CC BY
3
Эпистемология и философия науки 2022. Т. 59. № 2. С. 81-84 УДК 167.7
Epistemology & Philosophy of Science 2022, vol. 59, no. 2, pp. 81-84 DOI: https://doi.org/10.5840/eps202259225
Other points of view:
replies to comments
Emar Maier - PhD, Assistant Professor.
University of Groningen. 9712 CP Groningen, The Netherlands; e-mail: e.maier@rug.nl
Эмар Майер - доктор философии, доцент. Университет Гронингена. 9712 CP, Гронинген, Нидерланды; e-mail: e.maier@rug.nl
In my target paper, Unreliability and Point of View in Filmic Narration, I tried to argue that the familiar notion of an unreliable narrator, as discussed by literary scholars, linguists and philosophers, does not carry over well into the domain of filmic narration. The key assumptions in this argument were that unreliable narration requires a personal narrator, and that the filmic analogue of a narrator is typically impersonal. I further suggested that, to achieve the kind of ambiguous story-telling (is this really happening or is she just imagining this?) movie directors rely on other conventionalized techniques for reporting the contents of mental states.
I am deeply grateful to the scholars who took up the gauntlet and seriously engaged with my argument - despite its perhaps obvious flaws -, expertly dissecting and dismantling every part of it. Reading these comments was a humbling experience, and it has already significantly changed my thinking about narration and representation in text and film. Though, for reasons of time and space, this note records just some preliminary first impressions, I will be revisiting this thought-provoking collection of papers and mine it for original ideas in the years to come.
Dorit Abusch starts by offering a new characterization of unreliability. Unreliable narration occurs when a text gives rise to two distinct propositional contents - a literal content (where bats are swooping down on Duke, say) and a secondary content (where Duke is hallucinating that
© Emar Maier 81
Другие точки зрения:
ответ оппонентам
EMAR MAIER
bats are swooping down on him). This pragmatic characterization no longer presupposes first-person narration.1 Applied to film we can now analyze movies shot with impersonal viewpoints as unreliable if they license the derivation of these two clashing levels of content. Next, Abusch exploits a linguistic analogy to argue that conveying de se attitudes need not involve first-person morphology (i.e., first-person shots). While languages like Amharic report Jack's self-ascription of having a cold with an embedded first-person morpheme (lit: 'Jack said that I (=Jack) have a cold'), English just uses third-person morphemes ('Jack said that he (=Jack) has a cold'). Blended shots in film then are simply more like (the complements of) English attitude reports than Amharic ones. Indeed, I have also likened blended shots to English indirect discourse constructions. However, note that the English and Amharic reports are not in fact synonymous: only the Amharic variant forces a de se interpretation, the English variant is compatible with de se and de re situations. I would suspect an analogous difference between blended shots and more Amharic-like (or rather, more quotation-like?) non-veridical point of view shots.
Marc Champagne questions my hasty dismissal of Ann Banfield's famous position that in some narratives there is no narrator at all and the events "narrate themselves". Interestingly, Champagne thinks my dismissal may be defensible for linguistic narratives, which arguably do presuppose a speech act, which in turn presupposes an agent. But for visual narratives, he suggests, Banfield's suggestion is spot on. For Champagne the iconic nature of the discourse units in filmic narration implies a fundamentally different type of narration, which he aptly calls 'nonpersonal'. This mode of storytelling then yields a new type of unreliability where it's not the narrator but the viewer who unreliably infers the narrative from the shots. I've always thought there's something attractive about Banfield's narratorless view, but I never quite understood how events can narrate themselves, and applying it to the iconic mode of representation does not help me make sense of it.
Elena Dragalina-Chernaya re-analyzes the technique of the 'blended perspective shot' in terms of a clash between the objective "view from nowhere" (which she likens to Dziga Vertov's Kino-Eye style of filmmaking) and the subjective representation of the unreliable narrator (which she compares to the fake paper-eyes depicted in the 1989 movie Prishvin's Paper Eyes). Like Bucking, she draws attention to a crucial and admittedly neglected difference between verbal free indirect discourse and blended perspective shots, viz. that the latter (almost) never come with clear grammatical or otherwise conventionalized and detectable markers. In response I would again say that verbal free indirect
For a related argument based on an alternative definition of unreliability, see Sebastian Bücking's "Unreliable narration in film: A comment on Emar Maier's 'Unreliability and Point of View in Filmic Narration'". https://tinyurl.com/4bdkxwfa
OTHER POINTS OF VIEW: REPLIES TO COMMENTS
discourse is frequently unmarked at the morphological level as well, but admit that the context usually does disambiguate much more immediately and more precisely in the case of verbal free indirect discourse.
Kristina Liefke provides an intriguing analysis of blended perspective shots as a natural way of showing a personal narrator's first-person re-lived experiences. Zooming in on Fear and Loathing she first notices that in the movie's opening scene, Duke, through first-person voice-over, narrates that he remembers driving to Vegas and taking drugs. What follows is then best interpreted as depicting what Duke remembers. Liefke suggests we can think of the movie as a filmic representation of Duke's actual, first person memories. Even blended shots (such as her manipulation of a wide shot showing both Duke and the receptionist, with hallucinated eel's head, are direct, first-person representations of Duke's memory. This is because episodic memory is known to apply a mental perspective shift operation on itself, especially over time: we literally remember scenes in blended perspective. Fear and Loathing then provides us with an example of a genuinely first-person filmic narrator, giving us visual access to what they themselves are mentally experiencing in remembering a sequence of events. I find this a very attractive analysis, and would quibble only with a brief remark about another overlooked difference between filmic and verbal representation: while verbal statements express possible worlds propositions, a shot, says Liefke, expresses but a single situation, its truthmaker. I'm not exactly sure why it's necessary to assume such a fundamental difference if we can just as easily treat pictures (and shots) as expressing classical sets of (centered) possible worlds (those worlds that, when projected from some viewpoint onto a 2D surface would yield that picture (or shot).
Julian Schloder challenges two central claims of my paper. First, he argues that unreliable verbal (or filmic) narration need not always be personal. To show this he distinguishes different types of unreliability. Unreliability by omission, for instance, involves merely leaving out information. If so, every narration - personal or impersonal - is to some degree unreliable, as every narrator leaves out some (fictionally) true facts. The second challenge builds on the first. If even in third person narratives the narrator can - or must - be unreliable, then they are not that impersonal after all. According to Schloder, third person narratives like Harry Potter actually present the story "from the perspective of a character... without this character explicitly being the narrator". Schloder's prime example, A Song of Ice and Fire, seems to be a typical third person omniscient type of narrative, but according to Schloder each chapter is told from the perspective of a single character. As I understand 'perspective', a focus on the actions of a specific character, combined with an abundance of (free indirect discourse) reports of their stream of consciousness does not constitute a telling from that character's perspective. A fully reliable, detached, omniscient, impersonal narrator can tell a story that way.
EMAR MAIER
I wonder to what extent the style Schloder has in mind overlaps with phenomena known as protagonist projection or focalization, which include the blending of a protagonist's characteristic word choices in a seemingly impersonal narrative.
Daniel Tiskin continues the theme of highlighting crucial differences between film and verbal narration. For instance, he starts by observing that in a verbal first-person narrative the reader need not imagine themselves being in the shoes of the first-person, while in film the camera forces a degree of identification between the viewer and the camera viewpoint (this identification, I would add, is exactly what is exploited in firstperson shooter video games and VR experiences). Furthermore, he notes, in line with for instance Liefke, that verbal statements are by their (finite, conventional) nature necessarily non-exhaustive. The content of a shot, by contrast, is completely causally determined by what is in front of the camera. The camera is thus assumed to be inherently reliable, and the only artistic freedom concerns setting up the camera and what's in front of it (e.g., the actors). Though perhaps an oversimplification (there is a lot of content creation in editing, sound design, music, camera choice and focus, lighting, digital post-production), this presumed reliability of moving images may help explain why for instance the rise of deepfakes is so much more troubling than the spreading of fake news tweets.
Suren Zolyan, finally, agrees with me that unreliable narration, as applied to film adaptations of novels like Lolita or Fight Club, is a misnomer. But he goes much further, questioning the applicability of more basic narratological notions like personal and impersonal narrators. In the end, he is on board with my assumption that filmic narration presupposes a filmic narrator - not to be confused with the director, author, or voice-over. He then appeals to David Lewis' possible worlds account of truth in fiction to characterize the narrator as bifurcated: in some possible worlds they reliably tell (or show) the truth, while in the actual world, they tell a fictional story. Along the way, Zolyan, like some other com-menters, points out more overlooked differences between film and verbal narration, for instance quoting Pasolini's observation that somehow, "cinema, or the language of im-signs, has a double nature. It is at the same time extremely subjective and extremely objective".
