Mimesis as Make-Believe (6): Depiction vs Narration

In considering the tremendous diversity among representational art, Walton identifies two principal divisions: depictive representations, which are sensory in nature, and verbal representations, which are presented in language and often involve or imply a narrator. This week, we examine the nature of props of these distinct kinds.

Let us begin with depictive representations. Rather than providing a formal definition of what constitutes a depiction, Walton frames the issue via the example of a Dutch landscape painter (pictured above):

The viewer of Meindert Hobbema’s Water Mill with the Great Red Roof plays a game in which it is fictional that he sees a red-roofed mill. As a participant in the game, he imagines this from the inside. And this self-imagining is done in a first-person manner: he imagines seeing a mill, not just that he sees one, and he imagines this from the inside. Moreover, his actual act of looking at the painting is what makes it fictional that he looks at a mill... he imagines of his looking that its object is a mill. We might sum this up by saying that in seeing the canvas he imaginatively sees a mill. Let’s say provisionally that to be a “depiction” is to have the function of serving as a prop in visual games of this sort.

Walton asserts that “the phenomenal character of the perception is inseparable from the imagining which takes it as an object.” This is to say that one who looks at something like Hobbema's painting does not first see paint on canvas and only then interpret it as a watermill via imagining, but that seeing and interpreting the painting is one mental process. The same idea can be found in the philosophy of both Heidegger and Wittgenstein, and Walton in fact makes reference via a footnote to Philosophical Investigations, where Wittgenstein talks of “seeing as” as being an “amalgam” of seeing and thinking, noting that “the flashing of an aspect on us seems half visual experience, half thought”.

Sharpening the boundaries of what constitutes a depiction, Walton suggests that it is not enough merely for a representation to have the function of serving a prop in a visual game of make-believe. If it were, clouds would qualify as depictions when we imagine them as dragons and so forth. Rather, the games to be played with a depiction “must be sufficiently rich and vivid visually.” By rich, Walton refers to the extent a representation allows “for the fictional performance of a large variety of visual actions”, and by vivacity he refers to the intensity with which a participant is able to imagine “performing the visual actions which fictionally he performs”. In other words, a depictive representation in Walton's terms must support fictional worlds that are something more than trivial in nature, otherwise they may still qualify as props (recall Gregory and Eric's tree-stump bears) but they will not qualify as a depiction in the way he means the term to be understood.

Inherent to depictions appears to be the process of translation, thus three dimensional objects are shown in two-dimensions in a picture, and both sketches and sculpture render colourful subject matter monochromatically. Size, motion, time... all these elements can be altered in the process of depiction, which at its heart seems to be involved not in perfect reproduction (although some modern art has leaned in this direction) but in a kind of planned distortion of the world. This idea is, in many respects, tied up with the term mimesis, which since the ancient Greek philosophers has meant imitation not duplication. If a craftsman produces a perfect copy of Michelangelo's David it is a forgery of (or tribute to) the original. But if they produce a lithograph of the same artwork, it is a depiction of the original rendered in ink in the same way that the statue itself is a depiction of an ancient king of Israel rendered in marble, regardless of what the historical king may have looked like.

In mentioning that depictions were sensory, I alluded to Walton's willingness to extend the notion of a depictive representation beyond the merely visual. He writes:

Let us broaden our understanding of “depiction” to include representations that are auditory or tactile or otherwise perceptual in the way paintings are visual. A depiction, then, is a representation whose function is to serve as a prop in reasonably rich and vivid perceptual games of make-believe.

But this may at first seem strange, since it is clear that – sound effects not withstanding – representation via sound is not a great deal akin to painting, and certainly music does not generate fictional truths in anything like the manner of visual props. Walton makes the point that while a visual representation prescribes visual imaginings, representational music generally doesn't depict auditory imaginings. He suggests that music involves not imagined perceptions, per se, but rather imagined experiences: “The listener imagines experiencing excitement, passion, fervour, despair, conflict, feelings of exuberance, of striving, of determination, of well-being, of trepidation, of repose.” Walton suggests this music is well suited to this kind of  “fictional introspection... because introspecting is in some way more like hearing than seeing.” As a result, musical depiction may generate game worlds, but it does not obviously have a work world, or if it does, it will be much harder for us to agree as to its contents.

Let us move onto the use of words to create props. Like the sensory representations Walton terms depictions, verbal representations also involve a translation – but here the transition is from words into specific sensory and experiential imaginings. It may well be the case that 'a picture is worth a thousand words', as the saying goes, but using just a thousand different words we are able to imagine almost infinite numbers of different things. Walton notes that “words are well suited for use in make-believe” precisely because their combination is so versatile in what can be prescribed to imagine. But of course, not all written words are representations, and Walton suggests that the boundary of qualification in the case of language is that any collection of words that appears to issue prescriptions to imagine qualifies as a prop, and thus “if it is its function to be a prop, it is a representation in our sense.”

Note that this concept of verbal representation does not extend to the words spoken by actors of stage or screen:

Words uttered by actors are depictive. In hearing words pronounced on stage by Sir Laurence Olivier, the playgoer fictionally hears Hamlet speaking. In these cases the utterances or marks are reflexive depictions; they depict themselves.

(This makes, for the most part, plays and movies into depictions and not verbal representations, except where a verbal representation is embedded within the story – as in the opening crawl of a Star Wars movie, for instance).

Although Walton concedes that verbal representations have nothing in common but their use of words, he notes that there is one particular form that is especially central to the form: narration. He suggests this is the “historical ancestor” of many other forms of verbal representation, and even goes so far as to suggest there may be utility in treating almost all verbal representations as forms of narration, distinguishing between those works with an explicit narrator (such as all first person novels) and those in which the narrator has been effaced or de-emphasised (such as most third person novels). He acknowledges that this is a step of convenience, but pragmatically one cannot avoid the fact that all words have been written by someone and as such imply at the very least a situation which is a close analogue to narration.

Of course, talking of a single narrator is itself a simplification:

There need by no such thing as the narrator of a narrated work. Different narrators can replace one another in quick succession. In most plays (written or performed) there is a new narrator for every several lines of text; each of the speaking characters is the narrator, in our sense, of the lines attributed to it. Novels which, in representing conversation, omit the “he said”s (and perhaps some which do not as well) can be understood similarly...

Whether there are one or many narrators, the role of the narrator is to mediate our access to the events of the story being told, usually presented indirectly. When the narrator has been effaced, omniscience comes into play to soften this mediating effect and provide “immediacy” to a degree which approaches the norm for depictions. The use of omniscience can be misleading, however, for in a third person story it is generally not fictional that the narrator is supposed to be literally omniscient (nor “godlike or telepathic or clairvoyant or disincarnated or supernatural” – at least, not in the work world nor in any authorised game). It is simply a convention to enable access to every aspect of the fictional world for the smoother presentation of the story.

Walton's concept of the role of a narrator is thus that:

...fictionally, the narrator speaks as though he himself were, in many respects, in the epistemological position he attributes to the character, reporting what he takes the character to know and remaining silent about what he takes the character not to know.

We are dependent upon the narrator (whether literal or effaced) for all our information concerning the fictional world – and this is true even when the narrator is unreliable, and we disbelieve what the narrator says. In the games we play with a verbal representation as a prop, it is even fictional that our access to what happens comes via the narrator, a situation which is markedly different from depictions. The artist creates the depiction, but thereafter we do not generally need to include the artist in the fictional world we enter into with the work as a prop.

Walton goes on to explore many of the complexities of written narrative, using a great many novel examples that sadly are too detailed to explore in such a brief serial. Interested readers must take up the book itself if they wish to learn more of this. For now, we have seen in broad strokes the elements of the make-believe theory of representation – props, fictional worlds, principles of generation, quasi-emotions and lastly distinctions between depictive and narrative representational forms. All that remains is to see how Walton's theory can be applied beyond the world of art, to the interpretation of reality itself.

Next week, the final part: Truth vs Fiction