(1) A work is a work of fiction iff its claims are false, non-fiction iff its claims are true.
Though this might seem right at first blush--after all, non-fiction works contain(among other works) such works as reference works, works comprised of facts--whilst works of fictions, such as The Adventures of Huckleberry Finn, do not have "facts"--after all, Huckleberry Finn wasn't real.
The first stab doesn't pass even a first inspection, though. Both sides of the biconditional (A work is a work of fiction iff its claims are false) seem problematic. . First, it could be argued that the claims of fiction are not false because the proposition expressed by each fictive claim is of the form "According to work W, X" as opposed to "X." On such a view, the utterance "Huckleberry Finn was a friend of Jim" is true because the proposition it expresses, ["In The Adventures of Huckleberry Finn, Huckleberry Finn was a friend of Jim"], is true. This aligns with our pre-theoretical thinking that "Huckleberry Finn was a friend of Jim" is correct in a way that "Huckleberry Finn was a friend of Nurse Ratched" is not. Along similar lines, the conditional (if a work is a work of fiction, its claims are false) could be rejected on the grounds that works of fictions contain claims that are not truth-apt and so not false. The idea could be that fiction, by nature, is non-representational discourse and so is capable of truth or falsity no more than commands or expressions of attitudes are. So it's false that: if a work is a work of fiction, its claims are false. The other side of the biconditional can be challenged as well: a work can be comprised of false claims and not be a work of fiction. Suppose someone wrote a very poorly researched bibliography of former U.S. President John Adams. Suppose that each sentence in the biography was nearly true (verisimilitudious) but strictly speaking false. The biography for instance says that Adams came in second to Washington in the 1789 Presidential Election vote by a margin of 69-35 votes in the electoral college. Adams actually lost the 1789 Presidential Election by a margin of 69-34 votes. So this claim is false. Suppose our biography is full of claims like this. This work is a work whose claims are false, but it doesn't seem to follow that this rubbishly researched biography of Adams counts as fiction, as a book that belongs next to novels and story books, for instance. It seems instead to be just a very bad example of non-fiction. This consideration counts equally against the suggestion in (1) that a work is non-fiction iff its claims are true. The bad Adams biography is an example of non-fiction whose claims are false. So truth and falsity seem to be a poor litmus test for distinguishing between fiction and non-fiction.
A prominent philosopher of fiction Kendall Walton has proposed that we can individuate what counts as fiction by appealing to the concept of imagination rather than to the concepts of truth and falsity. Roughly, what individuates fictive utterances Walton thinks is that they (unlike non-fictive utterances) are a prescription to imagine. This account seems to capture something important about fiction: fiction involves a sort of make believe, and to read (in the case of written fiction) and enjoy fiction is to participate in the author's invitation to imagine. An author of non-fiction extends a different invitation: an invitation to believe, and perhaps to know (on testimony). Does Walton get things right then? Maybe. There are quite a few objections to his view. Here's one I'd like to consider. Suppose Dostoevsky had an uncle named Raskalnikov, and who seemed to Dostoevsky to be shrouded in mystery, leaving only a trunk full of documents in an attic that detailed his life. Suppose Dostoevsky wished to write a biography of Raskalnikov and so read through the documents in the attic, finding all sorts of court documents, diaries and other evidence that suggested Raskalnikov had murdered his landlady as well as a visitor Lisabetta. Dostoevsky then writes up a large biography and entitles it "Crime and Punishment*." Suppose further that Crime and Punishment is word-for-word identical with Crime and Punishment. Crime and Punishment* is not a prescription to imagine. Crime and Punishment is. So Walton's view suggests that fictionality must be a relational property; two works with all the same intrinsic properties can be such that only one is a work of fiction. Perhaps this is not problematic? I think this result is awkward and so counts against an otherwise very good account of fiction. Any thoughts?
This may be a unthought out idea, but could it be the facts in non-fiction are mind independent facts and in fiction they are not?
ReplyDeleteFor example stick insects eat leaves is a fact independent of whether it is believed by anyone.
But winnie the pooh likes honey is not. It depends on the belief that winnie the pooh exists and is a certain way.
I wonder if this would work for facts about theories though....
Naomi =)
Aren't the librarians, for purposes of convenience, creating a false dichotomy out of something that is basically a continuum, at least when referring to longer written works. Even a fantasy novel contains elements representing faithfully recorded observations by the author or a secondary trustworthy source ("fact"), and dry academic histories still contain conjectures and extrapolations, couched in language that invites the reader to exercise his imagination ("fiction").
ReplyDeleteI write both fiction and nonfiction though most of my published work is the latter, and I participate in writer's groups where the vague, shifting line between fiction and nonfiction is often a topic of discussion.
Let us consider an actual example, and its hypothetical counterpart. Some years ago, for purposes of a historical novel I was writing, I introduced a debate in Irish Parliament in 1796, which I subsequently discovered had actually taken place in roughly the form I had represented it. I can readily imagine the converse, writing a scholarly article on Irish parliamentary proceedings relying on secondary sources which quoted the speakers in this debate as having said things they did not actually say. In the first case, did the identical work morph from fiction to nonfiction, and in the second, did the article become fiction when it developed that the sources used were unreliable? I don't think so. I think the overriding concerns here are (a) the intent of the author when he wrote the passage in question, and (b)the overall character of the work in question.
Hi Adam, this is really interesting. Something in Walton’s account appeals to me, and I was wondering whether it would still seem as problematic when we adapt it to Lewis’s possible world’s framework? In this case, Crime and Punishment, as you mentioned, would have a counterpart, Crime and Punishment*, in W*. As well as this, though, the Dostoevsky of our world would have a counterpart, Dostoevsky*, in W*, who would in fact have had an axe-wielding murdering uncle named Raskalnikov, and an aunt, Avdotya. In this case it seems that Crime and Punishment really would be an invitation to share in the author’s imagination, while Crime and Punishment* would be an invitation to believe or to know. Thus, your example seems like a straightforward case of ‘if things were otherwise’. The reason I think this is unproblematic is that it seems, in this case, that we can accept that Dostoevsky is not the same person as his counterpart Dostoesvsky*, even though they share the same intrinsic features. Indeed, as described, it seems strange to suggest that Crime and Punishment is not fiction, or, that Crime and punishment* is not non-fiction, and it seems that the reason why is because of the relational properties that the books have. Therefore, it would seem more awkward, to me, to suggest that these relational properties should not play an important role in the classification of the book. However, I am aware that, without reading Walton, I am probably expressing views that he would not be prepared to accept.
ReplyDelete