Universal Fictions

Comments on Wildman and Folde's "Fiction Unlimited"

Nathan Wildman and Christian Folde, “Fiction Unlimited,” The Journal of Aesthetics and Art Criticism, Vol. 75, No. 1 (Winter 2017), pp. 73-80.

…truth is limited, in the sense that only a proper subset of the totality of all propositions is in fact true—for example, it is true that snow is white and false that snow is black. Yet fiction might be unlimited—that is, it is an open question whether there is (or could be) a universal fiction within which every proposition is true. (p. 73)

Is it possible to have a fiction in which everything is true? Even the possibility of such a fiction is a philosophically interesting one. Wildman and Folde argue that these fictions, called universal fictions, can exist. They set aside any questions about the particular account of what it means to be a fictional truth, and instead focus on whether anything could be a fictional truth. In order to consider this, they make a distinction between the primary content of the fiction (what is explicitly true in the story) and the secondary content of it, and then divide the latter into two kinds, imported content (which is brought into the fiction from outside, e.g., by being assumed by it) and entailed content (which nontrivially follows logically from the primary content and imported content). (These we have already discussed in looking at Wildman’s argument for the opposite of universal fictions, empty fictions.) This lets them break their question into two:

1. Are there possible fictions whose primary content includes all propositions?

2. Are there possible fictions that include all propositions, whether in primary or in secondary content?

To answer these questions, they make a few assumptions. First, they assume that there are inconsistent fictions. Second, they reject the Routley route to universal fictions, which posits a fiction that has a proposition like, “Everything is true,” because we cannot be sure the quantifier, ‘everything’ actually includes all propositions. Quantifiers in stories generally only quantify over what is relevant to the story, so to assume that ‘everything’ already ranges over all propositions requires assuming that they are somehow in the story (whether in a primary or secondary way). Likewise, they don’t want to assume that authorial intention can somehow make ‘everything’ range over all propositions. So there needs to be some other route to universal fiction.

They therefore propose a fiction as a universal fiction, “Monsieur Impossible”:

In the Kingdom of Classicalia, where Classical Logic holds, the most famous and wondrous of the King’s Musketeers is Monsieur Impossible. (Of course, if one is a member of the King’s Musketeers, then one is employed by the King!) Rumored to be from the far-away land of Australia, Monsieur Impossible is the very epitome of the Musketeer ideal. But what is so impressive about him is that he has exactly two hands and does not have exactly two hands. And this incredible power—to have and not have exactly two hands—makes him the [most] deadly sword fighter around, the elite among the elite! Of course, it also makes him the worst swordfighter around, as well as the best (and worst!) card player. (p. 75)

It is part of the primary content of the fiction that Classical Logic holds; it is part of the primary content of the fiction that Monsieur Impossible has exactly two hands and does not have exactly two hands. Therefore it must be part of the entailed comment of the fiction that every proposition is true. This arises, of course, by the principle of contradiction explosion in classical logic; in classical logic, everything follows from a contradiction. Since it is the principle of contradiction explosion that does the work here, they propose a second story in which it explicitly occurs, “Ohle’s Amazing Adventure”:

One day, Ohle the wonder dog set out on a wander through Explodiberg, a land governed by the principle of explosion (which, as we all know, states that from a contradiction, anything follows). During his adventure, Ohle ate exactly three treats and not exactly three treats (rather,exactly four). Doing so, he brought about the greatest calamity Explodiberg ever saw, since everything followed in his wake. (p. 76)

It is part of the primary content of “Ohle’s Amazing Adventure” that Explodiberg is governed by the principle of explosion; it is part of the primary content of it that Ohle ate exactly three and not exactly three treats. Therefore the entailed content includes every proposition. What is more, of course, you could have endlessly many such universal fictions; just make both the principle of contradiction explosion and some contradiction a part of the primary content.

Both “Monsieur Impossible” and “Ohle’s Amazing Adventure” give us a ‘yes’ answer to the second question above; they include every proposition in their secondary content. From this we learn a few interesting points:

1. Fictions are not necessarily incomplete; the possibility of universal fictions shows that you can have fictions such that, for all propositions, either that proposition or its negation are part of the fiction. (In universal fictions, of course, both are.)

2. Universal fictions are different despite the fact that they have all the same propositions, so simply having the same propositions is not enough to identify fictions with each other. The reason for this, of course, is that, while they have each have all the same propositions, they do not ‘cut’ them the same way. They put different propositions in their primary content.

3. Universal fictions also establish the truth of the principle of poetic license, “Every proposition is such that it can be in the content of some fiction” (p. 73). Since the principle of contradiction explosion reaches every proposition, the explosive character of universal fictions based on it guarantees that the entailed content of the fiction includes them all.

4. Some people have argued that certain propositions cannot be fictionalized. This is usually discussed under ‘imaginative resistance’; Wildman and Folde call it the alethic puzzle. Wildman and Folde argue, given universal fictions and the principle of poetic license, that every proposition can in fact be fictionalized, but their argument (as they recognize) only establishes this if we are including secondary content. It would still be possible to argue that some propositions cannot be found in the primary content.

There are some limitations and possible objections that have to be considered. Their argument for universal fiction does require some kind of closure principle. “Monsieur Impossible,” for instance, has to be closed under classical logic. Someone could block the route to universal fiction by arguing that it is not, in fact, closed in this way. If this were the case, it seems “Monsieur Impossible” would still be an inconsistent fiction, but not a universal one. Wildman and Folde respond by suggesting that they only need the principle of fictional modus ponens:

FMP If (p -> q) and p are both part of fiction f’s content, then q is also part of f’s content. (p. 77)¹

A potential difficulty is that non-universal inconsistent fictions often seem not to require, or even to reject, FMP; but Wildman and Folde suggest that they do in fact include FMP, just lacking other elements to make them universal. We ourselves are not convinced of this; it seems that we could have an inconsistent fiction in which FMP was explicitly denied in the primary content. In any case, they consider a further, and perhaps more serious objection, namely, that perhaps fiction does not have a single logic. We won’t go into their discussion here, although it is interesting in its own right. To address it, though, they give a third example of universal fiction, “Clara’s Crazy Caper”:

Exploring the castles, creeks, and crags of the canton of Concorida, where the principle of explosion—which states that, if p & ~p, then q (for every q)—holds, Clara discovered a conclave of carrots. Feeling hungry,Clara consumed exactly three and not exactly three (but rather four) carrots. (Of course, if one has consumed exactly three and not exactly three carrots, then some carrots have been consumed.) Consequently, filled to the brim on crispy carrots, Clara cavorted away, creating chaos. (p. 78)

A possibility that Wildman and Folde do not adequately consider arises from the fact that we cannot simply identify primary content with what is explicitly stated in the story, because not everything explicitly stated in a story is true in the story. Some of the things explicitly stated in some fictions are false in the story; this is clear in cases like lying narrators. Recognizing this creates a bit of a puzzle. Suppose someone says that contradictions in primary content are always false (e.g., they should always be taken to be stated by a lying or mistaken narrator), so their presence shows that the statement is not part of the primary content. The usual Wildman-Folde approach to addressing objections is to add some explicit statement to the primary content. No explicit statement one could add would address this problem. Thus we see that there is an interpretive element being presupposed in the background, one that at least does not seem to be necessary. They do in a footnote (p. 80n7) recognize that one of the ways we handle contradictions in fiction is by attributing them to unreliabilities in the internal narrator, but they only answer this very weakly by saying that it is unclear that every fiction has an internal narrator, and noting that there are fictions without narrators, and that the fact that this is an interpretation does not mean that it is the best interpretation. That fictions can fail to have explicit narrators, however, is irrelevant; if Wildman and Folde can allow entailed content, people can have implicit narrators. And even if there weren’t always internal narrators, one could very well take inconsistency to be attributable to one. Further, even if there are fictions that happen to be without internal narrators, we know that there is every fiction that can be stated is such that it could have an internal narrator; just imagine it being narrated by one. Thus we can’t evade narrator-based objections in this way.

Nonetheless, while there are a few issues that require further study, we are inclined to think that Wildman and Folde have proven the point, at least as far as making it a reasonable presumption until proven otherwise: It is indeed possible for there to be universal fictions. All of the objections seem not so much to prove that there can’t be universal fictions, but at most to add possible further conditions for their existence. However, we will need to discuss one of the most important arguments against universal fictions, as well: Michel-Antoine Xhignesse’s “Exploding stories and the limits of fiction,” which we will discuss another time.

Tsukioka Yoshitoshi, Masakiyo’s Challenging Battle (1866), triptych of prints from the Taiheiki chronicle, courtesy of The Met Open Access Collection. The triptych depicts soldiers caught up in an explosion.

1. They do note, however, that they also have to include in the primary content of a story something like, “If p & ~p, then q (for every q).” This can be granted in classical logic, but the potential difficulty is that someone might object that being entailed does not necessarily imply being true in the fiction.