Flee: Sounds of Fright and Flight

Abstract

Animation and live-action are two closely related media, which are foremost distinguished by the ideas and conventions surrounding them. The diverging discourses around animation and live action have tended to focus on animation as something constructed to represent characters and settings and on live action as something capturing actors and sets representing characters and settings. This difference between constructing and capturing, along with the perceived indexicality of the photo, is what seems to suggest live action as the preferred medium for documenting real events. Sound effects, in the form of recorded and edited sounds of objects, actions and environments, are of particular interest here, as they can be considered to balance somewhere between these poles of construction and capture, between the non-indexical and indexical, and ultimately between representation and reproduction. In this article, I will focus on aspects of ‘truth’ (understood as corresponding to some external reality) and ‘realism’ (understood as a representation of external reality) and how something comes to be perceived as truthful or realistic in animated documentaries in relation to the role played by sound effects. By discussing the Danish film Flugt [Flee], I will show how sound effects can aid in creating representations of truth.

Animated documentaries have existed since the early days of film, with the first animated documentary considered to be The Sinking of the Lusitania by Winsor McCay in 1918 (Honess Roe 2013: 6). In the years since, animated documentaries and autobiographies have only grown in popularity, particularly since the 1990s (Honess Roe 2013: 1), with notable and award-winning examples being the French Persepolis (Satrapi and Paronnaud 2007) dealing with the history of Iran, the Israeli film Waltz with Bashir (Folman 2008) about the 1982 Lebanon war, and more recently the Danish Flugt [Flee] (Rasmussen 2021) dealing with a flight from the war(s) and regime in Afghanistan. Still, the idea that you can present a documentary or autobiography in the form of an animated film seems almost paradoxical, and the live-action medium is still the dominant mode for audiovisual non-fiction.

Animation film and live-action film are two closely related media, which are foremost distinguished by their qualifying aspects (cf. Elleström 2021) – by the ideas, discourses and conventions surrounding them, more than by their basic media (Jensen 2021: 74–77). Animation uses drawings, dolls, clay or computer graphics to generate images which are then presented in quick succession, 24 frames per second, and typically with added sound, whereas live action uses live actors and sets which are recorded and presented as images in quick succession, 24 frames per second, typically with added sound. ^a The diverging discourses around animation and live action have tended to focus on animation as something that is constructed to represent characters and settings, and on live action as something that captures actors and sets representing characters and settings (Jensen 2021: 75–76). This difference between constructing and capturing, along with the perceived indexicality of the photo (although we very well know a photo might be manipulated), is what seems to suggest live action as the preferred medium for documenting real events. Sound effects, in the form of recorded and edited sounds of objects, actions and environments, become an element of particular interest here, as they can be considered to balance somewhere between these poles of construction and capture, between the non-indexical and indexical, and ultimately between representation and reproduction.

When it comes to the soundtrack, animated documentaries hold the same potential (or lack thereof) for indexicality and reproduction as live action. ^b A film, whether it be animated or live-action, is a complex medium integrating moving images with several auditory media. A film’s soundtrack typically consists of three such auditory ‘submedia’: music, auditory text (in the form of dialogue, voice-over, etc.) and sound effects (all sound that is not speech or music) (Jensen and Salmose 2022: 30). ^c While auditory text (often, somewhat confusingly, simply referred to as sound) in the form of interviews has been associated with ‘realism’ in animated documentaries (Ward 2005: 98) ‘truth’ (Honess Roe 2013: 28), and ‘credibility’ (Ehrlich 2021: 55), animation studies often highlight sound effects for their ability to both enhance the sense of realism and to withdraw from it depending on the style (Taberham 2018; Thom 2013; Whittington 2012).

In this article, I will focus on these aspects of ‘truth’ (understood as corresponding to some external reality) and ‘realism’ (understood as a representation of external reality) and how something comes to be perceived as truthful or realistic in animated documentaries in relation to the role played by sound effects. By discussing two different examples from the Danish film Flugt [Flee] (Rasmussen 2021), I will show how sound effects aid in creating representations of truth by two means: (1) by materializing and anchoring animated segments in a physically experienced reality, and (2) by directing attention and underscoring thematic and emotional underpinnings of the characters and the story.

Flee is an animated documentary which tells the story of Amin (pseudonym) and his family, who flee from Afghanistan during the 1990s. The film is based on audio interviews where Amin unfolds his memories. Mimetic animation illustrates most of the story, but the mimetic animation is also intersected with found footage of news coverage relating to the memories, and with sections consisting of stylized, abstract visuals. As such, the relationship between fact and fiction, and between animation and sound effects, varies through different sequences, building a complex and compelling case for a study of sound in animated documentaries.

The Animated Documentary

Definitions of the documentary and animated documentaries vary and several different documentary genres (or ‘modes’, cf. Nichols 2017) exist, but a commonality in discussions of documentaries is a focus on ‘truth’ or ‘reality’ (which I define and discuss in the section below) on the one hand, and some element of creativity, interpretation or representation on the other. Among the most influential writers on documentaries is film scholar Bill Nichols, who notes that there are several overlaps between fiction films and documentaries regarding style and production, but ultimately the two media can be distinguished because ‘documentaries address the world in which we live rather than a world imagined by the film maker’ (Nichols 2017: xi). Similarly influential has been documentarist John Grierson’s definition, where documentary is considered in terms of a ‘creative treatment of actuality’ (Grierson 1933, cited in Nichols 2017: 5). Nichols goes on to stress this implied tension between reality and creativity, arguing that documentary ‘draws on and refers to historical reality while representing it from a distinct perspective’ (Nichols 2017: 5) – a distinct perspective showcasing the voice and style of a director.

This distinction between reality and the creative representation of reality is particularly pertinent in an animated documentary such as Flee, which by no means attempts to conceal its creative styling of the non-fiction story. Building on this inherent tension and on work by both Nichols and Grierson, Annabelle Honess Roe has suggested defining animated documentary as an audiovisual work that:

(i) has been recorded or created frame by frame; (ii) is about the world rather than a world wholly imagined by its creator; and (iii) has been presented as a documentary by its producers and/or received as a documentary by audiences, festivals or critics. (Honess Roe 2013: 4)

The inspiration from Nichols is clear in the second requirement proposed by Hones Roe, while the first requirement, that the work should be ‘recorded or created frame by frame’ (my emphasis) encompasses all imaginable kinds of animated film. The last requirement showcases the importance of the institutional context and reception and perhaps implies what animation scholar Nea Ehrlich refers to as a ‘visual paradigm shift’ (Ehrlich 2021: 3), in which animation, as a medium of representation, is ‘poised between fact and fiction’ (Ehrlich 2021: 8), in the sense that animation is becoming more and more pervasive in non-fiction contexts, its ability to represent reality being increasingly acknowledged, while also still retaining associations with fantasy and children’s entertainment. At the same time, the credibility of the photographic image has seen a crisis with the rise of digital imagery and the ability to manipulate and construct fake content (Ehrlich 2021: 2–10). It is at this intersection between ‘fact and fiction’, between reality and representation, that Flee is operating, gaining institutional acknowledgement for its credible construction of real events through being nominated for the best documentary feature in both the Academy and Bafta awards in 2022, while also, at times, highlighting its own artificiality and artistic voice, for example through highly stylized sequences, which only portray grey colours and simple line drawings of people running, leaving the mimetic representations behind to focus on the embodied, subjective experience of fleeing.

This ability of animation to represent subjective experiences has often been highlighted as one of the strengths of animation (see, for example, Ehrlich 2021: 9), and Honess Roe has suggested that instead of trying to classify animated documentaries according to different schemes or modes based on historical or stylistic characteristics, one should focus on the functions that the animation in animated documentaries fulfil, functions arising from the freedom in not being indexically tied to the reality represented, like the live-action documentary (Honess Roe 2013: 22). Roe suggests three such functions: ‘mimetic substitution’ for non-existing ‘real’ footage, ‘non-mimetic substitution’, e.g., to anonymize interview subjects (Honess Roe 2013: 23–24), and finally ‘evocation’, representing ‘concepts, emotions, feelings and states of mind’ (Honess Roe 2013: 25). These functions are all evident in Flee, which uses mimetic and non-mimetic substitution to represent the events taking place in Amin’s story, and, more importantly, the animation protects Amin’s anonymity, securing his privacy when sharing a highly personal and somewhat controversial story (part of Amin’s flight involves lying to the Danish authorities to secure asylum). The evocative function is particularly prominent in the stylized sections, which I will discuss in more detail in the analysis below. These functions of animated documentary are not in contrast with the documentary’s aim to balance truth/reality and creativity, but rather allow for different aspects of the truth (substitution), and different kinds of truth (evocation), to be represented.

The Problem of ‘Truth’ and the Indexical Link in Animated Documentaries

The idea of ‘truth’ is strongly tied to ‘reality’ and ‘realism’. For something to be true, it needs to correspond with reality, which can only, to some degree, be represented by a realist aesthetic. As stated by sound designer and scholar Leo Murray:

Realism often means ‘the illusion of reality’ – realism as a representation, without limiting the representation to truthfulness or authenticity. […] Reality on the other hand is real life – actuality – as we experience it. […] Realism then is the product of creative choice, like any other in film-making or documentary making. (Murray 2010: 132)

To Ehrlich, realism is ‘the articulation of “the real”, or what is accepted as a believable delivery of “the real”’ (Ehrlich 2021: 27). Both scholars highlight the aspects of creativity, interpretation and representation in documentaries, rather than seeing documentaries as reproductions of something real. Despite this tension between reality and creativity, it is still common to discuss the relation between documentaries and reality as a matter of the photographic proof of a profilmic reality and, by extension, of the indexicality of the photographic medium. I will not recount the debate and arguments relating to documentaries, photographic imagery and indexicality, which has been done extensively by Honess Roe (2013: 27–40) and Ehrlich (2021: 27–83); suffice it to say that photo-realistic live action is still the dominant norm for documentary filmmaking and that the credibility associated with this film-making mode is largely associated with the idea that what is shown on the screen is in some way linked with a profilmic reality. It is important to note, though, that in discussions of the evidentiary status of photography, it is both the iconic resemblance to the real world and the indexical trace carried by a photograph that is highlighted, often in connection to each other (Ehrlich 2021: 56; Honess Roe 2013: 29). One enhances, or validates, the other, so to speak:

by virtue of the indexical relationship with reality, photographs are forced, through the process of light from an object refracting through the camera lens and hitting film, to correspond directly with what was in front of the camera and thus what the photographic image shows us is what we would have seen ourselves. This mutually dependent relationship of causality and analogy has maintained dominance in the interpretation of moving images, especially documentary ones. (Honess Roe, 2013: 29)

Following this, the perceived paradoxicality of animated documentaries persists in the fact that in most types of animation, no direct indexical connection between the visuals and the external world exists. In those types of animation that do have a clear indexical link, i.e. rotoscope, stop motion and motion capture, the ‘real’ source is blurred by layers of animation (rotoscope, motion capture) or the source is obviously manipulated and doesn’t directly link with ‘actuality’ (stop motion) (for an extensive discussion of indexicality in these and other forms of animation, including deictic indexicality, see Ehrlich: 2021). Furthermore, even though photo-realistic animation technology is available today, many animated documentaries, such as Flee, do not take advantage of this possibility, but deliberately construct images that are clearly distinguishable from photorealism, even while maintaining a high degree of resemblance to the real world (for a discussion of photorealism in animated documentaries, see Honess Roe: 2013). In these types of animated films, the lack of visual indexicality and perfect iconic resemblance thus leads to a different perceptual contract with the viewer. When it comes to sound effects, however, the iconic resemblance to real life is often maintained to a higher degree, which, I will argue, is part of constructing what I will refer to as perceptual indexicality, maintaining part of the evidentiary status associated with real indexicality.

As stated above, several scholars on animated documentaries acknowledge the importance of sound for maintaining a link to reality, with Honess Roe stating that:

At the same time, animated documentaries often retain the same oral/aural link with reality as conventional documentaries, by using audio recording of interviews with documentary subjects for example, and ‘there is a “realism” or indexicality to the sound that does not reside in the image’ (Ward 2005a: 98). (Honess Roe 2013: 27)

Whereas both Honess Roe and Paul Ward, quoted by Roe, mainly discuss ‘auditory text’, i.e. the functions of the voice in association with verbal statements, and only fleetingly mention other kinds of ‘diegetic’ or ‘location’ sounds (i.e. sound belonging to the represented world, in contrast with, for example, underscore music), I would argue that sound effects also hold an important ‘realism’ and (perceptual) indexicality, even in cases where the sound effects are added library sounds or Foley recorded in a studio. Furthermore, while non-verbal aspects of the voice have been discussed, e.g., by Honess Roe in association with interviews in the animated documentary (Honess Roe 2013), and by Irina Leimbacher in association with experimental documentary (Leimbacher 2017), the effect created when a voice is detached from the verbal mode altogether and used as a sound effect, an important feature in Flee, deserves more attention.

Although Flee is largely based on recorded interviews, it also uses a substantial amount of sound effects added in postproduction to create an auditory rendering of Amin’s story. As such, the film takes a similarly creative approach to the sound design as to the visual animation, once again blurring the boundaries between live-action cinema and documentary expectations, where the contemporary documentary tends to be associated with unmanipulated location sound (i.e. sound recorded along with the images, or in this case, along with the interviews) (Murray 2010; Rogers 2020: 137). However, as several scholars working on sound in documentaries have shown, the attitude towards sonic manipulation in documentary varies throughout history and between different documentary genres, and in contemporary documentary, creative treatment of sound design, including sound effects, is by no means unheard of, particularly in nature documentaries (Collins 2017; Murray 2010) and in artistic or experimental documentaries (Kara and Thain 2015; Rogers 2020; Strachan and Leonard 2015), where sound effects are used to varying degrees to create affect, empathy, narrative, or aesthetic effect.

A film typically makes use of three types of sound effects: location/production sound recorded on-site, which might later be edited; library sound, i.e., sound effects recorded, or created from midi, stored in an audio library; and Foley, i.e., sound effects created in a studio specifically for the film in question. It follows that of these different possibilities, only the production sound actually holds an indexical relationship to the sources presented in the images (for live-action) or the interview situation. Nonetheless, the use of ‘fake’ sound effects is so ubiquitous in audiovisual media, and conventions of their use so ingrained in most audiences of Western media, that we tend to accept ‘fake’ sound effects as ‘real’. As Michel Chion has argued (for fiction film):

sound that rings true for the spectator and sound that is true are two very different things. In order to assess the truth of a sound, we refer much more to codes established by cinema itself, by television, and narrative-representational arts in general, than to our hypothetical lived experience. (Chion 1994: 107)

Also writing about fiction film, film-music scholar Birger Langkjær has argued that Foley sounds add to the perceptual reality (cf., Prince 1996) of a film:

In general, the lifelike appearance of the moving images and the impression of reality they trigger can be coined ‘perceptual realism’. This perceptual realism is different from realism in terms of content, as the basic lifelike qualities of persons, objects and movements are not related to whether persons, objects and movements occur in a realistic or a non-realistic fictional world. (Langkjær 2010: 6)

As Langkjær further argues, Foley sounds add to the perceptual realism of films by being synchronized with seemingly realistic sound sources represented visually and by adhering to the same perceptual processes that real-life, everyday sound does (Langkjær 2010: 10–11). Furthermore, Foley sounds seem perceptually real because they, in general, replicate some characteristics of the sounds they are representing (Langkjær 2010: 10–12). Whether footsteps are recorded on-site or in a studio later, there are several aspects of the sounds that will correlate. Likewise, when Foley artists create Foley, for example for galloping horses, not all materials will do, only materials such as coconut shells that can produce a sound with an iconic (note that in Peirce’s sign theory, iconic relates to similarity and not necessarily to visuality, see Bruhn and Schirrmacher 2022: 23–24) relationship to the sound of hooves.

This impression of perceptual reality on the soundtrack happens despite the fact that many Foley sounds are designed to enhance certain qualities and diminish others in comparison with real-life sound. Echoing the quote by Chion above (who doesn’t specify what kind of sound effects he is talking about), I would argue that since the use and design of Foley sounds live up to filmic and aesthetic conventions for the construction and decoding of perceptual reality, these sounds add to the sense of perceptual reality precisely by adhering to these conventions – a point which Langkjær also hints at (Langkjær 2010: 10).

These arguments about Foley sound and perceptual realism would seem to apply to library and edited production sounds as well, in documentary as well as in fiction cinema. As Murray has stated in a discussion of documentary sound:

Audience expectation plays a role in determining the elements chosen for use in a soundtrack – we do not usually notice artifice in the soundtrack when we are presented with what appears to be a reasonable representation, which matches what we are seeing. Further, we can become aware of, and even distrustful of the soundtrack when it presents a real and authentic synchronous, if it does not meet with our preconceptions of what things should sound like. (Murray 2010: 132)

In a similar vein, Karen Collins, writing about exaggerated sound effects in David Attenborough’s nature documentaries, refers to the sound design as ‘cinematic realism’, stating that ‘[t]his is not the realism of everyday life, but a kind of expected believability imposed after years of media consumption’ (Collins 2017: 70). What is perceived as realistic is thus always a matter of convention. What I am aiming at here is that if sound effects are perceptually real to audiences, then audiences will, subconsciously, react to these sounds as they would to any real sound outside the screen, that is, to any sound functioning as an indexical referent of its real material source.

This is important for two reasons: (1) because this means we can analyse sound effects in a film such as Flee as if they were real, everyday sounds; (2) because, as indicated above, indexical signs hold a relatively higher truth value than any of the other types of signs.

Intermedial scholar Lars Elleström has argued that indices, due to their basis on contiguity, enhance a media product’s truth claim because they:

create external truthfulness in linking the intracommunicational and extracommunicational domains. In contrast to other sign types, which may indeed correctly correspond to the extracommunicational, indices are those signs that actually establish truthfulness. (Elleström 2018: 444)

The intra-communicational and extra-communicational domains can, for the subject of this article, be understood as the world constructed within the film and the world external to the film, respectively. This would mean that sound effects not only enhance the perception of realism in a film because they imitate real-life sounds (i.e., function as icons of real-life signs), but because they imitate, and are experienced as, ‘true’ indexical signs of an external reality. This argument can again be related to Chion’s theory of the audiovisual and his notion of ‘materializing sound indices’ as ‘the sound’s details that cause us to “feel” the material conditions of the sound source, and refer to the concrete process of the sound’s production’ (Chion 1994: 114). The difference between Chion’s concept of materializing sound indices and my concept of perceptual indices is that Chion only discusses elements of sound that directly point to, and make the listener aware of, the sound’s production process, whereas I take a broader perspective by including sounds that subconsciously and perceptually function as indices. When Gregg Barbanell recorded the breaking of celery stalks to use as Foley for breaking bones in The Walking Dead (Collins 2020: introduction), then the materializing sound indices are to the breaking of celery stalks, but the perceptual index we as audiences are tricked into hearing, points to the breaking of bones. In a similar vein, Murray has stated that although sound recording is never a reproduction of reality (because sound recording is always framed and partial) and in cinema rarely a true index (Murray is recounting the arguments made by Rick Altman: 1992) ‘we can more accurately describe cinematic sound as retaining, reconstructing or creating the illusion of an indexical link’ (Murray 2019: 86, emphasis added).

A sound is always an index of its source – it bears the trace of the materials and interaction creating it. As argued in ecological psychological research, this means that a sound carries with it information about this source and interaction, ‘a given sound provides information about an interaction of materials at a location in an environment’ (Gaver 1993: 6, emphasis in original) (see also Carello et al. 2005 for a review of empiric acoustic ecological research). It follows, that, as Chion’s quote above implies, an important function of sound effects might be to ‘materialize’ objects, environments and events in film, which could be particularly important in an animated film. In a live-action film, we may have visuals that, through colour and texture, give us an impression of the material world the characters are moving in, features that will often be lacking in non-photorealistic animation but, nonetheless, a large part of the materiality of the ‘intra-communicational’ (cf. Elleström), will remain invisible to the audience and can only be supplied by the sound effects, which provide this materiality by creating a link to the ‘extra-communicational’ domain, i.e., reality. As such, the materializing function and the function of ‘truth’ are interrelated and co-dependent.

The idea of the perceptually indexical link created through sound effects that I have argued for above is important for two reasons: (1) it aligns with the discourses above surrounding the necessity of indexicality for the representation of ‘truth’ and ‘reality’ and thus suggests an explanation of the mechanics underlying arguments surrounding sound effects’ role in creating ‘realistic’ representations; (2) it highlights how sound effects become meaningful in animated documentaries and in film more generally. With this in mind, I will now move on to the case study Flee and provide a general introduction to the film, before going into more detail with two examples, focusing on sound effects’ ability to materialize and evoke auditive memories, and on how voice can be treated as a sound effect to highlight thematic threads and increase empathy and identification.

Flee

Flee (Rasmussen, 2021) tells the true story of Amin (pseudonym) who flees Afghanistan during the civil war in the early 1990s ^d (21:06–23:50 min) together with his mother, two sisters and one older brother. Initially, the family flees to Russia, which is the only country that would grant the family a tourist visa (24:52 min). The film depicts the struggles of being an illegal immigrant on an expired visa in Russia with corrupt police, and how the family attempts to flee even this country. The two sisters are the first to be smuggled out to Sweden where an older brother already lives, and they nearly die on the way in a locked container onboard a ship. Next, the rest of the family tries to escape, with a group of barbaric smugglers who lock them underneath the deck of a small ship, and again, the flight nearly kills them during a storm where the ship starts to take in water. Eventually, the family is picked up by the Estonian border police and placed in a closed-off refugee camp, from where they are sent back to Russia. Finally, the young teenage boy Amin (around 15 at the time, perhaps a little younger) escapes to Denmark with a false passport and a story about being the only surviving member of his family, in order to convince the Danish authorities to take him in and give him asylum.

The different segments of the film can be divided into five different categories based on the visual style: (1) studio segments showing Amin answering questions and narrating in a studio; (2) everyday segments based on participatory observation/interviews taking place in Amin’s everyday life; (3) memory segments with animation that supports Amin’s narration, sometimes with added scripted dialogue (e.g. by Amin’s family); (4) stylized animated segments based on various blended grey colours as a background and simple line drawings; and (5) found footage. The soundtrack consists of auditory text in the form of Amin’s narration and dialogue between Amin and Rasmussen, naturally occurring dialogue in the everyday segments, scripted dialogue in the animated memory segments, naturally occurring sound effects in the everyday segments (maybe production sound, but it could also be added in postproduction), and sound effects created in postproduction for the memory and the stylized sequences. Finally, the soundtrack includes non-diegetic atmospheric music for creating mood and diegetic pop music for characterizing Amin at different stages, as well as sound of the found footage.

Sound for Materializing and Evoking Memories of Flight (44:27–46:13 Min)

In this memory sequence, Amin, his older brother, and his mother are trying to flee Russia on a boat intended for Sweden. They have been cramped together with several other refugees below deck and seemingly left alone. At 44:27 min the film transitions from showing the surroundings of the boat to only showing the darkness below deck. The transition is marked by a sound effect of a creaking hatch being slammed and latched, signifying Amin and his fellow refugees being locked in. After this, the dominant sound effects are of a squeaking hull, making the hull sound metallic, but also old and unreliable due to the high creaking sounds. We soon hear a low rumbling of an engine, and when a flashlight is lit to light up the otherwise completely dark screen, the flashlight is marked with a little click. Rumbling sounds are then mixed with the engine sounds, like sounds of waves crashing around the hull. The soundscape (i.e., the auditive ‘landscape’) is dominated by dark, vibrating sounds with a lot of reverb (reverb is a kind of echo effect which creates auditive spatiality, here signifying the metallic space of the hull surrounding Amin and the others). All of this creates an uncomfortable, eerie atmosphere. The images are dark and shift around between the faces of the refugees, overlaid with shadows. All of the information we get about the ship and the waters it is running through is through the sound.

At 44:58 min we see a darkened image of Amin in the studio where he is interviewed, his eyes are closed and the sounds of the ship and water continue, now with higher sounds of waves crashing. The images quickly turn back to the ship and the sound of waves keeps intensifying, now with darker metallic rumbling, as if the ship is being torn by the waves. At 45:17 min, the soundscape is turned down a bit to allow for Amin’s voice, which is soft with a melancholic tone. He says people started to throw up, which is then marked by sound effects of people coughing and maybe throwing up. We also hear unintelligible whispers from the people on the boat. At 45:29 min, the sound of the engine is intensified and raised to a higher pitch, before a halting sound signals that the motor is setting out and then stopping. The visuals show the scared faces of Amin and his mother as a reaction to the sound. People begin shouting and we hear the sound of running water, something we don’t see. People start slamming at the hatch, which is again marked with metallic sounds, until the hatch gives way at 46:01 min and the visuals shift to portray the outside of the ship. The soundscape now changes to be dominated by the sound of heavy rain and wind, punctuated by creaking from the ship mast and crashing waves.

In this sequence, the sound functions to materialize (cf. the discussion of materializing sound indices above) the boat engine, the creaking hull of the boat, and not least the storm and the water, which are all dominant in the soundscape. The sound effects are creating a link to the extra-communicational domain where these objects exist and have real implications, a link to ‘real’ life with boats, storms and threatening water. The sound here is a mimetic substitution for the lack of real footage, but it is also evocative (cf. Honess Roe). The sound is not just a transmediation and interpretation of a memory, it is creating a very real auditive experience of what it is like living this memory. An experience which the audience is forced to live with Amin as he relives it through the sound effects and accompanying animation.

Voice as Sound Effect: Empathy and the Feeling of Fright (21:27–21:43 Min)

This very short, stylized sequence occurs between different clips of found footage from news reports about the breaking out of the war in Afghanistan. The voice of a news reporter carries on into the sequence, which is dominated by grey colours and dark lines. After a few seconds, when the ‘camera’ zooms out, we realize the dark lines are forming the shape of a running boy. Dark atmospheric music (drone-like in character but shifting a bit) is playing over the sequence. Sounds of heavy breathing begin along with the atmospheric music and the grey colours at 21:27 min. At 21:31 min, we also begin to hear running footsteps. At 21:38 min, the footsteps disappear and the images blur and shift to show steps leading to a door. The heavy breathing continues until the images cut to another news report.

I would argue that this short sequence is a highly concentrated audiovisual image of fright and flight, perfectly encapsulating the theme and emotional strands of the story. The music aside, which is of course important for the eerie atmosphere of this clip, a lot of its effect comes down to the sound effects of the running footsteps and the breathing. Just as in the previous example, these sound effects are anchoring even these highly stylized segments in a physical experience of reality, while highlighting and underscoring the central theme of fright and flight.

Most importantly, however, the sounds of breathing and running come to signify, or evoke, a bodily experience. We breathe heavily when we are frightened, we run away when scared. The voice here becomes a non-verbal variant of the ‘acousmêtre’ (Chion 1999: 21), a voice which is present in the narrative but without a visual representation, because even though the visuals portray drawings of a running human, the drawings are too stylized and the sound too close, too intimate, and too detached from the abstract visual representation of a human. The voice becomes a generalized ‘I voice’, a voice ‘speaking’ in first person and so ‘close up, precise, immediate’ (Chion 1999: 52) it demands bodily identification from the audience:

We might call this an effect of corporeal implication, or involvement of the spectator’s body, when the voice makes us feel in our body the vibration of the body of the other, of the character who serves as a vehicle for the identification. The extreme case of corporeal implication occurs when there is no dialogue or words, but only closely present breathing or groans or sighs. We often have as much difficulty distancing ourselves from this to the degree that the sex, age, and identity of the one who thus breathes, groans, and suffers aren’t marked in the voice. It could be me, you, he, she. (Chion 1999: 53)

The intimate and close sounds of the breathing voice in this segment become, in a sense, a floating signifier, not because we don’t know that the source, the signified, is a human, but because the human could be anybody, it could be me, it could be you.

Conclusion

It is a common trend in discussions of documentaries, animated as well as live-action, to focus on the relation of the visuals to the represented ‘reality’, and secondarily to focus on the role of the verbal mode. The role of sound effects, which I argue are highly important in this regard, is much less often discussed. Interestingly, sound effects both in animated film and in documentary practices are sometimes discussed in terms of realism, treating sound effects as a way to either enhance or subtract from the realism of an animated film (Taberham 2018; Thom 2013; Whittington 2012) and as a potentially problematic post-production manipulation of documentaries (Collins 2017; Murray 2010). Despite this focus on sound effects and realism, the underlying mechanics of this perceived realism and the significance of sound effects for the listening audience still deserve more attention. Regarding the first point, I have aimed to provide a suggestion for how these mechanisms might work by referring to the concept of the perceptual index and the materializing capability of sound effects. Regarding the second point, I have aimed to show, through my analysis of Flee, how sound effects can convey different kinds of ‘truths’ as a consequence of this materializing function, and their ability to function evocatively by directing attention and constructing an auditive experience mediating thematic and emotional content.