Inclusivity and exclusivity in the use of Cantonese ngo 5 dei 6 (‘we’) in evasive replies in Hong Kong political discourse

Abstract

Politicians frequently face adversarial questions during election time. They often provide evasive replies to veer away from the controversial issues, but such equivocation also distances them from the audience. To deal with this problem, politicians often use the inclusive ‘we’ to identify themselves with the interest of the general public when they equivocate, or they sometimes use the exclusive ‘we’ to shift the responsibility of controversial policies to their political parties. The choice of inclusive versus exclusive ‘we’ in equivocation is not random but is governed by contextual factors, for example, the speech topic, the politician’s affiliation (if any) and the political system within a given culture. In Hong Kong, the Chief Executive Election candidates often do not belong to any political party. In this article, we examine how this unique contextual factor affects the choice of inclusive and exclusive ‘we’ in the evasive replies of politicians in the 2012 Hong Kong Chief Executive Election debates.

Keywords

Adversarial questioning evasion exclusive “we”footing shift inclusive “we”

Introduction

Evasion is a common strategy used by public leaders to handle adversarial questions in political discourse. Although evasion can help politicians to distance themselves from controversial issues, it also distances them from the audience. Previous studies show that politicians use a high frequency of first-person plural pronoun ‘we’ to align themselves with the audience in their evasive replies (Chen, 2007; Inigo-Mora, 2013). The use of the first-person pronoun was generally found to be influenced by a number of contextual factors, among them the topic and conversational goals of the political discourse, as well as the venue of the interaction (see, for example, Proctor et al., 2011; Van Dijk, 2002). In this study related to political discourse in Hong Kong, we will also include localized contextual factors such as non-partisanship of Chief Executive Election candidates.

The contextual influence on pronominal choice is well illustrated in Inigo-Mora’s (2013) study of television interviews with American President Bush and Spanish President Aznar about the US-Europe coalition against terrorism in the Middle East, which was found to lack justification for the war, that is, no discovery of weapons of mass destruction associated with terrorist groups at the time. Bush was interviewed on a British TV channel and had poor popularity ratings among the British public. In the interview, Bush frequently used the exclusive ‘we’ when referring to his American-led coalition against terrorism; in this way, he aligned himself with the European Union and took ‘political shelter’ from their support. On the other hand, Spanish President Aznar used equal amounts of the inclusive and exclusive ‘we’ in a local Spanish TV interview to solicit nationalistic support to justify Spanish participation in the American-led warfare. Inclusive ‘we’ was used to emphasize that Spanish security was being threatened by terrorism, and exclusive ‘we’ was used to highlight the importance of Spanish participation in the European Union’s cooperation with Americans in the counterterrorist movement.

The above examples illustrate how the inclusive and exclusive uses of the pronoun ‘we’ are often deployed by politicians to align or disalign themselves for political gain. While there have been a number of studies on these uses of ‘we’ in political discourse, including contexts involving adversarial questionings and evasive replies, the distribution of exclusive and inclusive ‘we’ among different types of evasive replies seems to be under-researched. Nor has there been a study that has identified contextual factors that have given rise to noticeable differences in the patterns of use of the inclusive and exclusive ‘we’ in different cultures. In this study, we will address these research questions by analyzing the use of the Cantonese first-person plural pronoun ngo⁵dei⁶ ‘we’ in electoral discourse in Hong Kong.

In 2012, Hong Kong experienced its first fiercely contested Chief Executive (CE) Election since the return of this former British colony to Chinese sovereignty in 1997. There was no prime candidate during this election campaign, and rumors of personal scandals were widely reported in the media. A lot of adversarial questions were raised in the election debates and evasive replies were commonly found. Unlike the political system in the countries with longer democratic traditions such as Britain and the United States, the CE candidates in Hong Kong generally did not belong to any political party and they had to face the aggressive questions alone. This study will examine how the pronoun ‘we’ was used in evasive replies in the CE Election debates in 2012, and whether the contextual factor of non-partisanship of CE candidates affected the ratio of inclusive and exclusive ‘we’ in their evasive replies.

Background on political discourse in Hong Kong

Hong Kong was once a British colony from 1841 to 1997. When it was returned to Chinese sovereignty, a unique ‘one country, two systems’ arrangement was stipulated in its mini-constitution known as ‘the Basic Law’. Under this special arrangement, Hong Kong became a Special Administrative Region with autonomous status, and its CE was chosen by an election committee. The initial years after the changeover were remarkable for their political stability, and for the first three terms the CE post was filled by prime candidates, who were favored by the central government in Beijing.

However, political stability was gradually eroded by unstable economic conditions and the social grievances that followed. The Asian financial turmoil in 1997 and the severe acute respiratory syndrome (SARS) epidemic in 2003 severely affected the economy, and polarization of wealth became a major public concern. In the wake of another financial crisis in 2008, various political forces began to criticize the government for lack of effective actions to address these economic and social ills. In the fourth CE election in 2012, criticism shifted to the CE candidates, especially when there was no clear favorite among them. Two candidates who had previously served in the government for extensive periods were heavily criticized for failing to deal with these problems.

Since these two CE candidates were not members of any political parties, the media tended to criticize them on personal grounds. During the election period, adversarial questions did not just focus on the candidate’s competence to govern but also targeted their moral integrity after various rumors were reported in the media. It was under this emotionally charged situation that each CE candidate needed to deploy some strategies to mitigate the damage from the adversarial questions during the 2012 election. One of the strategies that politicians often deploy in these challenging situations is evasion. However, avoiding to answer a question is not without risk for a politician, because he or she will be perceived to be hiding the truth, and thus seen as an irresponsible and untrustworthy candidate for high office. In this article, we will examine how politicians use the inclusivity and exclusivity functions of the Cantonese first-person plural pronoun ngo⁵dei⁶ ‘we’ to create solidarity with the audience and distance themselves from controversial issues.

Previous studies on evasion in political discourse

Adversarial questioning

In periods of political tension or scandals, criticism of politicians from the media and their rivals is often harsher (Bitiniene, 2007). They tend to ask adversarial questions to discredit or embarrass the politicians (see Clayman and Heritage, 2002a, 2002b; Greatbatch, 1988; Harris, 1991; Sun, 2010; Yip, 2003). The news media will not only cover political issues, but will also include controversial reports and rumors on partisan and personal scandalous or corrupt behavior (McGraw and Hubbard, 1996). Rather than presenting information in a neutral way, the news media tend to focus attention on certain problems and issues while ignoring others (Lippmann, 1920, 1922, 1925). The general public then begins to evaluate the politicians not just on their performance generally, but also their integrity specifically (Miller and Krosnick, 1996).

Adversarial questions are usually not simply asked for the purpose of seeking information, but are constructed to put politicians in a difficult position, often with a focus on the veracity and consistency of their message with reference to the present or previous context (Woods, 2006). This type of question is sometimes called a ‘communicative conflict question’, in which all possible replies will lead to negative consequences, yet a reply is still expected (Bull, 2003, 2008). It is also often referred to as an ‘avoidance-avoidance’ type question, where the politician prefers to avoid answering but at the same time wishes to avoid being seen as not answering. A possible reply strategy under such an ‘avoidance-avoidance’ situation is to say ‘something true but hurtful’, which then would work against the interest of the politician, or to say ‘something false but kind’, which then could damage the integrity of the politician (Bavelas et al., 1990). Clearly, neither strategy is a good choice for the politicians.

Evasion

Since answering questions is a basic moral obligation for politicians (Heritage, 1985; Raymond, 2000; Schegloff, 1968), politicians need to adopt ‘damage control’ strategies to reap the benefits of not answering while at the same time minimizing the costs associated with this risky action (Clayman, 2001). Evasion is one such strategy. It is a form of ‘non-straightforward communication’ (Bavelas et al., 1990: 87) and involves ‘intentional use of imprecise language’ (Hamilton and Mineo, 1998: 3). As Mehdipour and Nabifar (2011) put it, evasion is generally seen as ‘a way to satisfy both the obligation of telling the truth and of keeping secrets from those not entitled to know the truth’ (p. 57).

Bavelas et al. (1990) proposed that evasion can be understood along four dimensions: sender, content, receiver and context. Simply put, a reply is evasive if the sender does not fully acknowledge a previous statement as his or her own opinion, or attributes it to someone else. A reply is also seen to be evasive if its content is unclear. If a reply is not addressed directly to the receiver in the situation, or if a reply is not relevant to the immediate context of the question, it is also considered to be an evasive reply. Harris (1991) further added that evasion is done not just by answering a question indirectly in terms of the above four dimensions, but also by challenging the presuppositions or the illocutionary force of the question. For instance, the politician denies that he has even said or done anything related to the propositional content of an aggressive question.

Subsequent studies on evasion focused more on the content and context dimensions of Bavelas et al. (1990). Using clarity of the content and relevance of the topic in the politicians’ responses as a measure, Bull (1994) identified a range of intermediate responses which fall somewhere between ‘a full reply’ and ‘a non-reply’ in the question-answer format in political interviews. These intermediate responses include implications and incomplete replies. Implications are replies in which the speaker makes his or her views known in a subtle way, ‘without explicitly stating them’ (Bull, 1994: 127). Incomplete replies are responses in which the politician answers only part of a single question, or one of the two questions or a fraction of multiple questions. For this study, we will regroup these various evasion strategies into five main types, namely, challenges (following Harris, 1991), implications, half-replies, partial replies and non-replies (following Bull, 1994, 2003; Bull and Mayer, 1993).

Pronominal usage in evasive replies

In addition to different evasion types, scholars have also analyzed the relationship between pronominal use and evasion in political discourse (O’Connell and Kowal, 2002). Below, we will review some studies on the use of the pronoun ‘we’ in political discourse, with special focus on the use of this pronoun as evasion strategies.

Exclusive and inclusive ‘we’ in political discourse

Pronominal choice reflects the speaker’s intention and its influence on the hearer (Wilson, 1990). The pronoun ‘we’ may include the hearer or may not include the hearer in different utterances. For example, ‘We go together’ includes the hearer in the utterance, while ‘We will show you the way’ excludes the hearer. Therefore, the ambiguity in the use of the pronominal choice enables the speaker to show his or her identification with a particular group or his or her differences with another group (Tabakowska, 2002; Tannen, 1993). We need contextual knowledge, such as the topic or the speech event, to infer the changeable meaning of ‘we’ (Proctor et al., 2011).

Previous studies have noted that politicians often exploit the changeable meaning of ‘we’ in their strategic use to distinguish themselves from their rivals, to construct positive identities for themselves and to establish a positive relationship with the general public (Goffman, 1974, 1981; Mühlhäusler and Harré, 1990; Putri and Kurniawan, 2015 inter alia). For example, Chilton and Schaffner (1997) noted that politicians often use the exclusive ‘we’ to align themselves with tangible achievements (e.g. an economic boom under their governance), and to distance themselves from controversial policies or unfavorable political actions (e.g. by denying responsibility). On the other hand, Putri and Kurniawan (2015) found that politicians used the inclusive ‘we’ to align with the audience (and by extension, the general public) and convince them that ‘together’ they can solve the problems that confront them. Such alignment and disalignment strategies have been found to be widely used in various types of political discourses, such as speeches (De Fina, 1995; Kowal et al., 1997; Wilson, 1990), debates (Blas Arroyo, 2000; Jaworski and Galasinski, 2000) and interviews (Bull, 1994; Bull and Fetzer, 2006; Suleiman, 2000).

From a linguistic perspective, the pronominal ‘we’ is widely used in different political discourses because it has a wide referential reading in the interactional context (Strazny, 2005). For example, when using the pronominal ‘we’ to construct a positive identity for themselves, politicians can make reference not only to themselves but also to exophoric referents (Halliday, 1994), such as their institutional groups and the overhearing audience in the political discourse. In other words, in campaign speeches, election debates and political interviews, politicians can draw on the support from their political party and from the audience, both those physically present during the event and those following the event through various media channels (Helmbrecht, 2002). Essentially, the use of inclusive ‘we’ provides politicians with a rhetorical strategy to create shared feelings with the audience and help them gain more votes (Allen, 2007; Beard, 2000).

Pragmatic functions of exclusive and inclusive ‘we’ in evasive replies

As is evident from the above previous studies, the grouping of different referents with the pronoun ‘we’ is usually divided into two major categories, namely, inclusive ‘we’ and exclusive ‘we’ (see also Leech and Svartvik, 1994; O’Keeffe et al., 2011; Richards and Schmidt, 2002; Yule, 1996). Inclusive ‘we’ refers to a group including the speaker and the persons being addressed, which may be the audience in the speech event or the overhearing audience watching the media broadcast at home. It creates positive effects such as enhancing solidarity with the audience (Fairclough, 1995). Exclusive ‘we’ refers to a group including the speaker and a specific group of people, but excluding the addressee, who may be their opponents or the audience in the interactional context. It either creates positive effects such as highlighting one’s achievements, or avoids potential negative effects arising from the need to deny responsibility for wrongdoing or the need to distance oneself from controversial issues by saying ‘We do not do X, but others do’ (Harris, 2001).

Another strategic use of the pronominal ‘we’ when engaging in evasive replies is through ‘shift of footing’ (Goffman, 1981), that is, a shift in positioning, whereby the speaker uses inclusive ‘we’ to shift from a personal standpoint to a collective standpoint when adversarial questions are targeted at some political or social policies. Goffman noted that politicians downplay their involvement in controversial policies by anchoring themselves as a member of their political party when adversarial questions are targeted at them. To answer a question like ‘What is your opinion about issue X?’, they often provide a response like ‘We always take the position that …’, thus indicating that it is a party decision to adopt the controversial policy. In such situations, politicians tend not to provide answers with the subject pronoun ‘I’; instead, they try to distance themselves from the controversial policy and deflect responsibility by using the exclusive pronoun ‘we’ to stand for a collective decision, thus attenuating the threat to their individual face when confronted with the need to evade an aggressive question.

Quantitative and qualitative studies of pronominal ‘we’ as evasive strategies in different cultural contexts

The significance of evasion is not just reflected by its various forms of construction, it is also indicated by the findings of cross-cultural quantitative studies. In UK political interviews, Harris (1991) found that 61% of the politicians were evasive in answering questions, and Bull (1994) found a similar trend with 54% of the politicians’ replies being evasive. In the 2005 Taiwan presidential election debates, Chen (2007) found that the pronoun wǒmen ‘we’ was used with a mean frequency of 71% by the two leading candidates. Exclusive ‘we’ appeared with a frequency of 53% and was used to highlight the candidates’ past achievements and to show the future plans of their political parties in order to present themselves as strong leaders. Inclusive ‘we’ occurred with a frequency of 47% and was used to build rapport with the voters. The candidates also used inclusive ‘we’ to create a sense of shared knowledge and common ground with the general public, and to also create a sense of collective opinion to attack their election rivals in the debates.

In a similar vein, in a quantitative analysis of political interviews with British Prime Minister Tony Blair, American President George Bush and Spanish President José Aznar on the controversial topic of the coalition war against terrorist groups in the Middle East, Inigo-Mora (2013) found that 53% of the questions in the interviews were adversarial and the three political leaders combined equivocated in their responses to approximately 80% of the adversarial questions.¹ On average, these leaders used 3.3 tokens of the pronoun ‘we’ per evasive reply. Blair and Bush frequently used exclusive ‘we’ (to refer to the coalition of American and European countries), averaging 66% in their replies, to support their warfare against the alleged terrorists in the Middle East. Meanwhile, Aznar used both inclusive ‘we’ (when emphasizing the safety of the Spanish people) and exclusive ‘we’ (when highlighting Spain as a supportive ally of the United States and a responsible member of the European Union), averaging 49% for both inclusive and exclusive uses of ‘we’, to support their participation in the coalition war. Aznar tended to use more inclusive ‘we’ than Bush did because he needed to solicit nationalistic support to legitimize Spanish participation in the American-led warfare, essentially by emphasizing that Spanish security was also under threat from terrorism masterminded from beyond its borders. Although the topic may be the same, different speech contexts may influence the pronominal choices of the politicians. Similar findings on the use of inclusive and exclusive ‘we’ as alignment and disalignment strategies were also found in other studies on political speeches and political debates in Spain (Blas Arroyo, 1998, 2000; Carranza Márquez, 2012), as well as in Mexico (De Fina, 1995) and the United States (Silvestre-López, 2004).

Research questions

Over the past few decades, there has been growing interest in understanding how politicians deploy inclusive and exclusive readings of first-person plural pronouns for political gain. Previous studies have largely focused on the use of the English pronominal we in the context of American and British political discourse, with some recent studies particularly examining how contextual factors influence the pronominal choices in political discourse. This article will examine how the Cantonese first-person plural pronoun ngo⁵dei⁶ (‘we’) was used in the 2012 CE election debates in Hong Kong, which was the first fiercely contested election with no prime candidate since the changeover of sovereignty in 1997. So far, previous studies have not analyzed the use of the inclusive and exclusive ‘we’ in relation to the five main types of evasion strategies, namely, challenge, implication, half-reply, partial reply and non-reply. This study will particularly analyze the clusivity uses of ngo⁵dei⁶ (‘we’) in terms of these five main types of evasion strategies in Hong Kong political discourse. Since Hong Kong has a unique political system in which CE candidates generally have no affiliation with political parties, this study will also consider this contextual factor in analyzing the clusivity use of ngo⁵dei⁶ (‘we’) in electoral discourse within a non-parliamentarian-style context.

Our specific research questions are as follows:

How frequent were evasive replies used in the 2012 Hong Kong CE election debates?

What type(s) of evasion strategies was/were used in response to the adversarial questions in these election debates?

What was the usage frequency and ratio of inclusive versus exclusive first-person plural pronoun ngo⁵dei⁶ (‘we’) in each type of evasion strategy used in these election debates?

What were the socio-pragmatic functions of inclusive versus exclusive ngo⁵dei⁶ (‘we’) in the evasive replies in these election debates?

Was there any difference in the use of the first-person pronoun ‘we’ in Hong Kong electoral discourse compared to those in other political systems elsewhere? If yes, what contextual factor accounts for this difference?

Data and methodology

Our data come from two CE Election debates in Hong Kong televised on 16 and 19 March in 2012. They were chosen because a wide range of aggressive questions were raised against two of the candidates, Henry Tang and CY Leung, who often responded with various types of evasive replies. The debates were transcribed to form a corpus of 90,833 words. Acronyms will often be used in this article to refer to the three candidates in the 2012 CE elections, that is, HT for Henry Tang, CY for CY Leung and AH for Albert Ho. Our reason for doing so is to highlight the evasive strategies rather than the individual candidates.

In this study, we adopted a mixed quantitative and qualitative analysis. For the quantitative analysis, we first culled all the adversarial questions in the two debates. As noted in Pishwa (2015), adversarial questions in political discourse are often prefaced by some unfavorable background information, with a focus on the veracity and consistency of the message with reference to the present or previous context (Ekström and Tolson, 2017; Woods, 2006). To point out that their rivals are making contradictory statements or doing something that deviates from what they previously said or promised (Pishwa, 2015), politicians often use evidentiality-marking strategies such as ‘(hear) say’ verbs (Hanlon, 2015; Vukovic, 2014). These evidentiality-marking strategies help mark or imply non-firsthand information (Aikhenvald, 2004), and discreetly distance politicians from their face-threatening act (Ahn and Yap, 2015). The discrediting statements in adversarial contexts are then often rounded up by an evaluative closing segment (Thornborrow and Haarman, 2017), which could involve negative evaluative markers and question tags. In the case of Mandarin or Cantonese, the question tags may take the form of an aggressive A-not-A question construction (Holmberg, 2015; Huang et al., 2009).² In Excerpt 1, we illustrate how these rhetorical devices are used to identify the aggressive questions in this study.

During the election campaign, the media reported that one of the candidates, CY, had a dinner meeting with a gang leader. In the election debate, another candidate, AH, tried to discredit CY for violating the social norms and used the ‘say’ verb in line 003 to pin down CY for having denied knowing the gang leader and claiming that the gang leader had simply walked in and sat down in his high-level closed-door electioneering meeting. AH then escalates his challenge in lines 006 and 007 through the use of negative evaluative adjective ‘ridiculous’ to describe this excuse within the aggressive A-not-A question construction ‘Now, isn’t that ridiculous?’ (literally in Cantonese, ‘Now, is that ridiculous, or not?’).

Excerpt 1

Chief Executive Election Debate 16/3/2012 (Part II, 10:21 – 10:40)

001 AH: 梁生，呢個係一個私人嘅飯局

loeng⁴ saang¹ ，ni¹ go³ hai⁶ jat¹ go³ si¹ jan⁴ ge³ faan⁶ guk⁶

‘Mr. Leung, this was a private meeting.’

002 係你嘅競選辧呢，核心人員，

hai⁶ nei⁵ ge³ ging⁶ syun² baan⁶ ne¹， hat⁶ sam¹ jan⁴ jyun⁴，

同埋選委，一齊坐埋傾選舉。

tung⁴ maai⁴ syun² wai²， jat¹ cai⁴ co⁵ maai⁴ king¹ syun² geoi²

‘Core members of your election team met with the Election Committee members and promoted your election manifesto.’

003 閂埋度門，竟然有人拍門入嚟，

saan¹ maai⁴ dou⁶ mun⁴ ，ging² jin⁴ jau⁵ jan⁴ paak³ mun⁴ jap⁶ lai⁴，

唔識佢係邊個。

m⁴ sik¹ keoi⁵ hai⁶ bin¹ go³

‘The door was closed but someone just knocked on the door and entered the room. No one knew who he was.’

004 原來係江湖人物，大佬嚟嘅，

jyun⁴ loi⁴ hai⁶ gong¹ wu⁴ jan⁴ mat⁶ ，daai⁶ lou² lei⁴ ge³ ，

可以坐得低，同你哋傾。

ho² ji⁵ co⁵ dak¹ dai¹ ，tung⁴ nei⁵ dei² king¹

‘Actually, he was a gang leader and was allowed to sit down and join the meeting.’

005 喂，你又話唔識佢，

wai³ ，nei⁵ jau⁶ waa⁶ m⁴ sik¹ keoi⁵

‘Now, you said that you didn’t know him.’

006 喂，係咪天方夜談呀咁樣？

wai³ ，hai⁶ mai¹ tin¹ fong¹ je⁶ taam⁴ aa¹ gam³ joeng⁶

‘Now, isn’t that rather ridiculous?

[Question tag hai⁶ mai¹ ‘Isn’t it’ was used to challenge his rival’s unusual behavior of letting someone he didn’t know to join a meeting with him and his team, which was described by a negative evaluative marker ‘ridiculous’.]

007 喂，大家老實啲喇好唔好 ?

wai³ ，daai⁶ gaa¹ lou⁵ sat⁶ di¹ laa¹ hou² m⁴ hou²

‘Now, let’s be honest on this issue, shall we not?

[Question tag hou² m⁴ hou² ‘shall we (do so) or not?’ and evaluative adjective lou⁵ sat⁶ ‘honest’ were used to discredit the speaker’s rival and it was suggested that his rival should admit he just denied knowing the latecomer because this person was a gang leader.]

Having culled all the aggressive questions, we then culled all the evasive replies used in response to these adversarial questions. Each evasive reply was then analyzed and classified in terms of the type of strategy used: challenges, implications, half-replies, partial replies and non-replies.

Challenges are evasive replies in which the speaker disputes the presupposition of the question asked (see Harris, 1991).

Implications are evasive replies in which ‘the speaker makes his or her views clear but without explicitly stating them’ (see Bull, 1994: 127).

Half-replies are evasive replies in which the speaker answers only one of the two (i.e. ‘double-barreled’) questions directed to him or her (Bull, 1994).

Partial replies are evasive replies where the speaker answers only part of the question (Bull, 1994, 2003; Bull and Mayer, 1993).

Non-replies are evasive replies in which ‘the speaker fails to provide any of the information requested in the question’ (see Clayman, 2001: 412; Day, 1991).

Each evasion type was initially tagged by the first author and then reviewed by the second author. Then, all tokens of pronoun ngo⁵dei⁶ ‘we’ in the evasive replies were identified through the corpus software WordSmith. The inclusive and exclusive readings of pronoun ngo⁵dei⁶ ‘we’ were likewise initially coded by the first author and then reviewed by the second author to ensure high accuracy of analysis.

Analysis and discussion

Quantitative analysis

During the two election debates, 168 questions were raised and 107 questions (63.7%) were adversarial. In all, 73 (68.2%) replies to these adversarial questions were evasive (see Table 1). In their evasive replies, the candidates tended to choose implications (56 tokens; 76.7%). Non-replies were the next most frequent ones (14 tokens; 19.3%). The candidates did not favor using other evasive strategies, with two challenges (2.7%) and one half-reply (1.3%), and no partial reply at all (see Table 2).

Table 1.

Frequency of evasion in the 2012 Hong Kong Chief Executive election debates.

	No. of aggressive questions	No. of evasive replies	No. of direct replies
Frequency	107	73	34
Percentage	100%	68.2%	31.8%

Table 2.

Frequency of different types of evasion in the 2012 Hong Kong Chief Executive election debates.

	Types of evasive replies
	Challenge	Implication	Half-reply	Partial reply	Non-reply
Frequency	2	56	1	0	14
Percentage	2.7%	76.7%	1.3%	0.0%	19.3%

As seen in Table 3, altogether 95 tokens of the pronominal ‘we’ were used in the 73 evasive replies. That is, there were on average 1.3 tokens of pronominal ‘we’ per evasive reply. In terms of type of evasive reply, the candidates tended to use more pronominal ‘we’ when deploying the implication strategy (1.5 tokens per reply) and fewer pronominal ‘we’ for the non-reply strategy (0.4 token per reply). There were too few tokens to calculate a meaningful mean frequency per evasive reply for the challenge strategy and the half-reply.

Table 3.

Frequency of pronominal ‘we’ in different types of evasive reply in the 2012 Hong Kong Chief Executive election debates.

Type of evasive reply strategies	No. of evasive replies	No. of pronominal ‘we’	Ratio of pronominal ‘we’ per evasive reply in each type of evasive strategy
Challenge	2	1	0.5
Implication	56	86	1.5
Half-reply	1	3	3.0
Partial reply	0	0	n/a
Non-reply	14	5	0.4
Total	73	95	1.3 (on average)

In using the pronominal ‘we’ in their evasive replies, the election candidates tended to use more of inclusive ‘we’ than exclusive ‘we’ (see Table 4). Nearly two-thirds of the ngo⁵dei⁶ ‘we’ tokens (63.2%) were deployed with an inclusive ‘we’ reading, nearly double those with an exclusive ‘we’ reading (36.8%). Similar high ratios of inclusive ‘we’ were found in the two major types of evasive replies. For implications, approximately two-thirds of the pronominal ‘we’ tokens were inclusive (65.1%). For non-replies, the percentage of inclusive ‘we’ even went up to 80%. On the contrary, an opposite pattern of pronominal choice was found in the two minor types of evasive replies. Inclusive ‘we’ was not used at all in the challenge and half-reply formats. Instead, exclusive ‘we’ was preferred for these two minor strategies, although the small number of tokens preclude generalization and further study with a larger dataset may be needed.

Table 4.

Frequency of inclusive and exclusive ‘we’ in different types of evasive reply in the 2012 Hong Kong Chief Executive election debates.

Type of evasive reply strategies	No. of evasive replies	Inclusive ‘we’		Exclusive ‘we’
Type of evasive reply strategies	No. of evasive replies	Frequency	Percentage	Frequency	Percentage
Challenge	2	0	0.0%	1	100.0%
Implication	56	56	65.1%	30	34.9%
Half-reply	1	0	0.0%	3	100.0%
Partial reply	0	0	0.0%	0	0.0%
Non-reply	14	4	80.0%	1	20.0%
Total	73	60	63.2%	35	36.8%

The relatively low percentage of exclusive ‘we’ in these election debates may be explained by the fact that the CE candidates did not belong to any political party. Since they did not have past or current achievements linked to a political party, they tended to rely largely on some personal achievements during their service in the previous governments. Moreover, without a political affiliation, they also could not shift their personal responsibility to a political party when responding to questions on some unpopular policies.

Most of the exclusive ‘we’ tokens produced by the candidates were used instead to refer to their election team, for example, to highlight their team’s ability in formulating future policies. This was often done when the candidates tried to shift topics in response to adversarial questions. In doing so, most of the time the candidates would also use inclusive ‘we’ to highlight that their teams formulated their proposed government policies based on public opinion as well. These efforts to not only distance themselves from blame but to also enhance common ground with the general public explained to a great extent why election candidates tended to use inclusive ‘we’ more than exclusive ‘we’ in political debates.

Qualitative analysis

We will now examine the strategic use of the pronominal ‘we’ in each type of evasive reply in greater detail using contextual analysis. In particular, we will focus on the use of inclusive and exclusive ‘we’ in terms of alignment and disalignment strategies (i.e. shifts in footing; see Bull and Fetzer, 2006; De Fina, 1995; Goffman, 1981; Hutchby, 2013; Rendle-Short, 2007).

Exclusive ‘we’ in challenge-type evasive replies

There was one instance of challenge-type evasive reply in the election debates. Exclusive ‘we’ was used in this instance. In the CE Election debate on 16 March 2012, HT asked his rival CY why he suddenly changed his housing policy after his nomination was endorsed by the Election Committee. HT claimed that CY was scaling back on his earlier promise related to public housing in his election manifesto after he succeeded in gathering enough votes to stand as a CE candidate. CY had originally planned to supply 35,000 units of public housing in the first year of his governance, if elected, but he later revised the schedule and promised instead to achieve the goal as soon as possible in the initial years. HT questioned the integrity of CY, asking him if he was dishonoring his promise of ensuring a quick and adequate supply of public housing after he was nominated because he was trying to avoid damaging the interests of the property developers.

CY responded to the first part of HT’s aggressive question by implying that HT’s accusation was factually inaccurate and that he had not been following up with recent news. CY then went on to defend his revised manifesto with updated information regarding the shortage of land and used exclusive ‘we’ to highlight that he had a team to help him work out an improved policy to solve the housing problem. His reply is transcribed as follows:

Excerpt 2: ‘The Land Supply Problem’

Chief Executive Election Debate 16/3/2012 (Part I, 07:28 – 07:46)

008 HT: 梁振英，喺你政綱初期嘅時候，

loeng⁴zan³jing¹， hai² nei⁵ zing³gong¹ co¹kei⁴ ge³ si⁴hau⁶

‘Mr. Leung Chun Ying, in your initial election manifesto,’

009 你曾經提議過起三萬五千個公屋。

nei⁵ zang¹ging¹ tai⁴ji³ gwo³ hei² saam¹maan⁶ng⁵cin¹ go³ gung¹nguk¹

‘you had suggested constructing 35,000 public housing units.’

010 然後跟住又縮咗水嘅。

jin⁴hau⁶ gan¹zyu⁶ jau⁶ suk¹zo²seoi² ge³

‘Then you reduced the number of units afterwards.’

011 你係咪為咗招攬啲地產商

nei⁵ hai⁶mai¹ wai⁴zo² ziu¹laam⁵ di¹ dei⁶caan²soeng¹

012 去向佢招降呢？

heoi³ hoeng³ keoi⁵ ziu¹gong³ le¹

‘Are you going to surrender to the property developers in order to gain their support in the Chief Executive Election?’

013 CY: 你大概喺呢段時間無乜睇報紙，

nei⁵ daai⁶koi³ hai⁶ lei¹dyun⁶ si⁴gaan¹ mou⁵mat¹ tai² bou³zi²,

‘Perhaps you haven’t read the newspaper recently,’

014 現在特區政府有嘅地，

jin⁶zoi⁶ dak⁶keoi¹ zing³fu² jau⁵ ge³ dei⁶

‘Regarding the land supply of the current government,’

015 現在係無可能喺短期裡面

jin⁶zoi⁶ hai⁶ mou⁵ ho²nang⁴ hai² dyun²kei⁴ leoi⁵min⁶

016 增加土地供應黎到增加呢個公屋數量。

zang¹gaa¹ tou²dei⁶gung¹jing¹ lai⁴dou³ zang¹gaa¹ lei¹go³ gung¹nguk¹sou³loeng⁴

‘it is impossible to increase the land supply to build such (a huge) quantity of public housing units in a short time.’

017 我嘅講法喺記者會度講得好清楚。

ngo⁵ ge³ gong²faat³ hai² gei³ze²wui² dou⁶ gong dak¹ hou² cing¹co²

‘I have clearly stated in the press conference,’

018 就係喺七萬五千個單位裡面

zau⁶hai⁶ hai² cat¹maan⁶ng⁵cin¹ go³ daan¹wai² leoi⁵min⁶

019 攞三萬五千個出黎喺早年裡面完成。

lo² saam¹maan⁶ng⁵cin¹ go³ ceot¹lai⁴ hai² zou²nin⁴ leoi⁵min⁶ jyun⁴sing⁴

‘that 35,000 public housing units will be finished initially from a total of 75,000 units.’

020 → 喺中期呢 (.) 我地呢 (.)

hai² zung¹kei⁴ le¹ (.) ngo⁵dei⁶ le¹

021 再覓新地黎興建更多嘅公屋單位，

zoi³ mik⁶ san¹dei⁶ lai⁴ hing¹gin³ gang³do¹ ge³ gung¹nguk¹ daan¹wai²

‘After an interim review, we will source more new land for public housing construction,’

022 請唐英年先生返去翻查報紙。

cing² tong⁴jing¹nin² sin¹saang¹ faan¹heoi³ faan¹caa⁴ bou³zi²

‘Mr. Tang, please check the relevant news report accordingly.’

In his reply, CY challenged the validity of HT’s criticism. He first undermined his opponent’s ability to gather updated information (line 013), before going on to explain that lack of land supply was the reason for the reduced number of public housing units to be constructed in his revised election manifesto (lines 014–016). He pointed out that matters related to his revised manifesto had been reported in the media (line 017), and further undermined HT’s positive image by highlighting his lack of awareness of new developments. He went on to talk about his updated plan on how he would solve the public housing supply problem (lines 018–021), without addressing the question of whether he was trying to protect the interests of the property developers. He then closed his evasive reply with a rebuke to HT to ‘please check the relevant news report’ (line 022). In effect, through a series of challenges, CY was constructing a negative image for his rival while at the same time paving the way for him to shift the topic away from the aggressive questioning. It allowed him to turn to topics where he could promote his own manifesto and construct a positive image for himself.

In terms of his use of pronominal ‘we’ (line 020), CY used the exclusive reading to relieve himself and his team of culpability by pointing to the insufficient land supply in the current government’s land allocation system, which made it impossible for him to keep his original promise of building 35,000 public housing units within the first year of office. At the same time, he also used this exclusive reading to promote himself when talking about his revised future plan to make up for the shortfall in his earlier promise to build 35,000 public housing units at the earliest possible time. His use of exclusive ‘we’ was a strategy to improve his and his team’s image as a responsible and hard-working political group that would work for the benefit of the society. Such uses of exclusive ‘we’ were also frequently found in other evasive strategies, with 30 tokens attested for the implication-type evasive strategy.

Inclusive and Exclusive ‘we’ in implication-type evasive replies

In the CE election debates, the candidates also needed to answer a lot of adversarial questions that involved the conflicting interests of different groups in the society. As noted earlier, most of the replies from the CE candidates were evasive and took the form of implications, representing about three-quarters of all evasive replies (76.7%). In implications, the candidates tended to use more inclusive ‘we’ tokens (65.1%), which was double those of exclusive ‘we’ tokens (34.9%). Implications were often used because the candidates did not want to commit themselves to certain controversial policies, which might be against the interests of certain groups among the potential voters. However, lack of commitment would undermine the credibility of the candidates. Thus, they would often use inclusive ‘we’ to express their commitment to the society ‘as a whole’, and to show that they would handle the issue appropriately with regard to the overall benefit of the society.

In the CE Candidates Forum on 19 March 2012, the candidates faced this type of situation when a member of the audience asked all three candidates how they would balance the profit of the capitalists and the basic needs of the working class, and also how they would handle the social contradictions caused by the wealth polarization in Hong Kong. One of the candidates, CY, replied evasively using implication and inclusive ‘we’ as follows:

Excerpt 3: ‘The Wealth Gap’

Chief Executive Candidates Forum 19/3/2012 (Part IV, 01:53 – 03:33)

023 AUD: 如何唉處理資本家嘅合理嘅利潤，

jyu⁴ho⁴ e⁶ syu³lei⁵ zi¹bun²gaa¹ ge³ hap⁶lei⁵ ge³ lei⁶jeon⁶

024 以及受薪階層嘅基本權益，

ji⁵kap¹ sau⁶san¹ gaai¹cang⁴ ge³ gei¹bun² kyun⁴ jik¹

‘How will you balance the reasonable profit of the businessmen and the fundamental benefit of the working class?’

025 減少社會矛盾?

gaam²siu² se⁵wui² maau⁴teon⁵?

‘And how will you minimize the social conflict as well?’

026 CY: 至於社會貧富矛盾嘅問題，

zi³jyu¹ se⁵wui² pan⁴fu³ maau⁴teon⁵ ge³ man⁶tai⁴

‘Regarding the issue of the wealth gap and social contradictions …’

027 → 我地要注意呢個問題。

ngo⁵dei⁶ jiu³ zyu³ji³ lei¹go³ man⁶tai⁴

‘We have to deal with it.’

028 如果工商界要

jyu⁴ gwo² gung¹soeng¹gaai³ jiu³

029 有搵錢嘅機會嘅話，

jau⁵ wan²cin² ge³ gei¹wui² ge³waa²

030 一定要社會穩定。

jat¹ding⁶ jiu³ se⁵wui² wan²ding⁶

‘The businessmen can only have profit when there is a stable society.’

031 我亦都提出就話經濟發展嘅成果，

ngo⁵ jik⁶dou¹ tai⁴ceot¹ zau⁶waa² ging¹zai³faat³zin² ge³ sing⁴gwo²

032 我地要合理分配等等。

ngo⁵dei⁶ jiu³ hap⁶lei⁵ fan¹pui³ dang²dang²

‘I have also proposed that we need to redistribute wealth more reasonably …’

In his reply, CY just provided a broad principle to handle this problem without any detailed solution, such as a progressive tax rate or social security measure. It seemed that he would handle this problem but did not commit himself to a particular solution. He understood that a laissez-faire policy would protect the businessmen but the grassroots would be exploited, while vigorous ‘welfarism’ would protect the grassroots but hurt the business environment. Thus, through the use of pronominal ‘we’, CY tried not to align himself with either side. Rather, he used ‘we’ in line 027 in an inclusive sense and referred to the society as a whole to deal with this problem collectively. At the same time, he used exclusive ‘we’ when pointing out in line 032 that his election team would deal with this issue appropriately. In this way, he could align himself with the shared values of various sectors of the society and promote his election team as well.

Exclusive ‘we’ and ‘our’ in half-replies

There were three half-replies in the two debates. In one of these tokens, AH criticized CY for his inconsistent welfare policy during the election campaign. Before being nominated by the Election committee, CY had proposed a comprehensive welfare policy and gained the support from the Social Welfare constituency. After the nomination, CY narrowed the scope of his social welfare policy following a meeting with the business sector and tried to gain the support from the businessmen. He also labeled the other candidates, AH and HT, as advocates of ‘welfarism’ during the meeting. Using a double-barreled question, AH asked CY to explain why he changed his social welfare policy and why he labeled his rivals as ‘welfarists’. CY only answered the first question and avoided answering the second question in the following way:

Excerpt 4: ‘Letters from the Public’

Chief Executive Candidates Forum 19/3/2012 (Part IV, 07:24 – 08:01)

033 AH: 梁振英先生，喺佢報名之前，

loeng⁴zan³jing¹ sin¹sang¹， hai²keoi⁵ bou³ming⁴ zi¹cin⁴，

034 佢講咗好多政策贏倒

keoi⁵ gong²zo² hou²do¹ zing³caak³ jing⁴dou²

035 唔少社福界嘅人士對佢嘅一啲嘅欣賞。

m⁴siu² se⁵fuk¹gaai³ ge³ jan²si¹ deoi³ keoi⁵ ge³ jat¹di¹ ge³ jan¹soeng²

‘Before the nomination, Mr. Leung Chung Ying proposed a lot of welfarist policies and won some support from the social services sector.’

036 但係入咗閘之後呢，

daan⁶hai⁶ jap⁶zo² zaap⁶ zi¹hau⁶ le¹

‘However, after he was nominated,’

037 對住啲工商界呢，

deoi³zyu⁶ di¹ gung¹soeng¹gaai³ le¹

‘when he had a meeting with industrial, commercial and financial sectors,’

038 就話我同唐英年搞福利主義。

zau⁶ waa² ngo⁵ tung⁴ tong⁴jing¹nin² gaau² fuk¹lei⁴ zyu²ji⁶

‘he labeled Henry Tang and me as welfarists.’

039 我唉真係有啲唔明白。

ngo⁵ e⁶ zan¹hai⁶ jau⁵ di¹ m⁴ ming⁴baak⁶

‘I don’t quite understand it.’

040 我真係好想聽吓佢嘅解釋。

ngo⁵ zan¹hai⁶ hou²soeng² teng¹haa⁵ keoi⁵ ge³ gaai²sik¹

‘I really want to hear his explanation.’

041 CY: 我嘅政綱咨詢稿出台之後

ngo⁵ ge³ zing³gong¹ zi¹seon¹ gou²ceot¹toi⁴ zi¹hau⁶

042 至到定稿嗰段期間，

zi³dou³ ding⁶gou² go²dyun⁶ kei⁴gaan¹

‘From promulgation to finalisation of my election manifesto,’

043 我收倒五六百封市民嘅意見

ngo⁵ sau¹ dou² ng⁵luk⁶baak³ fung¹ si⁵man⁴ ge³ ji³gin³

‘I received around 500 – 600 letters from the public.’

044 → 我地根據市民嘅意見 (.)

ngo⁵dei⁶ gan¹geoi³ si⁵man⁴ ge3 ji³gin³

045 → 黎制定我地最後嘅政綱定稿。

lai⁴ zai³ding⁶ ngo⁵dei⁶ zeoi³hau⁶ ge³ zing³gong¹ ding⁶gou²

‘Informed by public opinion, we finalized our election manifesto.’

046 → 我地係以市民嘅意見為依歸。

ngo⁵dei⁶ hai⁶ ji⁵ si⁵man⁴ ge³ ji³gin³ wai⁴ ji¹gwai¹

‘We (made the changes) based on the opinion of the general public as our mantra.’

In his half-reply, CY only explained why he had changed the welfare policy in his manifesto but did not answer why he had labeled his rivals as ‘welfarists’. When defending his policy change, CY emphasized that he and his team had revised their plan after receiving 500–600 letters from the general public. He intentionally used ngo⁵dei⁶ ‘we/our’ (lines 044 – 046) to denote he had a reliable team to review his policy in light of public opinion, and that he did not change the policy because of pressure from the business sector. That was a strategy to maintain his positive face while engaging in an evasive reply. It was also a means of projecting himself as a strong leader with a capable team to work for the new government if he was elected.

Although there may have been good reasons to support his policy change, CY’s decision to modify his policies during the election period was not without risk because electoral candidates are often regarded as unreliable if they keep changing their election manifesto. To mitigate the potential negative impact on his positive face, CY adopted a change of footing in his reply to this aggressive question about his policy change after the nomination. First, he used the first-person ‘singular’ pronoun ngo⁵ (‘I’) to take direct responsibility for his policy change when he explained that the changes were prompted by hundreds of letters received from the general public (lines 041–043). Then he skillfully shifted to the first-person ‘plural’ pronoun ngo5dei6 (exclusive ‘we’/’our’) when he described the action of changing his welfare policy after the nomination (lines 044–046). The shift of footing from ‘I’ to ‘we’ helped reduce CY’s involvement in this controversial policy change and portrayed the change as a collective action with his election team. In this way, his use of the exclusive ‘we’ helped to minimize the potential threat to his public image.

Inclusive ‘we’ in non-replies

As pointed out earlier (see Table 4), inclusive ‘we’ was used more often than exclusive ‘we’ in non-replies during the election debates. As noted in Bull (2008), politicians often shifted the topic and went on to say something predetermined on their agenda when they chose not to answer an aggressive question. They often either promoted themselves or attacked their rivals with inclusive ‘we’, conveying the impression that they were representing public opinion in doing so.

In one of the non-replies in the CE Election Debate on 16 March 2012, HT deployed the inclusive ngo⁵dei⁶ ‘we’/’our’ in the expression ngo⁵dei⁶ hoeng¹gong² ge³ hat⁶sam¹ gaa³zik⁶ ‘the core value of our Hong Kong society’ (line 054) to align himself with the general public to attack his rival when he sidestepped an aggressive question about his contradictory statement during the election campaign. HT was reported to have blamed the government’s poor performance for his low rating in a pre-election poll, but in another situation, he indirectly praised himself for his achievements during his past government service. CY questioned why HT had made such contradictory statements on these two different occasions. HT sidestepped the question and went on to attack CY on another issue.

Excerpt 5: ‘The Broadcast License Controversy’

Chief Executive Election Debate 16/3/2012 (Part III, 22:14 – 23:12)

047 CY: 我想問唉，唐英年先生…

ngo⁵ soeng²man⁶ e⁶， tong⁴jing¹nin² sin¹sang¹

‘I want to ask Mr. Henry Tang,’

048 點解佢一方面有政績，

dim²gaai² keoi⁵ jat¹fong¹min⁶ jau⁵ zing³zik¹

‘Why, on the one hand, he (says he) has achievement,’

049 一方面又話佢嘅民望

jat¹fong¹min⁶ jau⁶waa² keoi⁵ ge³ man⁴mong⁶

050 受低民望嘅政府拖累。

sau⁶ dai¹man⁴mong⁶ ge³ zing³fu² to¹leoi⁶

‘but on the other hand he says his poll rating is adversely affected by the poor government performance.’

051 佢呢方面係咪自相矛盾呢噉?

keoi⁵ lei¹ fong¹min² hai⁶mai¹ zi⁶soeng¹maau⁴teon⁵ le¹ gam²?

‘Isn’t he self-contradictory?’

052 HT: 梁振英先生，

loeng⁴zan³jing¹ sin¹saang¹

^‘Mr. Leung Chun-ying,’

053 你都知道言論自由

nei⁵ dou¹ zi¹dou³ jin⁴leon⁴ zi⁶jau⁴

054 → 係我地香港嘅核心價值。

hai⁶ ngo⁵dei⁶ hoeng¹gong² ge³ hat⁶sam¹ gaa³zik⁶

‘you should know that freedom of speech is the core value of our Hong Kong society.’

055 你係不斷咁樣係呃緊市民呀你知唔知？

nei⁵ hai⁶ bat¹dyun³ gam²joeng² hai⁶ aak¹ gan² si⁵man⁴ aa¹ nei⁵ zi¹m⁴zi¹

‘Do you know that you keep deceiving Hong Kong people?’

056 喺商台續牌嘅時候

hai² soeng¹toi⁴ zuk⁶paai⁴ ge³ si⁴hau⁶

‘When (Commercial Radio) applied for renewal of its broadcast license,’

057 你曾經講過，

nei⁵ cang⁴ging¹ gong² gwo³

058 就係你 …

zau⁶hai⁶ nei⁵

‘you had said that you …’ (being interrupted by CY, he stopped speaking)

HT did not explain his contradictory statements at all but directly criticized CY’s credibility on another issue. HT referred to a rumor that CY had threatened not to renew the broadcasting license of the highly popular Commercial Radio in a closed meeting of the Executive Council because its phone-in program always criticized the government. Although his verbal attack on CY was interrupted before he could fully utter the reason, the audience could still recover the missing part because it had been widely reported in the media. HT’s counter-attack was aligned with the general public by bringing up the issue of freedom of speech as a core value of Hong Kong society, and was critical of CY for not supporting this core value, as evidenced by his refusal to renew Commercial Radio’s license. This non-reply strategy via topic shift was deployed by HT in the hope that the audience would forget the embarrassing question asked earlier as he directs their attention to his rival’s controversial behavior.

Conclusion

In election debates, politicians often face adversarial questions from debate hosts, their rivals and the audience, and they need to skillfully deploy various damage control strategies. Evasion is among the most common strategies used in such circumstances. But evasion is not without risk or cost to the politicians. It tends to give the impression that the politicians are irresponsible. To close the social distance created by evasion, politicians often adopt both alignment and disalignment strategies to maintain their positive self-image. Inclusive and exclusive uses of ‘we’ are some of the rhetorical strategies that politicians rely on to establish a sense of solidarity with the general public and to distance themselves from negative publicity.

In this study, we examined how electoral candidates in the 2012 Chief Executive (CE) Election in Hong Kong used inclusive and exclusive ngo⁵dei⁶ (the Cantonese pronoun ‘we’) in different types of evasive replies in two election debates. On average, more than one pronoun ‘we’ (1.3 tokens) was used in each evasive reply. The use of inclusive ‘we’ was much higher than exclusive ‘we’ (60 and 35 tokens, respectively) – almost a 2:1 ratio. The lower frequency of exclusive ‘we’ was to some extent due to contextual factors, in particular the non-partisan background of the candidates. Two out of three candidates – in fact, the leading candidates – did not belong to any political party, and thus had fewer opportunities to use exclusive ‘we’ as a means of ‘taking shelter’ from a larger-than-themselves political entity. Unlike politicians in other countries with a parliamentary election system (e.g. Britain and Taiwan), the CE candidates in Hong Kong could not shift their footing to attribute responsibility to a political party when aggressive questions were raised in relation to controversial policies or issues.³ Our analysis reveals that the CE candidates in Hong Kong tended to use exclusive ‘we’ in evasive replies when referring to constructive measures undertaken by themselves and their election team (e.g. revising their policies after receiving public feedback) or future plans that they had proposed for the post-election period (e.g. sourcing more land to alleviate the housing shortage problem).

On the other hand, when dealing with the conflicting interests between different sectors of the society, these CE candidates rely more on inclusive ‘we’ in their evasive replies to construct themselves as leaders who align themselves with the whole society, and are not partial to particular groups. For example, as seen in Excerpt 2, when asked how to balance profits for the business sector and benefits for the working class, CE candidate CY avoided choosing sides through his use of inclusive ngo⁵dei⁶ ‘we’. Instead he aligned himself with the general public and highlighted the overall interest of the society as a whole. To enhance solidarity with the audience and not to anger any sector in the society appeared to be the best way to handle controversial issues in this type of situation. As seen in Excerpt 4, the CE candidates also sometimes used inclusive ‘we’ to align with the general public to attack their rivals while side-stepping aggressive questions.

In the data from the 2012 election debates, there was also one example where one of the electoral candidates, CY, chose to use exclusive ‘we’ rather than inclusive ‘we’ to explain why he changed one of his policies during the election campaign. Being questioned by his rival about why he made a sudden policy change, CY used exclusive ‘we’ to skillfully shift the topic away from charges of individual culpability (‘Why did you suddenly change your housing supply policy?’) to an implicit promotion of his leadership style which embraces shared responsibility with an election team that responds to feedback from the general public (‘We make changes based on public opinion’). This example shows that even when exclusive ‘we’ was sometimes chosen in the evasive replies by Hong Kong politicians, the exclusive reading could still be cleverly aligned with the audience. Unlike the political parties in Western countries, CE candidates in Hong Kong and their election teams are not beholden to any political party. When these CE candidates shift partial responsibility to their election teams, it is often advantageous for them to align themselves and their election teams with the general public. Thus, we see that the use of exclusive ‘we’ need not always disalign the speaker from the audience.

Overall, our analysis reveals that the CE candidates preferred to use the implication strategy when equivocating in the face of aggressive questions. Compared with the results for British politicians (34%) and Taiwanese politicians (47%) reported in Inigo-Mora (2013) and Chen (2007), respectively, Hong Kong politicians tended to use inclusive ‘we’ with a higher frequency when vying for the highest office (63.2%). One reason why the implication strategy was greatly favored is that this strategy is generally more indirect and more polite compared with other evasive strategies (e.g. challenges, half-replies and non-replies), since the speaker makes an effort to clarify his views, albeit without explicitly stating them. Among the implicit evasive replies, inclusive ‘we’ was found to be used more frequently than exclusive ‘we’ – more specifically, 56 tokens (65.1%) to 30 tokens (34.9%), respectively (see Table 4). Unlike in previous studies which showed that presidential and prime ministerial candidates elsewhere often used exclusive ‘we’ to promote the collective accomplishments of their political parties and themselves, no tokens of exclusive ‘we’ with this function were used in the evasive replies of the two leading candidates in the CE debates in Hong Kong. This difference can be explained in terms of contextual factors, in particular the common scenario in Hong Kong where CE candidates tended not to belong to any political party, which meant that they could not rely on party affiliation to provide them some degree of political shelter when aggressive questions are raised on controversial policies or issues. Their use of exclusive ‘we’ in response to aggressive questioning mainly focused instead on highlighting the collective efforts of their election team to work for the betterment of Hong Kong.

On the other hand, inclusive ‘we’ was often used by the election candidates as a means to express their affiliation with the audience and thus enhance their solidarity with the general public – for example, by aligning themselves with the core values of the society as a whole, a common practice also found in earlier studies related to other cultures (e.g. Britain, the United States, Spain and Taiwan). Interestingly, our analysis also reveals an affiliative function of inclusive ‘we’ that has not been previously reported in earlier studies, namely, that election candidates also sometimes used inclusive ‘we’ to make their verbal attack on a rival appear less personal, since the criticisms being voiced by the speaker could be couched as a rebuke uttered in the public’s interest. This inclusive ‘we’ strategy helps to attenuate potential damage to the speaker’s positive face when criticizing their opponents. In our data, we have an example of ‘we’ being used by a CE candidate to criticize an opponent by aligning himself and his election team with the core values of society. In this way, we see how first-person pronominal ‘we’ can serve double duty for both exclusive and inclusive uses – by simultaneously distinguishing himself from his rival and further consolidating common ground with the general public.

Our study has helped to shed more light on how contextual factors affect politicians’ use of inclusive and exclusive ‘we’ when equivocating in response to aggressive questions. We have seen, for example, how non-affiliation with a political party could contribute to a noticeable reduction in the use of exclusive ‘we’ among the election candidates in the 2012 Hong Kong CE election debates compared with their counterparts in other countries studied thus far. More studies on the effects of contextual factors would be useful, especially from culturally diverse contexts. For example, future studies could examine how clusivity of the first-person plural pronoun ‘we’ is used in the evasive responses of political leaders in societies with a more hierarchical and more complex (and possibly more rigid) social structure. To what extent and in what contexts, for instance, would the political leaders in a status-conscious social system use inclusive and exclusive ‘we’ to establish common ground with their supporters and the general public? Another interesting question, from a diachronic perspective with implications for studies on socio-political change across time, is whether there has been any noticeable change (say over the past 50 years), in the way leaders in a given society construct political identities for themselves and others, and by extension how they engage with the general public, through their deployment of inclusive and exclusive ‘we’. Thus far, we have seen how politicians manage to deploy the same first-person plural form (e.g. English we, Spanish mí, Mandarin wǒmen, Cantonese ngo⁵dei⁶) to express either inclusive or exclusive reference for political gain when equivocating in response to aggressive questions in public discourse; it would be interesting in future studies to examine whether similar or different usage and frequency patterns are found in similar situations involving languages with distinct inclusive and exclusive forms (e.g. Malay/Indonesian kita vs kami, respectively). Greater understanding of how pronominal clusivity contributes to how speakers in different cultures align and disalign themselves with/from others will be valuable in helping to enhance our sensitivity to the many ways in which political leaders often subtly construct their positive image, even in adversarial public discourse.

Footnotes

Acknowledgements

We wish to gratefully acknowledge the funding support from the Hong Kong Polytechnic University (HKPU G-YBGG, 2015–2017) for the research project entitled ‘Addressee-oriented Agreement Marking and Politeness in East Indo-Aryan Languages’, for which this project on clusivity paves the way for cross-linguistic and cross-cultural comparisons on politeness strategies from the perspective of person and social deixis. Earlier versions of this article have received useful feedback from participants from the following conferences: the 2013 Annual Research Forum (ARF) of the Linguistic Society of Hong Kong, the 2nd American Pragmatics Conference (AMPRA 2013), and the 20th International Conference on Yue Dialects (2015). We also wish to thank the editor and the anonymous reviewers for their valuable feedback.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Research grant from the Hong Kong Polytechnic University: ‘Addressee-oriented Agreement Marking and Politeness in East Indo-Aryan Languages’ (HKPU G-YBGG, 2015-17).

Notes

Author biographies

Brian Lap-ming Wai is a Research Associate at the Hong Kong Polytechnic University and a Lecturer at the City University of Hong Kong. His teaching areas cover the topics of semantics, pragmatics, research design and methodology, and English for specific purposes. His research interests include conversational analysis, discourse analysis, political discourse and corpus linguistics.

Foong Ha Yap is an Associate Professor at the Chinese University of Hong Kong, Shenzhen. Her research areas include language typology, language change (grammaticalization studies), language acquisition and language processing, functional cognitive linguistics, sociolinguistics and discourse analysis.

References

Ahn

Yap

(2015) Evidentiality in interaction: A pragmatic analysis of Korean hearsay evidential markers. Studies in Language 39(1): 46–84.

Aikhenvald

(2004) Evidentiality. Oxford: Oxford University Press.

Allen

(2007) Australian political discourse: Pronominal choice in campaign speeches. In: 2006 annual meeting of the Australian Linguistic Society (eds Mushin

Laughren

), Brisbane, QLD, Australia, 7–9 July. Brisbane, QLD, Australia: The University of Queensland, pp. 1–13.

Bavelas

Black

Chovil

et al . (1990) Equivocal Communication. Newbury Park, CA: SAGE.

Beard

(2000) The Language of Politics. London: Routledge.

Bitiniene

(2007) Publicistinis Style. Vilnius: Vilnius Pedagogical University Publishing House.

Blas Arroyo

(1998) Estructuras triadicas en el discurso politico-electoral [Triadic structures in the political-electoral discourse]. Cahiers de l’institut de Linguistique de Louvain 24(1–2): 85–108.

Blas Arroyo

(2000) Mire usted Sr. Gonzalez … Personal deixis in Spanish political-electoral debate. Journal of Pragmatics 32(1): 1–27.

Bull

(1994) On identifying questions, replies, and non-replies in political interviews. Journal of Language and Social Psychology 13(2): 115–131.

10.

Bull

(2003) The Microanalysis of Political Communication. London: Routledge.

11.

Bull

(2008) ‘Slipperiness, evasion, and ambiguity’: Equivocation and facework in noncommittal political discourse. Journal of Language and Social Psychology 27(4): 333–344.

12.

Bull

Fetzer

(2006) Who are we and who are you? The strategic use of forms of address in political interviews. Text & Talk 26(1): 3–37.

13.

Bull

Mayer

(1993) How not to answer questions in political interviews. Political Psychology 14(4): 651–666.

14.

Carranza Márquez

(2012) Pragmatic implications of expanded deictic structures: Deixis as an expression of the concept of face in the Spanish parliament. Philologia Hispalensis 26(1–2): 95–106.

15.

Chen

LCL

(2007) A study of first-person pronouns in Chinese political discourse. PhD Thesis, National Tsing Hua University, Taiwan.

16.

Chilton

Schaffner

(1997) Discourse and politics. In: Van Dijk

(ed.) Discourse as Social Interaction. London: SAGE, pp. 206–230.

17.

Clayman

(2001) Answers and evasions. Language in Society 30(3): 403–442.

18.

Clayman

Heritage

(2002a) Questioning presidents: Journalistic deference and adversarialness in the press conferences of U.S. Presidents Eisenhower and Reagan. Journal of Communication 52(4): 749–775.

19.

Clayman

Heritage

(2002b) The News Interview: Journalists and Public Figures on the Air. Cambridge: Cambridge University Press.

20.

Day

(1991) Grand Inquisitor. London: Pan MacMillan.

21.

De Fina

(1995) Pronominal choice, identity and solidarity in political discourse. Text: Interdisciplinary Journal for the Study of Discourse 15(3): 379–410.

22.

Ekström

Tolson

(2017) Political interviews: Pushing the boundaries of ‘neutralism’. In: Ekström

Firmstone

(eds) The Mediated Politics of Europe: A Comparative Study of Discourse. Cham: Palgrave Macmillan, pp. 123–149.

23.

Fairclough

(1995) Media Discourse. London: Edward Arnold.

24.

Goffman

(1974) Frame Analysis: An Essay on the Organization of Experience. Cambridge, MA: Harvard University Press.

25.

Goffman

(1981) Forms of Talk. Philadelphia, PA: University of Pennsylvania Press.

26.

Greatbatch

(1988) A turn-taking system for British news interviews. Language in Society 17(3): 401–430.

27.

Halliday

MAK

(1994) An Introduction to Functional Grammar. London: Edward Arnold.

28.

Hamilton

Mineo

(1998) A framework for understanding equivocation. Journal of Language and Social Psychology 17(1): 3–35.

29.

Hanlon

(2015) Verbal aggression and neutrality in political interviews. Diffusion: The UCLan Journal of Undergraduate Research 3(2): 1–17. Available at: https://www.bcur.org/journals/index.php/Diffusion/article/view/156

30.

Harris

(1991) Evasive action: How politicians respond to questions in political interviews. In: Scannell

(ed.) Broadcast Talk. London: SAGE, pp. 76–99.

31.

Harris

(2001) Being politically impolite: Extending politeness theory to adversarial political discourse. Discourse & Society 12(4): 451–472.

32.

Helmbrecht

(2002) Grammar and function of we, us and others: Social identities across languages. In: Duszak

(ed.) Us and Others: Social Identities across Languages, Discourses and Cultures. Amsterdam; Philadelphia, PA: John Benjamins, pp. 31–49.

33.

Heritage

(1985) Analysing news interviews: Aspects of the production of talk for an overhearing audience. In: Van Dijk

(ed) Handbook of Discourse Analysis. London: Academic Press, pp. 95–117.

34.

Holmberg

(2015) The Syntax of Yes and No. Oxford: Oxford University Press.

35.

Huang

CTJ

YHA

(2009) The Syntax of Chinese, vol. 8. Cambridge: Cambridge University Press.

36.

Hutchby

(2013) Obama in the no spin zone. In: Ekström

Tolson

(eds) Media Talk and Political Elections in Europe and America. New York: Palgrave Macmillan, pp. 41–62.

37.

Inigo-Mora

(2013) Pronominal choice as an interpersonal strategy. American International Journal of Contemporary Research 3(3): 22–37.

38.

Jaworski

Galasiński

(2000) Vocative address forms and ideological legitimization in political debates. Discourse Studies 2(1): 35–53.

39.

Kowal

O’Connell

Forbush

et al . (1997) Interplay of literacy and orality in inaugural rhetoric. Journal of Psycholinguistic Research 26(1): 1–31.

40.

Leech

Svartvik

(1994) A Communicative Grammar of English. London: Longman.

41.

Lippmann

(1920) Liberty and the News. New York: Harcourt Brace Jovanovich.

42.

Lippmann

(1922) Public Opinion. New York: Macmillan.

43.

Lippmann

(1925) The Phantom Public. New York: Harcourt Brace Jovanovich.

44.

McGraw

Hubbard

(1996) Some of the people some of the time: Individual differences in acceptance of political accounts. In: Mutz

Sniderman

Brody

(eds) Political Persuasion and Attitude Change. Ann Arbor, MI: University of Michigan Press, pp. 145–170.

45.

Mehdipour

Nabifar

(2011) Critical analysis of evasion techniques in American political news interviews. Journal of Academic and Applied Studies 1(5): 56–70.

46.

Miller

Krosnick

(1996) News media impact on the ingredients of presidential evaluations: A program of research on the priming hypothesis. In: Mutz

Sniderman

Brody

(eds) Political Persuasion and Attitude Change. Ann Arbor, MI: University of Michigan Press, pp. 79–100.

47.

Mühlhäusler

Harré

(1990) Pronouns and People: The Linguistic Construction of Social and Personal Identity. Oxford: Basil Blackwell.

48.

O’Connell

Kowal

(2002) Political Eloquence. In: Ottati

Tindale

Edwards

et al (eds) The Social Psychology of Politics. Chicago, IL: Springer, pp. 89–103.

49.

O’Keeffe

Clancy

Adolphs

(2011) Introducing Pragmatics in Use. London: Routledge.

50.

Pishwa

(2015) Information source as persuasive power in political interviews: The case of Obama. In: Schulze

Pishwa

(eds) The Exercise of Power in Communication: Devices, Reception and Reaction. Hamsphire: Palgrave Macmillan, pp. 130–160.

51.

Proctor

Lily

(2011) The 1st person plural in political discourse – American politicians in interviews and in a debate. Journal of Pragmatics 43(13): 3251–3266.

52.

Putri

Kurniawan

(2015) Person deixis in USA presidential campaign speeches. English Review: Journal of English Education 3(2): 190–200.

53.

Raymond

(2000) The structure of responding: Type-conforming and nonconforming responses to yes/no type interrogatives. PhD Thesis, University of California, Los Angeles, CA.

54.

Rendle-Short

(2007) Neutralism and adversarial challenges in the political news interview. Discourse & Communication 1(4): 387–406.

55.

Richards

Schmidt

(2002) Dictionary of Language Teaching and Applied Linguistics, 3rd edn. London: Pearson Education.

56.

Schegloff

(1968) Sequencing in conversational openings. American Anthropologist 70(6): 1075–1095.

57.

Silvestre-López

(2004) John Kerry’s political rhetoric: An account of the main rhetorical features of his oral delivery. Fòrum de Recerca 10: 1–20.

58.

Strazny

(2005) Encyclopedia of Linguistics, vol. 1. New York: Fitzroy Dearborn.

59.

Suleiman

(2000) Pronoun use in television interviews: Social interaction and the Middle East peace process. PhD Thesis, Georgetown University, Washington, DC.

60.

Sun

(2010) Adversarial questioning and answering strategies in Chinese government press conferences. Taiwan Journal of Linguistics 8(2): 131–162.

61.

Tabakowska

(2002) The regime of the other: ‘Us’ and ‘them’ in translation. In: Duszak

(ed.) Us and Others: Social Identities across Languages, Discourses and Cultures. Amsterdam: John Benjamins Publishing, pp. 449–461.

62.

Tannen

(1993) The relativity of linguistic strategies: Rethinking power and solidarity in gender and dominance. In: Tannen

(ed.) Gender and Conversational Interaction. Oxford: Oxford University Press, pp. 165–188.

63.

Thornborrow

Haarman

(2017) Genre and cultural style in TV news coverage of the European Elections 2014. In: Ekström

Firmstone

(eds) The Mediated Politics of Europe: A Comparative Study of Discourse. Cham: Palgrave Macmillan, pp. 93–122.

64.

Van Dijk

(2002) Political discourse and political cognition. In: Chilton

Shaffner

(eds) Politics as Text and Talk: Analytic Approaches to Political Discourse. Amsterdam: John Benjamins Publishing, pp. 203–237.

65.

Vukovic

(2014) Strong epistemic modality in parliamentary discourse. Open Linguistics 1(1): 37–52.

66.

Wilson

(1990) Politically Speaking: The Pragmatic Analysis of Political Language. Oxford: Basil Blackwell.

67.

Woods

(2006) Describing Discourse. Oxford: Oxford University Press.

68.

Yip

(2003) Adversarial questioning in news interviews: How Hong Kong journalists exert pressure on public figures on the air and the deployment of politeness (face) strategies. MA Thesis, The Chinese University of Hong Kong, Shenzhen, China.

69.

Yule

(1996) Pragmatics. Oxford: Oxford University Press.