Open-access FICTIONAL MINDLESSNESS AND THE PROBLEM OF UNREPORTABILITY

Abstract

When we engage with fictions, we are, in effect, pretending to deal with reports of actual events. After all, numerous fictional works are explicitly designed to facilitate this kind of pretense. This was the prevailing understanding of fiction in both analytic philosophy and classical narratology for decades. However, there is a significant problem with this view: many fictional narratives routinely portray scenarios that could not possibly be the subject of anyone’s reporting. Currie’s 'mindless fictions' are one such example. This issue can be referred to as the problem of unreportability. The standard solution to this challenge has been to argue that fictions are not pretend reports, but rather direct authorial stipulations to imagine specific scenarios. This paper contends, however, that such an explanation fails to provide a satisfactory account of fiction. By drawing on Walton’s notion of ‘silly questions’, it instead argues for a revised version of the report model-one that doesn’t necessarily depend on ubiquitous fictional reporters.

Keywords: Fictional Report; Mindless Fictions; Narrator; Unreportability; Silly Questions

1. The ‘Fictional Report’ Theory (FRT)

When engaging with fiction-let's say we are reading Frankenstein-it appears that we “keep track” of “two mental files” (Antonsen 2019, p. 5684) at once. On the one hand, we are aware that we are dealing with a story invented by Mary Shelley. However, it seems that we also pretend that we are reading an account of actual events. Shelley's novel invites such an imagining since it’s structured as a series of letters produced by an 18th-century ship captain, documenting his experiences with a man rescued by his ship. Indeed, many fictional works are explicitly formatted as accounts of real events, such as journals (as in Robinson Crusoe) or autobiographies (like Moll Flanders). These observations gave rise to what was considered to be “the orthodox account” (Matravers 2014, p. 124) of fiction: fictional works have the form of “feigned reality statement[s]” (Hamburger 1957, p. 311), and to properly engage with them means to “imagine [...] that they are reports of actual events” (Alward 2006, p. 452). This idea-that can be dubbed ‘the fictional report theory’ (FRT)-gained wide traction in the philosophy of fiction after David Lewis employed it in his account of truth in fiction: the storyteller simulates reporting “known fact[s] when he is not doing so” (1978, p. 40). For instance, Conan Doyle invented Sherlock Holmes stories while inviting readers to imagine that they are published “memoir[s] of events” reported by “a doctor named Watson” (Lewis 1978, p. 40).

The FRT seems convincing when applied to first-person narratives (the situations in which authors create fictional characters who report the events from the storyworld, like Huck Finn in Twain's novel). However, it becomes less straightforward in the cases of ‘third-person narration’ (when there is no identifiable fictional reporter). This is the case with novels like Pride and Prejudice or Anna Karenina which disclose the inner thoughts of characters without any clear-cut fictional reporter. However, Lewis argues that such situations are “not essentially different” (1978, p. 40) from the unproblematic first-person ones. They are also best understood as fictional reports.

When we read Anna Karenina, we do not experience this novel as inviting a radically different imaginative attitude than, for example, Jane Eyre (a fiction framed as a first-person report). In other words, we do not read Jane Eyre as if it were a report of true events but then treat Anna Karenina as a mere authorial invention. Our reactions seem similar in both cases. When reading Tolstoy's novel, we pretend that what the novel says about characters is somehow also reported to us “as known fact rather than fiction” (Lewis 1978, 40). “Otherwise, why would [we] be moved by [Anna's] predicament?” (Walton 1990, p. 6). Now, an obvious difficulty with this approach would be the issue of the “grounds on which the narrator [reports] the inner acts of characters” (Kuroda 2014, p. 77). However, the proponent of FRT could simply brush off this concern by pointing out the fact that what is true in fiction often bends the rules of reality. If there are successful fictions portraying time travel or talking animals, then one could very well assume the fictional existence of some exceptionally insightful reporters. As Lewis points out, the narrator of (e.g.) Anna Karenina “purports to be telling the truth about matters he has somehow come to know about, though how he has found out about them is left unsaid” (1978, p. 40). This has become the dominant view in narratology, and most readers likely encountered it already in high school literature classes where they were routinely trained to read third-person novels as fictional reports (by adopting to their vocabulary phrases like ‘third-person narrator’ or ‘omniscient teller’); that is, by postulating some reporting instance with exotic epistemic privileges. The basic idea is that first-person reports function as the default model of fictional narration, which is then applied to third-person narratives as well, in order “to build a general account of [fiction]” (Culler 2021, p. 41).

The FRT rests on two general ideas. The first is that the act of fiction-making is based on pretense. When writing his novel, Tolstoy was aware that there was no such a person as Anna Karenina, so he was merely producing pretended assertions about her (Recanati 2021, p. 20). To immerse ourselves in fiction is to play along and pretend that we are reading reports of actual events.

The idea of pretense seems to lead to a position known as the ‘ubiquity thesis’, the view “that there is necessarily a fictional narrator [distinct from the author] in every narrative” (Kania 2005, p. 47). If we accept that narrative fictions are pretend reports, then it’s “inadequate to assume that the author” (Eckardt 2021, p. 157) himself is the reporter. We must further imagine a fictional reporter who “believes the events exist and reports them as fact” (Curran 2019, p. 99). Such fictional reporters would be ‘ubiquitous’ in all fictional narratives. The argument runs like this:

  • (P1) When reading narrative fictions, we pretend that we are dealing with reports.

  • (P2) A report entails the existence of a reporter. (This is an a priori claim.2)

  • (P3) Reporters are engaged in ‘fact-telling’.3 (This is a semantic claim about the meaning of the term ‘reporter’.)

  • (P4) Authors of narrative fictions are not engaged in fact-telling. (This follows from the standard definition of fiction.)

  • (C) If the propositions (1-4) hold, we must assume the existence of a fictional fact-teller4 (distinct from the actual author) as an integral part of fiction.

Thus, it follows that (for every x) if x is fiction, there is a fictional narrator for x.5 “To establish a [...] frame within which the narrative discourse may be read as report rather than [invention]” (Walsh 1997, p. 496), we need to presuppose a fictional narrator. As a result, readers find “the rationale” (Walsh 1997, p. 496) for the pretense that they are dealing with reports.6

It’s no surprise that the FRT has such a strong appeal. It aligns well with our intuitive understanding of what happens when we, e.g., read a novel. A lot of ordinary readers would probably agree that it accurately describes their experiences with narrative fiction. When we enjoy a story, we imagine that we are reading about events that “have happened” (Matravers 2014, p. 124). The theory also nicely accounts for the paradox of fiction. We are emotionally invested in stories we know are made up because we pretend that they are reports of real events. Overall, the FRT has the ambition to provide “a general [framework] for all stories” (Culler 2021, p. 47). However, there seems to be one significant class of fictional situations that the FRT cannot explain, which suggests that it may not be the accurate and comprehensive model of fiction. Let's unpack them.

2. Mindless Fictions

In July 1816, during one of the coldest summers on record, Lord Byron wrote a poem called “Darkness”, describing the heat death of the universe. The Sun and stars are extinguished, and the world becomes a “manless and lifeless” void. The actual poem is “delivered under the similitude of a dream”, but let's imagine it's narrated in a matter-of-fact manner. In such a story, every conscious being disappears, but the narration goes uninterrupted. Such tales, the genre Currie calls “mindless fictions” (1990, p. 126), would create a problem for the FRT. Accepting the FRT here will make us commit to a contradictory position. Namely, (i) the fictional situation is reported, and (ii) there can be no one to report it.

There are three main strategies one may employ here to preserve the FRT.

  1. The first is to accept that mindless fictions are contradictory but to disregard them as borderline cases that do not affect the general theory of fiction.7

  2. The other way around the problem is to treat mindless fictions as unreliable (or unserious) fictional (pseudo-)reports. Such stories describe certain situations only to discard them in the final step. When fiction presents contradictory claims: ‘Everyone has vanished’ and ‘Someone remained to tell us what happened’, the second claim simply cancels the first.8

  3. The third venue is that, in cases of mindless fictions, we only need to imagine that the events are reported by narrators who are part of fiction, but “do not belong to the fictional world they are telling us about” (Zipfel 2015, p. 57). That is, we are to postulate narrators who somehow have access to and describe events from worlds they do not inhabit. Mindlessness would be contained “only to the world on which [the narrator] reports” and would “not extend to the world [...] from which she reports” (Alward 2009, p. 359). “[T]he threat of contradiction” would, thus, be “forestalled” (Alward 2009, p. 359). We only need to entertain the idea of a reporter from another non-mindless realm, endowed with some mysterious access to the fictional world, and the paradox will evaporate.

The issue with (A) is that one can convincingly argue that mindless fictions are not isolated corner cases. There seem to be many well-known narratives to which one cannot reasonably attribute a fictional reporter. We regularly engage with such fictions without finding them particularly paradoxical. For example, Walton notices that many films include scenes with characters “getting [...] undressed, bathing or making love” (1990, p. 238). We do not experience problems with such portrayals until we are reminded that, according to the FRT, we are to imagine some hidden or invisible observer covertly recording these events. Such a supposition would generate various inappropriate imaginative consequences. To presume a fictional reporter in cinematic narratives would often lead to unwarranted genre recalibrations. For example, Bridget Jones' Diary (with a secret observer) would not be a romantic comedy but a disturbing horror. This is certainly not the imagining the film calls for. Furthermore, as Byrne (1993, p. 29) has claimed, the same reasoning can be applied to literary fictions with the so-called ‘omniscient’ narration. In the first chapter of Anna Karenina, Anna's brother is alone in his study, sleeping on a sofa, “vigorously embracing the pillow” (ch. 1) as he dreams of attending a lavish dinner party somewhere in the United States. According to the FRT, we make-believe that we are dealing here with a report of events. It would follow that (as part of the required pretense) there has to be someone within the story reporting the scene. This cannot be Tolstoy as he is simply inventing the story. But who else could it be? Only some supernatural agency that pries upon Anna’s brother, telepathically aware of the content of his dreams. Tolstoy's novel and Bridget Jones' Diary are realistic fictions that do not invite the supposition of invisible, ubiquitous entities with superpowers. Adding such cognitively enhanced narrators to these stories in order to accommodate the FRT would lead to all kinds of creepy and “absurd imaginings” (Curran 2019, p. 98) that are not prescribed by the fictions in question.

The problem is that we cannot simply disregard mindless fictions as uninteresting exceptions from the general model of fiction represented by the FRT. They share an important similarity with various works that depict intelligent life: it seems only adequate to suppose that many such fictions present situations “unobserved by any [reporting] subject” (Iversen 2018, p. 745). It is irrelevant that Anna Karenina (unlike mindless fictions) has sentient characters, as none of them can hold the proper epistemic position of fictional reporter. In various films and novels, certain affairs indeed take place, but it seems unreasonable to assume that they are reported by anyone.

The solution (B) to the issue of mindless fictions was to treat them as instances of unreliable narration. Let’s consider a mindless story (F1): ‘All life evaporated from the universe. My consciousness was forever extinguished. A few moments later, the Moon burst into pieces’. This first-person witness account suggests that the narrator is conscious of events even after his consciousness is obliterated, which is contradictory. In such cases, the natural reaction is to consider the narrator unreliable.

Let’s now imagine the same story narrated in the third person, without any trace of the reporting ‘I’. (F2): ‘All life evaporated from the universe. After every consciousness was extinguished, the Moon burst into pieces’. This version seems different from F1. While F1 involves imagining an explicit contradiction (what it means for someone to lose consciousness and still be conscious), F2 doesn't necessarily prescribe any such thing. We are merely invited to imagine the end of the world and its mindless aftermath, which we can easily do. However, someone might still object that, according to the FRT, there must always be someone reporting the scene (albeit covertly) in every fiction, which would produce a contradiction even in the case of F2. However, as we’ve seen, such insistence on clandestine reporters in all narratives would commit us to ridiculous consequences down the road. We would need to populate most third-person novels and films with paranormal entities to ensure that there exists someone in the story who can (reliably) report the events (like the innermost thoughts of the characters) which would throw our general understanding of fiction (and its genres) into disarray. The price for dismissing F2 on the grounds of unreliability seems too high.

The third attempt (C) to reconcile mindless fictions with the FRT was to assume that they are reported by an agent from another non-mindless world. The narrator reports a mindless situation but is not part of it. How to make sense of this supposition? Predelli (2020) introduces an interesting distinction between the storyworld and the narrative periphery.9 It's a reconceptualization of the classic narratological distinction between homodiegetic and heterodiegetic narration. Homodiegetic fictions are narratives reported from the ‘inside’. The reporter is a character inhabiting the fictional world of the story (like Lockwood in Wuthering Heights). In a heterodiegetic narrative, however, “the narrator is absent from the story he tells” (Genette 1980, p. 244). In a way, he reports the events from the ‘outside’. The narrator is not part of the storyworld but is, nevertheless, fictional. According to Predelli, any fictional census of Anna Karenina's storyworld would count Anna, Vronsky, Dolly, Oblonsky, but not “a reporter who dutifully describes their everyday lives” (Predelli 2020, p. 46). The reporter and the characters “fictionally inhabit different domains” (Predelli 2020, p. 46). The domain occupied by the reporter is what Predelli christens the periphery. What happens in the periphery is not bound by the rules of the storyworld. The storyworld of Anna Karenina excludes ‘omniscient’ reporters, but the same is not necessarily true of its periphery. It is “fictionally true that someone is speaking” (Predelli 2020, p. 48) to us about characters' thoughts, but such a fictional reporter is not part of the storyworld. He is not there on the spot watching, let’s say, Anna’s brother sleeping. He occupies a separate fictional domain.

Now, one may ask how the reporter occupying the peripheral domain could report things happening in the storyworld (as he is not connected with it). Predelli invokes the so-called “principle of poetic license” (Deutsch, qtd. in Xhignesse 2016, p. 149): “What actual semantics and epistemology forbid, fiction allows” (Predelli 2020, p. 54). Suddenly, our intuition that F2 fictions did not involve paradoxes seems validated. That’s because “there surely is no paradox” there, and the description of a mindless situation (Predelli himself constructs one called Tales from a Barren World) “smoothly and non-paradoxically occurs in the mouth of a peripheral teller” (Predelli 2020, p. 51). In the storyworld, no one indeed remains to report the events, but the same is not true for the periphery. What is false for the storyworld is not necessarily so for the periphery. The narrator survives the collapse of the storyworld, and the FRT with him.

Although the “story world-periphery divide” (Predelli 2020, p. 51) appears to be the most promising FRT preservation strategy in the face of fictional mindlessness so far, one may still not be sure “what to make of this third realm” (Patron 2021, p. 8), the fictional space that's neither part of the characters' nor the author's world, but is ‘a secret third thing’ populated by theoretical constructs. Would it not be preferable if we could discard such robust suppositions of additional fictional (or even ontological) levels, assuming instead something more straightfroward (for example, that the author is inviting us to simply imagine the mindless situation without intermediaries arising from strange fictional limbos)? We should leave Predelli's storyworld-periphery divide for now and search for simpler solutions. We will return to this topic again in the penultimate chapter.

3. The Stipulation Theory of Fiction (STF)

It seems that the idea of the mandatory reporter unnecessarily strains the fiction-making process. Can't authors simply invite us to imagine various situations without considering what some fictional insider might know and report? If so, we would no longer need to assume supernatural reporters or additional fictional realms in realistic narratives, nor would we encounter paradoxes in the case of mindless stories. The actual author would suffice. He would be the one establishing what happens without the need to “delegate the narrative to a fictive representative” (Patron 2008, p. 51). The general model for fictional discourse would not be a report but the authorial stipulation.10 According to this position, fictional utterances are to be understood as “prescriptions, or invitations, to imagine certain things” (Maier, Stokke 2021, p. 4). This theory, which we can refer to as the stipulation theory of fiction (STF), is based on Walton’s famous idea that fiction-making is essentially the activity of manufacturing props in “games of make-believe” (1990, p. 88). That is, fictional utterances are props that stipulate certain imaginings. In producing a story, the author may simply invite the readers to make-believe that, let's say, a character dying alone in his chamber is entertaining particular thoughts without simultaneously inviting them to make-believe that there is someone capable of reporting this.11 Hence, one could easily create a fiction in which all cognizant characters perish without thus generating a paradox. The author only needs to ask us to imagine such a scenario, which we can easily do. Of course, sometimes the authors can also invite us to imagine that we are engaged with reports, but this is far from being a necessary prerequisite for fiction-making. Often, the creator of fiction does not “[authorize] imaginings about a fictional [reporter at all]” (Bücking 2022, p. 51). Many stories are simply stipulated imaginings that are reported by no one.

According to the STF, it's always the author who stipulates fictional events either directly or by postulating a reporter within the fictional framework. Narratorial reporting is just “a particular possibility” the author can employ in stipulating fictional facts and certainly “not a given” (Culler 2021, p. 43). The advantage of the STF is that it covers the cases of unreportable fictional events that the FRT failed easily to account for. The STF, thus, appears to be a better candidate for the general model of fiction than the FRT. However, this is only so at first glance since, upon closer inspection, the STF seems to face several critical difficulties.

Phenomenological Objections. One of the earliest influential criticisms of the STF was formulated by Matravers (1997), who argued that the FRT holds certain phenomenological advantages over the STF because it aligns more “natural[ly]” (Matravers 1997, p. 81) with how we engage with fictions. For instance, when reading a novel (Matravers provides an example from Emma, where the main character is described grooming herself), we seem “to imagine reading this sentence as a reported fact” (Matravers 1997, p. 81). But it is unclear, he argues, “what does the second model mandate us to do? Imagine seeing” the event directly? “If that were the case, such an imaginative project would surely interfere with our continued reading” (Matravers 1997, p. 81).

According to Matravers, if STF were true, our experience with fictions would be phenomenologically different-and likely more demanding-than it actually is. Take, for instance, the act of reading a novel or watching a film (such as Aronofsky’s Noah). If the director does not intend for us to imagine engaging with a report, then what else are we meant to make-believe? That we are somehow witnessing the events firsthand? Such a notion would conflict with our actual experience of watching a film. We are seated comfortably, unable to interact with the events, as they have already ‘happened’ fictionally long ago. This overall phenomenological and emotional stance, as Matravers observes, “bears little resemblance to seeing” (Matravers 1997, p. 82) some catastrophic event-such as the near-total destruction of the world by the wrath of God-unfolding directly before us.

One might almost conclude, then, that the FRT, with all its limitations and problems, better captures our experience with narrative fiction than the STF. Watching Noah, for example, feels more like pretending to view a documentary than pretending to witness real events firsthand. Similarly, reading Anna Karenina resembles reading a report of past events rather than imagining we are somehow ‘present’ in the scenes as they unfold. In this way, the FRT seems to align more closely with the ‘natural’ way we engage with fiction.12Semantic Objections. Another trouble with the STF is that many fictions that initially raised the problem of unreportability are explicitly not arranged as authorial stipulations but (strangely enough) as reports. Many third-person narratives “contain elements that are best [understood] as simulations of factual [...] statements” (Schaeffer 2014, p. 21), that is, as pretend reports. There are semantic reasons to believe that the distinction between third-person and first-person character narration is not as clear-cut as the STF would suggest.

For example, in third-person ‘omniscient’ narratives, one often comes across “epistemic modals” (Eckardt 2015, p. 171) such as ‘it seemed’, ‘perhaps’, ‘probably’, or ‘maybe’, indicating some degree of uncertainty about the described events. This frequently occurs (to stick with the familiar example) in Anna Karenina. Amid detailed descriptions of the characters’ inner thoughts, one suddenly encounters markers of epistemic ‘vagueness’: “Feeling probably that the conversation was taking a tone too serious for a drawing-room, Vronsky made no rejoinder” (ch. xiv). Or: “Levin listened [...] but could think of nothing to say. Nikolay probably felt the same” (ch. xxxi). The same thing occurs (to provide a random example) in Larsson's The Girl with the Dragon Tattoo. The novel that offers detailed descriptions of confidential conversations and private ruminations of the characters sometimes also signals uncertainty about particular fictional goings-on: “[Martin] was aware that his guest was a financial reporter whom he hardly knew, yet he discussed the internal problems of his company so openly that it seemed reckless. Perhaps he assumed that Blomkvist was one of the family” (ch. 10). The problem is that the presence of lexical markers denoting epistemic uncertainty only makes sense if the author wants to stipulate the imagining that we are dealing with a report (i.e., that someone is telling us the story as a known fact). Narratorial uncertainty would be out of place if the author (with the authority of someone inventing the entire situation) directly invites us to imagine the story.

Furthermore, it can be argued that even the widespread use of the past tense in fictional narration suggests a report model rather than the pure authorial stipulation. This is because such verb inflections are generally employed to inform readers about “events that have happened in the past” (Zipfel 2015, p. 68). If, when engaging with fiction, we are to pretend that we are reading about events that have happened somewhere, it’s difficult to imagine that we are dealing with anything other than reports.

The “Oversimplification” Objection. Additionally, the STF seems to oversimplify the notion of truth in fiction. The STF suggests that any fictional scenario φ is considered true in fiction iff the author invites us to imagine that φ. However, what if the author invites us to imagine a character who committed atrocious war crimes as the noblest of heroes? Are we forced to accept this as true in the story just because the author invites us to do so? Often, we do not blindly accept authorial stipulations about what is true in fiction but instead engage in “a process of ‘reflective equilibrium’” by measuring “hypotheses about [...] the story against the evidence [...] presented in the story” (Curran 2019, p. 105). We treat stories more like reports to be interpreted than authorial commands to be obeyed. In other words, we often do not “passively accept what [we] are told, but rather actively test [the] assertions against [the wider] body of information” the text provides, looking for the interpretation that “best fits the [overall] evidence” (Alward 2009, p. 357). To summarize, the initial argument was that the FRT doesn't function due to the problem of unreportability (that is, the fact that many fictions routinely depict events that no one within the story could witness and report). A popular solution to this issue was to argue that these fictions are not pretended reports but authorial invitations to imagine certain scenarios, a stance we have termed the STF. However, this approach seems to introduce more problems than it resolves. Is there a way to address the issue of unreportability without abandoning the FRT altogether? We should now return to Predelli's periphery.

4. ‘Silly Questions’ and the ‘Blocked Fictional Inference’ Principle (BFI)

Suppose (once again) that we are reading Anna Karenina. While doing so, we make-believe that fictitious events are related as known facts. The narration seems styled as reporting: the epistemic modals are rhetorically used to create the effect of reporting, and the events are described in an assertoric manner. (As we’ve seen, it would be difficult to see this fiction merely as a set of stipulations to directly imagine fictive scenarios.) So we play along with the pretense that we are reading about something that has happened somewhere. Like hundreds of Tolstoy's readers, we encounter no particular problems with such an imaginative attitude. But, suddenly, a theoretician asks us: “Who speaks [here]?” (Genette, qtd. in Birke, Köppe 2015, p. 1). We are immediately confronted with something that appears to be a new and genuine problem (‘How can anyone report the thoughts of others?’), and so we come up with a quirky answer: ‘There has to be an omniscient teller’. But this is a forced ad hoc move, a far-fetched assumption that unwarrantedly complicates the ontology of the storyworld by postulating shady mind-reading “monsters” (Patron 2021, p. 4). It's simply an explanatory overkill.

Predelli’s solution was to relocate the reporter to the fictional periphery, outside the storyworld. However, as we have seen, this may be yet another unwarranted complication. What are we to make of this strange world-so different and utterly disconnected from the characters’ reality-inhabited by lonesome reporters who are not fully realized personalities but rather incomplete constructs, “disembodied voice[s]” (Predelli 2020, p. 54), often devoid of “any gender, any name, any age or ‘I’” (Behrendt, Hansen 2011, 227),13 with no distinguishable personal attitudes or only with shadows of them? This, at least, is how the major argument against heterodiegetic narrators sounds, as voiced by the so-called 'Optional-Narrator' theorists.14

However, reactions like these ultimately seem to miss the most interesting point of storyworld-periphery distinction. We should be cautious about viewing the periphery as a distinct world (in the Lewisian sense), a fictional realm with its own reality that somehow stands on par with the storyworld. Instead, the periphery should be reframed as a functional narrative tool, rather than as a parallel fictional world of its own right. Unlike the storyworld, the periphery is best understood as a purely instrumental construct-a backdrop that serves solely to convey certain information about the storyworld that cannot be (convincingly) provided otherwise. It functions as a practical support for fictional communication, a component of the narrative mechanism that doesn't hold any real fictional space.

Walton, similarly to Predelli, makes a “distinction between work worlds and game worlds” (Walton 1990, p. 215). The work world is a realm “in which fictional events occur” (Terrone 2022, p. 204). This is where Anna Karenina meets Vronsky and ultimately takes her own life. The game world, on the other hand, is the fictional space in which readers hear and get to know about these events. It is a dimension of fictional communication where the information about the storyworld is conveyed. This is the ‘world’ of the ‘omniscient’ reporter who provides information to us. However, we should be wary of understanding the terms ‘omniscience’ and ‘world’ here as anything other than metaphors covering up the purely artificial, functional nature of these constructs. This second fictional realm is purely illusory. Were we to treat the narrator's world in Anna Karenina as an actual separate fictional world (no matter how structurally different from the storyworld), it would require all sorts of elaborate imaginings that most readers simply do not experience, lending some credit to the criticisms of 'Optional-Narrator' theorists.15

Questions regarding how character-narrators within the storyworld acquire knowledge are meaningful and can yield insights about the narrative. For instance, when Ishmael in Moby-Dick describes Ahab's behavior inthe privacy of his cabin, it opens meaningful questions about the nature of Ishmael's insights. These inquiries could reveal aspects of the storyworld (potentially hinting at Ishmael's unreliability, suggesting that he is merely hypothesizing, or even possessing telepathy). However, such questions about peripheral actors (like the narrator of Anna Karenina) seem futile. They simply know by virtue of knowing. It seems unwarranted to say that they know things by some kind of epistemic privilege. Such narrators are merely background assets, conventions that make the particular kind of fictional communication possible.16

Inquires about the peripheral narrator, such as how they acquire information they report or what kind of entities they are, lead only to undecidable speculations. Is the narrator a person, an incorporeal entity, or some form of telepathic machinery? None of these questions could yield determinate answers. They are somehow “inappropriate to ask” (Curran 2019, p. 109) within the context of fiction. Walton refers to these as “silly questions” (1990, p. 174) because there is no real answer to them within fiction. They require us to speculate about issues that the actual fiction never invites us to consider. Such questions are not just ‘silly’ but also ‘embarrassing’ since they tend to expose the “artifice of fiction” (Dawson 2013, p. 210) if pursued too far. It seems that fiction itself invites us to ignore these issues, since they can interfere with the pretense. For example, when watching Petersen's Troy, one might question if it is fictionally true that the heroes of the Trojan War spoke contemporary English. Or whether it is fictional that Shakespeare's characters converse in “superb verse”, even when they are “immensely distraught” (Walton 1990, p. 175). These fictions invite us to exhibit a “sudden loss of interest” (Kania 2005, p. 49) towards such issues. They are “non-questions” (Curran 2019, p. 102). They do not pertain to the fiction itself, but to the methods employed by the author in producing it. They point out rhetorical choices that do not belong to the actual content of the storyworld. In this sense, they are peripheral and purely instrumental. Achilles’ modern English in Troy merely serves to engage the wider contemporary cinematic audiences (which would be less feasible if he were speaking in ancient Greek).

‘Silly questions’ are those that do not pertain to the fictional content of the story but rather to the mechanisms by which fiction is constructed and presented, particularly in heterodiegetic instances, mistaking these mechanisms for the fictional content of the storyworld. These questions address the artifice of narrative communication rather than the fictional reality itself, leading to inquiries that seem to belong to the content of fiction but actually distract us from it. ‘Silly questions’ also emerge when we see the periphery as more than a metaphor for particular mechanisms of narration and view it instead as a realm that is somehow ontologically and structurally similar to the storyworld.17 Once we understand the periphery not as an actual fictional world-like realm but as a metaphor for a set of authorial choices aimed at facilitating the delivery of information about the storyworld, the periphery-storyworld divide can help us define the notion of 'silly questions'.

Let’s try to make this even more intuitive. Beyond examples of certain first-person fictions (formatted as journals, correspondences, or autobiographies), certain narratorial “incoherences are at the very base of fictional works” (Kania 2005, p. 49). They are so deeply ingrained in our concept of fiction that “we don't even notice them” (Walton 1990, p.176). Such inconsistencies are basically linked to how fictional information is presented to us. Authors will often deliberately introduce some incoherence in fictional narration to support certain crucial pretenses.

They may sacrifice basic realism in “purchase[ing] realism of another sort” (Walton 1990, p. 362). In such cases, we are invited to ignore such incongruities.

We can formulate a principle that stems from the notion of ‘silly questions’ that may be termed the ‘blocked fictional inference’ principle (BFI).18 Works of fiction, generally, block imaginative inferences pertaining to the so-called periphery. That is, the essential part of engagement with certain fictions is that readers are discouraged from drawing specific conclusions about, for example, the narrative source of the presented fictional information. We somehow feel that what happens in the storyworld is disconnected from such issues. For example, if Anna Karenina were a completely coherent fictional report, we would have a particular fictional character reporting the events (by, e.g., making behavioral inferences or trading in gossip). The narrator would belong to the storyworld and we would have a clear case of pretend report. However, the author wanted to maximize “the immersive process” (Schaeffer 2014, p. 35) by generating “a rich[er] array of fictional truths about the characters” (Walton 1990, p. 362). We get to know the characters' thoughts, conversations taking place under four eyes, etc. At the same time, the author didn’t want us to make-believe that such narration comes with the cost of mind-reading reporters that would turn the realistic narrative into a specimen of science fiction. So, he simply allows an anomaly in narration that we are invited to ignore.19 Someone or something is telling us this but we are meant to ignore this source of information. The identity of the narrator is utterly “deemphasized” (Walton 1990, p. 365) to prevent any questions from being raised about the particular nature of such instances. This is the cost the author is willing to pay to produce deeper immersion and more fictional details.

When an intimate conversation between Anna and Vronsky is recounted in Tolstoy's novel, any potential ontological or epistemic inferences that we might make from such a report are blocked in fiction. No data about what makes such reporting possible is provided nor can it be provided without reframing the fiction drastically. To ask any specific question about the teller (concerning their age, gender, or “epistemological credentials” (Matravers 2014, p. 128)) is simply “unauthorized” (Walton 1990, p. 60) under the BFI principle. To imagine anything specific about such reporting would be to go further than the fiction allows. As readers, we tacitly understand that this is the very convention of fiction, and that, fictionally, there is nothing more to know about such reporting instances. The events are reported, but the issue of how this could be is framed, not as an epistemic or “ontological question, [but] as a pragmatic aspect of strategic reading” (Aczel 1998, p. 494).

Thus, our initial ‘pre-theoretical’ reading of Anna Karenina bypassed the issue of who is speaking, not due to our theoretical naivety, but because we were engaging with the fiction on its own terms. In such engagements, the question of access-how is it possible for us to receive certain fictional information-does “not arise” (Matravers 2014, p. 127).

Now, we can more easily understand why we do not encounter substantial problems with the F2 version of mindless fictions. In a way, they function exactly like many other heterodiegetic narratives. A mindless situation is narrated, but the presence of a particular narrator is completely deemphasized. The narrator seems almost effaced, remaining in the shadows and never drawing attention to themselves. It’s an anomaly in narration for sure, but it relates to the general artificial nature of many fictional works. It's all about creating the effect of some kind of phantom telling or a reporterless report-like the events are merely captured by some unmanned recording device, a language model that transforms visual inputs into descriptions. But nothing allows for any kind of specific inquiry.20

One may recall Bradbury’s “There Will Come Soft Rains”. Set in August of 2026, in the immediate aftermath of a large-scale catastrophe, the story is presented as a matter-of-fact report of scenes of utter desolation, emptied of all human activity. It's not part of the fiction that anyone is observing the scene moments after the nuclear explosion. The crucial effect of the story is the absence of any observing entity. The narration is marked by a completely “abstract speech position” (Bücking 2022, p. 43). This story is not so different from a mindless fiction since they both prompt us to disregard how the sentences could possibly emerge. It appears that it is part of the story not to ask how do we know this and who is telling us this since this would ruin the effect the fiction tries to produce. The sentences “point to [...] a phantom speaker, [...] an empty subject position” (Bal 2008, p.18). The events are reported, but the inference that someone is reporting them is simply blocked under the BFI principle. We pretend that we are engaged with phantom reports.

The same ‘phantom’ effect is regularly achieved in cinematic narratives. Often, when watching a film, we are invited to pretend that the report merely appears. That is, we usually see things “from an unoccupied perspective” (Iversen 2018, p. 744). “The camera imitates the eyepoint of a [reporting] character” (Kuhn, Schmidt 2014, p. 29) but no reporting character is part of fiction. The shot is “subjective, yet subjectless” (Iversen 2018, p. 745). The effect thus produced is sometimes referred to as “the invisible observer [position]” (Kuhn, Schmidt 2014, p. 26) but it would be more apt to describe it as ‘the absent observer position’. In such scenes, there appears to be “only a [reporting] function and not a particular [reporter]” (Aczel 1998, p. 490).

Authors may, in fact, build their stories on an incoherent assumption that would expose the fictitious nature of fiction if closely scrutinized. We are merely invited to ignore these “signposts of fictionality” (to use Dorrit Cohn’s phrase). In the case of fictional reports, we routinely overlook certain foundational contradictions that would be deal-breakers in our engagements with actual reports. Imagine that a novel invites us to pretend it’s “a diary that must have been destroyed in the story-culminating fire” (Kania 2005, p. 49). How the story is told often involves all sorts of anomalies that are not to be pursued or scrutinized (according to the BFI principle) in proper engagement with fictional content. Fictions could invite us to imagine reports but not particular reporting agencies. This may be paradoxical, but the paradox is not part of fiction itself but only pertains to the periphery-to the artifice of how fictional access is granted to us.

5. Conclusion: The Fiberboard Horizon

A few final theses about how fictional facts are communicated to the readers are due here.

  1. The standard idea was that when we engage with fiction, we pretend to interact with reports of actual events. We refer to this position as the FRT. The FRT assumes the presence of a ubiquitous fictional entity, commonly termed 'the narrator'. If we are indeed pretending to engage with reports, then each fiction would require fictional reporters who convey information as facts.

  2. However, the FRT, as traditionally conceived, encounters what we might call 'the problem of unreportability', which poses two challenges. First, there are so-called mindless fictions (stories set in worlds devoid of sentient characters) which appear to be unreportable. Second, postulating a reporter (a hidden, observing, telepathic entity) as the integral part of the storyworld (in the cases of the so-called ‘omniscient’ narration) would involve us in various absurd imaginings (Curran 2019, p. 98) that are clearly not intended by these fictions.

  3. A potential solution to the problem of unreportability is to argue for the author’s central, unmediated role in establishing fictional facts. The author does not need to 'delegate' narration to some fictional entity. He can act as the direct source of fictional information by stipulating certain imaginings. If such a supposition works, then the problem of unreportability evaporates since what the author may stipulate as a required fictional imagining doesn't necessarily depend on the point of view of any fictional entity. We've termed this the STF. The author issues “a prescription to the effect that it is to be imagined” (Walton 1990, 61), and such prescriptions constitute what is true in a particular fiction. He could, for example, unproblematically invite us imagine a world where events take place without sentient characters. It's important to recognize that the FRT and the STF are not entirely incompatible. They often intersect. Notably, the FRT is also based on an authorial stipulation. That is, the author stipulates the imagining that we are dealing with reports. We may call this the initial stipulation. But for subsequent “framing of fictional discourse”21 he relies on the report model. This is where the ‘report model’ theorist diverges from the ‘stipulation model’ theorist. The STF also accommodates the fact that many fictional works are obviously structured as reports, such as Robinson Crusoe or The Catcher in the Rye. In these cases, when there is a clear-cut character narrator, the author stipulates that readers are to imagine they are engaging with a report. However, the author can also bypass this and directly stipulate specific imaginings. According to the STF, this is precisely what happens in cases of mindless fictions or heterodiegetic narration.

  4. The STF, however, faces at least three major problems. The first is termed the phenomenological objection. Originally raised by Matravers (1997), it argues that direct authorial stipulations of fictional facts (beyond the initial stage) would entangle us in experiences that differ substantially from our actual interactions with fictional works (such as imagining ourselves as somehow directly witnessing fictional events). The second problem, the semantic objection, points to the fact that many instances of third-person narration that the STF attempts to explain are, curiously enough, formulated as narratorial reports rather than direct authorial stipulations. The STF struggles to account for this not so uncommon phenomenon. The third and perhaps strongest objection (the oversimplification problem) argues that the STF would force us to immediately accept contradictory or otherwise extremely problematic claims as fictional truths (which is something we typically do not do in our ordinary engagements with fiction).

  5. The most convincing way to save the FRT from the unreporability problem draws on Predelli's storyworld-periphery divide. This approach elegantly sidesteps the problem of fictional mindlessness by introducing a peripheral narrator who reports events occurring in the storyworld while remaining disconnected from it. Such a narrator would not be part of the storyworld (so the mindlessness remains intact), and we would not need to explain how they can report events since the peripheral world they inhabit doesn’t need to resemble the actual world in any relevant sense. We can simply invoke the concept of poetic license here. Additionally, the potential for weirdness is downplayed, as we no longer need to posit bizarre ‘omniscient’ characters roaming the ‘realistic’ storyworlds.

  6. However, a somewhat revised version of the ‘storyworld-periphery’ account (that does not significantly depart from Predelli’s) could be offered. This version might better align with our actual experiences of reading fictions (like Anna Karenina or Emma). Critics, such as Optional-Narrator theorists, might argue that the notion of a distinct narratorial presence in some nebulous space called ‘the periphery’ could still lead to overly elaborate and implausible conceptualizations that are not part of our ordinary reading experiences. Is it really necessary to think of a second fictional world (in non-metaphorical sense) in heterodiegetic novels?

The idea here is that ‘the periphery’ is a useful concept, but we should adopt a more functional view of it, treating it as a ‘backdrop’ rather than an ontologically distinct realm. The periphery can be seen merely as a metaphorical term for a set of artificial and instrumental narrative tools that provide essential information about the storyworld without inviting (or even allowing) further inquiry into how these pieces of information are provided. Rather than creating a fictional world of its own, the periphery merely facilitates storytelling by providing information that cannot be convincingly offered otherwise. This approach dismisses the need to always assign the narrator some actual and clear fictional status.22

The function of heterodiegetic narration is not to draw attention to itself but to subtly obscure and mask the anomaly of its very presence. Readers are encouraged to disregard the implausibility of such narrative voice and instead engage directly with the information it provides. Indeed, any inquiry into the nature of such a voice could be considered a ‘silly question’, a misplaced curiosity that confuses the content of the storyworld, the “fiction proper” (Schaeffer 2014, p. 19) with conventionalized narrative procedures that ease the conveying of fictional information. Making inquiries about these procedures as if they were part of the storyworld is implicitly discouraged in ordinary interactions with fictions. (This is a phenomenon we've termed the BFI principle.) There is simply no proper fictional answer to any such question.

In the movie The Truman Show, there is a famous scene where the main character discovers that his entire life has been broadcast on TV without his knowledge. In an attempt to gain his freedom, he sails across an artificial ocean, only to eventually crash into a large fiberboard wall painted to look like the horizon, revealing his entire world as a fake reality TV set. If we start seriously asking ‘peripheral’ questions about heterodiegetic narration as if they pertain to the storyworld (Who is telling us this?), we risk hitting the artificial wall of fiction. When we pose questions about fictional mechanics as if they pertain to the fiction itself, the fictitious nature of the narrative is revealed, the spell is broken, and the pretense is canceled. The BFI principle keeps us within the fiction proper and maintains this fictional pretense by deemphasizing certain foundational incongruities. When engaging with certain fictions, one is not intended to push the process of naturalization too far.

References

  • ACZEL, R. “Hearing Voices in Narrative Texts”, New Literary History, 29, pp. 467-500, 1998.
  • ALWARD, P. “For the Ubiquity of Nonactual Fact‐Telling Narrators”, The Journal of Aesthetics and Art Criticism, 65, pp. 401-404, 2007.
  • ALWARD, P. “Leave Me Out of It: De Re, But Not De Se, Imaginative Engagement with Fiction”, The Journal of Aesthetics and Art Criticism, 64, pp. 451-459, 2006.
  • ALWARD, P. “That’s the Fictional Truth, Ruth”, Acta Analytica, 25, pp. 347-363, 2009.
  • ANTONSEN, P. F. “Fictional reports”. Synthese, 198, pp. 5675-5688, 2019.
  • BAL, M. “Phantom Sentences”. In R. Kawashima, G. Philippe, and T. Sowley (eds.) (2008), pp. 17-42.
  • BATISTELA, I. M. “Uniformity in the dynamics of fiction-making”. Philosophia, 52. 475-500, 2024.
  • BEHRENDT, P., HANSEN, P.K. “The Fifth Mode of Representation: Ambiguous Voices in Unreliable Third-Person Narration” In by P. K. Hansen, S. Iversen, H. Skov Nielsen, and R. Reitan (eds.) (2011), pp. 219-252.
  • BIRKE, D., KÖPPE, T. “Author and Narrator: Problems in the Constitution and Interpretation of Fictional Narrative”. In D. Birke and T. Köppe (eds.) (2015), pp. 1-12.
  • BIRKE, D., KÖPPE, T. Author and Narrator: Transdisciplinary Contributions to a Narratological Debate. Berlin, Munich, Boston, Mass.: De Gruyter, 2015.
  • BÜCKING, S. “Narration Without Narrating”, Zeitschrift für Literaturwissenschaft und Linguistik, 52, pp. 35-64, 2022.
  • BYRNE, A. “Truth in fiction: The story continued”, Australasian Journal of Philosophy, 71, pp. 24-35, 1993.
  • CARROLL, N., DI SUMMA, L.T., LOHT, S. The Palgrave Handbook of the Philosophy of Film and Motion Pictures. Cham: Palgrave Macmillan, 2019.
  • CULLER, J. “Some Problems Concerning Narrators of Novels and Speakers of Poems”. In S. Patron (ed.) (2021), pp. 35-52.
  • CURRAN, A. “Silly Questions and Arguments for the Implicit, Cinematic Narrator”. In N. Carroll, L. T. Di Summa, and Shawn Loht (eds.) (2019), pp. 97-117.
  • CURRIE, G. The Nature of Fiction, Cambridge: Cambridge University Press, 1990.
  • DAWSON, P. The Return of the Omniscient Narrator. Columbus, OH: Ohio State University, 2013.
  • ECKARDT, R. “In search of the narrator”. In E. Maier and A. Stokke (eds.) (2021), pp. 157-185.
  • ECKARDT, R. “Speakers and Narrators”, In D. Birke and T. Köppe (eds.) (2015), pp. 153-186.
  • GARCÍA‐CARPINTERO, M. “Normative Fiction‐Making and the World of the Fiction”, The Journal of Aesthetics and Art Criticism, 77, pp. 267-279, 2019.
  • GENETTE, G. Discours du récit. Paris: Seuil, 1972. Transl. by J. E. Lewin and J. Culler, Ithaca, NY: Cornel University Press, 1980.
  • HAMBURGER, K. Die Logik der Dichtung. Stuttgart: Klett, 1957. Transl. by M. J. Rose, Bloomington, In: Indiana University Press, 1973.
  • HANSEN, P.K., IVERSEN, S., SKOV NIELSEN, H, REITAN, R. Strange Voices in Narrative Fiction. Berlin: De Gryter, 2011.
  • HJORT, M., LAYER, S. Emotion and the Arts. Oxford: Oxford University Press, 1997.
  • HUDSON HICK, D. Introducing Aesthetics and the Philosophy of Art, London: Bloomsbury Publishing, 2017.
  • HÜHN, P., C. MEISTER, J. PIER, W. SCHMID. The Living Handbook of Narratology. Hamburg: Hamburg University, 2014. https://www-archiv.fdm.uni-hamburg.de/lhn/index.html
    » https://www-archiv.fdm.uni-hamburg.de/lhn/index.html
  • IVERSEN, M. “The World without a Self: Edward Hopper and Chantal Akerman”, Art History, 41, pp. 742-760, 2018.
  • KANIA, A. “Against the Ubiquity of Fictional Narrators”. Journal of Aesthetics and Art Criticism, 63, p. 47-54, 2005.
  • KAWASHIMA, R., PHILIPPE, G., SOWLEY, T. Phantom Sentences. Essays in Linguistics and Literature presented to Ann Banfield. Berne: Peter Lang, 2008.
  • KÖPPE, T., STÜHRING, J. “Against pan-narrator theories”, Journal of Literary Semantics, 40, pp. 59-80, 2011.
  • KUHN, M., SCHMIDT, J. N. “Narration in Film”. In Hühn, Peter, et al. (eds.), pp. 1-39, 2014.
  • KURODA, S. Y. “Reflections on the foundations of narrative theory, from a linguistic point of view”. In S. Patron (ed.) (2014), pp. 71-101.
  • LEWIS, D. “Truth in Fiction”. American Philosophical Quarterly, 15, pp. 37-46, 1978.
  • MAIER, E., STOKKE, A. “Introduction”. In E. Maier and A. Stokke (eds.) (2021), pp. 1-12.
  • MAIER, E., STOKKE, A. The Language of Fiction. Oxford: Oxford University Press, 2021.
  • MATRAVERS, D. “The paradox of fiction: The report versus the perceptual model”. In M. Hjort and S. Layer (eds.) (1997), pp. 78-92.
  • MATRAVERS, D. Fiction and Narrative. Oxford: Oxford University Press, 2014.
  • PATRON S. Optional-Narrator Theory: Principles, Perspectives, Proposals. Lincoln, Ne.: University of Nebraska Press, 2021.
  • PATRON, S. “Introduction”. In S. Patron (ed.) (2021), pp. 1-34.
  • PATRON, S. “On the Epistemology of Narrative Theory: Narratology and Other Theories of Fictional Narrative”. In R. Kawashima, G. Philippe, and T. Sowley (eds.) (2008), pp. 43-65.
  • PATRON, S. Toward a Poetic Theory of Narration: Essays of S.-Y. Kuroda, Berlin, Boston, Mass.: De Gruyter Mouton, 2014.
  • PIEPER, V. “Author and Narrator: Observations on Die Wahlverwandtschaften”. In D. Birke and T. Köppe (eds.) (2015), pp. 81-98.
  • PREDELLI, S. Fictional discourse: A radical fictionalist semantics. Oxford: Oxford University Press, 2020.
  • RECANATI, F. “Fictional reference as simulation”. In E. Maier and A. Stokke (eds.) (2021), pp. 17-36.
  • SCHAEFFER, J-M. “Fictional vs. Factual Narration”, In Hühn, Peter, et al. (eds.), pp. 1-39, 2014.
  • STOKKE, A. “Fictional force”. Philosophical Studies, 180, 3099-3120, 2023.
  • TERRONE, E. “Observers and Narrators in Fiction Film”. Croatian Journal of Philosophy, 22, pp. 201-215, 2022.
  • VOLTOLINI, A. “Real Authors and Fictional Agents (Fictional Narrators, Fictional Authors)”. Organon F, 28, pp. 60-75, 2021.
  • WALSH, R. “Who Is the Narrator?”. Poetics Today, 18, pp. 495-513, 1997.
  • WALTON, K. L. Mimesis as Make-Believe: On the Foundations of the Representational Arts. Cambridge, Mass.: Harvard University Press, 1990.
  • XHIGNESSE, M. “The Trouble with Poetic License”. The British Journal of Aesthetics, 56, pp. 149-161, 2016.
  • ZIPFEL, F. “Narratorless Narration? Some Reflections on the Arguments for and Against the Ubiquity of Narrators in Fictional Narration”. In D. Birke and T. Köppe (eds.) (2015), pp. 45-80.
  • 2
    Cf. Patron 2021, p. 14; Kania 2005, p. 47; Köppe, Stühring 2011, p. 63.
  • 3
    See, Alward 2007, p. 402-3.
  • 4
    They are usually referred to as ‘narrators’.
  • 5
    See, Alward 2007, p. 404.
  • 6
    There are fictions that seem to be directly narrated by real-life authors (e.g., Dante's Divine Comedy). However, the FR theorist could argue that in such cases, the author essentially “fictionalize[s] about himself” (Walton 1990, p. 366) by taking on a reporting eponymous avatar. Fictional pretense implies that the author is not “speaking in propria persona” (Pieper 2015, p. 90). There must be (at least) some “minimal narrating agency” (see, Voltolini 2021, p. 68), or otherwise we would have no proper way to distinguish between fiction and insincerity.
  • 7
    E.g., Currie treats them as exceptions from general rules, and thus as “flawed games” (1990, p. 126). They are rare narrative anomalies that cannot reveal how most fictions function.
  • 8
    A similar position seems indicated in García-Carpintero 2019, p. 271. See also, Voltolini 2021, p. 69.
  • 9
    I thank the reviewer for highlighting this issue.
  • 10
    The term ‘stipulation’ is employed in this sense in Kania 2005, p. 52.
  • 11
    See, Byrne 1993, p. 35.
  • 12
    Additionally, one further issue for the STF could be the fact that the same authorial stipulation may prompt significantly different imaginings for various readers. Is Anna genuinely in love with Vronsky, or is she merely trying to escape an apathetic marriage by projecting all kinds of hopeless desires on her lover (which is Vronsky but could very well be someone else)? The text can support both imaginings. It would follow that different readers (when imagining according to Tolstoy's prompts) would activate different fictions. Anna Karenina would not be one single fiction anymore. This would significantly depart from “how we normally talk about [fictions]” (Hudson Hick 2017, p. 125).
  • 13
    Predelli (2020, p. 54), for instance, refers to the narrator of Emma as ‘he’ to distinguish this figure from the female author.
  • 14
    ‘Optional Narrator Theory' is a recent development in narratology that, generally, asserts that whenever narration cannot be attributed to a fictional character within the storyworld, it should instead be attributed to the actual author. According to this position, heterodiegetic reporters are weird theoretical constructs that do not play a role in our actual imaginative engagements with fiction. Such concepts were, ‘optional-narrator’ theorists argue, virtually absent from discussions about the nature of fiction and narrative prior to the late 19th century and only gained wide prominence after the 1960s, largely due to the influence of theorists like Barthes and Todorov. For a more general discussion, see, e.g., the volume edited by Patron (2021).
  • 15
    Terrone, for example, explores a line of inquiry that suggests that, in order to bridge the ‘ontological gap’, the periphery has to be populated not just with the reporter but also with a figure known in narratology as ‘the narratee’, a fictional “counterpart of the reader [...] who pays attention to the Narrator’s assertions, thereby gathering information about the Story World” (2022, p. 206).
  • 16
    No wonder that many narratologists and literary theorists prefer to speak of 'voice' rather than ‘narrator’ or ‘reporter’ in heterodiegetic fictions. The ‘voice’ conveys a “disindividualized” (Bal 2008, p. 28), abstract quality, breaking away from the notion of a personalized character. ‘Voice’ is “steeped in technology” (Bal 2008, p. 25). That is, it's a ‘tool’ or ‘convention’ that points to the artificial nature of heterodiegetic narration that operates without the full individualized presence of the narrator.
  • 17
    See, Terrone 2022, p. 207.
  • 18
    The term ‘blocked inference’ is also used, in a somewhat different context, in Köppe, Stühring 2011: 65.
  • 19
    One could claim that the question of whether the narrator of Anna Karenina inhabits a different world (in some non-metaphorical sense) is also 'silly'. The narratorial instance here is merely an anomaly, serving a specific rhetorical function of creating the impression that someone is telling us known facts. Outside of that, it is to be ignored. It seems that the majority of readers of such novels do not even question the nature of the entity narrating the story. They simply play along. There seems to exist an almost tacit readerly agreement with the convention of heterodiegetic narration.
  • 20
    However, if the narrative instance suddenly begins to present signs of an engaged personality-by employing, let’s say, sarcasm or neurotic verbal behavior-we may start to ask questions about this particular narrator. The narratorial instance is no longer deemphasized, and the BFI no longer applies. The inquiry is ‘unblocked’ and called for. It would then make sense to ask, “Who is telling us this?” Perhaps it’s an unreliable narrator, a madman hallucinating the end of the world, or someone who survives the eclipse of all other beings, similarly to Melville’s Ishmael who survives the sinking of the Pequod but concludes his narration with an apocalyptic tone, excluding himself from his own story: “Then all collapsed, and the great shroud of the sea rolled on as it rolled five thousand years ago”.
  • 21
    That is, according to the FRT, the real author “stipulates, via a higher-order speech act, that a fictional character is reporting”. I thank the anonymous referee for this phrasing, which draws on the discussions in Stokke (2023) and Batistela (2024).
  • 22
    Suppose an author wishes to convey a character’s stream of consciousness or produce a mindless fiction. Using his own identifiable voice would disrupt the narrative pretense, while relying on character-narrators would either introduce explicit contradictions or produce fictional oddities. Thus, the author employs an artifice, an unobtrusive voice capable of expressing more than characters can. It is the artificial voice of narration that we can only metaphorically refer to as ‘the narrator’, but we cannot attribute to him any particular place, personality, or history.
  • Article info
    CDD: 801
  • Funding:
    No funding agency.

Publication Dates

  • Publication in this collection
    16 Dec 2024
  • Date of issue
    2024

History

  • Received
    26 Aug 2024
  • Reviewed
    31 Oct 2024
  • Accepted
    05 Nov 2024
location_on
UNICAMP - Universidade Estadual de Campinas, Centro de Lógica, Epistemologia e História da Ciência Rua Sérgio Buarque de Holanda, 251, 13083-859 Campinas-SP, Tel: (55 19) 3521 6523, Fax: (55 19) 3289 3269 - Campinas - SP - Brazil
E-mail: publicacoes@cle.unicamp.br
rss_feed Acompanhe os números deste periódico no seu leitor de RSS
Acessibilidade / Reportar erro