Modality and personal pronouns in political texts: Corpus linguistic analysis of two Canadian political manifestos

(1)

Modality and personal pronouns in political texts Corpus linguistic analysis of two Canadian political manifestos

Riku Kareinen 241720 Pro Gradu Thesis English Language and Culture School of Humanities Philosophical Faculty University of Eastern Finland September 201

(2)

Tiedekunta – Faculty Philosophical Faculty

Osasto – School School of Humanities Tekijät – Author

Riku Kareinen 241720 Työn nimi – Title

Modality and personal pronouns in political texts Corpus linguistic analysis of two Canadian political manifestos

Pääaine – Main subject Työn laji – Level Päivämäärä – Date Sivumäärä – Number of pages English language and culture Pro gradu -tutkielma x 5 September 2019 76 + References

Sivuainetutkielma Kandidaatin tutkielma Aineopintojen tutkielma Tiivistelmä – Abstract

This thesis analyses the use of personal pronouns and modality in Canadian election manifestos. Both parts of the English language can be used to subtly and strategically convey a chosen message. In certain cases, their use can be interpreted as manipulative. The parties whose manifestos have been chosen for the study are the Conservative Party of Canada (CPC, 2015) and the Liberal Party of Canada (LPC, 2015).

The aim of this study is to find out whether there are quantitative and qualitative differences in the use of personal pronouns and modality between the two manifestos. The underlying assumption is that there will be a difference because of the contrasting positions of the parties before the election. CPC was the governing party, while LPC was in the opposition.

The theoretical section is divided into two parts the first of which covers personal pronouns. Personal pronouns are looked at regarding their multifunctionality and their use in politics. The idea of pragmatic choosing is especially relevant (Wilson, 1990). The second part deals with modality. For the purposes of this study modality has been limited to central modal auxiliaries and a number of marginal and semi-modals. The different categories of modality are also explored in detail (Huddleston and Pullum, 2002).

This study uses the methods of corpus linguistics. Keyness analysis, concordance and collocation are all used to help analyse personal pronouns and modal expressions as they occur in the texts. The data consisted of the two election manifestos compassing 58 984 tokens. The results were compared between the two manifestos.

The analysis showed statistically significant differences in the quantitative use of both personal pronouns and modal expressions. The results were affected by the drastic difference in the use of “we” and “will”. They were used far more frequently in the LPC manifesto. The results of the qualitative section were inconclusive on the part of personal pronouns. The use of modality was similar. This would indicate that political manifestos might have a textual style of their own based around identity creation both for self and other (opposing political parties). Further studies in this field are required for a more comprehensive view of political manifestos as well as personal pronouns and modality in them.

Avainsanat – Keywords

Corpus linguistics, political manifestos, modality, personal pronouns, Canada

(3)

Tiedekunta – Faculty Filosofinen tiedekunta

Osasto – School Humanistinen Osasto Tekijät – Author

Riku Kareinen 241720

Modality and personal pronouns in political texts Corpus linguistic analysis of two Canadian political manifestos

Pääaine – Main subject Työn laji – Level Päivämäärä – Date Sivumäärä – Number of pages Englannin kieli ja kulttuuri Pro gradu -tutkielma x 5.9.2019 76 + References

Sivuainetutkielma Kandidaatin tutkielma Aineopintojen tutkielma Tiivistelmä – Abstract

Tässä pro-gradu -tutkielmassa analysoidaan persoona pronominien ja modaalisuuden käyttöä Kanadalaisissa vaaliohjelmissa. Sekä pronomineja, että modaalisuutta voidaan hyödyntää huomaamattomasti, oman viestin tehostamiseen strategisessa mielessä. Käyttötapoja voidaan pitää jopa manipuloivina. Tähän tutkimukseen valikoituneet puolueet ovat The Conservative Party of Canada (CPC, 2015) ja The Liberal Party of Canada (LPC, 2015).

Tämän tutkimuksen tavoitteena on selvittää löytyykö kahden puolueen vaaliohjelmien väliltä quantitatiivisia tai qualitatiivisia eroja pronominien ja modaalisuuden käytössä. Oletuksena on, että erot puolueiden asemissa aiheuttavat eroja näiden vaaliohjelmissa. Vaalien aikana CPC toimi hallituspuolueena ja LPC oppositiossa.

Teoriaosuus on jaettu kahteen osaan, persoonapronomineihin ja modaalisuuteen. Ensimmäisessä osassa käsitellään persoonapronominien monikäyttöisyyttä ja niiden käyttöä politiikassa. ”Pragmaattisen valinnan” käsite on erityisen oleellinen (Wilson, 1990). Toinen osa koskee modaalisuutta. Tässä tutkielmassa modaalisuuden käsite on rajoitettu koskemaan keskeisiä apuverbejä sekä muutamia yleisiä erityistapauksia (central auxiliary modals, semi- ja marginal modals). Tutkimuksessa käytetyt modaalisuuden kategoriat esitellään myös yksityiskohtaisesti (Huddleston and Pullum, 2002).

Tutkimuksen metodologia on korpuslingvistinen. Keyness analysis, concordance ja collocation ovat kaikki olennaisia tutkimukselle. Niiden avulla persoonapronomineja ja modaalisuutta päästään tarkastelemaan osana tekstiä ja siinä muodossa sekä funktiossa, kuin ne tekstissä esiintyvät. Tutkielman datana toimiva korpus koostuu kahdesta vaaliohjelmasta ja sisältää 58 984 sanaa. Tutkimus on vertaileva joten puolueiden vaaliohjelmia vertailtiin keskenään.

Tulosten analysointi paljasti tilastollisesti merkittäviä eroja persoonapronominien ja modaalisuuden käytössä.

Tuloksiin vaikutti erityisesti ”we” ja ”will” sanojen käyttö. Niitä käytettiin huomattavasti suuremmalla frekvenssillä LPC:n vaaliohjelmassa. Qualitatiivisessa osassa persoonapronominien käytön tuloksissa oli epäselvyyttä.

Modaalisuuden puolella laadulliset tulokset oli samankaltaiset puolueiden välillä. Nämä tulokset viittaavat vaaliohjelmien välillä olevan mahdollisesti yhtenäinen teksti tyyli. Tämä perustuu identiteetin rakennukselle.

Identiteettejä rakennetaan niin omalle puolueelle kuin vaalivastustajille. Lisätutkimus aiheesta on tarpeen, ennekuin on mahdollista muodostaa parempi kokonaiskuva vaaliohjelmista tekstityyppinä sekä persoonapronominien ja modaalisuuden käytöstä niissä.

Avainsanat – Keywords

Korpuslingvistiikka, poliittiset vaaliohjelmat, modaalisuus, persoona pronominit, Kanada

(4)

1. Introduction

“Canada was built around a very simple premise. A promise that you can work hard and succeed and build a future for yourselves and your kids, and that future for your kids would be better than the one you had.”

(Trudeau, 2014)

The essence of political texts entails argumentation and persuasion. The text supports a stance and tries to influence the recipient’s thoughts on the matter. Therefore, analyzing the text could uncover these underlying ideals. Some linguists have gone as far as to say that “politicians’

language does not merely convey the message but creates for the listener a controlled cognitive environment from which any interpretation is manipulated” (Wilson, 1990: 11). While this rather emotional and skeptical view on the matter would recommend the reader to look for manipulative constructions built into the text, this study will focus on analyzing the text from an objective point of view. This is achieved through a linguistic analysis focusing on the form and structure of the language instead of commenting on the message conveyed. To illustrate, the quote from Trudeau at the start of the chapter would be analyzed for its use of the modal auxiliaries can and would and their collocates. Pronominal choosing would also be looked at, in this case the use of you.

In this study I will be conducting a linguistic analysis of two political manifestos published for the Canadian federal election of 2015. With the help of corpus linguistics, I will study modality and the use of personal pronouns in these texts. These two approaches have been chosen because of their subtlety and because previous studies have shown that they are both used pragmatically and strategically in political texts. For example, Bramley’s study (2001) on political interviews and Nartey and Yankson’s study (2014) on political texts.

The analysis of pronouns is based on the idea of pragmatic choosing. The idea behind this is that pronouns are selected in order to advance the parties message in the most effective way possible. This analysis considers how they make statements about themselves, for example, the way they position themselves in different topics. Another important factor is how they refer to

(6)

their opponents. The final category in pronouns focuses on individuals and groups outside the political parties.

The parties whose manifestos have been chosen are: the Conservative Party of Canada (from now on referred to as CPC) and the Liberal Party of Canada (LPC). The election ended in LPC’s clear victory, and the CPC lost its position as the governing party. As parts of the election campaigns the manifestos had an effect in the eventual results. Though this thesis does not comment on the scale of these effects, it does state that they had one. Other factors including interviews, debates, advertisement may have weight a great deal more. However, the manifestos are still the first official release of a party’s policies for the future. Though the content of the manifestos might be the priority, the linguistic structure of it is also important.

Getting a point effectively across is not just about what is said but how it is said. Given the media lead political campaigns of today, how one articulates one’s message is more important than ever. Based on this and the results of the election, the hypothesis of this thesis is that there will be a difference in the use of both modality and personal pronouns between the texts.

Modality and personal pronouns were chosen for the analysis for two reasons. Firstly, studies in modality and pronouns show that they can be used to subtly shape a text. Pronominal choice and pragmatic use of modal expressions are tools to effectively convey one’s message.

Secondly, despite a certain number of studies in the area (cf. Bramley, 2001; Lillian, 2008 and Nartey and Yankson, 2014) there are not enough studies conducted on modality and pronouns in political texts. Therefore, there is a need to learn more about their use.

This study aims to answer two research questions:

(1) are there quantitative differences in the use of pronouns and modal auxiliaries between the governing party (CPC) and the opposition party (LPC).

(2) what kind of qualitative differences in the use of personal pronouns and modality are there between the two parties.

The null hypothesis (H0) is that there is no difference in the use of modality and personal pronouns between the CPC and the LPC. The alternative hypothesis (H1) is that there is a difference in the use of modality and personal pronouns between the CPC and the LPC.

I will start by presenting theoretical background essential to this study. Personal pronouns, modality and political manifestos each have their own chapter. The theoretical parts will be clarified with the use of examples, which whenever possible have been taken from the two

(7)

manifestos: CPC’s “Our Conservative Plan to Protect the Economy” (2015) (from now on referred to as the CPC Manifesto) and LPC’s “A New Plan for a Strong Middle Class” (2015) (from now on referred to as the LPC Manifesto). At the end of each these chapters a theoretical framework relevant to this study will be explained in detail.

I will then present earlier studies which are concerned with linguistic analysis of manifestos, personal pronouns in political texts and modality in political texts. Following the earlier studies, the methodology of this seminar paper is explained. In linguistics studies political manifestos are still a rarely studied form of data. Therefore, this study tries to form a more exhaustive view of the way pronouns and modality is used within them. Finally, I will present the results of this study and then analyze and discuss them.

(8)

2. Personal pronouns

2.1 Definition(s)

“Personal pronouns are function words, which make it possible to refer succinctly to the speaker/writer, the addressee, and identifiable things or persons other than the speaker/writer and the addressee.”

(Biber et al. 1999: 328)

Original definitions for pronouns come from Latin (pro-nomen) and Greek (anto-numia) and mean ‘standing for a noun’ (Cassel’s Latin dictionary, 1977: 480). Most grammar books use a narrow definition for personal pronouns as ‘substitute for a noun or noun phrase’ (from now on referred to as ‘NP’). Often this definition is then supplemented with a table (e.g. Table 1, p.5). Wales (1996: 1) argues that because of the effect of defining personal pronouns as

‘substitutes’, the third person pronoun (from now on referred to as ‘3PP’) can be seen as a

‘prototypical’ personal pronoun. The same can be said about personal pronouns in relation to other types of pronouns: possessive, reflexive, reciprocal, relative, interrogative, demonstrative, indefinite.

Defining personal pronouns as substitutes also implies that the substituted term, i.e. the antecedent, needs to be known from the co-text (so called endophoric reference). If the endophoric reference points ‘backward’ to the history of the unfolding text, it is called anaphora or anaphoric reference. In example (1) we refers back to “Conservatives”.

(1) “For Conservatives, protecting Canada’s economy is our number one priority. That’s why we have a positive plan to protect the economy and create jobs over the next four years, and beyond.

(CPC Manifesto: 3)

There are two ways to use anaphora, either by a personal reference item (personal pronoun or possessive determiner) or a specified noun. Using a personal pronoun is the unmarked strategy for anaphoric referencing, i.e. to use something else is to vary from the norm and needs a specific reason (Halliday, 2014: 625). Personal pronouns are often seen to have characteristic

(9)

anaphoric reference, i.e. referring back. This fulfills the stylistic function of avoiding the repetition of a substituted noun or NP (2).

(2) “Canada’s National Parks belong to all Canadians. They are also an important part of local economies, generating nearly $5 billion each year and helping to employ 65,000 people.”

(LPC Manifesto: 44)

Table 1. Prototypical pronoun paradigm (Standard English) (Wales, 1996: 13)

Personal pronouns Possessive pronouns Reflexive pronouns subjective

case

objective case

determiner function

nominal function 1p sg.

pl.

I we

me us

my our

mine ours

myself ourselves 2p sg.

pl.

you you

your yours yourself yourselves 3p

sg.

masc.

fem.

non- personal

he him she her it

his

her hers its

himself herself itself

pl. they them their theirs themselves

From this point of view the pronoun itself is low on informational value and its meaning needs to be clear from context. This is also called semantic content hierarchy, i.e. NPs are referred as full NPs and contain more information while pronouns are reduced (Wales, 1996: 37). Ehlich (1982, quoted here from Wales 1996: 24) writes that “an anaphora is related to a word in what was previously said which, in turn, refers to an entry of the real world and/or its mental

(10)

representation” (Figure 1). For example, in (2) Canada’s National Parks is the NP that represents the real-world entity, while the reduced personal pronoun they relates to the NP Canada’s National Parks. From the structural point of view, they might also refer to all Canadians, however the context makes this a less reasonable assumption.

Figure 1. (Wales, 1996: 24)

The semantic content hierarchy is demonstrated in Figure 1 above. Using Example 2 as the basis, the [X] represents the real-world entity (the national parks in Canada), which is referred to via the NP (Canada’s National Parks), which in turn is referred to via the pronoun they (3PP).

It should be mentioned that in this case 3PP is not actually correct as they is the plural form.

The figure supports the idea of pronouns as substitute taking them one step further from the real-world entry than the NP antecedent. In this way pronouns are used to provide ties or overt linkages between sentences or utterances and maintain coherence without the need to repeat the original NP.

Endophoric reference can also point forwards. This is called cataphoric reference (Wales, 1996). In these cases, the antecedent only appears after the pronoun that has been used to replace it. In example (3) the personal pronoun his refers forwards to Bernard Levin.

(3) “In “the Pendulum Years”, his history of the 1960s, Bernard Levin writes of the ‘collective insanity which seized Britain’.”

(cited in Wales, 1996: 37) Cataphoric reference is used less frequently than anaphoric reference and according to Wales (1996: 37) some prescriptive grammarians have gone as far as to condemn its use for reasons of clarity. Using cataphoric reference is also a problem for grammars defining personal pronouns as mere substitutes. As Wales (1996: 1-2) writes, the personal pronouns role as a

‘substitute’ usually presupposes that the NP it refers to has been mentioned before. However, cataphoric reference is a part of the English language and is recognized by grammarians.

3PP NP [X]

(11)

Another aspect challenging the monofunctional view of personal pronouns as mere substitutes is the pronoun it. The Longman Grammar of Spoken and Written English (Biber et al. 1999) presents three major non-referential uses of the pronoun it. Firstly, as an empty subject/object (4). The empty subject/object is used when there are no participants to fill the subject slot. It is mostly used in referring to weather conditions, time and distance.

(4) “It’s cold.” (empty subject) Secondly as an anticipatory subject/object (5).

(5) “It is not surprising that 90 per cent of the accidents are caused by excess speed” (anticipatory subject)

Thirdly as a subject in cleft constructions (6).

(6) “It was at this stage that the role of the DCSL became particularly important.”

(subject in cleft construction)

(Longman Grammar of Spoken and Written English, 1999: 332) Defining personal pronouns as mere placeholders and stylistic tools only represents a part of their actual use. An overview of the different functions of personal pronouns will be given in chapters 2.1.1 and 2.1.2 while a more in depth look at personal pronouns in political texts is discussed in chapter 2.2.

2.1.1 Personal pronouns as nouns

Personal pronouns are also defined as a subclass of nouns (Hudson, 1992; Greenbaum, 1991:

84; Fabb, 1994: 47) and can have noun like features such as: premodification by article (7) or adjective (8), compounding and plurality. They also share gender (3PP; he and she) and case (cf. subjective I vs. objective me) with nouns. Number is rather more complicated as seen from the difference between examples (9) and (10).

(7) “Is your dog a he or a she?”

(8) “Poor old you.”

(Wales, 1996: 10)

(12)

(9) “We’re the only party – the only ones that can be trusted – to keep taxes low on small businesses and their workers for the next four years.” (We = speaker + other).

(CPC Manifesto: 20)

(10) “Mills have closed, and Canadians have lost their jobs because of these actions.” (Canadians = Canadian + Canadian + Canadian…)

(CPC Manifesto: 44) From a semantic viewpoint this has meant that anaphora implies co-reference i.e. the pronoun and its antecedent agree in number and gender but also have identical reference (11).

(11) “Stephen Harper’s approach to trade, however, has failed. His Conservative government has recorded the largest trade deficit in Canadian history, and Canadians are paying the price in lost job opportunities.”

(LPC Manifesto: 67) In this sentence both ‘Stephen Harper’ and ‘his’ point to the same individual and can be said to be co-referential.

Pronouns also function grammatically as subjects of clauses, objects or complements, and complements in prepositional phrases. Unlike nouns these differences can be seen in morphological changes: from she to her and he to him (cf. Table 1 p.5 above). Because of these properties, while pronouns are not always defined as nouns, they must be seen to have at least nominal functions. (Wales, 1996: 11)

3PP is still mostly seen as a ‘function word’ used for referring back an antecedent or an entity.

Meanwhile, its deictic properties are less frequently mentioned in linguistic literature. Personal pronouns are deictic as they require contextual information in order to be properly understood.

This means that their collocates make a difference and they can be used to position different groups within texts. This is an important aspect of their use in political texts. According to Wales (1996: 22) personal pronouns have two different categories of deixis: textual deixis and situational deixis. In textual deixis the pronoun simply points syntactically towards its antecedent (12), while in situational (contextual) deixis the meaning of the pronoun is clear from a situation outside of the text (13).

(13)

(12) “Canadians want a government they can trust to protect the environment and grow the economy.”

(LPC Manifesto: 39) (13) “If the baby will not drink cold milk, boil it.”

(cited in Wales, 1996: 22)

2.1.2 Polyfunctional pronouns

Wilson (1990: 46) writes that “pronouns may be selected within interaction for reasons beyond those reflected at a purely formal or categorical level; they may function communicatively to reveal various aspects of the speaker’s attitudes, social standing, sex, motivation and so on.”

Therefore, it is reasonable to presume that in the context of promoting a political agenda the use of pronouns will be selective. As previously stated in chapter 2, traditional grammars mostly define pronouns as mere endophoric devices. This rather simplistic view works in some cases, but it does not take in to account the social and personal aspect of using and interpreting pronouns. Though pronouns are a closed word class i.e. new pronouns are rare to enter a language, their functions and references develop as languages change. Wales (1996: 8) states that pronouns do have polyfunctionality and some have even developed special connotations e.g. he and she which can be used generically and symbolically to refer to men and women, as in (14).

(14) “’She’ is indefinitely other in herself… not to mention her language in which ‘she’ goes off in all directions and in which ‘he’ is unable to discern the coherence of any meaning”

(cited in Wales, 1996: 8) In the example above “she” and “her” are used to refer to all women and the language they use.

He is used to refer to men in general, unable to understand the meaning of the language used by women. Furthermore, while there are three forms for 3PP (he, she and it) which mostly do correspond to their male, female and inanimate reference respectively, there is an issue with co-referencing the type of NPs where the sex of the referent is unknown. In most cases like this he is used as the generic pronoun (15).

(14)

(15) “Nobody likes his tutor”

(Wales, 1996: 5) Biber et. al. (1999: 316) note that even though masculine pronouns (he, his) may be intended to have dual reference, i.e. ‘use of the generic he’, readers may perceive the referent as male.

The use of ‘he’ as a generic pronoun has been both questioned and supported by different scholars. Quirk et. al. (1985: 45) supports the use of the generic pronoun by arguing that

“gender in English is ‘natural’ (semantic) as opposed to grammatical (formal). This point of view has been questioned by Cameron (1985: 26) who asks “why, if gender is ‘natural’ within English, the masculine form is chosen as the unmarked alternative and not the feminine”. She continues that gender in English is not merely fixed by sex reference but that it also reflects ideologically motivated prescriptive practices and folk-linguistic beliefs. Later, she argues that unmarked he is a feature of grammar, but of prescriptive grammar, reinforced by male grammarians for ideological reasons (Cameron 1985:68). Given that English for long periods has been taught through prescriptive practices, meaning that language is taught not as it is used but as it “should” be used, and that through these practices certain aspects of the language must have become folk-linguistic beliefs i.e. common sense within a society, Cameron makes a strong argument.

Another good example of the polyfunctionality of pronouns is the personal pronoun they, which can be used with specific (16) or general reference (17) (Biber et. al., 1999: 329-331). In specific reference they deals with a specific group while in general reference they deals with people in general. Especially in discourse situations a more selective function is much more common i.e. they meaning a specified group of people (Biber et. al. 1999: 329-331).

(16) “The Liberals and NDP believe that they can raise taxes, run deficits, and accumulate more government debt without killing jobs.”

(CPC Manifesto: 18) (17) “As they say, it gets tough at the top”

(British National Corpus, A89) In (16) they specifically refers to the Liberal Party and the NDP, openly suggesting that they are the ones raising taxes, running deficits and accumulating government debt. What this proposition also indicates, but does not directly say, is that they (the CPC) do things differently

(15)

i.e. they oppose the others. Used this way the pronoun system can be used to distinguish the in-group from the outsiders, or in other words us from them. In addition to they, we can also be used in a similar way. Wales writes about this phenomenon in the field of politics where it is used to create identities. Her reasoning is twofold. Firstly, she notes how socially and ideologically speaking ‘we’ can mean “more than one” of the same. By this she means that a group with shared identity can speak as if it is one speaker, but multiplied (Wales, 1996: 59).

This is a common aspect of political speak, where a representative of a party uses we to describe the party’s actions, values, beliefs etc. (18).

(18) “We will do our part to help Canadian children live healthier lives, with less exposure to known health risks.”

(LPC Manifesto: 20) Secondly, Wales remarks how the creation of an in-group simultaneously means that there is an out-group, i.e. the ‘them’ to our ‘us’. The in-group is distinguished from the outsiders, “the alienated” the “not like us”. Wales notes, that in some cases this can literally mean that they are alien (Wales, 1996: 60). However, even a completely human they can be made to look menacing (19). She gives an example of how they can be used as an agent, leaving the speaker as the passive ‘victim’ (20).

(19) “The Liberals and NDP believe that they can raise taxes, run deficits, and accumulate more government debt without killing jobs.”

(CPC Manifesto: 18) (20) “They pinched me for parking on a double yellow line.”

(cited in Wales, 1996: 60) Wales (1996: 60) continues, that “They, of course, are often useful to pigeon-hole as ‘other’

because they can usefully be held responsible for things ‘we’ do not want to be blamed for.”

This “blame game” is often made easy by the fact that they are not actually physically present.

The polarizing effect of using ‘we’ and ‘they’ has long been used politically, and Wales uses the Cold War rhetoric as an example e.g. The West vs. The East (1996: 61). In this case outlets in the West would write about “them” or “they” when referring to the actions of the eastern bloc. This type of adversarial juxtaposition was highly politically motivated and stoked by they vs. us type rhetoric.

(16)

This section has shown that inter-personal use of pronouns is not as Wales (1996:50) remarks, always ‘neutral’ in its reference. She continues that given that there are a wide variety of speech and written discourse situations beyond the canonical, they also display a richness of rhetorical and social connotations beyond the strictly denotational (they do not simply just point to the speaker or the attendee). This side of personal pronouns is often discarded in grammars that focus on form instead of function (Wales, 1996: 50).

2.2 Personal pronouns and politics

The previous chapter offered an overview of the polyfunctionality of personal pronouns. The fact is that different uses of pronouns reflect different social dimensions. This sociolinguistic aspect of the pronominal system affects matters such as: formality/informality, status, solidarity, power, class and sex (Wilson, 1990). In the field of political texts pronouns have their own uses. The use of pronouns in politics has been written about by scholars such as Goffman (1981), Wilson (1990) and Wales (1996). Wales’ work (1996: 58-63) on the use of

‘we’ and ‘they’ in politics was depicted in the previous chapter (2.1).

Wilson (1990) writes about the pragmatics of pronominal choosing. He suggests that in situations where pronominal choice affects or reflects certain social facts about the speaker, the speaker makes a conscience decision as to which pronoun to use. Therefore, using a certain pronoun might not be a consequence of the meaning of the particular form in context, but instead the result of paradigmatic choice. Paradigmatic choice entails choosing a unit of language from a group of suitable words. In this case it means choosing one pronoun over another because of the connotations involved. This choice can be exemplified with the following example (Wilson, 1990: 48):

(21) Due to the rising balance of payment deficit…

(a) it has been found necessary to increase interest rates.

(b) I have found it necessary to increase interest rates.

(c) we found it necessary to increase interest rates.

In this paradigmatic choice the three options presented above differ in the amount of personal involvement the speaker has with the decision. In (a) the actor is unknown evident from the use of a passive structure, while in (b) there is a clear case of personal involvement. In option

(17)

(c) the role of the speaker is not as clear cut. ‘We’ can be used both inclusively and exclusively.

When using inclusive we, the speaker is a part of the group i.e. we meaning the speaker and at least one other person. In exclusive ‘we’ the speaker is not a part of the group. So, in the case of (c) the speaker could be talking about himself/herself and a group of designated others (e.g.

the government of a country) or a larger more generic group e.g. the nation. In the latter option the involvement of the speaker is much more ambiguous. Goffman (1981, quoted here from Wilson, 1990: 48) writes about the way forms like exclusive ‘we’ can be used to distance the speaker from what it is that is being said.

In this study personal pronouns are looked at according to three categories defined by Wilson (1990: 61-68) but with a slight modification. In the self-referencing category (explained below) the use of inclusive and exclusive ‘we’ is also examined. The three categories used are:

1. Self-referencing: the way in which the speaker chooses to portray himself/herself in relation to the topic and addressee (22). In this category the use of inclusive (speaker and at least one other person) (23) and exclusive (the speaker is not part of the group)

‘we’ (24) is also examined. These constructs can be used to either involve or distance the speaker from the discussed subject (Wilson, 1990: 61-68).

(22) “As part of the TPP negotiations, we’ve defended and protected Canada’s supply management system, as we committed to.”

(23) “We applaud Minister Kenney and the government for making changes to our immigration policy and programs that support economic growth by focusing on our skilled labour needs.”

(24) “Canadian businesses and workers know instinctively that we can compete with the world’s best as long as the playing field is level.” (Canadian businesses and workers will be competing, not the conservative party)

(CPC manifesto: 21, 22 & 33) 2. Relation of contrast: this refers to the way in which speakers make use of the pronominal system to compare and contrast others on a negative/positive scale. For example, in political debate instead of referring to your opponent by name, you may simply pinpoint them as ‘him’ or ‘her’ (Wilson, 1990: 61-68) (25). This discredits the opponent as someone not worth mentioning by name. It can be seen as working similarly to semantic content hierarchy (cf. chapter 2.1. above) as it distances the

(18)

opponent from the real-life person. Wilson mentions that relation of contrast could be used as part of chauvinistic politics “reducing” female politicians as ‘shes’ and ‘hers’.

Another use of relation of contrast is a form of personalizing a larger organizations action. This way a certain person seems to be responsible for all of the actions e.g. a political party makes. In (26) the former leader of the British Labour Party Neil Kinnock refers to the conservative treasury ministers as Margaret Thatcher’s own.

(25) “Stephen Harper has had nearly a decade to take action on climate change but has failed to do so. His lack of leadership has tarnished Canada’s reputation abroad, making it harder for Canadian businesses to compete.”

(LPC Manifesto: 39) (26) “And in similar slanderous style her Treasury Ministers blame their failure.”

(Wilson 1990: 67) 3. Other referencing: this indicates the use of pronouns to refer to individuals and groups outside the roles of speaker and addressee (Wilson, 1990: 61-68) (27).

(27) “When it comes to infrastructure that will help keep Canadians safe and better prepared for emergencies, local leaders know what needs to be done.

What they need is a federal partner willing to invest to help build stronger, more resilient communities.”

(LPC Manifesto: 14) In this study, the first category will consider statements the parties make about themselves. For example, what kind of actions they promise to take were they to be elected or how they position themselves in different topics (27). The second category shows how they refer to their political opponents, as in (28). It shows for example how they are using the pronominal system to position the other parties. The third category focuses on how individuals and groups outside the political parties are mentioned in the manifestos (29).

(27) “When we took office, Canada had free trade agreements with only five other countries. We’ve now reached agreements with 44, countries, encompassing more than half the global marketplace, including historic trade

(19)

deals with South Korea and the European Union, with its 500-million person,

$18-trillion economy.”

(28) “The Liberals and NDP would take the country down a much different and more dangerous path. They would undermine Canada’s security by: Taking away the tools that we have granted our law enforcement agencies to protect Canadians from terrorist threats…”

(29) “The best way to ensure more and better jobs is to create the right conditions for our enterprising businesses to do what they do best.”

(CPC Manifesto: 18, 21, 101)

(20)

3. Modality

“The speaker’s judgement, or request of the judgement of the listener, on the status of what is being said… What the modality system does is to construe the region of uncertainty that lies between ‘yes’ and ‘no’.”

(Halliday 2014: 176)

In the above quote, the term “region of uncertainty” is used. This aspect of the English language seems to be ever more relevant in the field of politics. The rise of populist rhetoric in the so- called post truth era means understanding the actual message conveyed via political texts is more important than ever. Constructing functional and self-serving texts is a vital skill in the field of politics. For this purpose, the English modal structures are especially useful. In the following chapters the term modality is discussed in depth.

Huddleston and Pullum (2002:173) state that “modality is centrally concerned with the speaker’s attitude towards the factuality or actualization of the situation expressed by the rest of the clause”. Depraetere and Reed (2006: 269) offer a practical definition for modality that states that it is “a cover term for a range of semantic notions such as ability, possibility, hypotheticality, obligation and imperative meaning.” However, finding a theoretically useful definition proves to be much more difficult. Palmer (1990: 1) notes that modality is not as easy to define as e.g. tense or number, though most grammarians seem to agree that it is concerned with the ‘opinion and attitude’ of the speaker (Lyons, 1977: 452). In order to do this one needs to find the common element in all modal utterances. Depraetere and Reed (2006: 269) give the following theoretical definition:

“In what follows, we shall work on the basis that all modal utterances are non- factual, in that they do not assert that the situations they describe are facts, and all involve the speaker’s comment on the necessity or possibility of the truth of a proposition or the actualization of a situation.”

As is evident from the quote above defining what modality is, is not an easy task. In addition to being difficult to define, there are a multitude of ways of expressing modality. Huddleston and Pullum (2002:173-175) find that in the English language modality can be communicated with modal auxiliaries, lexical modals, past tense, other verb inflection, clause type,

(21)

subordination and parentheticals. However, this study focuses on the use of modal auxiliaries, marginal modals and semi-modals, all of which will be introduced in the following chapters.

Next a number of different classifications and categorizations of modality are introduced.

3.1. Literary review

A number of scholars have written about modality. As Huddleston’s and Pullum’s (2002) work is especially important to this study it is covered in more detail. They begin by describing three significant dimensions of modality: strength, type and degree. Palmer’s work (1990) has also been key in defining the categories of type.

Huddleston and Pullum (2002) recognize three typess of modality: epistemic, deontic and dynamic. Epistemic modality (30) involves qualifications concerning the speaker’s knowledge, deontic modality (31) concerns obligation and permission while dynamic modality (32) deal with the abilities and willingness of the subject (Huddleston and Pullum. 2002: 178-179). These categories each have their own chapters wherein they will be discussed in more detail (cf. 3.3, 3.4 and 3.5).

(30) “He must have been delayed.”

(31) “You may stay if you wish.”

(32) “We will accelerate and expand open data initiatives and make government data available digitally, so that Canadians can easily access and use it.”

The next dimension strength is the “degree to which the speaker commits to the factuality or actualization of the situation”. In the examples below necessity (33) involves a strong commitment and possibility (34) a weak one.

(33) John must be there.

(34) John may be there.

Individual modal expressions of various word classes (the list included auxiliaries, adverbs, adjectives, verbs) can therefore be categorized by their strength, for example:

Strong (Necessity): must, need, necessary, necessarily etc.

Weak (Possibility): can, may, possible, possibly, perhaps.

(22)

Medium: should, ought (to), probable, likely, appear, seem.

However, it should be noted that classifications such as these are not clear cut. The difference in strength between modal expressions is not always notable e.g. should and ought (to). In addition, the strength of an individual modal expression can change depending on the context.

Huddleston and Pullum (2002: 176-177) state that there is a difference between semantic strength and pragmatic strength. They use the terms pragmatic weakening (36) and pragmatic strengthening (38) to describe situations where an otherwise stronger or weaker modal expression is changed by the context. In the examples below, these phenomena are exemplified with the use of must, normally a strong modal expression, and may, normally a weak modal expression.

(35) “You must come in immediately.” (Strong modality)

(36) “You must have one of these cakes.” (Pragmatic weakening) (37) “You may take your ties off.” (Weak modality)

(38) “You may leave now.” (Pragmatic strengthening)

Thirdly, Huddleston and Pullum (2002: 179) write about the dimension of degree: “The degree of modality shows to what extent a clearly identifiable and separable element of modality can be found.” In other words, it is the degree to which a modalized clause (39) differs from its unmodalized version (40):

(39) “They know her.” (unmodalized) (40) “They may know her.” (modalized)

In the examples above there is a clear difference in the way the sentences are interpreted. In (36) the speaker states as a fact the they know her. In (37) this is not the case as may involves a modal uncertainty over whether or not “they” know her. It is also simple to identify the element of modal meaning separating the two sentences: may. In this case may has a high degree of modality as it clearly gives the entire clause a modal interpretation. The two examples below showcase a low degree of modality:

(41) “It’s odd that he is so late.” (unmodalised)

(42) “It’s odd that he should be so late.” (low degree modality)

(23)

The difference in meaning between these two sentences is minimal. Huddleston and Pullum (2002: 179) note that there is a correlation between a low degree of modality and a strong strength of modality. This seems logical as the stronger the strength of modality in a clause is, the less it leaves room for uncertainty and is therefore closer to a factual statement. An example of this is given in (43) where will shows both a low degree of modality and strong strength of modality:

(43) “She will be one-year old tomorrow.”

The low degree of modality means that the sentence is near a factual statement, as the ”she” of the sentence is a mere day away of being one-year old. At the same time the use of “will” shows a strong strength of modality as the speaker seems certain that the one-year mark will indeed be reached tomorrow. Huddleston and Pullum also use modal harmony as an explanation for a low degree of modality. It means that the unit of modality is in harmony with the modality expressed in the larger construct (45).

(44) “Strange as it seems, I believe you.” (unmodalised)

(45) “Strange as it may seem, I believe you.” (low degree modality) In (44) and (45) the interpretation of the sentence is nearly identical despite (52) being modalised with the use of “may”. Caused by the idiom “Strange as it seems” and the modal harmony “may” has with it. Simply put modal harmony means that another expression within the sentence structure is already implying modality and the modal auxiliary mainly reinforces or changes the strength or type of the modality.

In addition to Huddleston and Pullum’s work there are several relevant works of grammar dealing with modality. Next the works of Quirk et al. (1990) and Biber et al. (1999) are covered to compliment the work of Huddleston and Pullum (2002).

Quirk & Greenbaum (1990: 60) divide modals into intrinsic and extrinsic. Intrinsic modality involves some intrinsic human control over events and includes the semantic meaning of

‘permission’ (46), ‘obligation’ (47) and ‘volition’ (48).

(46) “We will recommit to supporting international peace operations with the United Nations and will make our specialized capabilities – from mobile medical teams to engineering support to aircraft that can carry supplies and personnel – available on a case-by-case basis.”

(24)

(47) “For Parliament to work best, its members must be free to do what they have been elected to do: represent their communities and hold the government to account.”

(LPC Manifesto: 29, 69) (48) “Many of these ventures could be economic game-changers for the Canadians living in the surrounding regions.”

(CPC Manifesto: 46) Extrinsic modality involves human judgement of what is or is not likely to happen and includes

‘possibility’ (49), ‘necessity’ (50) and ‘prediction’ (51).

(49) “While Stephen Harper may sign trade deals, he walks away from partnering with businesses and entrepreneurs to ensure they can succeed in new markets and create wealth and jobs for Canadians.”

(50) “Canadian families looking for caregivers to help family members with physical or mental disabilities must pay a $1,000 Labour Market Impact Assessment fee.”

(LPC Manifesto: 42 & 67)

(51) “The Liberals and NDP would take the country down a much different and more dangerous path.”

(CPC Manifesto: 101) Though some of the categories seem similar there is a distinctive separation within them, that of control. For example, intrinsic obligation and extrinsic necessity may seem similar, but in intrinsic there is an amount of human control over the obligation, as in (52). Meanwhile in extrinsic necessity the speaker judges the circumstances and concludes the “obligation” based on those observations (53).

(52) “Productivity must be improved if the nation is to prosper.” (necessity) (53) “You must be feeling tired.” (obligation)

Quirk & Greenbaum (1990:60) state that in some cases there is an overlap of the two uses; e.g.

“I’ll see you tomorrow then” can be said to combine the meanings of volition and prediction.

Quirk & Greenbaum (1990: 60) also note, that most of the modals can be paired into present

(25)

and past forms: can/could, may/might, shall/should, will/would. They continue that from a semantic point of view, the past forms are often merely more tentative or more polite variants of the present forms.

Biber et. al. (1999: 485) also write about intrinsic and extrinsic modality, which they see as equal to deontic and epistemic modality respectively (for deontic modality and epistemic modality see 3.4 and 3.3). In addition, they divide modals into three major categories according to their meanings:

a) Permission/possibility/ability: can, could, may and might

b) Obligation/necessity: must, should, (had) better, have (got) to, need to, ought to, be supposed to

c) Volition/prediction: will, would, shall, be going to

(Biber et. al., 1999: 485) In conclusion, modality is a much-researched area of linguistics. Several linguistics have given their own classifications and categorizations to offer a theoretical framework from which to approach the subject. Depraete and Reed (2006) present a table (Table 2 below), where the works of a few scholars are comparable according to their classifications. Of these frameworks this study utilizes the work of Huddleston and Pullum (2002: 180-202). Their work was chosen as it is clear cut in its categories, broad enough to cover the chosen areas of modality and it is compatible for at least some comparison with previous studies. In the next section different modal expression are introduced, starting with the central modal auxiliaries (3.2). In the final three sections of this chapter epistemic (3.3), deontic (3.4) and dynamic modality (3.5) are explained in more detail.

(26)

Table 2. Classification of (analytical) modal meaning (Depraete and Reed 2006: 280)

3.2 Central modal auxiliaries

Regardless of the system of classification, most linguistics agree that there are a number of modal auxiliaries at the heart of modality. These are called the central or core modal auxiliaries (cf. Aarts, 2011: 280, Biber, 1999: 483). Central modal auxiliaries include the following auxiliary verbs:

- can, could, may, might, shall, should, will, would and must.

They share similar qualities, most importantly they adhere to the NICE properties: negative (53) form consists of the auxiliary verb followed by ‘not’, inversion (54), they can precede the subject in subject-verb inversion, code (55), they can be used instead of a full lexical verb which has occurred in the context, emphasis (56), they can be used in emphatic affirmations without requiring the verb ‘do’.

(53) “John Brown will not be there.” (Negation)

Epistemic root

necessity

root

possibility ability obligation permission willingness or volition modality

Epistemic Root modality Coates

(1983)

Extrinsic Intrinsic Quirk et al.

(1985)

Epistemic n/a agent-oriented

Bybee and Fleischman (1985) Propositional

modality n/a n/a event modality Palmer

(2001)

Evidential Epistemic dynamic deontic dynamic

Epistemic dynamic deontic dynamic

Huddleston and Pullum et al.

(2002)

Epistemic

non-epistemic

n/a

Van der Auwera and Plungian (1998) participant-

internal

participant-

external participant- internal

participant- external

non-deontic deontic

(27)

(54) “Will he be coming back here?” (Inversion)

(55) “He won’t be coming back, and neither is she.” (Code) (56) “He will come back.” (Emphasis)

In addition to the NICE properties, the central modals have no third person singular s-form shalls, wills etc. do not exist and they have no non-finite forms. The latter can be seen from the lack of tensed forms such as musting, mayed.

Table 3 shows the frequencies of the different central modals from a British English corpus.

These frequencies give an indication to the “normal” use of central modals. Will, would, can and could are clearly the most frequent ones while shall is used the least. This might be due to the formal tone that shall infers. The table below can be used as a preliminary guideline when checking the results of this study. Any statistically relevant discrepancies between these frequencies and the results of this study will be checked. However, no valid conclusions should be made based on them alone given that the data of this study is Canadian English and the table British English.

(28)

Table 3. Frequencies of the English core modal auxiliaries per million words in the ICE-GB corpus (adapted from Aarts, 2011: 280)

Present form spoken written

will / ‘ll won’t

1,883 / 1,449 232

3,284 / 361 80 can

can’t cannot

2,652 792

80

2,533 222 316 may

mayn’t

646 2

1,898 - shall

shan’t

196 5

217 0 must

mustn’t 472

24 857

0

past form spoken written

would / ‘d wouldn’t

2,581 / 795 394

2,533 / 182 87 could

couldn’t

1,339 231

1,353 90 might

mightn’t 670

3 600

0 should

shouldn’t 861

71 1,192

26

Table 4 below showcases the way central modals are categorized in this study (Huddleston and Pullum: 180-202). The table also offers a good view of the multitude of ways each central modal can be interpreted.

(29)

Table 4. Meanings expressed by central modals (Huddleston and Pullum: 180-202)

Can Epistemic possibility Deontic possibility Dynamic ability

“She cannot have done it deliberately.”

“She can stay as long as she likes.”

“She can easily beat everyone else in the club.”

Could Epistemic possibility Deontic possibility Dynamic possibility

“She could not have done it deliberately.”

“She could stay as long as she liked.”

“She could easily beat everyone else in the club.”

May Epistemic possibility Deontic possibility

“He may have done it deliberately.”

“You may attend the lectures.”

Might Epistemic possibility “You said to me once you might come to London.”

Must Epistemic necessity Deontic necessity Dynamic necessity

“He must have done it deliberately.”

“You must attend the lectures.”

“Now that she has lost her job she must live extremely frugally.”

Will Epistemic possibility Deontic necessity

Dynamic willingness Dynamic ability

“They will have made the decision last week.”

“You will report back for duty on Friday morning.”

“I will be back before six.”

“Oil will float on water.”

Would Epistemic possibility

Dynamic willingness

“I prophesized that I would be rich and look at me now!”

“Would you get the CD for me please?”

Shall Epistemic possibility Deontic necessity

Dynamic willingness

“We shall be away on holiday next week.”

“The committee shall meet at least four times per year.”

“I shall do as she says.”

Should Epistemic possibility

Deontic necessity

“The next road on the left should be King street.”

“You should tell your mother.”

(30)

In addition to the central modal auxiliaries this study also includes certain semi-modal and marginal modal expressions. Semi-modals, also called quasi-modals or periphrastic modals, are modal expression that are similar to modal auxiliaries in their functions. They are fixed idiomatic phrases such as: Have to, have (got) to, be able to, be going to, be supposed to, be about to and be bound to.

Marginal auxiliaries are verbs with only some of the qualities of modal auxiliaries. In Aarts’

(2011: 298) definition, this class includes: dare, need, and ought (to). Dare is seen to instantiate dynamic modality (59), but it lacks definite modal meaning. Need, although clearly modal by meaning, cannot be included as a full modal auxiliary as it has no past tense, while ought cannot be included since it is syntactically followed by to.

As can be seen from the different classifications given above, modality is not a simple phenomenon. This study will focus on the central modal auxiliaries can, could, may, might, shall, should, will, would, must, marginal modals dare, need (to), ought (to) and the semi- modals have to, have (got) to, be able to, be going to, be supposed to, be about to and be bound to.

In the qualitative part of this study all these modal expressions will be categorized according to their type (epistemic, deontic and dynamic), strength and degree according to Huddleston and Pullum (2002). These categories were chosen as they are widely used (Palmer, 1990, Huddleston & Pullum, 2002, Facchinetti et. al., 2003, Aarts, 2011) and as they are suited for a study of this nature. In the following three sections epistemic (3.3), deontic (3.4) and dynamic (3.5) modality are examined more closely.

3.3 Epistemic modality

The word ‘epistemic’ derives from the Greek word for “knowledge”. In terms of modality epistemic modality deals with the speaker’s attitude towards the factuality of a past or present situation (Huddleston & Pullum, 2002: 178). In other words, expressions of epistemic modality show to what extend the speaker finds something to be true. According to Palmer (2003: 7), epistemic modality is concerned solely with the speaker’s attitude to the status of the

(31)

proposition. Epistemic modality can be exemplified by the following examples from Huddleston and Pullum (2002: 178):

(57) “He has been delayed.” (fact)

(58) “He must have been delayed.” (strong epistemic modality) (59) “He may have been delayed.” (weak epistemic modality)

In the first example (57), above, the statement is unmodalised. It is presented as a fact that he has been delayed. However, the two latter examples (58) and (59) show epistemic modality. In (58) epistemic modality is strong. The speaker shows certainty with the use of must and seems confident that he in the sentence has indeed been delayed. In (59) epistemic modality is weaker.

The speaker is not as sure that he has been delayed but puts it forward as a possibility with the use of may.

According to Palmer the two modal auxiliaries that best represent epistemic modality are may and must. Possibility is marked by may (60) while necessity is marked by must (61). (Palmer, 1990: 50)

(60) “High earners may attempt to use tax planning strategies to avoid higher taxes.” (Possibility)

(61) “Any incident of electoral fraud must be prosecuted.” (Necessity)

(LPC Manifesto: 81 & 27) Palmer (1990: 51) acknowledges that epistemic modality can also be indicated with the modal auxiliaries will (62), might (63), would (64) and should (65) and the semi-modals be bound to (66) and have (got) to (67). He adds that with will it is sometimes difficult or even impossible to distinguish between epistemic use and referring to future events (1990: 57).

(62) “Tell him Professor Cressage is involved – he will know Professor Cressage.” (Possibility)

(63) “You think someone might be watching us?” (possibility) (64) “I think it would be Turner as well.” (possibility)

(32)

(65) “You should be meeting those later on this afternoon” (necessity) (66) “You are bound to find it a change being back in London.” (necessity) (67) “You have got to be joking.” (necessity)

3.4 Deontic modality

Simply put, deontic modality presents the speakers attitude towards whether something should be done (obligation) or could be done (permission). Huddleston & Pullum (2002: 178) state that ‘deontic’ is derived from the Greek for “binding” and concerns the speaker’s attitude to the actualization of future situations. Similarly, Marin-Arrese writes that deontic modality derives from the Greek word ‘deon’ meaning “duty” (2009: 30) and is concerned with necessity as obligation (must) or possibility as some form of enablement (may). Palmer writes, that deontic modality relates directly with the potentiality of an event. The important distinction in deontic modality is that the event is controlled by circumstances external to the subject of the sentence. To rephrase whether the event occurs is not up to the subject (Facchinetti et.al. 2003:

7). Deontic modality can be exemplified via the following examples from Huddleston and Pullum (2002: 178):

(68) “You must pull your socks up” (obligation) (69) “You may stay if you wish” (permission)

In the first example (68) the speaker is telling the recipient (obligating) to pull his/her socks up. The source of the obligation, in this case the speaker, is called the deontic source. In the second example (69) the speaker is giving the recipient permission to stay. In both of these cases the speaker, not the subject, is in control of the situation. In certain cases, it is difficult to differ between epistemic and deontic modality (70).

(70) “You must be very tactful.”

(Huddleston & Pullum, 2002: 178) (70) can be interpreted in two ways. The deontic interpretation sees it as a command to “be very careful”. The epistemic reading has the speaker give his/her opinion on someone’s tactfulness, i.e. “From what I know about you, I conclude that you are very tactful.” Therefore,

(33)

such ambiguous cases require contextual information in order to be interpreted as the speaker means. (Huddleston & Pullum, 2002: 178)

Palmer (1990: 69) divides deontic modality between deontic possibility and deontic necessity.

Deontic possibility is marked by may (71) and can (72) while deontic necessity is marked with must (73) and shall (74).

(71) “You may attend the lectures.” (possibility) (72) “You can attend the lectures.” (possibility) (73) “You must attend the lectures.” (necessity) (74) “You shall attend the lectures.” (necessity)

(Huddleston & Pullum, 2002: 182) However, Palmer notes that there is a problem with deontic must as it is complicated to distinguish it from neutral dynamic must. The problem is twofold, firstly there are cases where there is little or no indication of the involvement of the speaker (75). Secondly in some cases the subject is either ‘I’ or ‘we’ (76), and as Palmer points out “generally speaking we do not lay obligations upon ourselves” (1990: 113).

(75) “If the ratepayers should be consulted, so too must the council tenants.”

(neutral dynamic modality)

(76) “Yes, I must ask for that Monday off.” (willingness – dynamic modality) (Palmer, 1990: 113)

3.5 Dynamic modality

Like deontic modality, dynamic modality also relates directly with the potentiality of an event.

In dynamic modality the event is controlled by the subject. Dynamic modality in fact relates with one’s ability (77) or willingness (78), i.e. whether someone has the ability and/or the will to do something. (Facchinetti et. al., 2003: 7)

(34)

(77) “To help the UN respond more quickly to emerging and escalating conflicts, we will provide well-trained personnel that can be quickly deployed, including mission commanders, staff officers, and headquarter units.”

(LPC Manifesto: 69) (78) “A re-elected Conservative Government will increase resources to the Youth Gang Prevention Fund by 25 percent, to allow it to help even more young people avoid a life of crime.”

(CPC Manifesto: 109) Example (77) showcases the ability-aspect of dynamic modality. In it “resource-based

projects” have the ability to create jobs and spur investment. In example (78) “a re-elected Conservative Government” has the willingness to increase the resources for the Youth Gang Prevention Fund.

Huddleston and Pullum write about the difficulty of defining the boundary between dynamic and deontic modality (2002: 179). Their definition is that deontic modality requires an identifiable deontic source while dynamic modality can be more ambiguous. This interpretation is similar to Palmer’s (2003: 7) definition of the divide between internal (dynamic) and external (deontic) control over the event.

Huddleston and Pullum (2002) also point out that “Dynamic ability is less central to modality than deontic permission in that it does not involve the speaker’s attitude to the factuality or actualization of the situation.” Their views are supported by Halliday’s (2014: 691)

explanation of modality. According to him modality means construing the region of uncertainty that lies between ‘yes’ and ‘no’. However dynamic ability seems to imply a clearer ‘yes’ or ‘no’ answer i.e. either someone has the ability to do something or they do not.

Therefore the “region of uncertainty” that Halliday (2014: 691) writes about, is more closely related to epistemic and deontic modality.

In the previous chapters the grammatical constructs relevant to this study have been introduced.

The next chapter deals with the source of this study’s data, political manifestos.

(35)

4. Earlier studies

Political texts are widely researched, especially in the field of communication studies and political sciences eg. Wilkerson and Casas (2017) study which reviews the recent changes in political studies using text as data. The studies introduced below will focus on three areas relevant to this study: political manifestos, pronouns in political texts and modality in political texts. Political Discourse Analysis, a form of Critical Discourse Analysis, is a popular tool in analyzing political texts both in the field of linguistics as well as political sciences (e.g. Chilton 2002, 2004; Bello, 2013). However, there is a lack of linguistic studies that concentrate on election manifestos. They have mostly been analyzed for the purposes of political sciences, concentrating on the rhetorical aspects. An example of this is Kiratli’s study (2016) where he analyzed election manifestos for their discourses towards Europe and European integration in Germany, UK and Netherlands. A few linguistic studies on manifestos do exist and for the purpose of this study the work of Hanks and Može (Manifestospeak: What can linguistic analysis tell us about politicians and their attitudes, 2015) is especially relevant. They used linguistic analysis to review the recent election manifestos of four political parties from the United Kingdom: The Conservative Party, the Labour Party, the Liberal Democrats and the Green Party. Their results show that linguistic analysis of political manifestos can reveal more detailed information about what the parties are trying to convey about themselves. For example, the Green Party’s use of would instead of will shows hesitance regarding winning the election.

This is shown below (Table 5) in a comparison between statements from the Green Party and The Conservative Party.

Modality and personal pronouns in political texts: Corpus linguistic analysis of two Canadian political manifestos