CCO, Works of Pure Fiction, and BFO's Realism

[gregfowlerphd]

I have an interest in the ontology of entertainment industries and, as some folks might know, I’ve just recently begun developing a BFO- and CCO-conformant ontology in this area. This interest has led me to wonder how CCO deals with works of fiction－in particular, with works of pure fiction. [1]

A work of impure fiction is a work of fiction that features one or more actually existing entities (persons, places, things, times, etc.), such as the Spiderman comics (NYC, etc.) and Harry Turtledove’s How Few Remain (Abraham Lincoln, 1881, and so on); such a work is partly about those entities. [2] Works of pure fiction, on the other hand, aren’t (at least prima facie) about any actually existing entity: No actual person, place, thing, time, etc. is featured in such a work. Examples include the Star Wars saga and Game of Thrones (the first book in George R.R. Martin’s A Song of Ice and Fire series).

Works of pure fiction seem to pose a puzzle for CCO, and perhaps for BFO more generally. Consider two copies of Game of Thrones, one in English and one in Chinese. There seems to be something shared in common between these copies, just as there is between an English and a Chinese copy of a nonfiction work. In the latter case, the shared entity is an Information Content Entity, and that seems to be the most natural option in the former case as well. [3] But an ICE must, by definition, be about some entity, and there appears to be no such entity in the case of works of pure fiction, ruling out that option. Are there others? The only other option that readily presents itself, within the framework of BFO, is that the shared entity is a pattern (as in the phrase ‘the content or the pattern that multiple copies would share’ from the definition for GDC). However, given the vast differences between the languages, it’s hard to see what pattern an English copy and a Chinese copy of Game of Thrones share in common, so this option seems unavailable as well. Hence, CCO seems to have nothing with which to identify that which is shared in common between the two copies of Game of Thrones, with similar remarks applying to other work of pure fiction as well.

(One manifestation of this puzzle involves the CCO class Television Broadcast. This class is a subclass－albeit indirect－of Act of Communication, which is defined as follows:

A Planned Act in which some Information Content Entity is transferred from some Agent to another.

Given this definition and the subclass relation, an ICE must be transferred in every television broadcast. However, some television broadcasts are of works of pure fiction, so－given the argument of the preceding paragraph－there appears to be no ICE transferred in such broadcasts. The existence of such broadcasts thus seems to reveal a problem with either the definition or the subclass assertion.)

A potential resolution to this puzzle (which would also solve the apparent problem involving television broadcasts) would be to postulate that while works of pure fiction aren’t prima facie about any actually existing entity, there in fact are actually existing entities that they’re about. In particular, they’re about fictional characters, [4] places, things, times, etc.

However, this resolution might seem to conflict with BFO’s realism. Indeed, I seem to remember seeing the fact that other top-level ontologies include fictional entities being used in an explanation of how BFO’s realism differentiates it from those TLOs.

That said, it might be the case that it's not the mere inclusion of fictional entities that conflicts with BFO's realism but rather the particular treatment of these entities in other TLOs. Consider a treatment that maintains, a la Peter van Inwagen and Amie Thomasson, that fictional characters (and other fictional entities) come into being in the process of creating a fictional work and generically depend on the copies of the fictional works in which they appear (as well as, perhaps, on the intentional states of agents, such as their authors). [5] On this treatment, fictional entities are quite similar to ICEs themselves, so since the inclusion of the latter doesn't conflict with BFO's realism, perhaps the inclusion of the former doesn't either. (I'm hoping folks can let me know whether it nonetheless does.)

Regardless of how it's resolved, however, works of pure fiction at least seem to pose a puzzle for CCO, as argued above.

1 - I should note that I’ve talked to a few people about this topic, but they haven’t been able to direct me to CCO projects covering works of fiction. Any pointers would be appreciated!

2 - It’s an interesting question whether, to take just one case, Lincoln－the actual man－should really be taken to be a character in How Few Remain, or whether we should instead say that it features a character based on Lincoln. (If the latter, then perhaps all works of fiction are works of pure fiction.) But I’ll put that issue aside for the sake of discussion.

3 - There might be reasons other than its naturalness to favor the ICE option too. For instance, while I’m no expert in psychology, I’m somewhat inclined to argue as follows: At least at first glance, the cognitive processes involved in consuming a work of pure fiction are similar to those involved in consuming a work of nonfiction. So, since ICEs are involved in the latter, it seems like they should be involved in the former as well.

4 - Fictional characters are basically fictional people, or at least fictional agents (in case we don’t want to count fictional aliens or fictional elves as fictional people).

5 - Just to spell things out a bit more: On this treatment, to take just one example, a fictional detective isn’t a detective: Detectives are material entities (who bear a role of a certain sort), whereas fictional detectives are generically dependent continuants. And similarly for other fictional entities－a fictional place isn't a place, a fictional time isn't a time, etc.; rather, they're all GDCs.

[gregfowlerphd]

Perhaps worth noting that when modeling in this area, the issues discussed in this post seem pertinent to more than just the categorization of works of fiction themselves (e.g., ICE or not?). For instance, the notion of the genre or category of a work of fiction seems, implicitly, to concern what the work is about.