a ‘thousand years’ of historical progress towards constitutional government and Britain’s obligations to the Commonwealth. Gaitskell further argued:

We must be clear about this: it does mean, if this is the idea, the end of Britain as an independent European state. I make no apology for repeating it. It means the end of a thousand years of history. You may say ‘Let it end’ but, my goodness, it is a decision that needs a little care and thought. And it does mean the end of the Commonwealth. How can one really seriously suppose that if the mother country, the centre of the Commonwealth, is a province of Europe (which is what federation means) it could continue to exist as the mother country of a series of independent nations? It is sheer nonsense. (1962: n. p.)

There is irony in the fact that Jacob Rees-Mogg, Conservative MP and pro-Brexit hardliner, born six years after Gaitskell’s death, should use precisely the same historical argument de la longue durée to celebrate the UK’s liberation from the prison which Gaitskell wanted to prevent it having to enter:

Leaving the European Union is a great liberation for the United Kingdom, as worthy for celebration as victory at Waterloo or the Glorious Revolution. […] it chimed gloriously with Kipling’s idea of the Saxon and the Norman:

  • The Saxon is not like us Normans. His manners are not so polite.

  • But he never means anything serious till he talks about justice and right.

  • When he stands like an ox in a furrow – with his eyes on your own,

  • And grumbles, ‘This isn’t fair dealing, my son, leave the Saxon alone.’

(Rees-Mogg, 2017: 259f.)

For Rees-Mogg, who may have to pursue a part of his considerable European business interests through Ireland after Brexit, the Commonwealth does not figure as an important economic partner any more, but the reference to Rudyard Kipling does raise at least two questions. Is it intended to honour the literary champion of the British Empire? Or is it meant to evoke the communicative constellation of the quoted poem, in which a young Norman upstart (the European Union) is admonished by his wise father to give the Saxon (the UK) his fair deal?

4. Conclusions

British Eurosceptic discourse since 1945 shows a number of continuities. In view of the gradual rise of representative democracy in Britain during the 19th and early 20th centuries, as opposed to the political turmoil which affected much of continental Europe during the same period, there is an understandable value placed on national sovereignty (‘a thousand years of history’) and the position of Parliament in Westminster as the ultimate source of legal and constitutional authority. Distrust of European bureaucracy and – real or perceived – deficits of democratic accountability in European institutions have united Eurosceptics of the right and left for the past half century. There has always been a strong current of Euroscepticism in the Labour Party, reflected today in the rift between largely pro-European Members of Parliament and an ambivalent party leadership. What could not have been predicted in 1975, however, was the extent to which the Conservative Party was captured by Brexiteers of the extremer sort.

The major change that took place between the two referendums in 1975 and 2016 is the relative weight of immigration and the economy. In 1975 immigration was a controversial issue, but largely in the context of immigration from the Commonwealth (then already past its peak). In regard to Europe, economic arguments dominated the discussion. This was clearly different in 2016, when concerns about immigration from European Union member states, sadly including a referendum-related increase in anti-immigrant hate crime (O’Neill, 2017), were widespread among sectors of the electorate. A disconnect between official political, media and expert discourse on immigration, with its emphasis on the cultural and economic dynamism likely to be unleashed by a growing and diverse population, and deep-seated fears and resentments among people untouched by this prospect, may well turn out to have been one of the discursive silences, breakdowns and taboos mentioned in the introduction above.

As a linguist, I would like to point out the obvious, but frequently overlooked role of the English language in directing currents of migration, in Europe as well as in the world at large. Given the role of English as the global lingua franca, some level of competence in English can be expected among migrants starting out anywhere in the world, and the English language thus also works as a pull factor among migrants in the European context (cf. also Adserà & Pytliková 2015, with extensive documentation with statistics from the OECD countries). Britain, Germany and Finland may offer comparable economic prospects for migrants, but with free movement among citizens within the European Union, Britain may often turn out to be the destination of choice, because migrants already have the necessary level of language competence in English, whereas they would be held back by lack of competence in German or Finnish.

The outlook from the present analysis must be muted. For Europe, the likely departure of the UK is a major political, economic and – not least – cultural loss. In Britain, the 2016 referendum has revealed a ‘Divided Kingdom’ – divided by differences in emphasis on national autonomy and sovereignty and different attitudes towards the desirability of supra-national integration. In negotiating the new relationship between the UK and the EU, the elephant in the room is contemporary economic and cultural globalisation (a term which should definitely be investigated once the Hansard corpus has extended coverage to the present). Some Brexiters hoped for a prosperous and truly global Britain breaking free from the fetters of membership in a European bloc – a positive vision. More pro-Brexit sentiment, however, was fired up by fears of the unwelcome consequences of globalisation, such as enhanced economic competition, loss of national privileges and increase in migration. The central historical fact which the 2016 referendum has revealed in stark clarity is the existence of several ‘rifts’ in the UK (Korte & Mair, forthcoming [2019]):

  1. a rift between London, a global city highly connected in global politics, the global economy and contemporary cosmopolitan culture, and the rest of England,

  2. a rift between England and Wales and Scotland and Northern Ireland.

Brexit will put up new obstacles for the movement of goods, capital and people between the UK and the European continent. This will directly affect the cultural sector – whether it is the trade in books or the organisation of musical and dramatic stage performances by European artists in the UK, or vice versa. It is to be hoped that, not least because of the continuing role of the English language as the world’s and Europe’s lingua franca, the impact of Brexit on intellectual debate and the free exchange of ideas will remain limited.


  1. See Ginsburgh, Moreno-Ternero & Weber 2017 and 2018. Obviously, this complication is more of a legal-technical issue than a practical one, and I expect experts to devise a solution recognising the inevitable continuing presence of English as the EU’s main working language. In a global linguistic ecology in which English is the world’s first choice as lingua franca in so many regions and communicative domains – in other words, the ‘hub’ of the World Language System (de Swaan 2002, 2010) – it will remain the most practical link between Latvia and Germany, Bulgaria and Denmark, or Spain and Finland – regardless of whether the UK is or is not a formal member of the ‘bloc’. See Modiano (2017) and Seargeant (2017) for further analysis. [^]
  2. While I value the research carried out by Wodak and collaborators in the framework of their ‘Discourse-Historical Approach’ (DHA, Reisigl & Wodak 2015), I use the term discourse-historical in its non-technical descriptive sense here. [^]
  3. Born-digital data play a minor role in the present study. As it aims at historical time-depth, it had to be based on digital versions of traditional text-types, such as the Hansard parliamentary transcripts and online newspapers and magazines. However, it is self-evident that social media have played such a central role in events that no analysis of Brexit discourses would be complete without them. [^]
  4. For the record, I refer to ‘Hansard at Huddersfield’ (https://hansard.hud.ac.uk/site/index.php), a follow-up project which offers enhanced query options for the same material and further formats of data visualisation. These are not used in the present study. [^]
  5. The Pub Landlord is a stage persona created by comedian Al Murray. The (English) Pub Landlord holds – and freely articulates – strongly nationalist and anti-European sentiments. [^]
  6. The collocational span was set at zero to the right of Brexit and 2 to the left, with the part of speech of the collocate defined as ‘any adjective’ (_j*), which captures combinations of hard and Brexit in hard Brexit and hard Tory Brexit, but not in hard Boris Johnson Brexit. [^]
  7. The corresponding – and statistically more precise – normalised frequencies confirm the picture: between 0.16 and 0.20 occurrences per million words for the four half-year periods between January 2015 and December 2016, a peak of 1.04 for the first half of 2017, declining to 0.35 for the second half of the same year and to values equal to or lower than 0.20 in 2018 (i.e. the original very low frequency band). What the statistics will not show is how many of the post-2017 uses are sarcastic references to May’s slogan. [^]
  8. The 2005 endpoint is, of course, not a matter of principle but of practical convenience. The corpus-linguistic search options provided for this period by the Hansard Corpus (and the related Huddersfield project) are much superior to those offered by the original Hansard web interface (https://hansard.parliament.uk/). [^]
  9. This made possible identification of passages in which the politics of the currency were in focus. No search was carried out for the symbol , which was deemed of lesser interest as it was rare and generally restricted to identifying sums of money. [^]
  10. Labour Peer Dr John Gilbert in the House of Lords on 24 November 2004. [^]
  11. One of the anonymous reviewers for this article asked whether the high frequency of the word Europe itself may not skew the findings in critical ways. This is not the case. Before the emergence of the highly productive combining form Euro- in the 1960s and 1970s there is stability in the sense that the three high-frequency items Europe, European and Europeans make up almost the entire totals, so that the trough for ‘Euro words’ in the heyday of the Empire in the late 19th and early 20th centuries reflects a real decline in the number of references to the topic. By the 1970s, the share of the word Europe is no longer as critical, as it never accounts for more than a third of all occurrences. At normalised rates, it occurs 205 times (per million words) in the 1970s, 191 times in the 1980s, 357 in the 1990s and goes back to 236 in the 2000s. Note that this development is broadly in line with the general trend, although not as pronounced. An item which shows a more drastic increase is the adjective European, as part of the compound European Union. [^]

Competing Interests

The author has no competing interests to declare.


Adserà, A and Pytliková, M 2015 The role of languages in shaping international migration. Economic Journal, 125(586): F49–F81. DOI:  http://doi.org/10.1111/ecoj.12231

Armstrong, K A 2017 Brexit time: Leaving the EU – why, how and when? Cambridge: Cambridge University Press. DOI:  http://doi.org/10.1017/9781108233385

Baker, H, Brezina, V and McEnery, T 2017 Ireland in British parliamentary debates, 1803–2005. In Säily, T, Nurmi, A, Palander-Collin, M and Auer, A (eds.), Exploring future paths for historical sociolinguistics. Amsterdam: Benjamins. pp. 83–108.

Baker, P 2006 Using corpora in discourse analysis. London: Continuum.

Baker, P, Gabrielatos, C, KhosraviNik, M, Krzyżanowski, M, McEnery, T and Wodak, R 2008 A useful methodological synergy? Combining critical discourse analysis and corpus linguistics to examine discourses of refugees and asylum seekers in the UK press. Discourse and Society, 19: 273–306. DOI:  http://doi.org/10.1177/0957926508088962

BBC 1975 UK embraces Europe in referendum, BBC On This Day. http://news.bbc.co.uk/onthisday/hi/dates/stories/june/6/newsid_2499000/2499297.stm. [Last accessed 6 August 2019].

Benn, A W 2016 [1975] Tony Benn on the 1975 referendum. Spectator, 24 May. https://blogs.spectator.co.uk/2016/05/a-lesson-from-the-1975-referendum/ [Last accessed 6 August 2019].

Buckledee, S 2018 The language of Brexit: How Britain talked its way out of the European Union. London: Bloomsbury.

Clarke, H D, Goodwin, M and Whiteley, P 2017 Why Britain voted to leave the European Union. Cambridge: Cambridge University Press.

Davies, M 2013 Corpus of News on the Web (NOW): 3+ billion words from 20 countries, updated every day. https://corpus.byu.edu/now/.

de Swaan, A 2002 Words of the world: The Global Language System. Cambridge: Polity.

de Swaan, A 2010 Language systems. In Coupland, N (ed.), The handbook of language and globalization. Malden MA: Blackwell. pp. 56–76. DOI:  http://doi.org/10.1002/9781444324068.ch2

Eaglestone, R (ed.) 2018 Brexit and literature. London: Routledge. DOI:  http://doi.org/10.4324/9781351203197

Eberlen, K 2017. A loss for words In Various authors Goodbye Europe: Writers and artists say farewell. London: Weidenfeld & Nicolson. pp. 69–76.

Fairclough, N 2010 [1995] Critical Discourse Analysis: The critical study of language. 2nd ed. London: Routledge.

Fontaine, L 2017 The early semantics of the neologism BREXIT: A lexicogrammatical approach. Functional Linguistics, 4(6). DOI:  http://doi.org/10.1186/s40554-017-0040-x

Gabrielatos, C, McEnery, T, Diggle, P and Baker, P 2012 The peaks and troughs of corpus-based contextual analysis. International Journal of Corpus Linguistics, 17(2): 151–175. DOI:  http://doi.org/10.1075/ijcl.17.2.01gab

Gaitskell, H 1962 Text of speech made at the 1962 Labour Party Conference by the Rt. Hon Hugh Gaitskell MP. London: Labour Party. pp. 3–23. https://www.cvce.eu/content/publication/1999/1/1/05f2996b-000b-4576-8b42-8069033a16f9/publishable_en.pdf [Last accessed 6 August 2019].

Germond, B, McEnery, T and Marchi, A 2016 The EU’s comprehensive approach as the dominant discourse: a corpus linguistics analysis of the EU’s counter-piracy narrative. European Foreign Affairs Review, 21(1): 137–156.

Ginsburgh, V, Moreno-Ternero, J and Weber, S 2017 Ranking languages in the European Union: Before and after Brexit. European Economic Review, 93: 139–151. DOI:  http://doi.org/10.1016/j.euroecorev.2017.01.009

Ginsburgh, V, Moreno-Ternero, J and Weber, S 2018 The fate of English in the EU after Brexit: expected and unexpected twists. CEPR [Centre for Economic Policy Research]. https://voxeu.org/article/english-language-eu-after-brexit. [Last accessed 6 August 2019].

Heffer, S 2016 Meet the Three Brexiteers: the men who could change how we exit the EU. New Statesman, 13 September. https://www.newstatesman.com/politics/uk/2016/09/meet-three-brexiteers-men-who-could-change-how-we-exit-eu [Last accessed 6 August 2019].

Katz, I 2016 Victory of the swashbucklers: Did the word ‘Brexiteer’ help the Leave campaign win? The Spectator Australia, 24 September. https://www.spectator.com.au/2016/09/victory-of-the-swashbucklers/ [Last accessed 6 August 2019].

Koller, V, Kopf, S and Miglbauer, M (eds.) 2019 Discourses of Brexit. London: Routledge. DOI:  http://doi.org/10.4324/9781351041867

Korte, B and Mair, C Forthcoming [2019] New-old rifts: Brexit, migration and Britain’s BAME [=Black and Asian Minority Ethnic] communities. Journal for the Study of British Cultures, 25.

Lalić-Krstin, G and Silaški, N 2018 From Brexit to bregret: An account of some Brexit-induced neologisms in English. English Today, 34(2): 3–8. DOI:  http://doi.org/10.1017/S0266078417000530

Lansdall-Welfare, T, Dzogang, F and Cristianini, N 2017 Change-point analysis of the public mood in UK Twitter during the Brexit referendum. 2016 IEEE 16th International Conference on Data Mining Workshops (ICDMW 2016). Institute of Electrical and Electronics Engineers (IEEE), ICDM 2016, Barcelona, Spain, 12 December. DOI:  http://doi.org/10.1109/ICDMW.2016.0068

Lutzky, U and Kehoe, A 2019 Friends don’t let friends go Brexiting without a mandate. In Koller, K, Kopf, S and Miglbauer, M (eds.), Discourses of Brexit. London: Routledge. pp. 104–120. DOI:  http://doi.org/10.4324/9781351041867-7

Modiano, M 2017 English in a post-Brexit European Union. World Englishes, 36(3): 313–327. [Special issue Forum on English in a Post-Brexit European Union]. DOI:  http://doi.org/10.1111/weng.12264

Mollin, S 2007 The Hansard hazard: Gauging the accuracy of British parliamentary transcripts. Corpora, 2(2): 187–210. DOI:  http://doi.org/10.3366/cor.2007.2.2.187

Moretti, F 2013 Distant reading. London: Verso.

O’Neill, A 2017 Hate crime, England and Wales, 2016/17. Statistical Bulletin 17/17. London: Home Office. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/652136/hate-crime-1617-hosb1717.pdf [Last accessed 6 August 2019].

Pullum, G K 2018 A pocket Brexicon for American academics. Lingua Franca post in The Chronicle of Higher Education. Online: https://www.chronicle.com/blogs/linguafranca/2018/11/26/a-pocket-brexicon-for-american-academics/ [Last accessed 6 August 2019].

Rees-Mogg, J 2017 What the wise men promised. In Various authors Goodbye Europe: writers and artists say farewell. London: Weidenfeld & Nicolson. pp. 257–262.

Reisigl, M and Wodak, R 2015 The Discourse-Historical Approach (DHA). In Wodak, R and Meyer, M (eds.), Methods of critical discourse studies. 3rd ed. London: Sage. pp. 23–61.

Saunders, R 2018 Yes to Europe: The 1975 referendum and seventies Britain. Cambridge: Cambridge University Press. DOI:  http://doi.org/10.1017/9781108340915

Seargeant, P 2017 The symbolism of English on the Brexit battleground. World Englishes, 36(3): 356–359. [Special issue Forum on English in a Post-Brexit European Union]. DOI:  http://doi.org/10.1111/weng.12273

Simaki, V, Paradis, C and Kerren, A 2018 Evaluating stance-annotated sentences from the Brexit Blog Corpus: A quantitative linguistic analysis. ICAME Journal, 42: 133–165. DOI:  http://doi.org/10.1515/icame-2018-0007

Smith, A 2016 Autumn. London: Hamish Hamilton.

Sutherland, L and Farrell, S 2017 The history, workings, and future challenges of Hansard. Public lecture, 11 December. https://www.youtube.com/watch?v=yaLWV9gfiS0 [Last accessed 6 August 2019].

Tombs, R 2016 The long and winding road to Brexit. Foreign Policy, 24 June. Available at: https://foreignpolicy.com/2016/06/24/the-long-road-to-brexit-history-european-union/ [Last accessed 6 August 2019].

Zappettini, F 2019 The official vision for ‘global Britain’: Brexit as rupture and continuity between free trade, liberal internationalism and ‘values’. In Koller, V, Kopf, S and Miglbauer, M (eds.), Discourses of Brexit. London: Routledge. pp. 140–154. DOI:  http://doi.org/10.4324/9781351041867-9

Zappettini, F and Krzyżanowski, M 2019 The critical juncture of Brexit in media & political discourses: From national-populist imaginary to cross-national social and political crisis. Critical Discourse Studies, 16(4): 381–388. DOI:  http://doi.org/10.1080/17405904.2019.1592767