What constitutes a prop? Any and all representations count as props in Walton's system, and as a result the boundaries of the term are relatively soft. For instance, when watching a play, one views many different props – including the furniture and objects on stage (things already termed 'a prop'), the backdrop painting, and even the actors and actresses themselves. When one watches a movie, one could consider the film to be a prop, or comprised of many props. It is unnecessary to distinguish between these states of affairs in order to use the make-believe theory.
What is important is not which objects are ascribed the role of prop, but that props are generators of fictional truths – things that by their very nature render certain propositions fictional. A child's doll of an infant makes it fictional that there is a baby present, so if a loutish child kicks such a doll it is fictional that they have kicked a baby, which is to say that it is prescribed that any observer imagines that a baby has been kicked. Props generate their fictional truths irrespective of what is or isn't imagined, but they cannot do so on their own: there must be a person or people to imagine, and thus props function primarily in a social context.
If we can imagine more or less anything, how can it be that representations must be understood at a social level? It is for precisely the same reason that Wittgenstein advocates understanding language as socially embedded, and consequently denies the meaningfulness of a 'private language'. We learn to interpret certain kinds of representations just as we learn to speak a language – and we learn both these skills from the people around us, although this is not to say that all interpretation of representations is learned. This seems surprising, and intuitively we might suspect that we would always be able to interpret a picture of a mouse as representing a mouse. In fact, anthropologists have found tribes who lack this form of representation in their culture, and they are incapable of interpreting such a drawing without some instruction.
Furthermore, social elements are involved in the authorisation of the games that are played with particular props. Walton notes:
…dolls and toy trucks are meant to be not just props but props in games of certain kinds, ones in which they generate certain sorts of fictional truths: dolls are intended to “count as” babies and toy trucks as trucks. I will call games of the kind a given prop has the function of serving in authorized ones for it.
Thus, for instance, nothing stops us imagining that a baby doll is a murderous robot but the fictional world of the game where we make-believe this is so is not authorised for the doll: the only (socially) authorised games for dolls of this kind are those in which it is fictional that the doll stands for a baby. (We will return to this notion of an authorised game in a few weeks).
Among the philosophers who have attempted to tackle the problems of fiction, a common approach has been that fiction should be considered an aspect of our actions; that when we make a fictional utterance this is the root of the fiction. Walton sees this issue differently, and contends that the notion of fiction (and indeed of a story) is better associated with objects – props – rather than actions. This may seem odd: if I tell a short story that begins “There once was a little red train” it may seem that there is no object here to draw upon, and we must turn to my illocutionary act of speaking the words. But the prop in this case is the sentence. The words constitute a prop which prescribes we imagine a little red train.
Note that any given person will imagine a different train in response to this sentence – which doesn't, after all, specify whether it's a steam train, an electric train, a subway train or indeed a wedding train. This may seem to give problems in respect of the idea that a prop can be associated with a particular fictional world, as what is imagined in each of the games people play in respect of this sentence could be wildly different. Walton proposes that we view what is fictional about any given prop to be solely that which would be fictional in any (authorised) game in which the prop was used to prescribe imaginings. We can say that the above sentence is associated with a work world in which it is fictional that there is a train, and the train is both red and little.
An example concerning a visual representation may serve to clarify this idea. Van Gogh's painting Starry Night (pictured above) can be seen to comprise a night sky with stars and a crescent moon, a small town, and a dark object that partially occludes both. It is fictional in the work world of this painting that there is a town, and that it is night, and the moon is crescent. Other features are more ambiguous. A particular appreciator of the painting may interpret the dark object to the left as a tree, in which case it is fictional in the game they play using this painting as a prop that there is a tree. But it is not fictional in the work world of the painting that there is a tree – it could also be interpreted as a crag, for instance. The same objection does not validly hold for the town: one could imagine in one's own game that there was no town, perhaps there is just a series of cardboard cutouts in the shape of a town. But this is not an authorised game for this painting, for few if anyone would doubt that Van Gogh was painting a town, whatever their interpretation of the dark object.
However, not all props have their own fictional worlds. It is possible to create ad hoc props, by playing a game of make-believe that follows a particular rule, for instance. Walton frequently draws upon the example of two children, Gregory and Eric, who have decided to imagine that all tree stumps they encounter are bears. In their game, tree stumps are ad hoc props – the stumps do not prescribe any imaginings in and of themselves, they are merely wooden stumps. It is only in the game that Gregory and Eric play that the stumps are props which make it fictional in their game that there is a bear present. Gregory and Eric may act startled when they suddenly notice that there is a bear right there that they didn't see before, but this prescription to imagine belongs to their game, and not to the stump they encounter.
Representations, however, always generate some fictional truths, and thus have a fictional world (a work world) associated with them. The tree stumps in the previous example, while certainly props in the relevant sense, are not representations. Walton writes:
Representations generate fictional truths by virtue of their features – the marks on the surface of a painting, the words of a novel, occurrences on stage during the performance of a play – in accordance with principles of generation… There is uncertainty and disagreement, in many cases, about what principles of generation are applicable to a given work.
These principles, by which the fictional truths are implied or generated, are the subject of next week's instalment.
Next week: Principles of Generation