1 Introduction

If we were to identify the beginning of the study of visual argumentation, we would have to choose 1996 as the starting point. This was the year that Leo Groarke published “Logic, art and argument” in Informal logic, and it was the year that he and David Birdsell co-edited a special double issue of Argumentation and Advocacy on visual argumentation (vol. 33 no. 1 & 2, 1996). Among other papers, the issue included Anthony Blair’s “The possibility and actuality of visual arguments”. It was also the year that Gail J. Chryslee, Sonja K. Foss and Arthur L. Ranney published their short text on “The construction of claims in visual argumentation” (1996), stating that even though theorists may know a great deal about the process of argumentation, “virtually none of this knowledge is applicable to visual argumentation […] because of the properties that distinguish visual imagery from discursive symbols” (: 9).

Texts on visual rhetoric have been written as far back as the early 1980s, so 1996 was not the first year that publications on multimodal argumentation appeared.Footnote 1 Sol Worth claimed already in 1975 that “Pictures can’t say ‘ain’t’” (cf. Worth 1981). In 1994 Michael Gilbert published his text “Multi-modal argumentation”, which he later developed into the book Coalescent argumentation (1997). In 1995 Bruce GronbeckFootnote 2 published his “Unstated propositions: Relations among verbal, visual, and acoustic languages”, and countered Worth, among others, when proposing that if

we think of meanings as called up or evoked in people when engaged in acts of decoding, then not only words but also pictures, sounds, and other sign systems certainly can offer us propositions of denial or affirmation, and can, as Locke understood trueness and falsehood, articulate empirically verifiable propositions. (Gronbeck 1995).

Even though some research had been conducted before 1996, more papers on visual argumentation were published that year than in all previous years combined. Prior to 1996, few other publications addressed this issue so directly as those from this seminal year. Three years later, Gerard Hauser edited a special double issue of Argumentation and advocacy on body argument: seeing the human body as a discursive and argumentative site (Hauser 1999a, b).

Only a handful of people worked on visual argumentation in the early 1990s, but the expansion of the field became obvious when in 2001 at the Alta Argumentation Conference Catherine Palczewski devoted her keynote address to visual argument and personal testimony (Palczewski 2002). Then, in 2007, 10 years after the first special issue of Argumentation and Advocacy, the publication of a new special issue illustrated the maturation of the field of visual argumentation. As in 1996, the issue was edited by Birdsell and Groarke, who also contributed with “Outlines of a theory of visual argument” (Birdsell and Groarke 2007).

The present special issue of Argumentation builds on the tradition established since the beginning of the study of visual argumentation. The principal issue has been whether there is such a thing as visual argumentation. However, as Blair (2015) points out in this issue, the burden of proof for the possibility and the existence of visual arguments has been reversed (cf. Groarke 2003). In this issue we have therefore attempted to take visual argumentation a step further in order to examine what visual and multimodal argumentation is and how it may work.

This introduction will firstly provide a brief overview of some of the most common approaches to the study of visual and multimodal arguments, summarily mentioning the genres and forms of expression most commonly studied, and will provide a summary of the main objections put forward by sceptics of visual argument. Finally, the contributions to the special issue are introduced. The introduction is concluded with an extended literature list within the field of visual and multimodal argumentation.

2 Theoretical and Methodological Departure Points

The theoretical and methodological departure points in the study of visual argumentation are varied, and researchers use theories, methods and insights from a wide variety of argumentation research; informal logic, pragma-dialectics, and rhetoric are some of the most common approaches.

Among the most quoted texts are Daniel O’Keefe’s papers “Two concepts of argument” (1977) and “The concepts of argument and arguing” (1982). Since research on visual or multimodal argumentation often challenges traditional conceptions of what an argument is, O’Keefe discusses, conceptualises and defines a notion of argument that encompasses both verbal and visual argumentation (e.g. Gilbert 1994; Blair 1996; Lake and Pickering 1998; Schwed 2002; Kjeldsen 2007).

Some studies apply Stephen Toulmin’s notion of argument fields (Balter-Reitz and Stewart 2005, 2007) or his argumentation model for the reconstruction of premises and claims extracted from the visual objects or constructed from audience studies (e.g. Kjeldsen 2007, 2012; Ripley 2008), while others have demonstrated how the key components of the Toulmin model can be expressed in images and function as an essential element of visual argument (Groarke 2009).

In addition, other studies have shown how the pragma-dialectical principles of communication can be applied to visual arguments and their components, understanding and interpreting the arguments as speech acts. Pragma-dialectics is thus well suited to analysis of visual arguments because its theory of implicit and indirect speech acts, provides an account that easily applies to images (Groarke 2002). Images, such as cartoons, are regarded as indirect speech acts functioning as visual arguments, advanced “as a means for defending a standpoint in the context of a critical discussion” (Feteris et al. 2011: 60). A cartoonist thus manoeuvres strategically to convince the reader of a particular standpoint. The four stages of a critical discussion and the reconstruction methods (cf. van Eemeren and Grootendorst 2004: 57 ff.) may be used analytically to understand how visual argumentation—by a cartoonist for instance—is used to manoeuvre strategically (Feteris et al. 2011; Feteris 2013; Plug 2013; Schwed 2002; van den Hoven 2012a; cf. van Eemeren 2010). In general, it is the view of argumentation as a social and rational activity—in principle independent of form of expression—that makes pragma-dialectics a suitable perspective to study of visual and multimodal argumentation as interaction (Tseronis 2013; cf. van den Hoven 2013a).

In a similar vein, when investigating visual argumentation, some argue for the use of other pragmatic theories, such as Searle and Austin’s theory of speech acts, Paul Grice’s work on implicature, intention and cooperative principles and Sperber and Wilson’s relevance theory (e.g. Slade 2002, 2003; Tarnay 2003; Schwed 2007; Barceló Aspeitia 2012; cf. Alcolea-Banegas 2009), and congruity theory (Rocci et al. 2013).

A frequently used concept in the literature is the enthymeme (e.g. Heeney 1999) In accordance with Aristotle’s definition of an enthymeme, we may say that visual enthymemes—or simply visual arguments—involve 1) probable premises and conclusions, 2) accommodating ethical (cf. ethos) and emotional dimensions of argument, 3) depending on agreement between messenger and audience (cf. Smith 2007). This last point in particular informs studies of visual argumentation since arguing visually requires the active participation of the audience in reconstructing arguments (cf. Bitzer 1959), a point that is supported by audience research and reader-response theory (Barbatsis 1996). Thus, the most “important and foundational feature of visual argument is its enthymematic nature” (McNaughton 2007: 136). Visual rhetoric and argumentation strikes a responsive chord that elicits, cues, or awakens arguments in the audience (Gronbeck 1995; Kjeldsen 2007: 131, 2012: 241).

In comparison to this view, Finnegan (2001) uses the enthymeme in a slightly different way when she applies it to the perceived truth value of photographs. Because we perceive photographs as fundamentally “realistic”, she suggests, “we make assumptions about their argumentative potential” (2001: 135). Finnegan calls this process “the ‘naturalistic enthymeme’: we assume photographs to be “true” or “real” until we are given reason to doubt them” (ibid.).

The audience-participation and enthymematic view on visual argumentation goes back to 1996, when Chryslee, Foss and Ranney assumed an “audience-centred perspective on the creation of meaning in images”, and situated the viewer “as the dominant factor in the construction of arguments from images” (: 9). This way of thinking has made some researchers encourage and conduct more reception-oriented studies on visual argumentation (e.g. Kjeldsen 2007, 2015, forthcoming).

Rhetorical perspectives require an understanding of context. Birdsell and Groarke (1996) have argued that at least three types of context are important in the evaluation of visual arguments: (1) Immediate visual context, (2) Immediate verbal context, and (3) Visual culture. The rhetorical situation (cf. Bitzer 1968) has been introduced as an important form of context (Kjeldsen 2007; Collins 2007). The concept of adherence (cf. Perelman and Olbrechts-Tyteca 1969) is often applied, since visual rhetorical argumentation necessarily occurs in a specific situation in which someone attempts to secure the adherence of an audience (Tarnay 2003). Among other perspectives from rhetoric are McGee’s (1980) concept of ideograph (Pineda and Sowards 2007; Palczewski 2005) and Perelman and Olbrechts-Tyteca (1958) concept of presence. The latter is naturally central since visual argumentation has the ability to actually and materially create presence before the eyes of the audience (Lake and Pickering 1998; Kjeldsen 2012). Another influence from rhetoric is the use of tropes and figures in order to describe and understand pictorial argumentation (Kjeldsen 1999, 2012; Pollaroli 2013; van Belle 2013, Pollaroli and Rocci forthcoming; Mazzali-Lurati and Pollaroli forthcoming). This is a type of research that often draws upon theories of tropes and figures either in the classical theories or as they are conceptualised and applied in consumer research (McQuarrie and Mick 2003). Literary or cognitive metaphor studies are also often used (Tseronis et al. forthcoming; cf. Lakoff and Johnson 1980; Forceville 1996), as well as notions of analogy from both rhetoric and informal logic (Dove 2011b; Kjeldsen 2013).

An important issue in defining visual argumentation has been the positioning of visual argumentation in relation to visual persuasion and visual manipulation. Opponents of the idea of visual argumentation have argued that visuals such as pictures do not argue; they only influence or persuade—often in a manipulative way. According to this position, pictorially dominated advertising, for instance, is not argumentation at all but rather a subconscious and irrational kind of psychological persuasion (e.g. Johnson and Blair 1994: 225). This view has compelled researchers of visual argumentation to examine and clarify the relationship between visual persuasion and visual argumentation (e.g. Blair 1996, 2004; Slade 2003; Ripley 2008; Roque 2012; Nettel and Roque 2012),

Another recurring theme in the study of visual and multimodal argumentation is what Godden (2015) has called “the oldest and most basic question of the interpretative research in visual argumentation”: the question of propositionality. Is visual argument propositional? Must it be propositional to be argumentation at all? This question is especially prominent in the debate on the possibility and actuality of visual argument (Fleming 1996; Blair 2004; Johnson 2003; Barceló Aspeitia 2012; van den Hoven 2012a; Patterson 2010), and will be treated in more detail below and in Georges Roque’s contribution to this special issue. It has been argued that even though images may not be propositional and thus, strictly speaking, lack the capacity to negate, they are nonetheless able to “refute” other images (Lake and Pickering 1998) by visually dissecting, or breaking down, the image being refuted, or by visually substituting or transforming it in a repudiatory manner (cf. Groarke 1996: 118ff.).

Perhaps the most central—as well as the most general—issue in the study of visual and multimodal argumentation is the significance of the phenomenological and material difference between verbal language and imagery. What difference does the difference make? Even though much research is concerned with identifying and reconstructing premises, there is widespread agreement in the fact that we can never fully translate images into text or verbal propositions (Tarnay 2003; Alcolea-Banegas 2009; Barceló Aspeitia 2012; van den Hoven 2012a, b; Kjeldsen 2015). Pictorial representations are “inherently unstable sets of signs”, and thus constitute a process that is impossible to capture in full “in a single Toulmin-like or syllogistic model. Rather, pictures become woven into complex argumentative discourses, and their places vary across different discursive practices” (Gronbeck 2007). It is a distinctive feature of visual argumentation that premises and standpoints are “not separate elements but are fused together in a holistic, inseparable unit” (McNaughton 2007: 138; cf. Barbatsis 1996: 79); images “communicate an (inextricable) mixture of mimetic elements and diegetic elements” (van den Hoven 2013a, b: 5).

Generally speaking, research has moved between two different—and seemingly contradictory—positions on this matter. On the one hand scholars has pointed to the fact that argumentation is cognitive or logical and hence independent of a specific form of expression (Schwed 2007; Kjeldsen 2007; van den Hoven 2015) On the other hand scholars has pointed to the fact that visuals are obviously a very different form of expression and therefore either cannot be arguments (Fleming 1996; Johnson 2003; Patterson 2010) or provide a special kind of argumentation (Gilbert 1997; van den Hoven 2012a; Kjeldsen 2015). This question pertains directly to the matter of assessment of probative merits: Does visual argumentation require norms or methods of evaluation that are new and different from the ones traditionally used for verbal argumentation (Blair 2015)? The position that visual arguments are not radically different and thus do not require special norms or methods has been called normative non-revisionism (Godden 2013). The opposite position, normative revisionism, suggests that there “are distinctive criteria for the evaluation of visual argument, which are independent of, and not reducible of, evaluative criteria for non-visual argument” (Godden 2013: 4). The view that “the visual and the verbal are irreconcilably distinct” (Groarke 1996: 106) and that visual argumentation is not a derivative or extension of verbal argument has been called the autonomy-thesis (Johnson 2010, Godden 2013; see part 3 below on “The resistance towards visual and multimodal argumentation”).

Apart from theories of argumentation, the study of visual argumentation also draws upon theories of visuality, images, pictures and representation. Semiotics, art history, and media studies are not surprisingly among the most widely used theoretical backgrounds when trying to understand visual communication.

3 Genres and Forms of Expression

The study of visual and multimodal argumentation touches upon a wide variety of genres and forms of expression, including image events (Delicath and DeLuca 2003), flag waving (Pineda and Sowards 2007), slide show presentations (Kjeldsen 2013), brain imaging (Gibbons 2007) folk arts (Roberts 2007), fashion (Torrens 1999) prison tattooing (2007), yarn bombing (Hahner 2014), needlework (Goggin 2003), place, architecture and memorials (Fleming 1998; Blair et al. 2011), museums (Balter-Reitz 2003), artworks (Groarke 1996; Chryslee et al. 1996), postcards (Palczewski 2005), news magazine covers (Tseronis 2015), and photography (Finnegan 2001, 2003; Gronbeck 2007; Hahner 2013). Exploring the body as a site for argument (Hauser 1999a, b) has resulted in studies on actio in debates (Gelang and Kjeldsen 2011), disabilities and argument (Cherney 1999; Kiewe 1999), as well as social protest (DeLuca 1999), woman suffrage and feminism (Palczewski 2005, 2010; Stormer 1999).

The three most dominant genres of visual and multimodal argumentation, however, are arguably advertising, cartoons and scientific communication. This is not coincidental, since these all aim at arguing and persuading. When Roland Barthes in his seminal essay “The rhetoric of the image” (1977) chose to study an advertising image, he did so because “in advertising the signification of the image is undoubtedly intentional” (Barthes 1977: 51). This is also the reason that many studies turn to advertising when exploring visual argumentation. In the beginning of the 2000s the purpose of several papers was to correct the view of advertisements as mere emotional and irrational appeals, end establish that image based ads can offer rational reasons and arguments (Slade 2002, 2003; Ripley 2008). Recently as mentioned above—papers have explored the workings of visual argumentation in advertising through tropes and figures (Kjeldsen 2012; Pollaroli 2013; van Belle 2013; Pollaroli and Rocci, forthcoming; Mazzali-Lurati and Pollaroli, forthcoming). These studies almost exclusively examine commercial advertising, where visual tropology is most frequent. The study of visual argumentation in political advertising, on the other hand, mostly deals with moving images, such as political commercials making characterological arguments about the candidate (Gronbeck 1993; Collins and Schmid 1999) or public service announcements (Hatfield, Hinck & Birkholt 2007). Studies dealing with moving images also look at movies, documentary and television coverage. In general a main purpose is to establish how these form of expressions offer arguments, and to develop perspectives and methods for understanding and analysing multimodal discourse (Barbatsis 1996; Lake and Pickering 1998; Tarnay 2003; van den Hoven 2012a, b, 2013a; van den Hoven and Yang 2013; Bloomfield and Sangalang 2014; Tseronis et al. forthcoming).

Like advertisements cartoons belong to an inherently argumentative genre, which is why they are so frequently discussed. In the first stages of the study of visual argumentation, the point was rarely to examine cartoons as a genre, but instead to use cartoons to argue for the actuality of visual argumentation (Birdsell and Groarke 1996, 2007; Groarke 1996). Later, the emphasis has turned to the application of theoretical perspectives, such as the Toulmin model (Groarke 2009), reconstruction of argumentation (Feteris 2013) or strategic manoeuvring (Feteris et al. 2011; Plug 2013).

Scientific communication has been an important object in the study of visual argumentation at least since the late 1990s, where different kinds of visual reasoning were explored. Visual abductive reasoning has been examined in archaeology (Shelley 1996a) and geology (Dove 2013). Rhetorical and demonstrative modes of reasoning have been illustrated through images of human evolution (Shelley 1996b) and representations of the March of Progress (Shelley 2001). The demonstrative mode represents the actual course of visual thought, while the rhetorical mode works associatively by evoking informal verbal arguments. The fundamental distinction is that “rhetorical visual arguments show patterns closely related to verbal thinking, whereas demonstrative visual arguments serve as portrayals of prominently visual thinking” (Shelley 1996a: 64). It has been argued that analogical reasoning in biology (and advertising) can operate analogously to visual evidence (Dove 2011b), and that the argumentative role of photographs in science and diagrams in mathematics is evidentiary (Dove 2012; cf. Dove 2002, 2011a). Other studies has illustrated how functional brain images argue differently, and sometimes misleading, as they move from the frame of a scientific to a popular contexts (Gibbons 2007). This kind of frame analysis has also been applied to photographic memes (Hahner 2013).

4 The Resistance Towards Visual and Multimodal Argumentation

As demonstrated in the brief account above, there seems to be general agreement that visuals and imagery can play an important role in argumentation. However, ever since the first studies on visual argumentation, there has been resistance to the idea that pictures can constitute arguments. In general, the attitude is that argumentation is closely related to the explicit use of words and therefore cannot be argumentation in any proper sense. This is not surprising since the study of argumentation has been a study of verbal communication for more than 2,000 years.

We may differentiate between two kinds of critical positions towards visual argument. The first is a general assumption that argumentation is a verbal phenomenon. We can call this passive resistance to visual argument since it does not involve any active argumentation against the existence of visual argumentation but rather assumes that all argumentation is verbal. The second position we can call active resistance to visual argument since its proponents actively argue against the existence of visual argumentation.

We find the position of passive resistance in many of the classic works of argument. In The Uses of Argument, for instance Toulmin analyses “the layout of arguments”, and studies “the operation of arguments sentence by sentence” (Toulmin 1958: 94, my italics). In such works, the connections between verbal language and argumentation is strengthened by the fact that premises and conclusions in an argument follow each other like words follow each other in speech or writing. Both argumentation and verbal creation of meaning are a matter of connecting thoughts with other thoughts in a sequential order. An argument is understood as a “train of reasoning”. It is a connection between premises and conclusions, “a sequence of interlinked claims and reasons that between them establish the content and force of the proposition for which a particular speaker is arguing” (Toulmin et al. 1984: 5).

In the same way, Perelman and Olbrechts-Tyteca define the subject for argumentation studies as “the discursive techniques allowing us to induce or to increase the mind’s adherence to the theses presented for its assent” (1969: 4). In an early edition of Fundamentals of Argumentation, van Eemeren, Grootendorst and Kruiger unequivocally defined argumentation as a “verbal activity”: Argumentation can be accompanied by the use of non-verbal means of communication, but without the use of language there can be no argumentation (1987: 2, 5). Later, however, pragma-dialectics warmed to the idea of non-verbal argumentation and several pragma-dialectic studies of visual argument appeared (e.g. Groarke 2002; Feteris et al. 2011; cf. the brief account on pragma-dialectics above).

The active resistance generally springs from the view that visual and verbal communication are fundamentally different. The art historian Gombrich, for instance, states that without aids the visual image “altogether lacks the possibility for matching the statement function of the language” (Gombrich 1982: 138). Similarly, the media scholar Paul Messaris argues that iconic representations such as pictures are characterised by a lack of so called “propositional syntax” (Messaris 1997: x). While spatial and temporal connections can be explicitly communicated through images, “visual communication does not have an explicit syntax for expressing analogies, contrasts, causal claims, and other kinds of propositions” (Messaris 1997: xi). As soon as we go beyond spatiotemporal interpretations, he says, the “meaning of visual syntax becomes fluid, indeterminate, and more subject to the viewer’s interpretational predispositions than is the case with a communicational mode such as verbal language, which possesses an elaborate set of explicit indicators of analogy, causality, and other kinds of connections between two or more concepts” (Messaris 1997: xiii).

Even though there might be a general resistance to acknowledging visual argument within the field of argumentation, it seems that only three researchers have published papers that actively argue against the possibility and actuality of visual argument. The first, and probably most quoted, antagonist is David Fleming who primarily bases his rejection on the traditional definition of argumentation as “reasoning towards a debatable conclusion.Footnote 3 It is a human act conducted in two parts (claim and support) and with awareness of two sides (the claim allows for and even invites opposition)” (Fleming 1996: 19). Fleming states that because we cannot distinguish between premise and conclusion in a picture, and because pictures cannot provide claims that can be contested (since a picture just shows something), per definition they cannot be arguments. Only if we stretch the word “argument” beyond recognition, he writes, will we be able to say that pictures may argue. However, then we are actually talking about something different. The only role that a picture can play in argumentation, according to Fleming, is as a support for verbal premises, and this is only possible if the visual propositions are verbally anchored (Fleming 1996: 19; cf. Barthes 1977).

Ralph Johnson is also an opponent of the idea of visual argument (2003, 2010). He rejects the possibility of visual arguments with the claim that visual argumentation is dependent on verbal argumentation, but there is no reciprocity (2003). Visual arguments, he claims, are either not visual or not arguments. The problem with any theory of visual argument, Johnson states, is firstly that the “notion of visual argument remains somewhat unclear”. Secondly, a theory of visual argument must deal with the problem of “converting” images into reasons that function as premises. As with verbal argumentation, we must be able to recognise the arguments, identify their components, and reconstruct and evaluate the arguments. The process of moving from an image to establishing propositions is neither clearly defined nor well understood, and to the degree we can understand it, the process is “heavily dependent on verbal reasoning and verbal expressions of reasoning, thus illustrating that ultimately the process of reconstructing visual images as arguments will depend on our ability to “translate them in words and that in doing so we are dependent on our experience with verbal argument” (Johnson 2003: 6).

While Johnson in one paper (2010) apparently suggests that he does not completely reject the possibility of visual arguments, he certainly believes that the theory of visual argument, at the time of writing, is incomplete both in relation to the “explication of the nature of visual argument and how the visual is converted; that the benefits of a theory of visual argument are unclear and finally that there exists an important asymmetry between verbal and visual argument” (2003: 9). Even though Johnson agrees that visual literacy is important, he does not believe that we should achieve this by broadening the domain of argument. On the contrary, he believes that argumentation theory should tighten the domain and focus on the central sorts of arguments: “arguments with a dialectical tier” (Johnson 2010; cf. Johnson 2000). Johnson has also argued against what he has termed “the autonomy thesis”. According to Johnson, the autonomy thesis claims that there is a specific definition of visual argumentation that is not derived from the one of verbal argument. This autonomy requires procedures and techniques for analysing and evaluating visual argumentation that differs from those applied to verbal argumentation. This leads to a strange and puzzling situation, Johnson states (2010: 10), because how can there be an autonomous domain of visual argumentation, but no definition or accepted criteria for analysis and evaluation?

A third opponent to the idea of visual argumentation is Patterson (2010). Even though he accepts that visual images may constitute elements of argumentation, he does not believe that anyone has yet satisfactorily shown that there are visual arguments (: 118). His main argument is that pictures are ambiguous, and therefore cannot be arguments. Because of the inherent ambiguity of images, they are generally open to many alternative interpretations. Verbal language has rules and communal criteria that “keep us from falling into humpty-dumptyism”, but there are “no such checks on picturing, or if there are, their effectiveness falls far short of those accompanying our usage of language” (: 112). Images, Patterson argues, may “spur cognitive feelings or associations—and so may be persuasive—but they lack an internal pattern, the recognition of which would allow them the presentation of an illative move from premises to conclusion” (: 122). Hence, images do not carry propositional content and cannot be arguments. A picture, Patterson maintains, “argues nothing. It is the user of the image who argues, using it to make or illustrate or emphasise a particular point”. And what “an image “says” is largely going to be a matter of what the viewer brings to it in terms of contextual knowledge and cultural or linguistic framing” (: 126).

This introduction is not the place to attempt a refutation of the sceptics’ idea of visual argumentation—hopefully the contributions in this issue speak for themselves—but a few comments are in place.

Firstly, we should not confuse the study of visual argumentation with simply the study of purely visual argumentation. Even though many of those who argue for the existence of visual argumentation believe that argumentation through purely visual means is possible, generally these researchers—and the field in general—are more concerned with studying the place and function of the visual in argumentation. This is in line with the most common definitions, such as this one: “Visual arguments forward premises and conclusions which are, wholly or partially expressed by (non-verbal) visual means” (Groarke 2009: 230). The aim is primarily to explore the role of the visual in argumentation.

Secondly, since argumentation, as illustrated in this issue, is a cognitive phenomenon, there is no theoretical reason that visual argumentation should be impossible. Because they are cognitive or logical operations, arguments can, in principle, be expressed verbally, visually or in many other modes (cf. Groarke 2015). Neither a picture nor a verbal text constitutes the argument itself; it is just a way of expressing or evoking argumentation. It is true, of course, that visual manifestations such as pictures and photographs do not have grammar or syntax in the same way verbal language does. However, as we often experience, verbal language is also ambiguous; and pictures—as well as other visual modes of communication—have the potential to argue because they can offer a rhetorical enthymematic process in which something is condensed or omitted, and, as a consequence, it is up to the spectator to provide the unspoken premises (cf. Kjeldsen 2015, this issue). When reconstructing the implied arguments, the viewer may draw upon knowledge of the context of the picture, such as in the circumstances of the current situation, or by certain structures within the picture.

Thirdly, the arguments against visual and multimodal argumentation seem to insist on a rather sharp distinction between the verbal and the visual—a divide upon which modern studies of visual culture do not agree. There is a difference between words and images, of course, but we cannot out of hand separate the discursive from the visual. Words evoke images, and our perception and understanding of visual representations are connected to and dependent on our verbal concepts, without which pictures would be incomprehensible (cf. Mitchell 1987, 2005a, b; Palczewski 2002: 6; Roque 2010, Hariman 2015, cf. Groarke 2002). As Mitchell (2005b: 343; cf. 2005a: 1) has poignantly pointed out: “All media are mixed media”.

Finally, it is worth mentioning that Ian Dove has proposed a middle position between sceptics and proponents of visual argumentation. Dove suggests that “the claim that the visual can serve as support for a linguistic claim [is] enough to make room for the visual in argumentation” (Dove 2012: 225; cf. Dove 2011a; Groarke 2013). The role of diagrams and images, especially in science, Dove (2012: 237) points out, is not one of premise, but of evidence.

5 The Contributions in this Issue

Many of the objections to visual and multimodal argumentation are countered in the following papers of this special issue—some only indirectly, though, since the primary aim has been to bring the study of visual and multimodal argumentation one step further by addressing some of the most pertinent issues in the field.

Georges Roque examines the propositionality of visual argument. If we take the view of some logicians, Roque explains, a proposition is something that can be asserted or denied and is either true or false. Even though this may seem to indicate that images cannot be propositional, they actually can be propositional when images do state something about the world—and often these visual statements will be true or false. Furthermore, even though we may be forced to translate the visual propositions and arguments into verbal language, this does not differentiate them from verbal argumentation since we do the same kind of translation here: first we convert words into a sentence, then sentences into propositions, and then finally propositions into arguments. Add to this the fact that a vast amount of argumentation does not concern truth or falseness because it belongs to the domain of choice and values such as the imperative sentence demanding a certain action. So, some arguments are not propositional—in the traditional logical sense—since they have no truth-value. The conclusion is that we cannot reject the possibility of visual argumentation by claiming that images are not propositional: Firstly, because images actually can be propositional; secondly, because much argumentation—verbal, visual or otherwise—does not concern propositional truth-value. We must therefore, writes Roque, dissociate arguments and truth conditions and abandon the concept of propositions when considering visual argumentation.

In his paper on probative norms for multimodal visual arguments, J. Anthony Blair examines which norms are appropriate for evaluating the probative merits of visual arguments. His point of departure is that arguments are not themselves verbal or visual. Such forms of expression are simply means of communicating arguments. When someone attributes an argument to a text or an image, the argument is a construct of the interpreter; entitlement to accept reasons or infer conclusions are not a function of the mode they are expressed but of the nature of the reasons and the inference. Even though a visual rhetorical stimulus may be important, it does not alter the probative merits of an argument, and therefore visual arguments do not introduce or need new criteria for the evaluation of probative merits. Blair thus finds that there is “no distinctive visual logic”, and concludes that he is “inclined to believe that the interesting properties of visual argument are rhetorical, not logical”.

In line with Roque and Blair, Paul van den Hoven argues that even though a verbal propositional text is generally considered the prototypical argument, argumentation is actually a cognitive category. Because argumentation is a cognitive category or process, arguments do not need to be conveyed or reconstructed verbally. Furthermore, because there are no general or systematic mapping procedures between text structure and the argumentation an audience mentally represents on the basis of that text, we cannot require that all audiences are guided in “the same way towards representing an argumentation”. Actually, some argument structures probably cannot be verbalised in a general and clear way. Instead of focusing on the argumentative text, researchers should therefore focus on the process of argumentation. van den Hoven introduces a theoretical and analytic distinction between mimesis and diegesis in argumentation theory. The former represents the describing or showing of the phenomena that the discourse is about (descriptions, narratives, images, etc.). The latter represents the elements that guide the audience towards evaluations—such as arguments—of objects or events in the discourse world. van den Hoven emphasises that images, as well as multimodal narratives, “communicate an inextricable mixture of mimetic elements and diegetic elements”. The crucial question, however, he says, is whether the mimetic elements need to be formatted as sets of propositions. van den Hoven concludes by asserting that argument theory needs to develop reconstruction procedures that can deal with the mimetics of argumentation.

Leo Groarke advocates an argumentation theory that not only makes room for visual arguments, but also other non-verbal modes. The problem of expressing the verbally inexpressible, which both texts of van den Hoven and Kjeldsen point out, is addressed by creating what Groarke calls a “Key Component” table (KC table). In a diagrammatical way a KC table way isolates the key components of an act of arguing by identifying the premises and conclusions they propose, and by providing a rationale for interpreting them in the proposed way. The KC table can be used to summarise the meaning of a verbal argument, but Groarke has primarily developed it to identify and distinguish between key segments of multimodal arguments in order to prepare the way for discussion and evaluation of such multimodal argumentation. According to Groarke, not only verbal and visual elements or modes can perform argumentation, scents, taste, sounds or touch can also be used argumentatively. The KC table is developed to dress—or to standardise—an argument in any mode by categorising the act of arguing, the argument, and the mode(s) of arguing. Groarke’s contribution, then, is also an attempt at developing an understanding of different modes and submodes of arguing, as well as a theory of multimodal argument.

When standardising visual arguments in tables or models, or reconstructing them as verbal premises, one sometimes runs the risk of giving the impression that there is no difference between verbal and visual arguments. However, Jens E. Kjeldsen argues in his paper that even though we may offer the same argument verbally and visually or reconstruct a visual argument verbally, it is the form of expression used that is decisive. We may verbally reconstruct the premises of a visual or multimodal argument; however, when doing so we will lose a substantial part of the characteristic virtues of visual argumentation. Pictures have the potential for a multiplicity of simultaneous codings (transcriptions), allowing them to perform thick representations of events and phenomena. This kind of semiotic thickness can provide a fuller understanding of an issue, and thereby render the importance and strength of an argument salient. Thus, an important rhetorical and argumentative function of pictures, Kjeldsen proposes, is to argue for the significance or severity of a certain problem or the urgency of a situation. The visual aesthetics of pictures are able to function as an integral part of an argumentation we often only reconstruct verbally as thin propositions. However, the visual aesthetics are important precisely because their thickness can create presence and evoke the importance and urgency of a situation. In this way, visual argumentation may help people understand the gravity and importance of the issue at hand. This is essential in argumentation, because we cannot truly evaluate the strength and value of a deliberative argument if we do not fully understand the character and consequences of the actions the argument proposes.

Any emerging and maturing field needs the perspective of foreign eyes. We have therefore asked two researchers to consider the contributions in this special issue and the literature on visual argumentation in general in order to provide a brief commentary on the possibilities and benefits, challenges and problems to the study of visual and multimodal argumentation. David Godden has written his comment as a representative for the field of argumentation and philosophy, and Robert Hariman has offered his comments as a scholar of rhetoric and visual studies.