Fiction as a Genre

Comments on Stacie Friend's "Fiction as a Genre"

Stacie Friend, “Fiction as a Genre,” Proceedings of the Aristotelian Society, New Series, Vol. 112 (2012), pp. 179-209.

We distinguish fiction and non-fiction, and it is clear that this distinction is relevant to how we use various kinds of texts. Of what, however, is this a distinction? A very common view is that fiction involves some kind of pretense, imagination, or make-believe in some way that nonfiction does not. We will be arguing against this kind of view at various points this year, as we argued against the view that fiction and nonfiction are different speech acts (an account which can overlap the pretense account) last year, but to tackle such a big and popular view it would be helpful to have some sense of the alternative. For this reason, we are going to look at Stacie Friend’s genre account of fiction, which we think is at least in the right ballpark.

Any account of the distinction between fiction and nonfiction must address two questions:

What places a work in each category?

What difference does such a placement make?

Friend argues that most theories of the distinction fail to do well in answering either question. Everybody recognizes that purely syntactic or semantic properties cannot answer them, so one of the most common views, the fictive utterance account argues that fiction works are works that involve an invitation to pretend, imagine, or make-believe. Friend notes that the fundamental problem is that there seems to be no account of pretense, imagination, make-believe, or whatever one wishes to call it, that can deliver both the distinction and give an adequate answer to both of these questions. A significant problem is that it seems that any kind of pretense, imagination, or make-believe can play a role in both fiction and nonfiction. Nonfiction tales can be grippingly told and the reader can be invited by them, even explicitly, to imagine the events they describe, or even put themselves in the shoes of historical characters facing various problems or decisions.(Friend suggests the idea, which seems plausible to us, that the invitation to imagine is in fact just a common feature of a wide variety of narratives, and that fictional works are saddled with it primarily because they are more likely to be narratives than nonfictional works. It is in fact a common problem with attempts at accounts of fiction that they just describe features that are common to, or at least common among, narratives of all kind.)

To deal with this problem, fictive utterance theorists usually propose an additional condition – perhaps a condition that what is said must be non-accidentally true, or that it has some end in view other than fidelity to the order of events. As Friend notes, however, fictional works often include non-accidentally true statements that are integral to the fiction, and they often have an aim of fidelity to the order of events. Historical fiction is often a good example of both, but this is quite general among different kinds of fictional works. In addition, nothing prevents nonfiction from having statements that are only accidentally true, or to have aims other than fidelity. Creative nonfiction is an example of a field of nonfiction that often has both, and, in fact, any nonfiction that is concerned heavily with presenting what people’s interpretations or opinions were of an event will have difficulty sticking strictly to these kinds of condition, because people’s interpretations of and opinions about such events often do not meet these conditions, either. And historically, people writing nonfiction have often explicitly not had the aim of fidelity to order of events, but some other end, like moral instruction or philosophical theorizing, in view. Tacitus invents speeches and events that he himself makes clear we are not supposed to take to be strictly true, in order to serve a larger end of giving a clear view of what was not necessarily clear at the time, speculatively exploring people’s thoughts, motivations, and expectations on what is sometimes excellent evidence and sometimes nothing more than rumor or even guesswork. Another possible case we could add to Friend’s: The history of physics one gets in many physics textbooks is famously careless about precise details and exact order of events and actual historical evidence, because the point of it is not to do history but to give students an idea of what goes into present theory. But the fact that it is intended and expected to be loose and not strictly accurate does not generally lead us to treat it as fictional.

In order to deal with this problem, fictive utterance theorists usually press for a more fine-grained analysis of works. Instead of treating whole works as the primary unit for applying the distinction, perhaps we should break works up into parts, and recognize that the same work could have fictional and nonfictional parts. Some take this so far as to argue that we should often break things down to single statements that are sorted into ‘fiction’ and ‘nonfiction’. If we do this, however, Friend argues that we end up struggling to make sense of the fact that we do, in fact, sort whole works into ‘fiction’ and ‘nonfiction’. It matters to people whether the overall narrative voice of a complete work is from a fictional perspective, for instance.

Another example, different from those used by Friend, might be popularizations. Both historical and scientific popularizations rely heavily on fictional techniques, but we distinguish consistently between these and historical fiction or science fiction. A considerable portion of Albert Einstein’s 1916 work, Relativity: The Special and the General Theory, is concerned with scenarios that would count as fictional under the fictive utterance view. Why then do we not put it in the fiction section of the library? On the other hand, a ‘hard science fiction’ short story might have a significant portion that is purely nonfictional by this view, describing in detail the exact mechanics or physics or biology of things in the real world.

Elsewhere (“Fictive Utterance and Imagining II,” Proceedings of the Aristotelian Society, Supplementary Volumes, Vol. 85 (2011), p. 167) Friend calls these patchwork problems, and there are versions relevant to each of the major questions we noted above. That is, works that we unambiguously treat as fiction or non-fiction on a finer-grained fictive utterance analysis end up being thoroughly patchwork, and this raises puzzles for both how we actually draw the distinction and how this distinction affects everything else. There she also notes that in dealing with these problems, imagination-based approaches often end up being one-sided in plausibility; that is, when someone with an imagination-based approach tries to address the patchwork problems, they often reach a plausible theory of fiction that results in a very implausible theory of nonfiction, or a theory of nonfiction that is plausible while the theory of fiction is very implausible. It’s not just that there are hard cases, which might be the case for any position; it’s that the solutions seem to create hard cases,to take cases you wouldn’t expect to be hard cases and make them hard cases – e.g., by drawing lines so that something everyone treats as nonfiction ends up as fiction or vice versa.

Friend suggests that the general problem is that imagination-based accounts are reductionist in character; they are trying to approach the distinction by breaking them down into more fundamental properties of the works themselves, or, alternatively, by starting with components and build the works up. Instead, we should take a contextualist approach to the distinctions. We get the distinction not from looking at the properties or parts of the works themselves but from looking at “whole work is embedded in a larger context, and specifically in certain

practices of reading, writing, criticizing, and so on” (“Fiction as a Genre” p. 187). This is the foundation for Friend’s genre account:

I propose instead that we construe fiction and non-fiction as genres. A genre, for my purposes, is a way of classifying representations that guides appreciation, so that knowledge of the classification plays a role in a work’s correct interpretation and evaluation. (p. 181)

A genre in this sense is not a rigid classification based on internal structure or on necessary and sufficient conditions, but on practical use and value. Classifying by genre can consider origins, linguistic devices, stylistic choices, structure, topic, as well as tropes that arise from the history of the genre itself. A work in a genre can bend the genre a bit by subverting our expectations for it, and any two works in a genre could be in that genre for rather different reasons. But what all genres have to do is establish a contrast class (i.e., help us recognize what is definitely not in that genre) which can be used to sort of whether features of a work are standard, contra-standard, or variable.

Standard genre features are features that are directly relevant to classification in a particular genre. If a story has a detective, that’s at least a reason to classify it as a detective story. Not every story with a detective is a detective story, but if you are trying to decide whether a story is a detective story, the fact that it has a detective is one point that is definitely relevant. Likewise, we often tend to use publisher’s classifications. These are not perfect (science fiction and fantasy fans are often complaining that publisher’s classifications often treat fantasy as a kind of science fiction or science fiction as a kind of fiction), but they are reasons to classify works in a particular genre – the publisher thinks that selling them as works in this or that genre will lead people to buy them. Contra-standard features are features that are directly relevant to something’s being classified as not being in that genre. For instance, limericks are contra-standard for physics textbooks; if you know of a work only that it consisted entirely of limericks, you would expect it not to be a physics textbook. Of course, nothing absolutely forbids the existence of a physics textbook entirely in limericks, but the limericks are a reason to think it might go outside the ‘physics textbook’ category. A single such reason on its own isn’t necessarily definitive, and what counts as contra-standard can change. Agatha Christie’s The Murder of Roger Ackroyd famously turns on a feature that was contra-standard for mystery stories at the time, although it is very definitely a mystery story. In fact, Christie was sometimes accused of cheating by violating the mystery-story expectations. Through its success, however, the work has made it so that the gimmick is no longer contra-standard. Variable features are features that don’t have any direct relevance to the classification. A novel can be written in very sparse or very florid prose, a play can focus more on stage-action or on dialogue, a painting can be realistic or abstract or allegorical. However, any of these could at any point become so important and interesting to enough people that it begins to form a standard for a new genre or subgenre.

Friend thus argues that fiction and nonfiction are something like supergenres. They have all the features of genres, but we are often as interested in using them to sort other genres as we are in using them to sort individual works. That we do in fact use them this way is a point in favor of the genre account – one of the phenomena you need to account for with a theory of the fiction/nonfiction distinction is the fact that we sort genres with it. Friend argues that this is not the only strength, but what we have said so far on this point is sufficient for our particular purposes, and we will not go through all of them. It is important to recognize, however, that the genre account not only answers the question of how we sort works into ‘fiction’ and ‘nonfiction’, it also meshes well with what we know of how this distinction is used and for what it is valued. If I tell you a story and say that it is nonfiction, you may well assume that any details are there because they are true, or at least relevant to what actually happened; if I say that it is fiction, you may well assume that the style of it is part of the point. These are common kinds of expectations; they are also not hard and fast. The genre account can accommodate both the commonality and the lack of strictness of those expectations. We read differently if we read something as fiction compared to reading it as nonfiction, but there is lots of variation in how this can work, and there’s no single way we read either. Friend likewise notes that the genre account, unlike all of its current rivals, can account for the fact that we sometimes classify the same work as both fiction and nonfiction, for different reasons.

As we have already said, we hold that some version of the genre account is almost certainly right, although there’s certainly room for variation in how exactly genres should be conceived.

Jean-Baptiste-Camille Corot, “Woman Reading in the Studio” (c. 1868), courtesy of the National Gallery of Art Open Access collection, Collection of Mr. and Mrs. Paul Mellon.