Do Narratives Require Narrators? I: The Analytic Argument
Given a narrative, is it a priori that it has a narrator?
Is a narrator essential to a narrative? One position you could take is that every narrative, in order to be a narrative, requires a narrator. Call this narrative positivism. The opposing position (we can call it narrative naturalism) holds that there are narratives that, in order to be narratives, do not require narrators. We want to look at this question a bit in a series of posts.
Some basic clarification is necessary here. The questions is not whether every narrative happens to have or could have a narrator. Rather, the question is whether having a narrator is essential to being a narrative. One possible narrative naturalist position, for instance, is that all narratives can have a narrator, but that this is not what makes the narrative a narrative, but something we attribute to them afterward for different reasons. That gets quite subtle, of course, but it’s important to distinguish between needing a narrator to be a narrative and being a narrative that happens to have one.
There are many arguments that narratives require narrators in order to be narratives at all. In this post, we are going to look at what is generally called the a priori argument or analytic argument.1 In many ways it is the simplest, since it is an argument that narratives just are narrator-based. A simple way to argue this would be something along these lines:
(1) Narratives are constituted by narration.
(2) Narration is an activity.
(3) Every activity requires an agent.
(4) The agent of narration is the narrator.
(C) Therefore narratives require narrators.
Narratives, therefore, require narrators because they are narrated. This kind of argument captures, we think, a lot of why people assume that narratives need narrators. There are several problems with it, however, that play a significant role in developing any theory of narrative at all. We’ll start with the most serious.
Problem One: The argument seems to trade on an ambiguity of ‘narrator’.
We would usually say that Ishmael is the narrator of Moby Dick. Ishmael, however, is clearly not engaging in the activity of narration that constitutes Moby Dick; Herman Melville is. That is, the sense in which premise (4) is true requires that N‘narrator’ just means the author (or perhaps at times the redactor). Thus it would be less misleading to make this an argument that narratives require authors, regardless of their narratorial status: Narratives are authored, authoring is an activity requiring an agent, namely, the author, so narratives require authors. The question of whether all narratives require an author is an interesting one, but it is distinct from the question of whether all narratives require a narrator.
If, on the other hand, we tried to claim that Ishmael is in the story narrating, this is true, as well, but then premise (1) makes the argument question-begging, because it is then just the position that every narrative requires a narrator to be a narrative.
Problem Two: Are all narratives constituted by an act of narration?
It’s obviously the case that if we start with acts of narration we will typically get narratives as a result. But is this the only process by which we can get a narrative? Combining premises (1) and (2) requires this, but there seem to be a few possibilities that first have to be eliminated.
The first is the possibility of things that become narratives not by an act of narration but by being received as narratives. The idea of a received narrative is that we could perhaps get a narrative not because of a narrative agent, of any kind, but because someone takes it to be a narrative. Imagine that someone makes a grocery list, and then, five thousand years later, some future Athanasius Kircher is trying to decipher this ancient relic. However, Future Kircher is not aware of the actual function of a grocery list, and so just assumes that the list is a brief story of some kind, perhaps a religious myth. Using an artificial key based on a lot of guesswork, Future Kircher translates the list about a pound of beef, a gallon of milk, and a block of cheese as a brief able about a cow hammering a helmet full of milk into cheese to make bricks for a house, perhaps with a lot of other minor details filling out the story that arise entirely from mistranslation of other items in the list. The author of the list did not narrate a narrative. Future Kircher is not narrating the narrative, but attempting to translate it. No one seems to have engaged in any act of narration. Yet at the end, we have a simple narrative.
With respect to the broader question, a received narrative like this is, of course, not conclusive on its own, because one could argue that Future Kircher, in order to translate the list into a narrative, has to posit there being a narrator. But if this is how we get the narrator into the mix, it does not appear to come by any activity of narration. It certainly does not come from any real activity of narration. Thus the narrative would not be constituted by an activity of narration, and therefore we get nothing that can be combined with premise (3).
Related to received narratives would be found narratives, in which narratives are formed by some non-narrating process outside of any receiver. That is to say, the idea of a found narrative is the idea that something could genuinely be narrative without a narrator, but could be a narrative even before being taken by someone to be a narrator. This is at least sometimes how particular events (lives, memories, historical sequences) are described. One reason why one might think these descriptions are correct and literal is that it would then explain why some events lend themselves so well to further narrative stories — we already start with a narrative in reality, and we then, as part of our narration, articulate a description of the already-existing narrative in words. However, one might take these descriptions to be simply figures of speech, perhaps indicating not that they are actually narratives already, but merely that they have other features that make them narratable. Then the question would be whether these found narratives are narratives in themselves or have to be ‘activated’ into narratives by someone’s further narration.
Received narratives and found narratives have some relevance to ethics, because one popular family of approaches to ethics takes the narrative character of our lives to play an essential role in ethics, and how this would work would depend considerably on whether our lives are already narratives or have to be turned into narratives by how we narrate them.
Problem Three: Does every activity have an agent?
The third problem that has to be considered is whether every activity requires an agent. We ourselves are inclined to think so, but it is worth noting that there are relatively common theories of action and causation that do not support this premise. For instance, David Hume’s account of causation in A Treatise of Human Nature can easily be adapted to argue that at least some things we classify as ‘activities’ have no agents. One also occasionally finds people holding something like the view that activities are more primitive or fundamental than agents. We are, again, not particularly inclined to think these positions succeed, but we mention them here because they are a further reason why one might reject the analytic argument for narrators.
****
Of these problems, we think that Problem One is insuperable, and we are inclined to think that there really are received and found narratives. (Problem Three, as we noted, we do not think a genuine problem, but whether it is regarded as one depends on broader metaphysical positions.) Thus the analytic argument does not appear to establish that all narratives are constituted by narration by narrators. At best, it is useful for clarifying some of the issues. There are other arguments for narrative positivism, however, which we will consider in the future.
In this series, our naming conventions, as well as (more loosely) a few of the argument-characterizations, will be following Sylvie Patron’s very useful “No-Narrator Theories/Optional-Narrator Theories: Recent Proposals and Continuing Problems,” Contemporary French and Francophone Narratology, John Pier, ed., The Ohio State University Press (Columbus, OH: 2020), pp. 31-53.