miscelánea 70 (2024): pp. 67-92 ISSN: 1137-6368 e-ISSN: 2386-4834

A CORPUS STUDY OF BREXIT POLITICAL

DISCOURSE: EXPLORING MODALITY THROUGH

LEXICAL MODALS

UN ESTUDIO DE CORPUS SOBRE EL DISCURSO

POLÍTICO DEL BREXIT: LA MODALIDAD

A TRAVÉS DE LOS MODALES LÉXICOS

https://doi.org/10.26754/ojs_misc/mj.20249837

ENCARNACIÓN ALMAZÁN-RUIZ

ealmazan@ujaen.es

Universidad de Jaén

<https://orcid.org/0000-0001-8724-6596>

AROA ORREQUIA-BAREA

aroa.orrequia@uca.es

Universidad de Cádiz

<https://orcid.org/0000-0003-1478-7847>

Abstract

This paper aims to analyse the lexical modals used in the political speeches given

by Boris Johnson and Jeremy Corbyn during the final months of the Brexit process.

This study explores whether lexical choice shows the politicians’ commitment to

their constituents, particularly to determine which lexical modals each politician

uses and which semantic implication(s) these modals convey. The study is

descriptive-interpretative and uses the corpus-assisted discourse studies approach.

It contributes to research on modality in the English language by examining parts

of speech other than (semi)auxiliary modal verbs. As the corpus analysis shows,

lexical modals are a prominent resource employed by politicians to present facts to

their audience.

Keywords: modality, lexical modals, corpus-assisted discourse studies, political

discourse, Brexit.

Resumen

El objetivo de este trabajo es analizar los modales léxicos utilizados en los discursos

políticos de Boris Johnson y Jeremy Corbyn durante los últimos meses del proceso

del Brexit. Este estudio se centra en investigar si la elección léxica muestra el

Encarnación Almazán-Ruiz and Aroa Orrequia-Barea

miscelánea 70 (2024): pp. 67-92 ISSN: 1137-6368 e-ISSN: 2386-4834

compromiso de los políticos con sus votantes, determinando qué modales léxicos

utiliza cada político y qué implicación o implicaciones semánticas transmiten. El

estudio es descriptivo e interpretativo y se ha realizado mediante la metodología

del análisis del discurso asistido por corpus. Este artículo, al examinar otras clases

de palabras diferentes a los verbos modales, contribuye a la investigación sobre la

modalidad en la lengua inglesa. Como muestra el análisis del corpus, los políticos

emplean los modales léxicos como un recurso destacado para presentar los hechos

a la audiencia.

Palabras clave: modalidad, modales léxicos, estudios del discurso asistido por

corpus, discurso político, Brexit.

1. Introduction

After the United Kingdom (UK) held a referendum on June 23, 2016, to

determine whether the country should remain a member of the European Union

(EU) or leave it, a process known as Brexit was initiated. Although the referendum

result was fairly close —52% vs 48%—1 more than half of voters chose to leave the

EU. From the time the referendum was held until the withdrawal took effect, a

few years went by in which the political confrontation between the leaders of the

country’s two main parties, Boris Johnson of the Conservative Party and Jeremy

Corbyn from the Labour Party, drew wide attention.

As language is the main instrument at the disposal of politicians, it is relevant and

worthwhile to analyse political discourse, since linguistic traces can uncover

political stance. Speakers may have different opinions on the same issue, leading to

the presentation of facts in one way or another; therefore, the speaker’s attitude

towards a given proposition is very significant from a linguistic point of view and

is related to so-called modality.

The semantic category of modality is mainly associated with the speaker’s attitude

towards the situation or facts expressed in the clause. Apart from including a

(semi)auxiliary modal verb in the verb phrase, there are other linguistic devices

that the speaker can use to express modality. According to Huddleston and Pullum,

“lexical modals” are other word classes (i.e. adjectives, adverbs, nouns, lexical

verbs) that can also convey the same meaning as (semi)auxiliary modal verbs

(2002: 173). Depending on the meaning expressed in the utterance, modality is

classified as either epistemic or deontic (Huddleston 1988a: 78-80).

Therefore, this paper aims to compare and analyse the lexical modals used in the

political speeches of both politicians during the last months of the Brexit process.

As most studies on modality have focused exclusively on (semi)auxiliary modal

A Corpus Study of Brexit Political Discourse: Exploring Modality

miscelánea 70 (2024): pp. 67-92 ISSN: 1137-6368 e-ISSN: 2386-4834

verbs, we intend to study the connections between modality expressed through

lexical modals and the political context in which it occurs, namely the Brexit

process (Paltridge 2012: 186). This analysis thus fills a gap in the research on

modality and seeks to shed light on the nuances and intricacies of modality, further

expanding the scope of modal research by exploring the concept beyond the realm

of (semi)auxiliary modal verbs. In this sense, the study extends a previous analysis

of (semi)auxiliary modal verbs in the Brexit political discourse (Orrequia-Barea

and Almazán-Ruiz 2021), as there is relatively scarce literature on the topic and a

lack of case studies. These two studies examine how modality might influence the

public’s perception of facts since politicians use it, for instance, to present facts as

possible or necessary.

This descriptive-interpretative study uses Corpus-Assisted Discourse Studies

(CADS) as its methodology (Baker 2020). It describes lexical modals using data

from political discourse on Brexit. The study seeks to answer three main research

questions: (RQ1) Which parts of speech —other than (semi)auxiliary modal

verbs— are more commonly used to express modality in the political speeches of

Boris Johnson and Jeremy Corbyn on Brexit? (RQ2) Are lexical modals more

frequently used than (semi)auxiliary modal verbs? (RQ3) Which politician uses

lexical modals more frequently in his discourse and what does this reveal?

After this introductory section, this paper is organised as follows. Sections 2 and 3

provide the political and theoretical background, focusing on political discourse

on Brexit, and on modality and lexical modals, respectively. Section 4 describes the

data and methods used to analyse the corpus. Sections 5 and 6 include the results

and the discussion. Finally, some conclusions are drawn in Section 7.

2. Language, Politics and Brexit

Language is a distinctive feature of human beings, which becomes particularly

relevant in politics. In political contexts, language is used to persuade the audience

to take a particular political position. According to Ananko, communication in

politics is essential as it helps political processes advance (2017: 128). Furthermore,

communication can contribute to politicians’ ability to influence society and make

it more cohesive. In this vein, Chilton and Schaffner state that “politics cannot be

conducted without language, and it is probably the case that the use of language

in the constitution of social groups leads to what we call ‘politics’ in a broad sense”

(in Dunmire 2012: 735). Likewise, Fairclough highlights that politics is language

since it “consists in the disputes and struggles which occur in language and over

language” (1989: 23).

Encarnación Almazán-Ruiz and Aroa Orrequia-Barea

miscelánea 70 (2024): pp. 67-92 ISSN: 1137-6368 e-ISSN: 2386-4834

Political language functions as an index of the speaker’s current view of reality, just

as the audience’s interpretation of the same language may indicate an entirely

different perspective (Edelman 1985: 10). Discourse interpretation and the

possible implicit message underlying the words uttered become crucial in the

political context, because expressing thoughts and conveying ideas is not neutral,

and language always carries a purpose and meaning. Still, language simultaneously

represents the intended meaning of the utterance (Ekawati 2019: 6). In this

regard, the context and the timing of political discourse must be considered, as

they will determine the interpretation of the particular language aspects analysed.

Furthermore, the linguistic devices used to reach the electorate can reveal how

politicians present facts.

As is well known, the UK has been notably ambivalent to EU membership. Only

two years after joining the EU, the country held its first referendum to determine

whether to remain in the European alliance, and two-thirds of voters chose to

maintain membership in 1975 (Somai and Biedermann 2016: 139). Several

decades later, the British nation again faced a choice between remaining or

embracing an uncertain future shaped by a new reality (Torrecuadrada García-

Lozano and García Fuente 2017: 5), as numerous socio-economic and political

aspects would have to be decided if the UK ultimately split from the EU. In the

so-called Brexit referendum on 23 June 2016, Britons voted to leave the EU.

The Brexit movement demanded the UK’s exit from the EU, primarily to restore the

nation’s control. In an attempt to please Euroskeptics, the Prime Minister at the

time, David Cameron, negotiated a deal with the EU to improve the UK’s economic

governance, competitiveness, sovereignty and control over the welfare state, and the

free movement of people (Torrecuadrada García-Lozano and García Fuente 2017:

14). The unexpected result of the referendum led to Cameron’s resignation, and a

particularly troubled period in the UK political scenario began. After two years of

negotiating with the EU to make Brexit take effect, his successor, the Conservative

Theresa May, also resigned as Prime Minister. Then, the former Mayor of the City

of London, Boris Johnson, clearly in favour of Brexit, became Prime Minister.

3. Mood, Modality and Lexical Modals

In a broad and traditional sense, the distinction between mood and modality is

established by defining the former as a grammatical category that, in English, can be

expressed in different ways. The latter, in contrast, is a semantic category related to the

different implications expressed in the sentence. As Nuyts states, these two concepts

have been the subject of numerous linguistic studies, both synchronic and diachronic

(2016: 2). However, although mood and modality often appear together, the scholar

A Corpus Study of Brexit Political Discourse: Exploring Modality

miscelánea 70 (2024): pp. 67-92 ISSN: 1137-6368 e-ISSN: 2386-4834

points out that while “mood is the older term”, “modality […] has become the

absolutely dominant concept in the last several decades” (Nuyts 2016: 1).

According to Huddleston (1988a; 1988b), mood is a grammatical property of the

verb phrase, similar to tense or aspect, and can be marked analytically by the

presence of (semi)auxiliary modal verbs. In addition, the subjunctive and the

imperative are considered inflectional mood-related forms. Nuyts (2016) remarks

that the term mood has been used in the literature to refer to three linguistic

domains: the domain of the grammatical characteristics and formal realisations in

the verb phrase, the dichotomy of indicative and subjunctive, and the domain

between the distinction of sentence types and their illocutionary classification.

Nevertheless, it could be said that mood is “the grammatical coding of modal

meaning in verb inflections” (Depraetere and Reed 2020: 207). As Khomutova

concludes, mood is also considered “the morphological means of expressing

modality” (2014: 400).

Modality is defined as a semantic category related to different meanings or

functions, which, as Huddleston et al. point out, “express certain kinds of alteration

in the relation of clause contents to reality” (2021: 70). The authors state that

“modality relates to the ways in which the possible situations described in a clause

can relate to reality” (2021: 69). In this regard, a given situation can be expressed

in various ways, which affect and determine the sense and meaning of what is

described in the clause. Although modality is mainly expressed in English by non-

inflectional resources such as modal verbs, other lexical categories such as adverbs,

adjectives or nouns can be related to the expression of modality.

Since modality relates to how events are presented in an utterance, the speaker’s

perspective and involvement in these events become particularly meaningful.

Accordingly, a distinction is made in the literature between what is considered

modal and non-modal, and this distinction is associated with the dichotomy

between factual and non-factual (Palmer 2001: 1). As Traugott states, “modal

utterances are non-factual and […] involve speaker’s comments on the necessity or

possibility of the state of affairs” (2011: 382).

The concepts of possibility and necessity become particularly relevant in the

context of modality, since the traditional and most widely used classification among

scholars (Quirk et al. 1985; Huddleston 1988a and 1988b; Bybee and Fleischman

1995; Palmer 2001) is to determine whether events are presented as something

that may happen or as something necessary. Although there is no agreement in the

literature when it comes to establishing a taxonomy of types of modality (Nuyts

2016), in a traditional and general sense, modality can be epistemic and deontic.

There is also a lack of consensus regarding terminology and the typology used to

distinguish modals. However, the two most widely accepted terms in the literature

Encarnación Almazán-Ruiz and Aroa Orrequia-Barea

miscelánea 70 (2024): pp. 67-92 ISSN: 1137-6368 e-ISSN: 2386-4834

are epistemic and deontic. Scholars (e.g. Rozumko 2019; Huddleston and Pullum

2002) use the etymology of both terms to clarify them. Both terms, epistemic and

deontic, are derived from the Greek knowledge and binding, respectively. That is

the reason why epistemic modality is related to “qualifications concerning the

speaker’s knowledge”, and deontic modality is connected to “a matter of imposing

obligation or prohibition, granting permission, and the like” (Huddleston and

Pullum 2002: 178).

Therefore, establishing the differences between the two types of modality2 entails

determining the speaker’s viewpoint on the events conveyed in the utterance.

According to Palmer, “with epistemic modality, speakers express their judgements

about the factual status of the proposition” (2001: 8), whereas “deontic modality

relates to obligation or permission, emanating from an external source” (2001: 9).

Palmer’s words can be used to determine a broad distinction between the two

types of modality. Nevertheless, more significant traits are worth mentioning.

As Rozumko states, it is necessary to consider that linguistic communication

involves not only the speaker but also the addressee and the social context in which

communication takes place (2019: 19-20). Hence, when analysing modality, it is

important to consider that events or what is said by the speaker are rarely intended

to express their viewpoint. Moreover, Nuyts points out that in both types of

modality, a scale can be established to determine the degree of certainty or

possibility in the case of epistemic modality and the level of obligation regarding

deontic modality (2016: 36-39).

Even though modality in English is mainly expressed by the presence of (semi)

auxiliary modal verbs in the utterance, Huddleston and Pullum state that there are

other “items expressing the same kind of meaning as the modal auxiliaries, but

which do not belong to the syntactic class of auxiliary verbs” (2002: 173). The

scholars use the term “lexical modals” to refer to these items and include “adjectives

like possible, necessary, likely, probable, bound, supposed, adverbs like perhaps, possibly,

necessarily, probably, certainly, surely, verbs like insist, permit, require, and nouns

like possibility, necessity, permission, and similar derivatives” (2002: 173).

Modal meanings are “understood to involve subjectivity or grounding in the

speaker’s perspective” (Traugott 2011: 390), and including modal words in the

proposition may reveal the speaker stance. Accordingly, Simon-Vandenbergen

claims that modal choices unveil “the speaker’s position in the discourse” and that

a high degree of commitment to a specific position reflects the speaker’s aim to

convince others of a questionable standpoint (1997: 353). Therefore, it is worth

studying the use of lexical modals to determine whether the political protagonists

of this study present facts or whether they try to avoid positioning themselves

clearly in revealing the truth value of a given argument.

A Corpus Study of Brexit Political Discourse: Exploring Modality

miscelánea 70 (2024): pp. 67-92 ISSN: 1137-6368 e-ISSN: 2386-4834

4. Data and Methods

4.1. Procedure of Analysis

The first step of our study was to compile the data. Both politicians’ speeches

were downloaded from the internet using the programming language R.3 Boris

Johnson’s corpus was extracted from the British Government’s official website.4

We restricted the query using the different options the website displayed.

Firstly, we selected the dates from 24 June, 2019, until 31 January, 2020.

Secondly, we set some keywords as filters, namely ‘PM’ (Prime Minister) and

‘speech’. Additionally, we restricted the query to search for speeches delivered

only by the Prime Minister. Once downloaded, we checked that the word

‘Brexit’ was mentioned to avoid those in which it did not appear. Eleven

speeches were retrieved in this case, which together made for a corpus of

16,869 tokens.

The Labour Party’s official website allows users to download their leader’s speeches

from “The Jeremy Corbyn Archives”.5 In this case, the same criteria were followed:

we set the dates from June 24, 2019 to January 31, 2020, downloaded all the

speeches retrieved by the query, and checked whether the word ‘Brexit’ was

mentioned in each speech, discarding the ones in which the word did not appear.

The corpus comprises 35,251 tokens from 28 speeches.

In the second stage, we compiled an exhaustive list of lexical modals using

Huddleston and Pullum’s (2002) classification as a foundation. Following their

criteria, our classification also incorporated “similar derivatives” (Huddleston and

Pullum 2002: 173) regarding morphology and semantics; the researchers’

proficiency in the English language, their intuition and the use of English

dictionaries were essential at this stage. Table 1 shows the classification of lexical

modals in terms of word class and ordered alphabetically.

Adjectives Adverbs Nouns Verbs

(un)bound

(un)certain

(un)clear

convinced

(un)likely

(un)necessary

(im)possible

(im)probable

supposed

(un)sure

actually

(un)certainly

(un)clearly

indeed

maybe

(un)necessarily

perhaps

(im)possibly

(im)probably

surely

truly

assurance

(un)certainty

intention

necessity

permission

possibility

promise

proposal

reassurance

request

requirement

allow

assure

convince

ensure

intend

let

permit

promise

propose

require

Table 1. Compilation of lexical modals classified by word class

Encarnación Almazán-Ruiz and Aroa Orrequia-Barea

miscelánea 70 (2024): pp. 67-92 ISSN: 1137-6368 e-ISSN: 2386-4834

The analysis was conducted using the software Sketch Engine (Kilgarriff et al.

2014). This tool was used to manage the two sub-corpora analysed in this paper.

We used the concordance function to query the items in Table 1 and checked

whether they appeared in the corpora. Secondly, the concordance lines were

downloaded, and the authors manually annotated these stretches of text for the

type of modality according to the distinction made in Section 2, namely epistemic

and deontic. The small size of the corpus also allowed us to annotate each lexical

modal in terms of degree of certainty: strong, medium or neutral modality. Then,

a comparison was made between the two annotations, and an agreement was

reached between the annotators in cases of doubt.

The statistical test used to carry out the analysis was the log-likelihood measure.

The cut-off point for the p-value was established at p<0.05 “to reject the null

hypothesis and conclude that the observed difference is unlikely to be due to

chance and therefore is statistically significant” (Brezina 2018: 12). Thus, if the

result is larger than 3.84 (the cut-off point for significance at p<0.05), it is

statistically significant (Brezina 2018: 84). To calculate such a test, we used Paul

Rayson’s online Log-Likelihood Calculator.6

The last step was to look up the lexical modals in a reference corpus. CPQWeb

software (Hardie 2012) was used to perform the comparison with a reference

corpus, that is, the British National Corpus (BNC). The use of this tool instead of

Sketch Engine was motivated by the part-of-speech tagging available in the former,

as CPQWeb allows the user to specify word class using tags, particularly the UCREL

CLAWS C6 tagset.7 The verbs were queried using the lemma search. Verbs and

nouns that coincide in form, e.g. promise, were looked up using tags, i.e. NN0 for

nouns, which refers to neutral for number, such as in aircraft; and VVB for verbs,

which stands for the base form of lexical verbs, except the infinitive, such as in take

or live. Finally, for the (semi)modal auxiliaries, two different tags have been

employed, namely VM0, which refers to modal auxiliary verbs such as can, could

or will and VMK for modal catenative (ought to, used to).

4.2. Corpus Description and Analytical Framework

The corpus consists of transcribed speeches delivered by Johnson, Prime Minister

of the Conservative Party, and Corbyn, the leader of the opposition at that moment,

of the Labour Party8. Table 2 shows the distribution of the two sub-corpora.

Tokens Types

Boris Johnson 16,869 2,926

Jeremy Corbyn 35,251 4,149

TOTAL 52,120 7,075

Table 2. Distribution of the corpora

A Corpus Study of Brexit Political Discourse: Exploring Modality

miscelánea 70 (2024): pp. 67-92 ISSN: 1137-6368 e-ISSN: 2386-4834

The methodology used to carry out this study is CADS, as it combines corpus

linguistics and discourse studies, which are two methods commonly considered

quantitative and qualitative, respectively (Arcano 2020; Partington 2004). CADS

aims to uncover meanings that are not obvious (Partington et al. 2013) but not in

a “particular language or linguistic variety but rather a particular situation, purpose

or function repeatedly enacted within a speech community” (Taylor and Marchi

2018: 61).

According to Morley, combining these two methods “increases the researcher’s

analytical capacity to an extent greater than would be predicted from the sum of

the two” (2011: 10). On the one hand, corpus linguistics allows researchers to

manage huge amounts of texts using digital technology to study language,

providing objectivity to the research since patterns that are not visible at first

glance can be uncovered (Taylor and Marchi 2018: 2). On the other hand,

discourse studies aims to analyse how language is employed, revealing subtle or

underlying discourse. The corpus becomes a valuable tool for discourse analysts,

allowing researchers to explain how language constructs discourses or reality

(Baker 2006: 183). Researchers must account for the patterns found in the text,

as “interpretation is also an essential step in any corpus-based analysis” (Biber in

Baker 2006: 2). In addition, to carry out the statistical analysis intended,

descriptive statistics have been employed, with a specific focus on frequency

distribution. This method aims to determine the frequency of occurrence in the

dataset.

5. Results

This section aims to answer the first two research questions. RQ1 explores how

modality is conveyed through lexical modals in the speeches of the two politicians

from a quantitative perspective. As mentioned previously, lexical modality is

expressed using items with the same meaning as (semi)auxiliary modal verbs but

that are not classified as such. Therefore, modality can be expressed through

nouns, adjectives, adverbs and other lexical verbs. To answer this first research

question, we thoroughly searched each sub-corpus for every lexical modal listed in

Table 1 of Section 4.1. The results are depicted in Figure 1.

As Figure 1 shows, both politicians rely on verbs and adjectives to express modality.

However, the least frequently used word class to express modality differs between

the two politicians. Johnson uses nouns less frequently (10.11 per cent), while

adverbs were the least used category by Corbyn (9.2 per cent). However, Corbyn

used more nouns (19.54 per cent vs 10.11 per cent) and verbs (46.36 per cent vs

40.43 per cent) than his counterpart.

Encarnación Almazán-Ruiz and Aroa Orrequia-Barea

miscelánea 70 (2024): pp. 67-92 ISSN: 1137-6368 e-ISSN: 2386-4834

In the following lines, the use of each word class will be explained.9 Table 3 shows

the results of modal verb use by Johnson and Corbyn.10

Boris Johnson Jeremy Corbyn

Verb Hits NF LL Hits NF LL

let 23 13.6 253.16 52 14.8 580.50

ensure 20 11. 9 250.68 24 6.9 221.23

allow 15 8.9 154.81 17 4.9 154.71

promise 6 3.6 70.77 10 2.9 113.43

propose 4 2.4 42.49 5 1.4 47.98

insist 21.2 24.52 0 0 0.01

assure 21.2 22.15 3 0.9 31.23

intend 21.2 1 7. 1 0 0 0.04

require 21.2 13.12 10 2.9 82.91

Figure 1. Percentages of lexical modals used by each politician

Table 3. Number of hits, normalised frequency and log-likelihood results of lexical modal verbs

A Corpus Study of Brexit Political Discourse: Exploring Modality

miscelánea 70 (2024): pp. 67-92 ISSN: 1137-6368 e-ISSN: 2386-4834

As shown in Table 3, let is the most frequently used verb in both sub-corpora

(253.16 in Johnson’s sub-corpus and 580.5 in Corbyn’s), followed by the verbs

ensure (250.68 and 221.23, respectively) and allow (154.81 and 154.71,

respectively). The next most significant verb in both sub-corpora is promise (70.77

in Johnson’s and 113.43 in Corbyn’s). Another frequent verb in Johnson’s sub-

corpus is propose (42.49), and in Corbyn’s, it is require (82.91). It is also

noteworthy that some verbs occur in one sub-corpus but not in the other. For

instance, the verbs insist and intend were employed in Johnson’s speeches but not

in Corbyn’s.

Adjectives are the second most prevalent grammatical category used by Johnson

(31.91 per cent) and Corbyn (24.9 per cent). Table 4 shows the adjectives retrieved

from both sub-corpora.

Boris Johnson Jeremy Corbyn

Adjective Hits NF LL Hits NF LL

possible 23 13.7 254.48 5 1.4 32.82

clear 14 8.3 144.69 34 9.8 361.57

sure 7 4.2 65.65 6 1.7 45.64

necessary 6 3.6 57.88 6 1.7 49.08

impossible 4 2.4 43.03 1 0.3 6.54

convinced 31.8 37.45 0 0 0.01

certain 21.2 14.14 1 0.3 4.28

likely 1 0.6 5.59 7 2 55.87

supposed 0 0 0.01 1 0.3 7. 1 2

bound 0 0 0.01 1 0.3 6.82

uncertain 0 0 0.00 1 0.3 9.01

unlikely 0 0 0.01 1 0.3 6.94

unnecessary 0 0 0.00 1 0.3 9.16

Table 4. Number of hits, normalised frequency and log-likelihood results of lexical modal

adjectives

The analysis reveals that possible is the most frequently used adjective by Johnson,

with a log-likelihood result of 254.48. In addition, Johnson employs other

adjectives with regularity, including clear (144.69), sure (65.65), necessary

(57.88) and impossible (43.03). In contrast, clear is the adjective with the greatest

log-likelihood result in Corbyn’s sub-corpus (361.57). It may be worth noting

that the use of likely (55.87), necessary (49.08) and sure (45.64) are also

Encarnación Almazán-Ruiz and Aroa Orrequia-Barea

miscelánea 70 (2024): pp. 67-92 ISSN: 1137-6368 e-ISSN: 2386-4834

significant. Another remarkable aspect of Corbyn’s speech is that certain

adjectives are absent from Johnson’s sub-corpus, such as supposed, bound,

uncertain, unlikely and unnecessary. In the same vein, impossible, certain and

convinced, though retrieving hits in Corbyn’s sub-corpus, are not statistically

significant, as the result is less than 6.63 in the first two cases and less than 3.84

in the last one.

Adverbs are the third most frequently used word class in Johnson (17.55 per

cent) but the least used in Corbyn (9.2 per cent). Therefore, Corbyn infrequently

uses adverbs to express modality in his speeches. The results can be seen in

Table 5.

Boris Johnson Jeremy Corbyn

Adverb Hits NF LL Hits NF LL

indeed 14 8.3 157.89 0 0 0.07

perhaps 6 3.6 50.36 3 0.9 16.70

truly 4 2.4 49.29 10 2.9 126.78

actually 5 3 42.9 7 2 54.49

necessarily 21.2 19.54 0 0 0.00

certainly 21.2 14.87 1 0.3 4.63

maybe 0 0 0.02 2 0.6 14.29

surely 0 0 0.01 1 0.3 6.78

Table 5. Number of hits, normalised frequency and log-likelihood results of lexical modal

adverbs

Table 5 shows that, although Corbyn uses this resource less, the variety of adverbs

is more extensive than in Johnson’s case. The most widely used modal adverbs in

Johnson’s sub-corpus are indeed (157.89), perhaps (50.36) and truly (49.29).

However, in the case of Corbyn, the most prevalent are truly (126.78) and actually

(54.49). As in previous categories, some adverbs do not retrieve hits in either

corpus. For instance, maybe and surely do not appear in Johnson’s sub-corpus,

whereas indeed and necessarily do not appear in Corbyn’s. The case of indeed is also

remarkable, as it is the most significant item in Johnson’s adverbs, but does not

occur in Corbyn’s sub-corpus.

The last word category to comment on is lexical modal nouns. As previously

stated, these nouns are the least used resource by Johnson (10.11 per cent), but

they are used by Corbyn more than his opponent (19.54 per cent). Table 6

summarises the results of both sub-corpora.

A Corpus Study of Brexit Political Discourse: Exploring Modality

miscelánea 70 (2024): pp. 67-92 ISSN: 1137-6368 e-ISSN: 2386-4834

Boris Johnson Jeremy Corbyn

Nouns Hits NF LL Hits NF LL

promise 6 3.6 77.85 12 3.5 154.63

proposal 5 3 51.24 18 5.2 203.98

permission 31.8 34.8 0 0 0.01

certainty 21.2 24.83 3 0.9 35.26

possibility 21.2 17.48 2 0.6 14.55

uncertainty 1 0.6 9.92 7 2 86.34

intention 0 0 0.01 4 1.2 37.93

assurance 0 0 0.00 2 0.6 21.11

reassurance 0 0 0.00 2 0.6 26.02

necessity 0 0 0.00 1 0.3 8.96

Table 6. Number of hits, normalised frequency and log-likelihood results of lexical modal nouns

As Table 6 shows, the noun proposal (203.98) is Corbyn’s most frequently used

noun, whereas promise is the most used by Johnson (77.85), making his

political commitment explicit. Although Corbyn also uses the noun promise

(154.63) to express modality, Johnson, on the other hand, favours the use of

proposal (51.24) and permission (34.8). As expected, Corbyn makes use of a

wider variety of nouns. For this reason, many nouns do not retrieve hits in

Johnson’s sub-corpus: intention, assurance, reassurance and necessity. Similarly,

the noun permission, which occurs in Johnson’s texts (34.8), does not appear

in Corbyn’s.

RQ2 compares the frequency of lexical modals to (semi)auxiliary modal verbs. For

this question, the results of each word class will be compared with a previous study

on (semi)auxiliary modal verbs using the same corpus (Orrequia-Barea and

Almazán-Ruiz 2021). In addition, results for each politician regarding the

expression of modality will be contrasted with how modality is expressed in a

reference corpus such as the BNC. The results of this analysis will provide

information about whether modality is a resource typically used by politicians in

their speeches.

Results were similar in both politicians in the case of (semi)auxiliary verbs. Whereas

Johnson’s most frequently used (semi)auxiliary modal verbs are will, can, going to,

would and should, Corbyn coincides in the first two, will and can, but the following

ones are would, should and could. Regarding the semantic implications, both

politicians use the modal verbs similarly: the epistemic readings of will and can and

Encarnación Almazán-Ruiz and Aroa Orrequia-Barea

miscelánea 70 (2024): pp. 67-92 ISSN: 1137-6368 e-ISSN: 2386-4834

the deontic use of will. It makes sense that the most recurrent meaning is epistemic

rather than deontic since the former implies possibility, whereas the latter,

obligation. Politicians are not expected to place obligations on their electorate if

they want to win votes. As previously stated, will is also used in this reading to

express promises regarding the deontic modality (Orrequia-Barea and Almazán-

Ruiz 2021).

In this context, (semi)auxiliary modal verbs are used more commonly by politicians

than lexical modals. Johnson’s use of (semi)auxiliary modal verbs represents 14.55

per cent of the words in the corpus, whereas the one referring to lexical modals is

1.11 per cent. In this regard, Corbyn uses (semi)auxiliary modal verbs slightly

more than Johnson, with a result of 15.2 per cent, while he uses lexical modals less

than his opponent, 0.74 per cent (Orrequia-Barea and Almazán-Ruiz 2021). As

can be seen, in the political arena of Brexit, (semi)auxiliary modal verbs are the

most widely used resource to express modality. This result was expected, as (semi)

auxiliary modal verbs are the words expressing modality par excellence in the

English language.

Nevertheless, the comparison is not complete unless we check how modality is

expressed in a reference corpus, that is, how frequently speakers use the different

ways of expressing modality. To do so, (semi)auxiliary modal verbs and lexical

modals have been queried in the BNC.

The results retrieved in the reference corpus show that (semi)auxiliary modal

verbs are used less frequently (1.3 per cent). As mentioned above, Johnson’s and

Corbyn’s percentages were 14.55 and 15.2 per cent, respectively. Therefore, we

can see that (semi)auxiliary modal verbs as a resource to express modality are

overused in the context of Brexit, as politicians used this strategy twice as much

as the language speakers, showing their stance, against or in favour, in the

discourse.

Concerning lexical modals, it is unsurprising that these items (1.11 vs 0.74 per

cent) are not used as frequently by politicians as (semi)auxiliary modal verbs. As

previously mentioned, the latter has been traditionally considered the resource

used to express modality par excellence. However, when comparing the lexical

modal results obtained in this study with the BNC occurrences, the BNC results

are closer to Corbyn’s use of lexical modals than Johnson’s. In fact, lexical

modals have a 0.28 per cent presence in the BNC, whereas Cobyn’s is 0.74 per

cent. In the case of Johnson, the use of lexical modals is 1.11 per cent. The data

indicate an overuse of lexical modals by Johnson. Table 7 summarises the

comparisons.

A Corpus Study of Brexit Political Discourse: Exploring Modality

miscelánea 70 (2024): pp. 67-92 ISSN: 1137-6368 e-ISSN: 2386-4834

Boris Johnson Jeremy Corbyn BNC

(Semi)auxiliary modal verbs 14.55 15.2 1.3

Lexical modals 1. 11 0.74 0.28

Table 7. Use of (semi)auxiliary modal verbs and lexical modals expressed in percentages

The data in Table 7 answer the third research question, which will be discussed in

more detail in Section 6. As can be seen, Johnson uses modality more frequently

in his discourse. Even though Corbyn uses (semi)auxiliary modal verbs more than

the Conservative leader, lexical modals are more recurrent in Johnson’s discourse.

It is noteworthy that modality occurs more frequently in the analysed corpus than

in the BNC. As shown in Table 7, lexical modals are a more prominent resource in

the political discourse than in general language. However, in the literature, more

attention has been given to (semi)auxiliary modal verbs than to this linguistic

device. In this sense, this study contributes to emphasising their importance in

presenting facts to the audience.

6. Modelling Lexical Modals

This discussion section presents concrete examples from both sub-corpora and

describes the modality types established in the theoretical framework used. Thus,

this section aims to answer RQ3 from a qualitative perspective, considering the

different degrees of modality and the various lexical modals used to express it. In

the political context under discussion, lexical modality can reveal framing strategies,

that is, the speaker’s stance and viewpoint in presenting events and arguing for or

against them (Cheng 2016).

As described in section 3 above, epistemic modality connects the speaker’s

knowledge and the referenced facts or actions presented in the discourse. According

to Rozumko, certainty is not an absolute category but can be expressed according

to a scale which establishes the degree of certainty (2019: 37); therefore, a

distinction can be made between strong, medium and neutral modality. Thus,

examples from the corpus will be used to discuss their semantic implications and

politicians’ viewpoints when presenting facts regarding Brexit.

The adverb necessarily mainly expresses the epistemic meaning of necessity.

However, the syntactic polarity changes when combined with not, and it is used to

express that what is said is possibly but not definitely true. This adverb falls within

the strong modality scale according to Huddleston and Pullum (2002: 207), and

in the corpus, only Johnson uses this formula, as shown in Examples 1 and 2.

Encarnación Almazán-Ruiz and Aroa Orrequia-Barea

miscelánea 70 (2024): pp. 67-92 ISSN: 1137-6368 e-ISSN: 2386-4834

(1) That is not necessarily the case in the digital age.

(BJ - epistemic necessity)

(2) When I fully intend to be around, though not necessarily in this job, we

will look back on this period, this extraordinary period, as the beginning

of a new golden age for our United Kingdom.

(BJ - epistemic necessity)

Other adverbs, such as certainly, are also used to express necessity. As seen in the

following examples, both politicians use the word to indicate that what is stated

must necessarily be so. Besides, it should be considered that even though “epistemic

certainly belongs with the strong modals”, it “does not suggest any reasoning

from evidence” (Huddleston and Pullum 2002: 207).

(3) It isn’t really the fault of the places and it certainly isn’t the fault of the

people growing up there.

(BJ - epistemic necessity)

(4) That must be followed by radical and decisive action that will only be

delivered by a Labour government. It certainly won’t come from the

Tories.

(JC - epistemic necessity)

As Rozumko points out, this epistemic adverb implies a restricted knowledge of

the facts on the speaker’s behalf (2019: 440). Despite this, it is used to emphasise

what appears true from their perspective, as demonstrated in Examples 3 and 4.

According to Simon-Vandenbergen, expressions of certainty are a “tactic strategy”

to “persuade the audience of the essential rightness of their claims” (1996: 408).

Through this adverb, both politicians establish the essential accuracy of their

assertions, communicating their confidence in the validity of their arguments.

Similarly, the speaker’s degree of conviction is made evident by the adjective

certain, which determines their position on the events described. The meaning of

Examples 5 and 6 are closely related to the use of the modal verb must when

expressing logical necessity, since the speaker expresses that given the evidence, no

other conclusion can be inferred from it.

(5) I’m absolutely certain there will be displays celebrating the dawn of a new

age of electric vehicles, not just cars or buses, but electric planes, made

possible with battery technology being developed now in the UK.

(BJ - epistemic necessity)

(6) That is why Labour will not support this Bill, as we remain certain there

is a better and fairer way for this country to leave the EU.

(JC - epistemic necessity)

A Corpus Study of Brexit Political Discourse: Exploring Modality

miscelánea 70 (2024): pp. 67-92 ISSN: 1137-6368 e-ISSN: 2386-4834

It is important to note that the use of these linguistic devices indicates a high level

of confidence in the speaker’s position, which is clearly emphasised in Example 5

through the adverb absolutely. Johnson uses this adverb to add force to his degree

of certainty. In fact, the degree of certainty becomes clearer and more assertive

when the noun category is used because the noun certainty formulates the facts in

the utterance, indicating confidence and assertiveness in the speaker’s perspective.

“It creates the image of a knowledgeable person who knows what he or she is

talking about and who therefore deserves public trust and political power” (Simon-

Vandenbergen 1996: 390). Thus, it inspires confidence, as the speaker seems to be

committed to the truth of their claims (Simon-Vandenbergen 1996: 390).

(7) I can assure them that under this government, they will have the absolute

certainty of the right to leave and remain.

(BJ - epistemic necessity)

Once more, the Conservative leader emphasises his speech by highlighting the

degree of certainty with the premodifying adjective absolute. This strategy makes his

discourse more assertive, making it easier for him to influence the audience,

persuade them to trust his arguments and establish a greater sense of credibility and

authority. Interestingly, the Tory leader is the only one who introduces the adjective

convinced in his speech, thus demonstrating that he is entirely certain about the

facts to which he refers, which must necessarily be so, as Example 8 shows.

(8) I am convinced that an overwhelming majority in this House, regardless of

our personal views, wishes to see Brexit delivered in accordance with the

referendum.

(BJ - epistemic necessity)

Likewise, the certainty of what is expressed in a statement can be shown using the

adjective clear and the adverb clearly. In any context, the speaker could denote a high

level of confidence in a statement when using them. This linguistic strategy can be

considered a means of emphasising one’s conviction and conveying a sense of clarity

in the message. The speaker can enhance the persuasiveness of their argument and

establish a greater sense of credibility and authority, as the following examples show:

(9) If an agreement is to be reached, it must be clearly understood that the

way to the deal goes by way of the abolition of the backstop.

Our vision is clear: to unite and level up across the whole United Kingdom.

(BJ - epistemic necessity)

(10) It is becoming increasingly clear that this reckless government only has

one plan, to crash out of the EU without a deal.

It is clear they have tried to hide the truth of a no deal Brexit.

(JC - epistemic necessity)

Encarnación Almazán-Ruiz and Aroa Orrequia-Barea

miscelánea 70 (2024): pp. 67-92 ISSN: 1137-6368 e-ISSN: 2386-4834

Nevertheless, the adjective necessary can also be interpreted from the perspective

of deontic modality, as it presents the facts as requirements. According to

Huddleston and Pullum, deontic necessity is very close to obligation, and the

adjective mentioned above usually “takes a mandative or infinitival complement”

(2002: 207). In Example 11, the Labour leader steadfastly commits himself to take

all necessary measures to avoid a no-deal departure from the EU.

(11) We will do everything necessary to stop a disastrous No Deal for which this

government has no mandate.

(JC - deontic necessity)

Deontic necessity is also expressed in the corpus by means of lexical verbs such as

require, which is more recurrent in Corbyn’s speeches than in Johnson’s. From

Corbyn’s point of view, the country requires and demands certain conditions

regarding Brexit, as shown in Example 12.

(12) The office of the Prime Minister requires integrity and honesty.

This inequality is unsustainable and immoral. Ending it requires

government action.

(JC - deontic necessity)

While necessity and certainty correspond to strong modality, the medium degree

corresponds to the meaning of probability (Huddleston and Pullum 2002).

However, in the corpus analysed in this study, there were no hits for uncertain

terms such as probably or probable. As a result, politicians presented their discourse

as factual. This absence indicates that both leaders were confident in their assertions

and did not feel the need to rely on speculative language. Nevertheless, it is worth

mentioning that weak modality does appear in the corpus through lexical modals

such as possible, perhaps or maybe.

The adjective possible presents events as likely to occur. Both politicians use this

item in the analysed corpus, though Johnson does so more frequently. When a

statement expresses the possibility of a fact, the speaker presents it as potentially

happening but not entirely certain of its fulfilment. In Example 13, Johnson

presents the facts as a probable occurrence that is further emphasised by the

comparative form. This construction makes him add emphasis and significance to

the possibility, lending greater credence to the argument presented. Using the

comparative form highlights the likelihood of the event, resulting in a more

compelling and impactful presentation than when using only the adjective in its

base form. Overall, Johnson’s linguistic choice in this example appears to be

discoursally effective in conveying the importance of the presented information.

(13) By 2050, it is more than possible that the United Kingdom will be the

greatest and most prosperous economy in Europe.

(BJ - epistemic possibility)

A Corpus Study of Brexit Political Discourse: Exploring Modality

miscelánea 70 (2024): pp. 67-92 ISSN: 1137-6368 e-ISSN: 2386-4834

Similarly, in Example 14, Corbyn expresses his factual appraisal. In his view, the

achievement of a positive goal, such as reducing carbon emissions, would imply

a necessary increase in public spending, which he perceives as an unfavourable

outcome, explicitly manifest with the premodification (massive) of the head

noun.

(14) Bringing our carbon emissions down [...] will only be possible with massive

public investment in renewable energy and green technology.

(JC - epistemic possibility)

The adverbs perhaps and maybe are used when the speaker intends to express that

events are possible despite their uncertainty about them (Rozumko 2019: 443).

According to the author, the use of perhaps “signals that an opinion or interpretation

is offered to the addressee(s) for consideration” (444). The following example

illustrates how the Tory leader presents the possibility that what is stated can be

considered from a different perspective.

(15) They can see that we have a clear vision for our future relationship with the

EU - something that has perhaps not always been the case.

(BJ - epistemic possibility)

However, this epistemic adverb is also used from a concessive perspective; when it

is uttered, the speaker presents a counterpoint to the facts stated. This use of

perhaps can be observed in Example 16 and, as Rozumko notes, “the concessions

marked by perhaps emphasise the positive aspects of the situations, and signal

solidarity with the addressee” (2019: 445).

(16) I don’t think I’ve heard a single member call for an ever closer union or

ever deeper integration or a federal destiny [...] perhaps I’ve missed it but

I don’t think I’ve heard much of it Mr Speaker.

(BJ - epistemic possibility)

The adverb perhaps is also used to express a polite request. In the corpus, Corbyn

uses it to indirectly invite the Prime Minister to explain some issues he does not

consider entirely clear, as the examples in 17 show. This adverb leaves open the

possibility that Johnson is not, in fact, capable of explaining what Corbyn is

demanding of him. Through this discursive strategy inherent to the engagement

system (Martin and White 2007: 105), Corbyn adeptly aligns with a section of the

audience that echoes his doubts about Johnson’s inability to offer a satisfactory

response.

(17) Perhaps he’d like to explain why these documents confirm the US is

demanding the NHS is on the table in the trade talks?

Perhaps he can tell us what has changed?

(JC - deontic possibility)

Encarnación Almazán-Ruiz and Aroa Orrequia-Barea

miscelánea 70 (2024): pp. 67-92 ISSN: 1137-6368 e-ISSN: 2386-4834

Huddleston and Pullum also observe that certain lexical verbs such as allow and

permit “can express deontic possibility, permission” (2002: 208). In the analysed

corpus, the verb ‘allow’ is one of the most frequent lexical modal items used by

both politicians. As Examples 18 and 19 show, both politicians use it to present a

plausible situation:

(18) A deal that allows us to create a new shared destiny with them.

(BJ - deontic possibility)

(19) We can create natural solutions to bring down emissions and allow our

wildlife to flourish, because currently wildlife is in crisis.

(JC - deontic possibility)

However, the verb allow is also used to clarify what both leaders are reluctant to

accept. The following examples show that the two politicians’ clashing positions

express a lack of permission from a deontic perspective.

(20) Leaving the EU is a massive economic opportunity - to do the things

we’ve not been allowed to do for decades.

(BJ - deontic permission)

(21) They’re not allowed to pledge not to sell out our NHS in a trade deal with

Donald Trump.

(JC - deontic permission)

In both cases, allow is used in passive structures, thus emphasising that the patient

subjects of the sentence receive the action of the verb and avoid mentioning the

agent subject. However, while Johnson focuses on things the UK has been unable

to do because of the EU, Corbyn highlights the aspects his party will not permit

the government to do.

As previously mentioned, deontic modality is also concerned with expressing other

meanings such as prohibition, desires or promises. In the political context analysed

here, Johnson’s most frequent noun is promise. As is known, a promise entails that

one will certainly do something for somebody else, and frequently, politicians

show the audience their intentions as promises, as Johnson does in Example 22.

(22) We are going to fulfil the repeated promises of parliament to the people.

We will not accept any attempt to go back on our promises or scrub that

referendum.

(BJ - deontic volition- promise)

As observed, the Prime Minister combines the lexical modal promise with (semi)

auxiliary modal verbs —be going to and will— which allows him to emphasise and

reinforce the idea of volition and his implication of acting. However, this noun

allows Corbyn to criticise his opponent by mentioning all the promises the Tory

A Corpus Study of Brexit Political Discourse: Exploring Modality

miscelánea 70 (2024): pp. 67-92 ISSN: 1137-6368 e-ISSN: 2386-4834

government has failed to keep regarding Brexit. The political confrontation is

evident in the Labour leader’s statements, as observed in Examples in 23:

(23) Johnson’s deal will be just the beginning of years more painful negotiations

and broken promises.

Who cannot tell the truth about the Brexit negotiations that he has so far

failed to deliver on, or one who keeps on making promises that turn out to

be a mirage the following day.

(JC - deontic volition - promise)

Although both politicians use the verb promise, it is worth mentioning that when

this happens, the Conservative leader commits to doing what they say they will do

and will give the people what they commit to. This is part of the framing strategy

of Brexit, as Examples in 24 illustrate:

(24) Mr Speaker, we promised the people we would get Brexit done.

So I have promised to find a long term solution to social care once and for

all. And that is what we will do.

(BJ - deontic volition- promise)

Once again, Corbyn uses this lexical modal verb —promise— to criticise the

Conservative government’s performance and directly accuses its leader of failing to

deliver on his promises. In the examples in 25, Corbyn highlights the Conservative

government’s failure to uphold their promises and fulfil their commitment. From

them, it can be inferred that such inconsistency can lead to a loss of confidence and

trust from the British people.

(25) His predecessor promised to end austerity but spectacularly failed to

deliver.

Friends, today is the 31st of October, the day Boris Johnson promised we

would leave the EU. [...] But he has failed. And that failure is his alone.

(JC - deontic volition - promise)

Interestingly enough, the Labour leader prefers to introduce the noun proposal in

his speeches and uses this noun more frequently than Johnson. The difference in

meaning between the two nouns —promise and proposal— is rather evident since a

proposal should be understood as a plan or idea that is suggested. Corbyn uses this

noun to criticise the Tory government’s actions again and question whether its

ideas for a successful Brexit could be effective, as shown by examples in 26.

(26) These proposals are nothing more than a cynical attempt by the Prime

Minister to shift the blame for his failure to deliver.

Mr Speaker, the Government’s proposals are neither serious nor credible.

(JC - deontic volition - suggestion)

Encarnación Almazán-Ruiz and Aroa Orrequia-Barea

miscelánea 70 (2024): pp. 67-92 ISSN: 1137-6368 e-ISSN: 2386-4834

In Example 27, the Labour leader shows his sharpest criticism when he emphasises

that the Conservative government’s proposals are unrealistic and will, therefore,

be rejected in Brussels. The tone of the speech becomes more assertive when he

alludes to the Prime Minister to criticise his lack of involvement even though he is

aware of the facts.

(27) The current proposals would damage the whole UK economy.

The proposals are unrealistic and damaging and will —as I think the Prime

Minister knows— be rejected in Brussels, in this House, and in the

country.

(JC - deontic volition - suggestion)

The empirical data and corpus analysis illustrate how modality is strategically

employed as a “linguistic and ideological tool” to “frame their argument in a purpose

of persuading the electorate and soliciting their support” (Cheng 2016: 172). In this

sense, it can be seen how lexical modals are as effective as (semi)auxiliary modal verbs

when conveying modality. The speaker reveals his position to the audience through

modality since epistemic modality presents the facts as necessary or likely to happen.

On the other hand, deontic modality, which is much more complex because it

involves the speaker’s perspective on what is to be done, allows them to suggest,

recommend or propose something. Either type of modality shows the speaker’s

perspective and degree of involvement in the events expressed in the statement.

7. Conclusions

This corpus-based study has analysed the Brexit political discourse of the then UK

Prime Minister, Boris Johnson, and the leader of the opposition, Jeremy Corbyn,

from the Labour Party. The results have been shown from two perspectives:

quantitatively, we have presented the data regarding the frequency of occurrences

of lexical modals of both politicians when discussing Brexit in the corpus. On the

other hand, qualitative interpretation has made it possible to show a semantic

analysis of the lexical modals found in the corpus, thus distinguishing between

epistemic and deontic modality.

As far as RQ1 is concerned, the corpus analysis has shown that lexical modals are a

prominent resource employed by politicians. Adjectives and verbs are the most

common words used by both politicians. However, nouns are Johnson’s least

frequently used word class, and adverbs are the least frequently used by Corbyn.

In the case of Brexit, as illustrated, this strategy has been used more frequently

than in other contexts of the English language, which reveals the significance of

lexical modals in political discourse.

A Corpus Study of Brexit Political Discourse: Exploring Modality

miscelánea 70 (2024): pp. 67-92 ISSN: 1137-6368 e-ISSN: 2386-4834

However, when comparing lexical modals with (semi)auxiliary modal verbs (RQ2),

the latter are still the items used par excellence when expressing modality, as the

answer to the second research question has shown. The quantitative analysis has

revealed that politicians employ lexical modals as a prominent resource. When

compared with a reference corpus, Johnson and Corbyn use lexical modals with a

higher frequency than in the BNC. (Semi)auxiliary modal verbs are used twice in

the discourse of Brexit. In contrast, lexical modals are used similarly in Corbyn’s

case but are far more frequent in Johnson’s corpus.

Finally, the qualitative analysis has focused on a semantic analysis of the lexical

modals found in the corpus. The discussion to answer RQ3 has revealed that

politicians employ lexical modals to present factual information with high

assertiveness. In a sense, lexical modals help politicians to establish a tone of

confidence in their statements. This is evident in Johnson’s use of the (semi)

auxiliary modal verb will or the lexical modal promise to discuss Brexit, which

effectively enhances the credibility of his claims to voters. The analysis demonstrated

that the use of lexical modals is more common in Johnson’s speech than in

Corbyn’s. This linguistic resource serves as an effective tactic to obscure a clear

position regarding the facts. Thus, it becomes an advantageous strategy for the

speaker to refrain from (not) subtly committing himself. As a result, this article

contributes to research on modality in the English language, as it delves into

examining parts of speech other than (semi)auxiliary modal verbs in political

discourse. This study sheds light on the complexity of modality and its expression

through various lexical words and their use in the political context.

So far, lexical modals have been neglected in the literature, as studies on this

phenomenon are rather scarce as opposed to (semi)auxiliary modal verbs. However,

as this study shows, lexical modals are another resource used by politicians, which

should also be explored since they serve to convey the meanings of modality. This

study is original not only in highlighting the importance of these particular words

but also in encouraging other researchers to look into modality from this

perspective and the implications these lexical modals may have when expressing

modality.

Contribution of the Authors

The present paper is the outcome of the coordinated and collaborative work by the

two authors. Author 1 was in charge of the theoretical part of the work, while

Author 2 was responsible for compiling the corpus as well as quantitative analysis

and processing. Both authors actively participated in the study planning, manuscript

writing, and qualitative analysis, as well as reviewing and revising the entire work.

Encarnación Almazán-Ruiz and Aroa Orrequia-Barea

miscelánea 70 (2024): pp. 67-92 ISSN: 1137-6368 e-ISSN: 2386-4834

Works Cited

AnAnkO, Tatiana. 2017. “The Category of Evaluation in Political Discourse”. Advanced Education 4

(8): 128-137.

ARCAnO, Clyde. 2020. “Corpus-Assisted Discourse Studies”. In Georgakopoulou, Alexandra and

Anna De Fina (eds.) The Cambridge Handbook of Discourse Studies. Cambridge: Cambridge

U.P.: 165-185. <https://doi.org/10.1017/9781108348195.009>.

bAkER, Paul. 2006. Using Corpora in Discourse Analysis. London and New York: Continuum.

bAkER, Paul. 2020. “Corpus-Assisted Discourse Analysis”. In Hart, Christopher (ed.) Researching Dis-

course: A Student Guide. London: Routledge: 124-142. <https://doi.org/10.4324/9780367815042-8>.

BNC Consortium. 2007. The British National Corpus, XML Edition. Oxford Text Archive. <http://

handle.net/20.500.12024/2554>. Accessed October 30, 2022.

bREzInA, Vaclav. 2018. Statistics in Corpus Linguistics: A Practical Guide. Cambridge: Cambridge

U.P. <https://doi.org/10.1017/9781316410899>.

bybEE, Joan and Suzanne fLEIsCHMAn. 1995. “Modality in Grammar and Discourse: An Introductory

Essay”. Modality in Grammar and Discourse 14: 503-517.

CHEng, Maria. 2016. “The Power of Persuasion: Modality and Issue Framing in the 2012 Taiwan

Presidential Debates”. Discourse and Society 27 (2):172-194.

DEPRAEtERE, Ilse and Susan REED. 2020. “Mood and Modality in English”. In Aarts, Bas and April

McMahon (eds.) The Handbook of English Linguistics. Oxford: Blackwell Publishing Ltd: 207-227.

Notes

1. https://www.bbc.co.uk/news/

politics/eu_referendum/results

2. In the literature (Palmer 2001;

Nuyts 2016), a third type associated with

ability is distinguished: dynamic modality;

however, since it is not related to the speaker’s

attitude to a given situation or event

(Huddleston and Pullum 2002; Gisborne

2007), this type of modality lies beyond the

scope of this study.

3. We used the script by Fradejas

Rueda (2019) that can be accessed at: https://

www.aic.uva.es/cuentapalabras/cosechar-

textos.html

4. Accessible at: https://www.gov.uk/

5. Accessible at: https://labour.org.uk/

6. Accessible at: http://ucrel.lancs.

ac.uk/llwizard.html

7. For more information see: https://

ucrel.lancs.ac.uk/claws6tags.html

8. Corbyn announced his plans to

resign as leader of the Labour Party on

December 12, after the 2019 election results.

https://www.businessinsider.com/jeremy-

corbyn-resigns-as-labour-party-leader-after-

election-defeat-2019-12

9. Results in the tables are ordered

according to Johnson’s result in the likelihood

test.

10. Items which did not retrieve any

hits in the corpora have not been included in

the tables due to space limitations.

A Corpus Study of Brexit Political Discourse: Exploring Modality

miscelánea 70 (2024): pp. 67-92 ISSN: 1137-6368 e-ISSN: 2386-4834

DunMIRE, Patricia L. 2012. “Political Discourse Analysis: Exploring the Language of Politics and

the Politics of Language”. Language and Linguistics Compass 6 (11): 735-751.

EDELMAn, Murray. 1985. “Political Language and Political Reality”. PS: Political Science and

Politics 18 (1): 10-19.

EkAwAtI, Rosyida. 2019. “Power through Linguistic Modalities in Indonesian Presidential

Speeches”. Discourse and Interaction 12 (1): 5-28. <https://doi.org/10.5817/DI2019-1-5>.

fAIRCLOugH, Norman. 1989. Language and Power. London: Routledge.

fRADEJAs RuEDA, José Manuel. 2019. Cuentapalabras. Estilometría y análisis de textos con R para

filólogos. <http://www.aic.uva.es/cuentapalabras>. Accessed May 25, 2022.

gIsbORnE, Nikolas. 2007. “Dynamic Modality”. SKASE Journal of Theoretical Linguistics 4 (2): 44-61.

HARDIE, Andrew. 2012. “CQPweb – Combining Power, Flexibility and Usability in a Corpus

Analysis Tool”. International Journal of Corpus Linguistics 17 (3): 380-409.

HuDDLEstOn, Rodney. 1988a. English Grammar: An Outline. Cambridge: Cambridge U.P.

HuDDLEstOn, Rodney. 1988b. Introduction to the Grammar of English. Cambridge: Cambridge U.P.

HuDDLEstOn, Rodney and Geoffrey K. PuLLuM. 2002. The Cambridge Grammar of the English

Language. Cambridge: Cambridge U.P.

HuDDLEstOn, Rodney, Geoffrey K. PuLLuM and Brett REynOLDs. 2021. A Student’s Introduction to

English Grammar. Cambridge: Cambridge U.P.

kILgARRIff, Adam, Vít bAIsA, Jan buŠtA, Miloš JAkubíčEk, Vojtěch kOVÁř, Jan MICHELfEIt, Pavel RyCHLý

and Vít suCHOMEL. 2014. “The Sketch Engine: Ten Years on”. Lexicography 1 (1): 7-36. <https://doi.

org/10.1007/s40607-014-0009-9>.

kHOMutOVA, Tamara N. 2014. “Mood and Modality in Modern English”. Procedia: Social and

Behavioral Sciences 154: 395-401.

MARtIn, James andPeter wHItE. 2007. The Language of Evaluation: Appraisal in English. London:

Palgrave Macmillan.

MORLEy, John. 2011. “Introduction. A Description of CorDis”. In Morley, John (ed.) Corpus-

Assisted Discourse Studies on the Iraq Conflict: Wording the War. London: Routledge: 1-11.

nuyts, Jan. 2016. “Surveying Modality and Mood”. In Nuyts, Jan and Johan Van Der Auwera

(eds.) The Oxford Handbook of Modality and Mood. Oxford: Oxford U.P.: 31-49.

ORREQuIA-bAREA, Aroa and Encarnación ALMAzÁn-RuIz. 2021. “Modelling through Modality: (Re)

shaping Brexit”. ES Review. Spanish Journal of English Studies 42: 127-153. <https://doi.

org/10.24197/ersjes.42.2021.127-153>.

PALMER, Frank Robert. (1988) 2001. Mood and Modality. Cambridge: Cambridge U.P.

PALtRIDgE, Brian. 2012. Discourse Analysis: An Introduction. London: Bloomsbury Publishing.

PARtIngtOn, Alan. 2004. “Corpora and Discourse, a Most Congruous Beast”. Corpora and

Discourse 1: 11-20.

PARtIngtOn, Alan, Alison DuguID and Charlotte tAyLOR. 2013. Patterns and Meanings in Discourse:

Theory and Practice in Corpus-Assisted Discourse Studies (CADS). Amsterdam / Philadelphia:

John Benjamins Publishing.

QuIRk, Randolph, Sidney gREEnbAuM, Geoffrey LEECH and Jan sVARtVIk. 1985. A Comprehensive

Grammar of the English Language. London: Longman.

Encarnación Almazán-Ruiz and Aroa Orrequia-Barea

miscelánea 70 (2024): pp. 67-92 ISSN: 1137-6368 e-ISSN: 2386-4834

RAysOn, Paul. “Log-likelihood and Effect Size Calculator”. <https://ucrel.lancs.ac.uk/llwizard.

html>. Accessed February 27, 2024.

ROzuMkO, Agata. 2019. Modal Adverbs in English and Polish. A Functional Perspective. Białystok:

Wydawnictwo Uniwersytetu w Białymstoku.

sIMOn-VAnDEnbERgEn, Anne-Marie. 1996. “Image-Building through Modality: The Case of Political

Interviews”. Discourse and Society 7 (3): 389-415.

sIMOn-VAnDEnbERgEn, Anne-Marie. 1997. “Modal (Un)certainty in Political Discourse: A Functional

Account”. Language Sciences 19 (4): 341-356.sOMAI, Miklós and Zsuzsanna bIEDERMAnn. 2016.

“Brexit: Reasons and Challenges”. Acta Oeconomica 66: 137-156.

tAyLOR, Charlotte and Anna MARCHI. 2018. Corpus Approaches to Discourse. A Critical Review.

London: Routledge.

tORRECuADRADA gARCíA-LOzAnO, Soledad and Pedro gARCíA fuEntE. 2017. “¿Qué es el Brexit? Origen

y posibles consecuencias”. Anuario Mexicano de Derecho Internacional 17: 3-40.

tRAugOtt, Elizabeth Closs. 2011. “Modality from a Historical Perspective”. Language and

Linguistics Compass 5 (6): 381-396.

Received: 02/11/2023

Accepted: 27/05/2024

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.