The Strategic Use of Reasonable
Accusations of Inconsistency in
Election Debates
Helge Linnert Graduate School of Humanities Rhetoric, Argumentation Theory & Philosophy supervisor Dr Corina Andone
second reader Dr Francisca Snoeck Henkemans
1 The Context of Political Election Debates 4 1.1 Political Election Debates are ”Counterfeit Debates” . . . 5 1.2 Election Debates: A Continuum of Genres? . . . 6 2 Evaluating Accusations of Inconsistency 9 2.1 Reasonable Accusations of Inconsistency . . . 10 2.2 Fallacious Accusations of Inconsistency: The tu quoque . . . 15 3 Accusations of Inconsistency and their Strategic Use in Election Debates 17 3.1 Persuasion by Unexpressed Premises? . . . 19 3.2 Effectiveness through Ethos . . . 22
4 Conclusion 24
4.1 Findings of the Study . . . 24 4.2 Discussion of the Findings . . . 26
References 29
Introduction
This study analyzes a specific kind of personal attack in election debates and aims to discover their strategic use as a means to further the discussion in the arguer’s favour. The personal attack under discussion is a so-called accusation of inconsistency and is conceptualized as a legitimate argument, i.e. as a productive contribution to the dispute at hand. Election debates are characterized as revolving around the claim concerning which candidate is more suitable for the office in question. To gain more insights into the strategic trajectory available to the political candidate initiating the charge, a specific case of an accusation of inconsistency from an election debate will be scrutinzed.
A typical example of an accusation of inconsistency is the ’smoker’s example’, where a smoking parent admonishes her child not to smoke. The child responds: ’What about you. You smoke. So much for your argument against smoking!’ (Walton, 1985, p. 71). Supposedly, the child is accusing his parent of taking inconsistent positions on smoking since she is advocating not to smoke but is a smoker herself.
Traditionally, accusations of inconsistency are associated with illegitimate arguments that are employed as a kind of diversionary tactic to denounce an opponent (Copi et al., 2011; Whately, 1827). In this case, scholars talk about the tu quoque ad hominem fallacy. When scholars talk about the strategies of arguing with fallacies, we might be reminded of Schopenhauer’s stratagems (’Kunstgriffe’) to which the ad hominem can be counted (Schopenhauer, 1932, p. 415). However, winning an argument by using fallacies is not the issue here. In fact, more recently, accusations of inconsistency have been found to be legitimate arguments if they can actually contribute to furthering the discussion (Johnstone, 1952; Hitchcock, 2017, amongst others). It might therefore be enlightening to discover which strategies arguers choose to do so.
”The argument against the person is, in fact, a very common form of argument, for example, in political debates. It could even be said to be the most powerful kind of argument in politics”, Walton (1989, p. 135) says. Accusations of inconsistency can be characterized as arguments ’against the person’. So when examining accusations of inconsistency, we are well-advised to turn to political debates. A pertinent example can be found in an election debate from Canadian campaign discourse.
On August 6th in 2015, against the backdrop of the Canadian federal election, the leaders of
four of the five leading parties held an anglophone debate which was lead by a moderator and broadcasted on national television. Participating in the debate were Conservative Party leader and then-incumbent Prime Minister Steven Harper, Liberal Party leader Justin Trudeau, leader of the New Democratic Party (NDP) Thomas Mulcair, and leader of the Green Party
Elizabeth May. The debate covered four topics of which one was the environment. In the environment segment Mulcair made a remark to which Trudeau was asked to respond. Trudeau did so by criticizing Mulcair for having taken inconsistent positions towards pipelines:
Justin Trudeau: ”Mr. Mulcair has been somewhat inconsistent on pipelines. In English he’ll say that he supports the Energy East pipeline; in French he said that it’s out of the question. And that kind of inconstancy, quite frankly, isn’t the kind of leadership we need for Canada. You can’t say one thing in English and its opposite in French. The fact is we need to restore public trust”.
(Macleans.ca, 2015, l. 39-43, see Appendix on page 34 for a transcript of the segment of this debate)
By accusing Mulcair of being inconsistent, Trudeau attempts to present his opponent in an unfavourable light. Specifically, he stresses that what Mulcair is doing is not what Canada ”needs”, thereby disqualifying him for the office of Prime Minister.
Trudeau’s criticism is a reaction to Mulcair’s prior remark directed at him personally and additionally the criticism is a remark about Mulcair’s lack of skills. In that vein, Trudeau’s criticism directly resembles the characterization of personal attacks postulated by van Eemeren. Specifically, van Eemeren (2015, p. 800) characterizes personal attacks as taking place in a discussion as follows. They are (i) directed at the opposing party and (ii) are about that party’s ”bad character, insufficient knowledge, low abilities, lack of skills et cetera, which disqualify him or her as a participant in the discussion”. Personal attacks are then, in his view, discursive elements that can surface in a discussion. This enables their ’analysis and evaluation’ in terms of reasonable or fallacious contributions to the dispute. In the pragma-dialectical framework of argumentation, for example, personal attacks are said to be fallacious if they are used to prevent a participant in the discussion from putting forward a claim and thereby violate one of the rules of a critical discussion (van Eemeren et al., 2014, p. 545). In that vein, accusations of inconsistency are considered reasonable if they contribute to the resolution of the difference of opinion at issue.
Walton (1998a, p. 2) concurs that personal attacks are directed at an opponent’s supposedly bad character. Moreover, he contends that personal attacks can be evaluated as either fallacious or reasonable discussion moves. In Walton’s (1998a, p.267, p. 293) view the evaluation takes place not with respect to a set of rules but with respect to the dialogue type that the participants are involved in. However, he asserts that most fallacious cases are fallacious due to the prevention of critical questions being asked. The accounts of both van Eemeren and Walton for fallacious personal attacks resemble the pragma-dialectical analysis; both approaches envision a preclusion of one of the participants from the discussion as the
decisive factor for the personal attack being a fallacy. Despite their crossover in opinion, their continued discussions on the reasonableness of personal attacks differ dramatically. When aiming to examine the strategies employed by arguers, we are essentially dealing with the question, how does reasonable argumentation achieve ’effectiveness’. With our understanding of reasonable argumentation we might be able to identify peculiar aspects that need to be taken into account when examining what routes are taken to achieve effectiveness. Aristotle (1984, Rhetoric, I.2, 1356a13) makes a strong point that character, or ethos, might be ”the most effective means of persuasion”. But how can an arguer make use of character appeals and achieve ’effective argumentation’? This is a question worth investigating. The aforementioned debate example between Trudeau and Mulcair is claimed to represent a typical case of a personal attack. Under the assumption that Trudeau’s contribution can be conceptualized as an argument, the question also arises what standpoint he is trying to substantiate with it. As Martel (1983, p. 119) observes, political debate rarely focuses on propositions, i.e. ”specific, debatable arguments”. In debate theory, propositions are the motions being debated which can be refuted or corroborated by debate participants.1 The example appears to be such a case as the moderator did not put forward a proposition, or a specific motion, to which arguers can define their positions in terms of affirmative or dissenting. Although the topic concerned the environment in general, no polarizing statement was formulated to require the participants to clarify their position. This is odd and requires further investigation. While we can assume that an argument was put forward, can we account for such controversies that lack an explicitly formulated point of discussion as an instance of argumentation? Under the assumption that it was a reasonable argument, what would make it an effective argument? And, ultimately, which strategies can be identified to achieve effectiveness?
In order to approach these questions, this study will first lay out some insights on the context in which the accusation of inconsistency is put forward. In particular, the first section deals with the controversial classification of election debates as a type of debate. Debate theorists, as much as they might differ about election debates, provide valuable characterizations that are helpful for the examination at issue. The second section will provide the theoretical background for understanding accusations of inconsistency as legitimate, or reasonable arguments. Moreover, a distinction is made from illegitimate, or fallacious arguments, that do not contribute to furthering the discussion and are therefore not suitable for the examination at stake. In the third section, insights from the preceding sections are gathered and an attempt
1In the words of van Eemeren (2015, p. 253), participants take a positive or negative standpoint vis-`a-vis
the proposition in the debate. Confusingly, he states that arguers try to justify the proposition (and not their positive or negative standpoint) expressed in the standpoint by putting forward argumentation. It seems that whatever the proposition in a debate is, it wil resurface in the particular standpoints taken by debate participants.
is made to achieve a greater understanding of the strategic use of reasonable accusations of inconsistency in election debates. Specifically, insights from argumentation scholars about unexpressed premises and character appeals (ethos) are taken into account as a means to achieve effective argumentation. The concluding section deals with two aspects. First of all, it provides a methodological recap of this study and lays out the findings. Secondly, a discussion is lead in which the findings are viewed in a critical light and prospects for further research are given.
1
The Context of Political Election Debates
Political election debates are subject to pervasive study in disciplines like critical discourse analysis, political discourse analysis, conversation analysis and argumentation theory, amongst others. Researchers may be interested in politolinguistic, discursive, social or argumentative insights when looking at political candidates in an election debate confronted with a myriad of challenges in, for instance, televised campaign debates. It might be helpful to make a distinction between political debates and political election debates (also called campaign debates). Political debates refer to controversies in the public sphere with varying contexts and denote debates with a general political subject matter, including political election debates, amongst others (Benoit, 2014, Preface, p. vii ). On the other hand, political election debates, or campaign debates, for that matter, are a specific type of political debate. Nowadays, they are mostly a televised encounter and are part of an election campaign where the candidates run for a particular office (Martel, 1983, pp. 1-2). An additional term is broadcast debates (or TV-debates) which applies to any one type transmitted on television (Hess-L¨uttich, 2007,
p. 1360). Political election debates are the central debate format in this examination and might henceforth be referred to as election debates or debates.2
The study of election debates, as Martel (1983, p. 1) writes, has gained ”historical significance because of the Lincoln-Douglas debates of 1858”3 and ”contemporary significance since the
1960 televised presidential debates between John F. Kennedy and Richard M. Nixon”. Political election debates nowadays surface as media events, either as presidential, non-presidential or leaders’ debates (as in Canada, Germany or the Netherlands, for example) (Benoit, 2014, p. 1). Presidential debates, for one, are regarded as the ”paramount example of applied debate” in traditional debate theory (Freeley & Steinberg, 2009, p. 368).4
2The difference between these different terms is that ’political debate’ is the generic term and ’political
election debate’ or ’broadcast/ TV-debate’ are specific terms.
3The Lincoln-Douglas debates have been studied from a rhetorical perspective, for example, by Zarefsky
(1986).
4Presidential debates are considered to be a distinct debate format (Benoit, 2014; Kraus, 2000), but for
But what is the purpose of political debates? Are they aimed at the candidates’ rational deliberation of particular policies? Are they maybe rather of an informative nature for the benefit of the voters’ decision-making process in the election? Or do they serve another purpose altogether? In fact, Myers (2008, p. 122) contemplated these questions and found that ”often the voters are assumed to be an isolated individual seeking information for a rational choice between two candidates, eager for clarification through argument [. . . ]. But voters may be looking for impressions rather than information [. . . ]”. Evidently, in order to obtain a more clear-cut understanding of election debates, more insights are necessary.
1.1
Political Election Debates are ”Counterfeit Debates”
The most notable sceptics of political election debates are scholars of traditional debate theory, distancing election debates from the traditional format for being ”counterfeit debates” (Auer, 1962; Freeley & Steinberg, 2009; Kraus, 2000). Prevalent criticism of election debates
concerns two aspects, as Martel (1983, p. 2) writes. I will refer to them as the constitutive and the contentual criticism, respectively.
The constitutive criticism is that, predominantly, (presidential) debates are not regarded as true debates due to their purported ”lack of adherence to [the] classical debate format” (Auer, 1962, p. 31). Specifically, Auer (1962, pp. 147-148) defines a debate as ”(1) a confrontation, (2) in equal and adequate time, (3) of matched contestants, (4) on a stated proposition, (5) to gain an audience decision”. Though in political debates participants are somewhat matched and face a confrontation in equally distributed time, they are not thought do so adequately in order to ”gain (or, perhaps, contribute to) an audience decision”, as Martel (1983, p. 2) points out. Most importantly though, there is hardly ever a stated proposition or contentious point in election debates. In other words, political debates are thought not to constitute the classical format. As will become clear later, the term ’proposition’ is more commonly understood as a concept from logic. Therefore, when referring to debate propositions, the phrase ’contentious point’ will henceforth be used.
The contentual criticism is that they are said to ”emphasize images over issues or substance” (Martel, 1983, p. 2). Kraus (2000, p. 152), along with Auer (1962), holds the opinion that (televised) ”presidential debates are counterfeit” in view of the ”essentially genuine” Lincoln-Douglas encounters. In fact, he stipulates that the format of (presidential) debates should be changed to resemble those in the Lincoln-Douglas debates so that ”voter-viewers will be impressed less by candidates’ personalities and more by their discussion of issues” (Kraus, 2000, p. 153). Not only does it become clear from this criticism that the Lincoln-Douglas encounters are idealized as the prototype of election debates in traditional debate theory. It also stands out that Kraus contends that in the current format voters are amenable by the
candidates’ personalities. Moreover, Zarefsky (1992, p. 412), for example, criticized election debates for their unfulfilled potential of identifying and deliberating issues. In other words, critics contend that election debates should rather orbit around specific content than around ”putting up a good front- in making an overall positive impression” (Martel, 1983, p. 2).
Examining various forms of political campaign messages, Benoit (2014, p. 34) concedes that political election debates might be more appropriately described as ”joint press conferences” (after Auer (1962); Jamieson & Birdsell (1988)) in view of the panel organisation of journalists
collectively asking the candidates questions. Thereby he dismisses the idea that campaign debates might be a form of debate entirely. This criticism is somewhat at odds with the categorization outlined above. Though Benoit relinquishes the idea that election debates are real debates, his criticism is notably based on the organisational discrepancy with classical debates.
It stands out that, constitutively or contentually, the criticism of political election debates concerns rather formal matters. Ultimately, debate theorists may want to guarantee that their typologies of debates are universally applicable which is why such criticism is definitely necessary to warrant the scope of their subject of study. However, researchers from disciplines such as argumentation theory are dealing with political election debates as well and their main objective is not the rehabilitation of election debates as instances of classical debate. From a descriptive perspective, the criticism outlined above might even serve the argumentative examination at issue.
1.2
Election Debates: A Continuum of Genres?
The subsumption of political election debates in a particular genre has somewhat given rise to controversy in scholarly research. An elegant way of conceptualizing them is to allow hybrid genres that integrate aspects of broadcast genres from both political interviews and political speeches (Halmari, 2008, p. 258, amongst others), situating election debates on a ”continuum” between both (Myers, 2008, p. 140). This way the oratorical, or rhetorical, characteristics of political speech and the conversational, or dialogical, properties of interviews can both be accounted for (Boyd, 2013, p. 299).
This is reminiscent of the approach taken by Walton (1998b, p. 223), who classifies election debates in light of several dialogue types.5 In the framework presented by van Eemeren (2010, p. 144) election debates are referred to as a ”communicative activity type”. Such activity types are thought to implement so-called ”genres of communicative activity”, such
5Question Time, a kind of political debate, is classified by Walton as a mixture of four types of dialogue:
information-seeking dialogue, partly eristic dialogue, partly negotiation and partly persuasion dialogue (van Eemeren, 2010, p. 134).
as deliberation as the genre implemented by the communicative activity type of election debates.6 Quite like Walton’s dialogue types, communicative activity types are granted to
implement several genres. Furthermore, even communicative activity types can overlap when describing a particular communicative practice, like the 1960 Nixon-Kenedy debate (van Eemeren, 2010, p. 144, footnote 34). Whatever classification election debates satisfy, there does not seem to be one distinct genre under which election debates can be analyzed. Instead a multitude of genres, situated on a ”continuum”, seems more useful. Consequently, the specific activity type, i.e. the election debate, needs to be examined with due diligence. Candidates who participate in political election debates have specific motives for their participation. As Breglio (1987) explains, candidates want to show why they are the more suitable candidate for the office in question and, consequently, they defend the claim that they should be elected. At the same time, they also criticize their opponent and provide reasons why voters should not vote for him (Kraus, 2000, p. 33). From an analytical standpoint, the interaction taking place between the candidates is essentially different from the one (implicitly) taking place between a candidate and the audience. Intriguingly, the candidates who debate with each other do not aim to settle their dispute.7 Rather than
persuading one another, the candidates want to persuade the audience (Lehnert & Ringle, 1982, p. 275). However, a settlement of a discussion with the audience does not take place either and, realistically, cannot even take place since the audience is not in direct contact with the contenders. Although a winner is oftentimes medially determined, this is by no means a reliable indication of who was the more persuasive candidate.
In terms of claims adhering to the debate, it could be said that election debates always orbit around one and the same implicit, yet dominant, issue, namely that ”candidate A would fit the role requirements or criteria perceived by the audience for the office being sought more impressively than candidate B, his opponent” (Martel, 1983, p. 2). However, a distinction should be made between two prevalent and interrelated claims, i.e. (i) voters should vote for candidate A and (ii) voters should not vote for candidate B, A’s opponent. They are prevalent, as just illustrated, because the candidates’ primary goal is to be elected. They are also interrelated because supporting the claim why the opponent should not be elected in turn supports the candidate’s own cause, namely that he is consequently the more suitable
6The genre of deliberation is formally associated with Auer’s (1962, p. 146) definition of debate but is also
thought to extend to television debates which oftentimes lack an explicit ’contentious point’ and an audience that makes a decision (van Eemeren, 2010, p. 142). In a later publication, van Eemeren et al. (2014, p. 558) attenuate their judgment of deliberation and leave it at stressing the importance of deliberation also referring to informal communication.
7Dascal (2008, p. 43) suggests a dichotomy of (i) discussions and (ii) disputes to differentiate between (i)
debates that yield an opinion change, that are about contents and are logical/ rational and (ii) debates that do not yield opinion change, that are about attitudes and are rhetorical/ irrational, amongst others. From such a perspective, election debates would probably qualify as (ii) disputes, even though ’rhetorical’ and ’logical are not per se mutually exclusive.
candidate. Ultimately, by supporting their claim, candidates do not so much try to persuade their immediate opponent, but rather the electorate (Kraus, 2012, p. 102).8
The main claims to be defended in an election debate thus concern the suitability of the candidates. The candidates’ suitability, or ”preferability”, as Benoit (2014, p. 13) calls it, can be demonstrated via two topics, policy and character. Under the assumption that candidates try to persuade voters of their preferability over their opponent, Benoit stresses that candidates do so by highlighting what they do (policy) and who they are (character) to gain the voters’ adherence (Benoit, 2014, p. 19). However, even statements about policy can reflect a candidate’s character. Consider the following example, provided by Martel (1983, p. 3), who points out that if a candidate has a rigorous position against a trade agreement between the United States and the Soviet Union, ”he may feel that the merits of the proposition warrant the stance, but he may also regard his position as a significant vehicle for projecting strength”. This is what Benoit (2014, p. 19) terms ”influence perceptions”.9 How policy statements support the implicit claims, however, can hardly be analyzed in a suitable way if they are regarded as being subject to perception because such an account is always liable to be subjective.
Character statements, on the other hand are made explicitly. And character matters. In fact, Benoit (2014, p. 21) stresses that this is the case because candidates must be trusted if voters want to believe that candidates stick to their campaign promises. Voters also need to know that candidates can enforce the pertinent policies in unexpected scenarios. Hence, trustworthiness is a character trait that candidates gladly ascribe to themselves to entertain their ethos of being a sincere and credible person (D’Errico et al., 2013, p. 181). Moreover, this is closely related to the candidates’ intention to always uphold their ”public self-image” or face (Brown & Levinson, 1987, p. 61). To achieve that end, the participants of election debates oscillate between positive-face and negative-face strategies (Chilton, 2004, p. 40). In other words, ”the candidates draw on rhetorical ’appeals’ for posivite self-presentation and negative other-presentation” (Boyd, 2013, p. 300, see also Myers 2008, Reisigl 2006 and van Dijk 2001).
Positive-face strategies concern acclaims of the candidate and negative-face strategies concern attacks on the opponent (Benoit, 2014, p. 13). Acclaims are considered to emphasize a candidate’s qualities and thereby can provide support for the claim that he or she is the preferable candidate. Attacks, on the other hand, are more eclectic in that they can be used both to denounce an opponent, thereby supporting the claim that the opponent in question
8The electorate, i.e. the addressee, is an additional entitiy in the study of election debates. Critics discuss
the informative and persuasive nature of debates and in turn raise the question whether they are an expedient format for helping the electorate make rational decisions (Kelley, 1960; Myers, 2008).
9Benoit (2014, p. 19) illustrates influence perceptions with an example from Bill Clinton’s 1996 Acceptance
is an unsuitable candidate. Or they can be used to increase the initiator’s preferability over his contender, thereby suppporting the claim that the initiator is the more suitable candidate (Benoit, 2014, p. 13).
One specific type of attack are accusations of inconsistency. Accusations of inconsistency are statements made by an arguer, a political candidate, for example, to point out a discrepancy between what his opponent has practiced and what he has preached (Walton, 2007, p. 165). Specifically, accusations of inconsistency can be understood as attacks on character, or personal attacks. Given that attacks can serve as support for claims in election debates, as explained above, they are conceptualized as arguments which can be analyzed and evaluated. Hence, understanding whether accusations of inconsistency are a reasonable contribution to the debate and how such arguments can be used to support a claim, in election debates in particular, is an endeavour worth investigating.
2
Evaluating Accusations of Inconsistency
This section deals with different approaches to evaluating accusations of inconsistency and presents them in a comparative fashion. Traditionally, accusations of inconsistency are associated with illegitimate ad hominem arguments. Newer insights have given rise to a reorientation towards such arguments which allows for legitimate accusations of inconsistency. Contrasting those approaches may highlight valuable insights about the conditions for evaluating accusations of inconsistency. Some scholars consider accusations of inconsistency to be illegitimate arguments, such as Copi et al. (2011). Others regard them to be legitimate depending on the context in which those arguments are used, like van Eemeren (2015) or Walton (2007). As a consequence, when scholars deem an argument illegitimate, they do not ascribe that argument any contributive value that would further the dispute at issue. Illegitimate accusations of inconsistency are considered not to contribute to a discussion, for example, if the argument amounts to a personal attack aimed primarily at discrediting the opponent.
Accusations of inconsistency can be used by arguers to undermine the credibility of their opponents (Walton, 2007, pp. 165-169).10 In election debates, when candidates want to
support the claim that their opponent lacks credibility, they can point out an inconsistency in two ways. Inconsistencies can be between (i) the opponent’s statements and actions, or (ii)
10According to van Laar (2008, pp. 171-172), there are three reasons why arguers may want to have
consistent positions, and why an antagonist may want to point out inconsistencies: 1. ”to be perceived as a credible arguer in order to persuade the antagonist of a proposition on the basis of his say-so”; 2. the arguer may want to remain consistent also as an antagonist; 3. ”the arguer may want to remain consistent in order to keep up the image of a sincere and capable arguer”.
between the opponent’s earlier and later claims.11 Hence, candidates are using accusations of inconsistency as arguments in support of a claim concerning their opponent’s credibility. If those arguments are further supported by additional evidence, then an accusation of inconsistency can be the substandpoint (van Eemeren, 2015, p. 430).
Arguments can be deemed either reasonable or fallacious. The evaluation of arguments depends on whatever a particular approach to argumentation stipulates as being ’appropriate’, or the ’standard of reasonableness’ of argumentation (van Eemeren et al., 2014, pp. 6-10). In any given approach, a reasonable argument thus corresponds to the standard of reasonableness that is envisioned. Accordingly, an argument that does not correspond to the standard of reasonableness of a particular approach is deemed unreasonable and hence fallacious. When it comes to evaluating accusations of inconsistency, venerable approaches to argumentation come to different results. This seems intriguing. Let us take a closer look at some of these approaches to find out how they come to diverging results.
The term ’accusation of inconsistency’ (van Eemeren, 2010, p. 248), or ’allegation of inconsistency’ (Walton, 2007, p. 165), has not been adopted in studies in this area. What is more, arguments fallaciously making use of inconsistencies are sometimes referred to as circumstantial ad hominem (Walton, 2007; Zarefsky, 2014) and sometimes as tu quoque ad hominem (van Eemeren, 2015).12 Copi et al. 2011, for example, stick to the term tu quoque
entirely and use it exclusively for describing fallacies. In this study, unreasonable accusations of inconsistency will be referred to as the tu quoque argumentum ad hominem fallacy, or simply tu quoque, in line with van Eemeren et al. (2014, p. 165).
2.1
Reasonable Accusations of Inconsistency
According to Walton (2007, p. 163), the first to question the mere fallaciousness of ad hominem arguments was Johnstone (1952). From then on, argumentation scholars started to contemplate the necessary conditions for the counterpart of ad hominem arguments to be considered reasonable. As mentioned before, the reasonable counterpart of tu quoque, in particular, are accusations of inconsistency. Venerable scholars of rhetoric and argumentation have investigated when accusations of inconsistency can be reasonable arguments. Amongst the scholars who study reasonable instantiations of ad hominem are van Eemeren (2010); Hitchcock (2017); Perelman & Olbrechts-Tyteca (1958); Walton (2007) and Zarefsky (2014).13 11See Andone (2013, p. 105) for the distinction between inconsistencies of type (i) and (ii). An illustration
of types of inconsistencies can be found in van Laar (2008, pp. 169-170).
12The term ad hominem is generally associated with John Locke (1959) although it can even be traced
back to Aristotle and Galileo (Finocchiaro, 1973; Hitchcock, 2017, p. 410). See van Eemeren & Grootendorst (2015) for an overview of the history of ad hominem arguments.
13Even Johnson & Blair (1977), in their Logical self-defense, concede that attacks on a person might be
One approach to reasonable accusations of inconsistency, which will be illustrated briefly, is the one provided by Walton (1998a,b, 2000, 2007). From the Waltonian perspective, allegations of inconsistency can be reasonable in specific ’dialogue types’, for instance in an eristic dialogue, or ’quarrel’ (Walton, 1998a, p. 183). Walton (1998a, p. 182) states that ”if one participant is quarreling or shows evidence of clearly contravening earlier commitments he had advocated, it could be appropriate for the other party to reply, ’Be serious!’, or ’You are not really honestly and sincerely taking part in the [. . . ] discussion here!’”. In such a case the critic is essentially calling out his opponent’s inconsistency with commitments the opponent made earlier. It could be stated that the concept of dialogue types allows for reasonable accusations of inconsistency because they are conceived as relevant contributions to the argumentation at issue. However, their relevance is ’dialogically’ dependent, i.e. it fluctuates per dialogue type.
Walton’s dialogue types turn out to be rather unwieldy when it comes to argumentation in political discourse. Walton himself attributes no less than five dialogue types to political debate, for example.14 In what framework of dialogue should accusations of inconsistency in campaign debates ideally be evaluated? Walton & Krabbe (1995) propose a solution: They envision dialectical shifts from one type of dialogue to another. Dialectical shifts allow for dialogues to change throughout the argumentative encounter. When this happens against an arguer’s prior commitment to stick to one specific dialogue type, then the shift is illicit, i.e. illegitimate (Walton, 1998a, p. 130). In the eristic dialogue, which is amongst others attributed to election debates, such illicit shifts can occur. However, Walton does not specify how illicit shifts can be identified other than by looking at what commitments speakers have explicitly made. Unfortunately, candidates in election debates rarely specify towards what type of dialogue they are going to shift. Therefore, illicit shifts can hardly be accounted for. In contrast to Walton’s ’dialogical’ approach to reasonable ad hominem arguments, Perelman & Olbrechts-Tyteca (1958) offer an audience-related perspective. They do not, however, mention accusations of inconsistency (let alone their fallacious counterpart) when talking about ad hominem arguments. Nonetheless, it is worth taking a closer look at how they motivate their judgment of ad hominem as reasonable arguments.
Perelman & Olbrechts-Tyteca (1958, pp. 110-111) claim that ad hominem arguments are ”valid only for particular groups ” and that ”[i]t would be possible to distinguish as many different types of arguments ad hominem as there are audiences to hear them”.15
they do not specify what type of argument that would amount to (cf. Hitchcock, 2017, p. 409).
14The types of dialogue are: critical discussion, negotiation, deliberation, information-seeking dialogue, and
eristic (quarrelsome) dialogue (Walton, 1998a, pp. 196-197).
15Perelman & Olbrechts-Tyteca (1958, p. 110) refer to ad hominem arguments that are accepted by
everyone as arguments ad humanitatem. Correspondingly, they call the entirety of an audience that accepts an ad hominem argument the ’universal audience’.
In particular, this means that an accusation of inconsistency (as a type of reasonable ad hominem argument) might be a valid argument for one particular person only. In their view, ad hominem arguments are equivalent to ex concessis arguments and depend on what the audience is ready to concede, i.e. to accept (Perelman & Olbrechts-Tyteca, 1958, p. 110). From this perspective, practically any acceptable argument could be considered ad hominem. It is then no surprise that Perelman & Olbrechts-Tyteca (1958, p. 111) underline that ”there is nothing improper in this mode of procedure”. Yet they constrain reasonable ad hominem arguments to those that are supported by accepted claims. ”[I]t’s sphere is not truth, but adherence [. . . ] When truth is in question, and not adherence, argumentum ad hominem should be rejected” (Perelman & Olbrechts-Tyteca, 1958, p. 114).
From the rhetorical perspective taken by Perelman & Olbrechts-Tyteca (1958), accusations of inconsistency can only be acceptable arguments if they draw on concessions made by the audience. However, against the backdrop of campaign debates, we can imagine that an accusation of inconsistency is directed at another candidate, rather than the electorate (i.e. the audience). It needs to be pointed out that by ’the audience’ rhetoricians mean at whom an argument is directed. The rhetorical perspective on acceptable arguments is somewhat relativistic. One and the same argument can be acceptable relative to one particular audience, whilst it can be unacceptable to another. As a result, from a theoretical perspective, the standard of reasonableness employed by Perelman & Olbrechts-Tyteca (1958) is rather unfathomable. For the analyst there is no reliable way to identify what is a reasonable and what is a fallacious argument. This is because it would be necessary to make assumptions about what an audience accepts, and assumptions are volatile and subjective. Moreover, the question arises at whom an accusation of inconsistency is directed. On the one hand, the accusation is quite clearly directed at a candidate’s opponent. On the other hand, however, candidates want to direct themselves at the electorate to defend their claim that they are the more suitable candidate (and should therefore be elected). Focus should be laid on the discussion taking place between the two candidates. It turns out that the rhetorical approach does not offer the necessary notions to differentiate between various discussions between multiple arguers.
The pragma-dialectical program, on the other hand, makes explicit comments on accusations of inconsistency and provides clear guidelines for determining whether they are employed as reasonable arguments (van Eemeren, 2010). It might therefore be enlightening to have a look at how they motivate their assessment of reasonable accusations of inconsistency. Argumentation, from the pragma-dialectical perspective, is the attempt to resolve a dispute between a protagonist and an antagonist based on mutually accepted concessions or starting points. This is usually conceptualized as an idealized model that serves as a normative framework for argumentation that aims to resolve a difference of opinion (van Eemeren,
2010, p. 28). Here, pragma-dialecticians refer to the effort to maintain reasonableness in argumentative discourse (van Eemeren, 2010, p. 29).16
Based on this ideal model, specific rules need to be adhered to in order to enable a resolution of a dispute. Once one of those rules is violated, the resolution of the dispute is impeded. Essentially, what contributes to resolving the dispute is also reasonable. Since the pragma-dialectical standard of reasonableness corresponds to the rules governing the model, a violation of those rules amounts to an unreasonable discussion move, i.e. a fallacy (van Eemeren, 2010, pp. 193-196). In that vein, tu quoque arguments are considered to be fallacious because they violate one of those rules. Specifically, the so-called ’Freedom Rule’ stipulates that participants ”must not prevent each other from advancing standpoints or casting doubt on standpoints” (van Eemeren, 2015, p. 623). Correspondingly, if no rule is violated, then an inconsistency allegation can be a legitimate argument. More importantly, consistency in argumentation is even considered a minimum requirement for the application of the rules (van Eemeren & Grootendorst, 2016, pp. 170-172). In that regard, accusing an opponent of not being consistent can even be understood as the arguer’s effort to underline that the inconsistent contributions will not advance the resolution of the dispute at issue.
Accusations of inconsistency need to be based on mutually agreed-upon concessions or starting points (which can involve the character of the agents at issue). Say, A and P agree on q. If P then claims ¬q, then A has reasonable grounds to accuse P of being inconsistent. What is curious, however, is that A then actually prevents P from advancing the standpoint ¬q. Such would then in turn amount to a violation of the Freedom Rule. However, the Freedom Rule, like all other discussion rules, can only come into effect, if the minimum requirements for the rules are fulfilled, i.e. consistent argumentation. Ergo, arguers are required to be consistent if a reasonable discussion should ensue. Say, P wants to maintain his standpoint ¬q. In that case reasonableness can be restored if A and P can agree to retract q.
Matters are complicated further if we look at accusations of inconsistency in context, like campaign debates, for example. It is absolutely possible that certain concessions are never made explicitly and can only be inferred by the antagonist. Inconsistency allegations might even refer to purported concessions made siginificantly earlier in time. Such is the case in campaign discourse, for example, which sometimes spans weeks, if not months. After all, participants in a campaign debate might allude to statements made earlier in the campaign when accusing their opponent of having committed an inconsistency.
In campaign debates, as established before, the electorate and the opponent both are the
16Moreover, van Eemeren (2010) offers a framework that integrates dialectical and rhetorical insights
called ’strategic maneuvering’. From this perspective, argumentation is also the attempt to resolve a given dispute in one’s favour. In that case, they talk about the aim for effectiveness in argumentative discourse (van Eemeren, 2010, p. 36).
addressees of a candidate. In that sense, a reasonable accusation of inconsistency would be directed at the opponent or the electorate. This, however, seems strange seeing that from a theoretical perspective a discussion always takes place between two parties. In argumentative reality disputes oftentimes arise between multiple arguers. Analytically speaking, however, this would amount to multiple simultaneous discussions. If the difference of opinion adhered between the accuser and the electorate, then the accused should not be concerned at all. However, we can easily imagine that a candidate accused of having committed an inconsistency will do anything but remain silent. As van Eemeren (2010, p. 248) concedes, accusations of inconsistency committed in public political debate involve all three parties. However, he does not make clear whether the starting points of public political debate would need to be accepted by all three parties. Assume that there are multiple discussions taking place. One discussion takes place between each of the candidates and the electorate while another between the candidates themselves. If van Eemeren (2010) is right and accusations of inconsistency in public debate involve all three parties, then there must be some kind of interference between the discussions at issue. However, it does not become clear how multiple discussions can impact each other.
In the above analysis multiple accounts for the reasonableness of arguments were given. Walton (2007) claims that the reasonableness of arguments depends on the type of dialogue in which those arguments are employed. Perelman & Olbrechts-Tyteca (1958), on the other hand, claim that the decisive question to ask should rather be what is the audience is ready to concede, or rather, what starting points is the audience willing to accept. Lastly, van Eemeren (2010) suggests that accusations of inconsistency can be reasonable as long as advancing such arguments does prevent the other party from maintaining their standpoint (or putting one forward). This condition is fulfilled if the accusation of inconsistency is based on mutually agreed upon starting points. The pragma-dialectical concept of ’starting points’ seems to be similar to the rhetorical concept of concessions in that both approaches use the concepts to describe the way in which arguments can be grounded. The difference is that in the rhetorical view the starting points (or argument grounds) vary per audience while in the pragma-dialectical view they vary per discussion. In the same vein, it could be stated that in Walton’s view the starting points vary per dialogue type.17
The approaches illustrated, although briefly, propose various theories as to why accusations of inconsistency can be reasonable arguments. Although the definition of a reasonable argument might differ per approach, what all approaches have in common is the ambition to establish a framework for evaluation. At any rate, the concept of reasonableness serves as a theoretical
17It should be mentioned that discussions and dialogue types are by no means similar concepts. The
pragma-dialectical concept of a discussion is based on conventionalized practices (van Eemeren, 2010, p. 129). On the other hand, dialogue types are a descriptive concept (Walton, 1998a, p. 130).
scaffold to determine the quality of argumentation.
2.2
Fallacious Accusations of Inconsistency: The tu quoque
According to van Eemeren et al. (2000, p. 421), the term tu quoque argumentum ad hominem is used for arguments that are put forward to illegitimately unveil an inconsistency on the part of the opponent. In fact, when an arguer is putting forward a tu quoque argument, he is indeed accusing his opponent of being inconsistent. In other words, the arguer is putting forward an ’accusation of inconsistency’. Various approaches discuss the tu quoque in view of its purported fallaciousness (e.g. Copi et al. 2011, van Eemeren et al. 2014). However, Whately (1827), for example, comments on ad hominem arguments in general and it was only later, that some of his insights were associated with tu quoque argumentation.
Generally, ad hominem arguments are associated with Locke (1959), who originally coined the term (Finocchiaro, 1973; Hitchcock, 2017; Walton, 2007).18 The Lockean type of ad hominem argument, however, denotes a somewhat different type of personal attack than the one at issue, because it specifically concerns the opponent’s commitments. In particular, the Lockean ad hominem argument describes arguments used ”to press a man with consequences drawn from his own principles or concessions” (Locke, 1959, p. 278; IV.XVII.21).19
Consider the example of the so-called ’sportsman’s rejoinder’, as it was termed by Walton (1985, pp. 53-59):20
A hunter is accused of barbarity by a critic for killing animals for his own joy. He replies to his critic: ’You are being inconsistent! On the one hand, you accuse me of killing animals, and on the other hand, you are a meat-eater and are therefore involved in killing animals yourself!’21
As Hitchcock (2017, p. 412) points out, the hunter’s accusation is based on the presumption that the critic is bound to ’flesh-eating’. As a consequence, the critic is also involved in the killing of unoffended animals and displays an inconsistency if he criticizes others for what he does himself. With this presumption in place, the critic is required to dissociate killing for nutritional purposes from killing for fun. According to Hitchcock (2017, p. 412), 21st century
logic textbooks claim that the fallacy here arises if the hunter points at an inconsistency
18In contrast to Locke, however, Aristotle, for example, regarded the ad hominem argument as part of an
acceptable argument (Wohlrapp, 2014, p. 156, fn. 41).
19This is somewhat comparable to the argument from commitment, which draws on the concessions made
by the opponent in a discussion (Johnstone, 1959; Zarefsky, 2014, p. 145-146).
20This is a reformulation of the original example provided by Whately (1827, p. 192).
21When refering to the tu quoque fallacy, pragma-dialecticians usually provide this example (van Eemeren,
while there is none. Moreover, it is claimed that a fallacy can also arise if the hunter wrongly portrays the critics past actions, say, if the critic is actually a committed vegetarian (cf. Copi & Cohen, 2002).
According to Copi (1972, pp. 75-76), the hunter is pointing out an inconsistency between the critic’s standpoint (that killing is reprehensible) and the critic’s behaviour (being carnivorous). The hunter’s standpoint therefore is something like ’The killing of animals is not reprehensible’. The assumption is that the means for eating meat consists of accepting the killing of animals. In order to be consistent, condemning the killing of animals would require not being involved in the killing oneself. The fallacy supposedly arises because the hunter’s argument (that the critic is involved in the killing of animals) is irrelevant to the hunter’s standpoint (that the killing of animals is not reprehensible). In other words, the hunter’s argument is not considered a suitable justification of his standpoint, because the argument is regarded as irrelevant.
Copi (1972) approaches the argument from a logical perspective and considers ad hominem arguments to be fallacies of relevance (cf. Hamblin, 1970). Copi et al. (2011, p. 111) specify that this is the case because ”the premises of the argument are simply not relevant to the conclusion”. In that vein, ’fallacies of irrelevance’ would actually amount to a more suitable term (Copi et al., 2011, p. 112). Copi’s perspective remains general and only takes into account what is formalizable from the perspective of logic. Why exclude informal properties of an argument like the character of the person putting forward an argument, for instance? This could play an important role in standpoints and arguments about the person’s character, for example. Interestingly, Aristotle made a distinction between judging a conclusion in view of its arguments, on the one hand, and judging a conclusion in view of its author, on the other.22 It can be said that Copi (1972) sticks to the former way and thus focuses on
what Aristotle calls absolute proof, rather than relative proof. In logic in general, the latter perspective does not seem to be available.23
Let us ponder on Aristotle’s distinction between absolute proof and relative proof a little longer. Aristotle envisioned two ways of proving a conclusion or claim. Either a claim is judged in view of the supporting arguments, or it is judged in view of the person putting forward that argument. This can be a bit confusing, as a person-related judgment can be formulated as an argument as well. Moreover, the question arises inhowfar the two kinds of proof are related to claims. Can we make an analogous distinction between absolute claims and relative claims? Absolute claims in this sense are statements made to hold
22Aristotle referred to two kinds of proof, namely absolute proof and proof ’ad hominem’, i.e. ’to the
man’(Aristotle, 1957, Metaphysics, XI.5.1062a3).
23Because logicians focus on absolute truth they refer to the ’validity’ of arguments when talking about
the appropriateness of an argument for a conclusion. In informal logic or pragma-dialectics, amongst others, scholars rather refer to the ’reasonableness’ of an argument.
universally, while relative claims are statements made to hold relative to a specific person, for instance. Moreover, the Aristotelian distinction between proofs is crucial for accusations of inconsistency. The distinction points at the two possible strategies for determining how an argument can be evaluated. The question then arises: how should accusations of inconsistency be evaluated. Should it be ’absolutely’ (using those arguments as support for universal statements), or ’relatively’ (using those arguments as support for statements related to a particular person)? Perhaps arguments related to a person’s character, such as accusations of inconsistency, need to be judged in view of that person himself, rather than in view of a universal claim. An accusation like ’You are a flesh-eater yourself!’ specifically refers to one person. Therefore, such an argument might support a claim which also refers to that person. For example, ’Be serious. You cannot believe that killing animals is reprehensible’. Conversely, a ’person-related’ argument might not be suitable to support a universal claim, like ’Generally, killing animals is reprehenisble’.
Regarding campaign debates, Leff (2007, pp. 859-860) urges his readers to focus not only on logic when evaluating ad hominem arguments. In his view, doing so would ”virtually exile campaign debates from the domain of rational argument”. Logic only takes into account the formal aspects of an argument, i.e. what can be formalized from the contentious point of a debate in terms of propositions. But, as mentioned in Section 1, campaign debates revolve by no means around a contentious point. Furthermore, the candidates’ character plays a crucial role. Character, or ethos, relates to the personality of a human being. An arguer may want to portray the ethos of being trustworthy, for example. The evaluation of accusations of inconsistency in campaign debates therefore needs to take into account the character of the agents in such encounters.
3
Accusations of Inconsistency and their Strategic Use
in Election Debates
In the political domain argumentation is a goal-oriented endeavour. Alongside the goal to resolve a dispute, some scholars ascribe the arguer the intrinsic goal to resolve the dispute in his or her favour (van Eemeren, 2010, p. 39). This is referred to as the ’effectiveness’ of argumentation. Others, based on ancient rhetoric, refer to the effectiveness of an arguer’s argumentation to achieve his or her audience’s adherence (Perelman & Olbrechts-Tyteca, 1958, p. 6). Either way, it should be stressed that the effectiveness of argumentation does not concern the empirical measurement of the effects of argumentation. Instead, what is meant is the description of the available means to achieve such effects, i.e. the means to
persuade the audience (Leff, 2000, p. 245).24 It might therefore be helpful to stress that what is meant here is the potential effectiveness of argumention. But what strategies are available to arguers committing accusations of inconsistency in election debates? In other words, what strategies for effectiveness can be identified? As has been pointed out above, political candidates will pursue positive- and negative-face strategies (cf. Chilton, 2004, p. 40). Personal attacks, like accusations of inconsistency, belong the the latter category. Essentially, the effectiveness of argumentation concerns the means of persuasion available to arguers to achieve their goal of persuading the audience, i.e. the opponent. In the context of election debates candidates would want to persuade their audience that they are the more suitable candidate for the office in question (and should therefore be elected). In order to find out what strategies can make accusations of inconsistency effective in election debates as a means to persuade the opponent, let us consider an example.
During the Canadian federal elections of 2015 several (televised) election debates were held, all of which were attended by the leaders of the running parties. The moderator had prepared a number of questions that he asked each candidate and to which his or her opponents could then individually respond.25 One of the moderator’s questions was about the Energy East pipeline project and how it would affect the carbon price. He first addressed Thomas Mulcair from the New Democratic Party (NDP). Subsequently, Justin Trudeau (Liberal Party) was asked to respond to what Mulcair said on carbon prices. In his response, Trudeau accused his opponent Mulcair of being inconsistent:
Justin Trudeau: ”Mr. Mulcair has been somewhat inconsistent on pipelines. In English he’ll say that he supports the Energy East pipeline; in French he said that it’s out of the question. And that kind of inconstancy, quite frankly, isn’t the kind of leadership we need for Canada. You can’t say one thing in English and its opposite in French. The fact is we need to restore public trust [. . . ]” (Macleans.ca, 2015, l. 39-43, see Appendix on page 34 for a transcript of the
segment of this debate)
Trudeau’s accusation relates to differing positions that Mulcair has suppposedly taken concerning pipelines. In fact, Trudeau specifically points out different (linguistic) contexts in which Mulcair has formulated contradictory opinions on the Energy East pipeline.26 The 24This understanding is reminiscent of Aristotle’s definition of rhetoric as ”the faculty of discovering the
possible means of persuasion in reference to any subject whenever” (Aristotle, 1937, Rhetoric, 1355a14).
25The debate at issue is the (anglophone) Maclean’s debate that took place on August 6th 2015 and which
was attended by four of the leaders of the five running parties; Justin Trudeau, the leader of the Liberal Party; Elizabeth May, the leader of the Green Party; Thomas Mulcair, the leader of the New Democratic Party; and Stephen Harper, the leader of the Conservative Party.
accusation manifests itself in Trudeau’s criticism that ”You can’t say one thing in English and its opposite in French”. Seen from this perspective, the accusation can be conceptualized as one pointing towards an inconsistency between the opponent’s claims (cf. Walton 2007, p. 163-169). What is remarkable, is that Trudeau directly associates Mulcair’s ’inconstancy’ with a lack of leadership (”That kind of inconstancy [. . . ] isn’t the kind of leadership we need for Canada”) and with a lack of trust or credibility (”[W]e need to restore public trust”). Leaders need to be credible and are expected to demonstrate consistency.
But how does putting forward the accusation of inconsistency contribute to turning the debate to Trudeau’s favor? And, more to the point, how can we describe the accusation of inconsistency as a potentially effective argument?
3.1
Persuasion by Unexpressed Premises?
Trudeau’s charge concerns Mulcair’s position on pipelines. From the example above, it stands out that this is not exactly what Mulcair had actually talked about, namely a carbon price internalized in the cost of the pipeline project. So could we say that the charge is related to what Mulcair had stated before? Is Trudeau’s charge relevant to Mulcair’s statement? From what is explicitly said, it looks like Mulcair’s position on the pipeline project is only remotely related to how Mulcair thinks the financing of the pipeline project should be composed. At first glance, it looks like Trudeau’s accusation of inconsistency is not really a response but a contribution of a different type. And indeed, as elaborated above, the entire debate is not really about a specific topic, like pipelines. Rather it is about individual candidates claiming suitability. From this perspective, we would expect Trudeau’s accusation to somehow support the implicit, yet overarching ’suitability’ claim. However, there seems to does not seem to be a direct connection between who thinks what about pipelines and who is the more suitable leader.
It may be that only looking at explicit statements does not suffice. It is possible that the overarching claim relies on implicitly made arguments. It is argued that both the overarching claim and the immediate argument supporting it are implicit. The reasoning mechanism, which makes such a connection possible, despite their implicitness, is called an inference. In particular, the inference is made from unexpressed premises.
But can we conceptualize the use of unexpressed premises as a strategy for effective argumen-tation? According to Zarefsky (2014, p. 208) we can. In his analyis of the Lincoln-Douglas Debates he comes to the conclusion that unexpressed premises (through inferences) have a
this ”does not automatically mean that his standpoint is incorrect”. Their advice to remedy the inconsistency is to retract one of the standpoints.
persuasive potential because they engage the audience, i.e. the voters. The audience then actively participates in the reasoning of the arguer. He bases his argument on his observation that inferences are the mechanism to complete, as it were, incomplete syllogisms.27 What is
inserted to complete the syllogism is the unexpressed premise. The audience then supposedly commits to the conclusion of the argument more readily because, in a way, they ’contributed’ to it themselves by adding that extra unexpressed premise. Strikingly, Zarefsky remarks, an opponent criticizing such arguments inadvertently also criticizes the audience. This seems a little too hasty since the audience might also be committed to arguments that the opponent has brought forward. After all, ’committing to arguments’ is what is meant by the audience’s adherence.28 And both the opponent and the proponent attempt to have the audience commit to their arguments as much as possible. However, it seems plausible that unexpressed premises can influence the effectiveness of arguments.
By taking a closer look at unexpressed premises, it may become clear how Trudeau’s accusation supports the overarching claim of the debate. As has been asserted above, election debates revolve around two interrelated claims regarding the suitability of the candidates (cf. Breglio, 1987; Martel, 1983). From that perspective, Trudeau would be maintaining the two claims (i) I am the more suitable leader, and (ii) Mr. Mulcair is the less suitable leader. Moreover, we can assume that Trudeau tries to undermine Mulcair’s credibility by putting forward an accusation of inconsisitency (cf. Walton, 2007, pp. 165-169). And we can assume that a lack of credibility makes a candidate less suitable for a political office. We can even assume that Mulcair is a less suitable leader because Mulcair lacks credibility. Moreover, credibility is a character trait necessary for political candidates (cf. D’Errico et al., 2013, p. 181). Ergo, a lack of credibility entails a lack of suitability as a political leader. In such a case, the main standpoint (the ’lack of suitability’ claim) is supported by the substandpoint (the ’lack of credibility’ claim). To illustrate, consider the reconstruction below:29
(1) Mr. Mulcair is the less suitable leader. (1.1) Mr. Mulcair lacks credibility.
(1.1’) Lacking credibility makes a candidate a less suitable leader.
So far, the entire argumentation structure consists of implicit standpoints and arguments. The main standpoint is supported by Mulcair’s purported lack of credibility. But what gives rise to these implicit standpoints is the explicit accusation of inconsistency from which the superordinated standpoints follow.
27The mechanism of inferring unexpressed premises is reminiscent of the one at play in Aristotelian
enthymemes. Here, the strategy of leaving certain premises unexpressed could also be referred to as an enthymemic strategy for effective argumentation.
28Moreover, Zarefsky’s idea would need to be corroborated through empirical research by cognitive
psychologists in order for the claim to be more profound.
Walton (1998a, p. 212) (after Govier, 1983, p. 24, fn. 20) points out that ”when someone fails to practice what he preaches, his credibility is undermined”. In the same vein, we can hold that if someone is being inconsistent between ’what he preaches’, his credibility is undermined. In other words, being inconsistent conveys a lack of credibility. This way it is possible to incorporate the accusation of inconsistency as support for the lack of credibility standpoint, as the following structure illustrates:30
(1) Mr. Mulcair is the less suitable leader. (1.1) Mr. Mulcair lacks credibility.
(1.1’) Lacking credibility makes a candidate a less suitable leader. 1.1.1 Mr. Mulcair is being inconsistent.
(1.1.1’) Being inconsistent makes a leader lack credibility.
1.1.1.1 Mr. Mulcair says one thing in English and its opposite in French. (1.1.1.1’) Saying opposite things (in different languages) is being inconsistent.
1.1.1.1.1a Mr. Mulcair says in English that he supports pipelines.
1.1.1.1.1b Mr. Mulcair says in French that pipelines are out of the question. (1.1.1.1.1’) 1.1.1.1.1a and 1.1.1.1.1b are opposite.
The number of implicit standpoints in this illustration is tremendous. The bridging premise of the ’lack of credibility’ claim must be inferred: being inconsistent makes a leader lack credibility. If Zarefsky (2014) is right, then the audience will draw this premise themselves and, more importantly, feel committed to it. It now becomes apparent, why committing to the (unexpressed) bridging premise would make such a big difference. The accusation of inconsistency requires the bridging premise in order to be a complete argument. What is more, the accusation as support for the ’lack of credibility’ standpoint is only cogent or relevant, if that bridging premise is there. To put it another way, only if it is made clear that inconsistency conveys a lack of credibility can the claim be upheld that who is being inconsisent also lacks credibility, Mr. Mulcair in this case. Note that this accusation of inconsistency would be considered a reasonable argument in support of the ’lack of credibility’ claim, at least from the perspective of logic.
However, it is not guaranteed that an audience will draw that inference, simply because the claim, that Mr. Mulcair lacks credibility, is not made explicit. Granted, the audience needs to actively listen and ’participate’ in order to follow the reasoning of Trudeau’s accusation. It is definitely possible that the audience infers the unexpressed premise which makes them adhere to the subsequent standpoint. But there is no reliable indication as to why the audience should put the bridging premise in place, seeing the ’lack of credibility’ claim is implicit. Consequently, claiming that Trudeau gains his audience’s adherence is not tenable. Nor, in
30The accusation of inconsistency illustrated here bears striking resemblance to the illustration of pragmatic
pragma-dialectical terms, can it be said that Trudeau resolves the dispute in his favor. Based on this example, specifically, we can only accept Zarefsky’s claim that unexpressed premises have a persuasive potential with reservations.
3.2
Effectiveness through Ethos
As has been made clear, election debates are about the candidates, their personalities and why people should vote for them. Election debates are essentially about the candidates’ character. In rhetorical terms, character is referred to by the Greek word ’ethos’, which was coined by Aristotle (Hitchcock, 2017, p. 466).31 Aristotle writes that personal character
”may almost be called the most effective means of persuasion” (Aristotle, 1984, Rhetoric, I.2, 1356a13; cf. van Eemeren et al., 2014, p. 118). So understanding Trudeau’s accusation from the perspective of character might reveal valuable insights about the potential effectiveness of his argument.
Aristotle already acknowledged the influence of an arguer’s character on the credibility of the arguer’s argument (Aristotle, 1937, 1356a, 4-5)32. It is claimed that this holds for people other than the arguer as well. As Tindale (2015, p. 165) writes, ethos ”can refer to a range of argumentation that also addresses the characters of others, from ad hominem reasoning to appeals to authority”. So when Trudeau (implicitly) appeals to the lack of credibility of Mulcair, he is essentially appealing to Mulcair’s (purportedly questionable) character. But does this mean that the strategy at play with accusation of inconsistency is simply their relation to character? Specifically, could it be stated that character appeals gain the adherence of the audience in election debates or further the discussion in any way? As mentioned in Section 1, according to Benoit (2014, p. 19), candidates can indeed gain the voters’ adherence by highlighting their character, i.e. who they are. One possible explanation might be that most voters are not as well-versed in political subjects as politicians and can therefore not verify the politicians’ claims as easily. This might be why the candidates sometimes resort to arguments that concern their character in order to portray themselves as the more suitable candidate; as ’good men’.
”[W]e believe good men more fully and more readily than others”, Aristotle writes, ”this is true generally whatever the question is, and absolutely true where exact certainty is impossible and opinions are divided” (Aristotle, 1984, 1356a6-8). Aristotle’s insight is vital for a discussion of election debates. ’Exact certainty’ is indeed impossible where several parties present themselves as the most desirable candidate for a position of power. There are
31The Latin word ’character’ in turn translates to personality, or disposition.
32van Laar (2008, p. 172) goes even further and claims that ”protagonist credibility” is a prerequisite for
just no probative means that will yield absolute ’truth’. Interestingly, the accusation put forward by Trudeau is such a case. Whilst it can be proven whether Mulcair said inconsistent things at different occasions, matters are more complicated when it comes to proving whether he therefore lacks credibility. The audience would simply have to believe Trudeau, provided that they trust him. Conversely, by accusing Mulcair of being inconsistent and thereby denouncing his credibility, Trudeau tries to reduce the level of trust that the audience has in Mulcair. Eventually, should Trudeau be successful, Mulcair’s claims that are not about exact certainties will not be believed.
Making a legitimate accusation could also back up Trudeau’s position because he thereby demonstrates a sense of productivity, as it were, by holding Mulcair accountable to his claims. In a way, by reducing the credibility of his contender, Trudeau’s own credibiltiy might even increase. It could then be claimed that Trudeau has a personal interest in putting forward a legitimate accusation because it strengthens his position as a candidate. Let us assume that putting forward legitimate arguments demonstrates that the arguer himself is reasonable, or rather has common sense. Then the argument could be made that Trudeau creates credibility for himself because he demonstrates reasonableness, or common sense, for that matter. In fact, Aristotle (1937, 1738a) considered ’good sense’ (phronesis, prudence) to be one of the aspects that create confidence in the arguer’s character (cf. Brinton, 1985, pp. 219-222). Generally, what a politician says must be assessed somehow by the audience. But because the majority of voters only have a limited understanding of the complex dynamics of political reality, they need to resort to simply trusting politicians. Yet blindly trusting someone would be foolish. Therefore, voters only trust those politicians that show the character traits from which they can infer that a person is trustworthy. If a politician reasonably shows that their opponent lacks a particular character trait, then the audience will make that inference straight away. In the same vein, participants of election debates can blacken or denounce their contender’s character in front of the audience. If the denunciation is a reasonable accusation of inconsistency, for example, then the accused politician’s character is defamed.33 If a succesful accusation of inconsistency decreases an opponent’s credibility (i.e. the audience’s trust), then anything the opponent will say afterwards (and anything he might have said before) might be tainted as being not credible. Evidently, the repercussions of accusations of inconsistency are immense. I call the tactic to allude to an arguer’s credibility, or credence, the credential component of the strategy under discussion.
From the perspective taken here, Trudeau’s accusation of inconsistency appeals to Mulcair’s negative character. Thanks to Aristotle’s understanding of ethos, the inference from an inconsistency allegation to negative character can be accounted for. Yet there might be
33For arguments in which ethos is used to transfer credibility (positively or negatively) from an arguer to
