CC BY
5
Эпистемология и философия науки 2022. Т. 59. № 2. С. 56-65 УДК 167.7
Epistemology & Philosophy of Science 2022, vol. 59, no. 2, pp. 56-65 DOI: https://doi.org/10.5840/eps202259221
The filmic representation
of 'relived' experiences*
Kristina Liefke - PhD, Assistant Professor. Ruhr Universität Bochum. 150, Universitätsstraße 44801, Bochum, Germany; e-mail: Kristina.Liefke@ ruhr-uni-bochum.de
This comment discusses Emar Maier's argument against the characterization of unreliable filmic narration as (first-)personal narration. My comment focuses on two assumptions of Maier's argument, viz. that the narrating character's mental states can be described independently of other mental states/experiences and that personal filmic narration can only proceed from a de se perspective (as captured by first-person shots). I contend that the majority of movies with unreliable narration represents an experientially parasitic mental state (typically, the character's remembering - or 'reliving' -a defining personally experienced event). Since these states are well-known to involve perspective-shifting and various kinds of semantic enrichment, unreliable filmic representation is perfectly compatible with the presence of a personal narrator. Keywords: unreliable narration in film, experiential attitude reports, experientially parasitic remembering, perspective-shifting, observer memory, semantic enrichment
К
кинорепрезентация «оживленного» опыта
Кристина Лифке - доктор философии, доцент. Рурский университет в Бохуме.
UniversitatsstraKe 150, 44801, Бохум, Германия; e-mail: Kristina.Liefke@ ruhr-uni-bochum.de
В этом комментарии обсуждается возражение Эмара Майера, направленное против понимания ненадежного кинематографического повествования как повествования от имени (первого) лица. Мой комментарий сосредоточен на двух предпосылках Майера, а именно, что ментальные состояния персонажа, выступающего в роли нарратора, могут быть описаны независимо от других ментальных состояний/переживаний и что личное киноповествование может происходить только в перспективе de se (будучи запечатлено кадрами от первого лица). Я утверждаю, что большинство фильмов с ненадежным повествованием репрезентирует переживаемое паразитическое психическое состояние (как правило, персонаж вспоминает -или «оживляет» - определяющее лично пережитое событие). Поскольку эти состояния, как известно, включают в себя сдвиги перспективы и семантические обогащения разного рода, ненадежная кинематографическая репрезентация совершенно совместима с присутствием личного рассказчика. Ключевые слова: ненадежное повествование в кино, сообщения о пережитых установках, паразитическое в отношении переживаемого воспоминание, сдвиг перспективы, память наблюдателя, семантическое обогащение
I would like to thank Emar Maier for his thought-provoking target article, and the guest editors, Ivan Mikirtumov and Ahti-Veikko Pietarinen, for inviting me to engage with this piece.The research for this paper is supported by the German Federal Ministry of Education and Research [BMBF] (through Kristina Liefke's WISNA professorship).
56
© Kristina Liefke
1. Introduction
In his paper Unreliability and point of view in filmic narration (this volume), Emar Maier argues against the characterization of unreliable filmic narration as a particular case of (first-)personal narration. Maier characterizes unreliable narration in film as "the occurrence of shots depicting characters and events that only exist in a fictional character's [non-veridical experience, e.g. in their] imagination, hallucination, or dream" (p. 30). His argument against unreliable personal narration is based on the assumption that, in movies, unreliable narration is typically effected by certain film conventions for attributing mental states to the movie's characters (esp. by (i) point-of-view sequences and (ii) blended perspective). Since these conventions involve either switching back-and-forth between perspectives (for (i)) or a 'third-person' point-of-view that includes the character as well as the objects of their non-veridical experience (for (ii)), Maier concludes that unreliable filmic narration is a special case of impersonal narration. The latter is narration by an abstract entity whose existence is only inferred from the perspective of the camera shots.
I am not convinced by Maier's argument. Specifically, I believe that two of its premises, viz. the assumption that the narrating character's mental states can be described without recourse to other mental states/experiences and the assumption that personal filmic narration can only proceed from a de se perspective, cannot be reasonably upheld. This is so since the majority of movies with unreliable narration represent a character's 'reliving' [=vividly and experientially remembering] or reflecting a personally experienced event (rather than the experience of this event itself), and since reliving is well-known to involve several constructive phenomena (incl. perspective-shifting and a semantic enrichment of the experienced content with hindsight information and general world knowledge).
To support my claim, I first show that point-of-view sequences in film have a close linguistic analogue, viz. 'bare experience' reports, whose formal-semantic treatment can serve as a template for the analysis of sequences (in Sect. 3). Following a presentation of the phenomenon of experiential parasitism [=the dependence of some mental states on experiences], I argue that Maier's examples of unreliable filmic narration (paradigmatically: Terry Gilliam's Fear and Loathing in Las Vegas) are not examples of 'bare experiences', but of experientially parasitic mental states (Sect. 4). I review relevant psychological work on experiential parasitism to argue that the most common filmic device for unreliable narration, i.e. blended perspective, in fact supports the adoption of a personal viewpoint (Sect. 5).
To set the stage, I precede my argument with an observation about the semantic difference between shots and statements (Sect. 2). This
observation will be relevant for my comparison of unreliable narration in film and literary texts (see Sect. 3), and for the question whether unreliable narration in texts typically is - or needs to be - first-personal (see Sect. 6).
2. The Linguistic Representation of Events
Maier attributes the difference between filmic and textual narration to the different way in which shots and statements convey discrete chunks of information, viz. iconically vs. through grammatically structured conventional symbols (pp. 26-27). However, in doing so, he neglects a crucial semantic difference between shots and statements: while statements can be supported by different situations/events (evidenced by the semantic analysis of statements as sets of situations [propositions]), shots display a single such situation (viz. the statement's truthmaker).
The difference between propositions and truthmakers already plays a role in Maier's description of the simplest [=unmediated and, hence, reliable] case of filmic narration, viz. documentaries (p. 27). There, a shot (e.g. (1c)) is intuitively denoted not by a declarative sentence (or its associated speech act, i.e. assertion; see (1a)), but by an event DP (see (1b)):
(1) a. A receptionist is standing at a reception desk.
b. (The specific event of) A receptionist's standing at a reception desk.
c. Figure 1: Still from Fear and Loathing in Las Vegas (Terry Gilliam, 1998 Universal Pictures).
3. The Linguistic and Filmic Representation of 'Bare Experiences'
Maier identifies several linguistic constructions that can be used to represent a characters' mental states (in (2): Duke's hallucinating). These include indirect and direct experience reports (below: (2a) resp. (2b)) as well as represented experience reports (e.g. (2c)). These reports differ with respect to their 'vividness' (see [Stephenson, 2010]) and with respect to whether they present the character's own subjective point of view.
a. that the receptionist is growing an eel's head.
b. this: The receptionist is growing an eel's head!
c. at the hotel. The receptionist is growing an eel's head.
Interestingly, Maier's list excludes 'bare experience' reports. The latter are reports like (3) whose non-finite [=infinitival or gerundive] complements intuitively denote events (see [Barwise, 1981; Higginbotham, 2003]). However, in contrast to (2a-c), (3) has a close filmic correspondence, viz. the point-of-view sequence in Figure 2. Given Maier's neglect of constructions like (3), his claim "[that t]he medium of film does not have obvious analogues for the use of mental state verbs with [...] embedded complements" (p. 32) is unsprising1 (see my Sect. 6).
(2)
Duke is hallucinating
(3) Duke is hallucinating the receptionist {grow, growing} an eel's head.
= Duke hallucinates the receptionist grow(ing) an eel's head in his hallucination .
Shot (event) 1: Maier's (3a)
Shot (event) 2: what Duke is hallucinating
LF: Duke hallucinates -in-wi [Aw2. the receptionist-in- W2 grows an eel's head-in- w2] Figure 2a Figure 2b
(4) (3e)[ hallucinate (e) A ag (e) = duke A
Shot (event) 2
th (e) = (ne e')[3x. receptionist (x) A ag (e1) = x A x grows-an-eel's-head (e')] ]
Stills from Fear and Loathing in Las Vegas (Terry Gilliam, 1998 Universal Pictures) with additional drawing from the author of this article
The event-semantic analysis of (3)'s LF (in (4); following [Champollion, 2015 and Liefke & Werning, 2021]) makes the correspondence between point-of-view sequences and bare experience reports explicit: the first two conjuncts in (4) describe the relevant event, e, of Duke hallucinating [=non-veridically watching sth. happen] (Shot 1); the last conjunct describes the theme (th) of Duke's hallucination (i.e. the receptionist's growing an eel's head; Shot 2). In the relevant shot from Gilliam's Fear and Loathing in Las Vegas [hereafter, Fear and Loathing] (Fig. 2a), the hallucinatory character
But see his observation (on p. 33) that "point of view sequences are not dissimilar to [...] represented perception". I will show below that - in contrast to bare experience reports - reports of represented perception (or, more generally, of represented experience) do not as straightforwardly capture cross-attitudinal dependence relations as bare experience reports.
of Duke's experience is suggested by his sweat-covered face, the joint hanging from his mouth, and by him speaking with a strong drug-induced slur (cf. the audio accompanying this shot).
4. The Filmic Representation of Experientially Parasitic Mental States
Maier identifies point-of-view sequences as a means "to give the viewer access to [the protagonist's] mental state" (p. 33). However, in his cinematic examples, the represented state seems to be more complex than what is captured by the sequence in Figure 2. Take again the hotel check-in scene from Fear and Loathing: In this scene, the events from Duke's hallucination (see Fig. 2b) are not self-contained, but are parasitic on the content/objects of Duke's veridical perception.2 The use of the definite article in the DP the receptionist in (3), the intuitive paraphrasability of (3) as (5), and the three (!)-element sequence from Figure 3 (all elements found in the original movie) all support this observation. To capture the dependence of Duke's hallucination on his perception, the LF in (5) uses different variables for the (fictional) world in which Duke's seeing/hallucinating occurs (W1), and for Duke's perception (W3) and hallucination alternatives (W2) (see: [Blumberg, 2018; Liefke & Werning, 2021]; following [Percus, 2000]):
Shot (event) 1
(5) Duke hallucinates r
Shot (event) 0
| the receptionist (from his veridical perception) grow an eel's head in his hallucination ^ Shot (event) 2 (dep't on event 0)
LF: Duke hallucinates -in-wi [Àw3. [aw2. the receptionist-in-w3 grows an eel's head-in- w2] ]
Shot 0 ('baseline')
Shot (event) 0 Shot 1: Maier's (3a)
Shot 2 (cf. Maier's (3b))
Figure 3: Figure 3a
Figure 3b
Figure 3c
Stills from Fear and Loathing in Las Vegas (Terry Gilliam, 1998 Universal Pictures) with additional drawing from the author of this article
The notion of attitudinal parasitism is itself due to Maier (2015) (see [Maier, 2017]). However, Maier largely restricts this notion to doxastic parasitism [= the dependence of attitude content on the content of a belief].
2
Note that Duke's phasing in and out of (fictional) reality already affords different - and differently reliable - first-person perspectives on Wi. The possibility of 'integrating' these perspectives accounts for the presence of "[some] events that the [hallucinating] protagonist is not privy to" (see [Maier, 2002, p. 33]). These include the fact that the woman who is growing an eel's head is, in fact, a normal [=non-eel] hotel receptionist.
The above notwithstanding, the consideration of Duke's visually parasitic hallucination does not (yet) account for Maier's reported observation of a neutral, abstract narrator.
5. The Filmic Representation of 'Relived Experiences'
In Fear and Loathing, the abstraction from the narrator can be plausibly assumed to be effected by Duke's voice over.3 In the opening of the movie, this voice over suggests that the portrayed journey through Las Vegas is a personally experienced past event which Duke is presently remembering or 'reliving'. Duke's voice over (in (6)) explicitly captures this mnemonic perspective by using the verb remember (underlined in (6)):
(6) Raoul Duke (V.O.): We were somewhere around Barstow, on the edge of the desert, when the drugs began to take hold. I remember saying something like:
Raoul Duke: I feel a bit lightheaded. Maybe you should drive...
The addition of another attitudinal 'layer' (i.e. remembering) changes the report in (5) to the report in (7):
Event 3 (V.O.)
Shot (event) 1
(7) Duke remembers PRO hallucinating Shot 0
the receptionist (from his veridical perception)
grow an eel's head in his hallucination
Shot (event) 2 (dep't on event 0) ,_1__
LF: Duke remembers-in-Wo [A <wt, y>. y hallucinates -in-wi
[Aw3. [ Aw2. the receptionist-in- w3 grows an eel's head-in- w 2] ] Shot (event) 0
Audio (3)
Duke (v.o.):
I remember [...] something like:
Figure 4:
Shot 0
Shot 1
Shot 2
Figure 4a
Figure 4b
Figure 4c
Stills from Fear and Loathing in Las Vegas (Terry Gilliam, 1998 Universal Pictures) with additional drawing from the author of this article
In fact, thescriptlab [Sinyard, 2011] lists Fear and Loathing as one of the top 10 voice overs in film, see https://thescriptlab.com/features/the-lists/1014-top-10-voice-overs-in-film/2/
In virtue of the above, the eel's head in Shot 2 can be interpreted either as an object in the memory alternatives (w1), as a figment of Duke's hallucination (w2), or both.
It is well-known in psychology and cognitive science that episodic [='event'] memory often involves perspective-shifting [Nigro & Niesser, 1983; Rice & Rubin, 2009; St. Jacques et al., 2017] and the semantic enrichment of the experience content with hindsight information and with general world knowledge [Intraub et al., 1992; Levine, 1997; McClelland et al., 1995]. Perspective-shifting changes the first-person [='own eyes'-, or field] perspective from which a remembered episode was originally experienced (e.g. veridically seen (see Fig. 4a), or hallucinated (see Fig. 4c)) to a third-person [='observer-like', or observer] perspective (see Fig. 5 for veridical visual perception; Fig. 6 for hallucination). Observer perspective is a 'blended perspective' [Meier & Bimpikou, 2019] that integrates the protagonist's (possibly distorted) view of the storyworld with an impersonal view of this world. However, in contrast to blended perspective, observer perspective takes a personal (!)4 (even if not de se) point of view.
Figure 5 (original movie shot)
Figure 6 (the author's manipulated shot)
Stills from Fear and Loathing in Las Vegas (Terry Gilliam, 1998 Universal Pictures) with additional drawing from the author of this article
In Fear and Loathing, observer perspective is made plausible by the emotional intensity of Duke's journey through Las Vegas, the (expected) temporal distance of the event, and Duke's high self-awareness with regard to this event. Work on episodic memory has identified all of these factors as triggers for perspective-shifts in memory (see, e.g., [Berntsen et al., 2003; McCarroll, 2018; Robinson & Swanson, 1990]).
I agree with Maier that, to the first-time viewer, "the blended shot is not always easily recognized as such" (p. 34). In Fight Club, this holds for joint shots of Jack and Tyler Durden. In these shots, there is nothing about Durden that would 'tip off' the viewer about Durden's non-existence in the storyworld. Things are different for my manipulated shot (i.e. Fig. 6) from Fear and Loathing. In this shot, the presence of an eel-headed receptionist is inconsistent with the (silent) assumption that Duke and Dr. Gonzo are interacting in an otherwise 'normal' environment.
4
Interestingly, perspective-shifting can even change the mode of the original experience (e.g. from feeling what it's like - or from hallucinating - to (veridically) seeing). This is achieved by a 'reshaping' of experience content on the basis of general semantic knowledge (for Fig. 4c/6: 'people cannot grow eel heads'; see [Roediger & McDermott, 1995]), by extending experience content to a wider perspective (including full objects and their background; see [Intraub et al., 1992]), by incorporating the content of other (preceding or following) events (see [McClelland et al., 1995], and/or by retrospectively distorting event content for consistency (see [Levine, 1997]). These different (re-)construction processes can even generate a close-to neutral observer-like perspective. The subjective nature of this process questions Maier's claim that "film is less suitable for first-person (unreliable) narration" (p. 29).
6. Conclusion
I close my comment with a remark on Maier's observation that movies use certain "film language analogue[s]" of linguistic report constructions (p. 35): Maier suggests that, in film, a distorted view of the storyworld is most easily presented by means of point-of-view sequences and blended perspective. He contends that, since blended perspective is inherently abstract and removed from a particular character's point of view, it disqualifies as an instance of (first-) personal narration.
I believe that Maier's argument against unreliable personal narration in film can also be used against (certain cases of) unreliable personal narration in literary texts: Some researchers [e.g. Anand, 2011; Walton, 1990] have argued that memory reports with subject-controlled gerundive complements (e.g. (8)) are ambiguous between an 'inside' (subjective, or field-perspective) reading (i.e. (8a)) and an 'outside' (objective, or observer-perspective) reading (i.e. (8b)). The 'inside' reading captures what it was like for the rememberer to undergo the described experience. The 'outside' reading captures what it would be like for the rememberer to witness this event from an out-of-body point of view (see [Liefke & Werning, 2021]).
(8) Duke remembers [hallucinating an eel].
a. Duke remembers what it was like to hallucinate an eel.
b. Duke remembers seeing/watching himself hallucinate an eel.
The 'inside' reading interprets the complement in (8) as the event from Figure 4c. The 'outside' reading interprets this complement as the event from Figure 6. The availability of an 'outside' reading of (8) suggests that, just like unreliable narrators in film - and pace Maier -, unreliable narrators in literary texts can also be (interpreted as) non-de se narrators.
References
Anand, 2011 - Anand, P. Suppositional Projects and Subjectivity. 2011 [http:// web.eecs.umich.edu/~rthomaso/lpw11/anand.pdf, accessed on 18.01.2022].
Barwise, 1981 - Barwise, J. "Scenes and Other Situations", The Journal of Philosophy, 1981, vol. 78 (7), pp. 369-397.
Berntsen et al., 2003 - Berntsen, D., Willert, M., & Rubin, D.C. "Splintered Memories or Vivid Landmarks?", Applied Cognitive Psychology, 2003, vol. 17 (6), pp. 675-693.
Blumberg, 2018 - Blumberg, K. "Counterfactual Attitudes and the Relational Analysis", Mind, 2018, vol. 127 (506), pp. 521-546
Champollion, 2015 - Champollion, L. "The Interaction of Compositional Semantics and Event Semantics", Linguistics and Philosophy, 2015, vol. 38 (1), pp. 31-66.
Higginbotham, 2003 - Higginbotham, J. "Remembering, Imagining, and the First Person", in: A. Barber (ed.). Epistemology of Language. Oxford University Press, 2003, pp. 496-533.
Intraub et al., 1992 - Intraub, H., Bender, R.S. & Mangels, J.A. "Looking at Pictures But Remembering Scenes", Journal of Experimental Psychology: Learning, Memory, and Cognition, 1992, vol. 18 (1), pp. 180-191.
Levine, 1997 - Levine, L.J. "Reconstructing Memory for Emotions", Journal of Experimental Psychology: General, 1997, vol. 126 (2), pp. 165-177.
Liefke & Werning (in press) - Liefke, K., & Werning, M. "Factivity Variation in Episodic Memory Reports", in: T. Yasufumi & K. Mineshima (eds.), New Frontiers in Artificial Intelligence. Lecture Notes in Computer Science. Heidelberg: Springer. (In Press)
Liefke & Werning, 2021 - Liefke, K., & Werning, M. "Experiential Imagination and the Inside/Outside Distinction", in: N. Okazaki, K. Yada, K. Satoh, & K. Mineshima (eds.), New Frontiers in Artificial Intelligence. JSAI-isAI 2020: Vol. LNAI 12758. Springer, 2021, pp. 96-112.
Maier, 2015 - Maier, E. "Parasitic Attitudes", Linguistics and Philosophy, 2015, vol. 38 (3), pp. 205-236.
Maier, 2017 - Maier, E. "Referential Dependencies between Conflicting Attitudes", Journal of Philosophical Logic, 2017, vol. 46 (2), pp. 141-167.
Maier & Bimpikou, 2019 - Maier, E. & Bimpikou, S. "Shifting Perspectives in Pictorial Narratives", Proceedings of Sinn und Bedeutung, 2019, vol. 23, pp. 1-15.
McCarroll, 2018 - McCarroll, C.J. Remembering from the Outside: Personal Memory and the Perspectival Mind. Oxford University Press, 2018.
McClelland et al., 1995 - McClelland, J.L., McNaughton, B.L., & O'Reilly, R.C. "Why There Are Complementary Learning Systems in the Hippocampus and Neocor-tex: Insights from the Successes and Failures of Connectionist Models of Learning and Memory", Psychological Review, 1995, vol. 102 (3), pp. 419-457.
Nigro & Neisser, 1983 - Nigro, G. & Neisser, U. "Point of View in Personal Memories", Cognitive Psychology, 1983, vol. 15, pp. 467-482.
Percus, 2000 - Percus, O. "Constraints on Some Other Variables in Syntax", Natural Language Semantics, 2000, vol. 8, pp. 173-229.
Rice & Rubin, 2009 - Rice, H.J. & Rubin, D.C. "I Can See It Both Ways: First-and Third-Person Visual Perspectives at Retrieval", Consciousness and Cognition, 2009, vol. 18, pp. 877-890.
Robinson & Swanson, 1990 - Robinson, J.A. & Swanson, K.L. "Autobiographical Memory: The Next Phase", Applied Cognitive Psychology, 1990, vol. 4 (4), pp. 321-335.
Roediger & McDermott, 1995 - Roediger, H.L. & McDermott, K.B. "Creating False Memories: Remembering Words Not Presented in Lists", Journal of Experimental Psychology: Learning, Memory, and Cognition, 1995, vol. 21 (4), pp. 803-814.
Jacques et al., 2017 - St. Jacques, P.L., Szpunar, K.K., & Schacter, D.L. "Shifting Visual Perspective During Retrieval Shapes Autobiographical Memories", Neurolm-age, 2017, vol. 148, pp. 103-114.
Walton, 1990 - Walton, K.L. Mimesis as Make-Believe. Harvard University Press, 1990.
