Can pictures be arguments?

• Author: Fleming, David
• Position: Visual Argument - Part 1

My purpose in the following paper is to consider whether the term "argument" should be extended to include pictures.(1) Clearly, a drawing or photograph independent of words can influence the thought and action of others; but can it, I wonder, argue? And if we say that it can, do we risk losing something important in our conventional understanding of argument? Below, I elaborate more fully what I mean by "conventional understanding of argument"; as for "picture," I define it here as any representation meant to look like the thing it represents. Thus, a sketch, drawing, painting, photograph, or three-dimensional model would be a "picture" if it were constructed to resemble imagined or actual objects in the visible world. Now, the concept of likeness has been much criticized of late; as Mitchell (1986) and others have pointed out, the picture is often (if not always) every bit as opaque and cultured as language. But without some notion of visible likeness, it is difficult to see how one would make sense of the everyday, and presumably useful, distinction between "showing" and "telling." I define "picture," then, as an artifact constructed to be iconic with the external world; and my question is, can such a thing, independent of language, be an "argument"?

Many scholars are now convinced that the application of the word "argument" needs to be broadened. Willard (1989, p. 109), for example, has rejected the traditional association of the term with propositional discourse. Argument is interaction around incompatibility, he has written, and any attempt to distinguish it from other forms of persuasive communication is driven by "a bureaucratic rationale." Similarly, for Hesse (1992), all discourse is argumentative because all discourse is productive of belief. Narrative persuades because its "emplotment" of events over time engenders belief in acceptable sequences of action. It involves, that is, some propositions causing other propositions. At the same time, Hesse writes, what we have traditionally referred to as "argument" is itself a kind of narrative, a believable sequence of propositions plotted over time in such forms as the enthymeme.

Fisher and Filloy (1982) agree. Although argument has typically been conceived in terms of inferential structures (claims and reasons), people also "arrive at conclusions based on 'dwelling-in' dramatic and literary works" (p. 343). The mode of arguing in such works is "suggestion": an immediate, emotional response becomes a reasoned belief through critical interpretation (p. 347). "Aesthetic" proofs, then, are those structures of suggestion which are invented by an author, experienced by an auditor, and used by a critic to substantiate a reasoned interpretation (p. 347). On this line of thought, it is hard not to see argument "everywhere," as Brockriede (1975, p. 179) had proposed. Couldn't a picture, by virtue of suggestion, engender belief in an argumentative sort of way? Yes, say Medhurst and DeSousa (1981), for whom political cartoons are a kind of enthymeme, relying on socially-sanctioned presuppositions to produce reasoned belief and action in others. Cartoons, that is, argue for political positions by adducing acceptable (albeit unspoken) reasons to hold those positions. Similarly, for Buchanan (1989, pp. 97, 107), arguments can inhere in things as well as words. When material objects solve problems in a reasonable manner, they are persuasive in the same way that verbal arguments are. Deliberation and use reveal the premises of such objects and arrange them into inferential structures. According to Buchanan, the Krups coffee mill, for example, is an "argument" for use and aesthetics over technology.

Paintings are also susceptible to this kind of analysis. For Varga (1989), a 13th century painting of St. Francis of Assisi is a kind of argument; the portrait in the center is claim, and the narrative episodes flanking it are evidence for that claim: he who lived thus, who did these things, should be revered. Similarly, Kessler (1993) has written that medieval images often contained arguments. Not only were such images used as evidence in verbal disputes; their visible nature was itself an argument for various contested beliefs. The painting of Christ over a sketch of the Jewish tabernacle, for example, was not just a pictorialization of a prior verbal text; it was an argument for the New Testament supplanting the Old, the Incarnation of God replacing the shadowy presence of the Law.

Others, however, have resisted the idea of non-linguistic argumentation. Perelman and Olbrechts-Tyteca (1969), for example, maintain that the theory of argument should concern itself only with "the discursive means of obtaining the adherence of minds" (emphasis in original, p. 8). An action "designed to obtain adherence falls outside [italics added] the range of argumentation," they write, "to the degree that the use of language is lacking in its support or interpretation" (p. 8). Van Eemeren, Grootendorst and Kruiger (1984) also consider argumentation to be necessarily verbal:

Argumentation requires the use of language. A person engaged in argumentation makes an assertion or statement, assumes or doubts something, denies something, and so on. For the performance of all these activities he must utter words and sentences (whether spoken or written). Besides these verbal means he can, of course - just as in any other verbal activity - employ non-verbal means (e.g., facial expression and gestures). To the extent that these fulfill an argumentative function, they can always be explicitized verbally. However, non-verbal means of communication can never completely replace verbal ones: argumentation without the use of language is impossible.

This means that a person engaged in argumentation has deliberately opted for the use of verbal means. He clearly prefers to use words rather than non-verbal means of communication; he speaks, rather than resorting to blows or other forms of violence. Here we have another excellent reason for calling argumentation a pre-eminently verbal activity. The arguer uses words to lend force to his words. (p. 3)

For Kneupper (1978), non-verbal signs may be used in argument, but they don't function as arguments unless linguistically translated. And Balthrop (1980, p. 185) claims that argument is "inherently discursive and linguistic." The notion that non-discursive or artistic forms can be arguments, he writes, "would preclude argument from fulfilling" its reason-giving or justificatory function (p. 188). Finally, Toulmin, Rieke and Janik (1979, p. 137) write that "reasoning could not exist in the absence of language. Both claims and all the considerations used to support them must be expressed by some kind of a linguistic symbol system."(2)

Let's try out a fairly conventional definition and see how pictures stack up. An argument is an intentional human act in which support is offered on behalf of a debatable belief. It is characterized first and foremost by reasonableness. Now, to say that an act or object is "reasonable" is, first, to assert that reasons can be adduced in order to make that act or object acceptable to some audience (what is acceptable is, of course, local and indeterminate, but the act of adducing reasons to make something acceptable is, I believe, cross-situational). An argument, in other words, involves a two-part relation, one part (evidence, data, proof, support, reason, etc.) supporting the other (position, claim, assertion, conclusion, thesis, point, argument, proposition, etc.). Second, to say that something is "reasonable" is to assert that it admits of improvement, is corrigible, refutable, accountable; it is an act or object which can be interrogated, criticized, and elaborated by others (and even invites interrogation, criticism, and elaboration). An argument exists, that is, in a specifiable context of debate, controversy, opposition, or doubt; its position is thus necessarily contestable.

Now, whatever else a picture can do, it cannot satisfy these two criteria. First, it lacks the requisite internal differentiation; it is impossible to reliably distinguish in a picture what is position, and what is evidence for that position. The distinction at the heart of argument, the difference between that which asserts and that which supports, is thus collapsed. Second, a picture cannot with reliability be refuted, opposed, or negated. It can be countered but only by introducing words into the situation; the picture itself makes no claim which can be contested, doubted, or otherwise improved upon by others. If I oppose the "position" you articulate in a picture, you can simply deny that your picture ever articulated that, or any other, position. The picture is only refutable if first translated into language - in which case we have left the realm of pictures altogether.

Now, we can ignore these problems and call pictures "arguments" anyway. But to do so would, I believe, deprive the term of its significance. Because if a picture can (self-sufficiently) be an "argument," then what do we call the linguistic act of asserting and supporting debatable claims? To say that a picture can be an argument is to leave individuals with the impression that they have argued for something when they have merely placed it in someone else's field of vision. Further, to claim that a picture can be an argument is to make it less likely that analysts will attend to the rhetorical functions that pictures can and do serve.

An Argument Contains Two Parts

Traditionally defined, an argument has two parts.(3) These go by many names, and there is some disagreement as to which of the two, if either, needs to be linguistically explicit (see O'Keefe, 1982). But the belief that an argument has two parts - "claim" and "support" - is a cornerstone of Western thought. The practice of bringing forth statements that explain, support, question, and comment on other statements...