Feeling the Child’s Voice: Sobs, Sniffs and Snuffles in Anatomy of a Fall

At the release of Justine Triet’s acclaimed film Anatomy of a Fall (2023), child actor Milo Machado-Graner was praised for an exceptionally emotive performance as Daniel, the adolescent son of an accused murderer. His magnetism is largely credited to his ability to convincingly, movingly cry onscreen. But what makes the crying child onscreen so moving? And how might films seek to move us, as spectators, through the child’s distress? Triet’s strong interest in Machado-Graner’s voice suggests that the child’s vocalised distress has a unique claim on spectatorial fascination and affectivity. In the tactile proximity of hiccups, sobs, sniffs and snuffles, we can examine how spectators experience the crying child in film as a particular cinematic body that aims to touch our own. Drawing on Jennifer Barker’s phenomenological concept of the film’s viscera,¹ I posit the crying child as a visceral cinematic body and analyse how sound technologies contribute to the effect of the “real” child’s presence. Keying our ear to the organs of this “real” cinematic body opens up avenues to explore how Triet structures bodily empathy between adult spectator and child, the human and non-human. 

Touching the Spectator

The past three decades have seen a steady increase in scholarship focusing on embodiment and film,² with foundational works including Vivian Sobchack’s The Address of the Eye,³ Laura Marks’ The Skin of the Film,⁴ Steven Shaviro’s The Cinematic Body⁵ and Linda Williams’ essay Film Bodies: Gender, Genre, and Excess.⁶ In her book The Tactile Eye,⁷ Jennifer Barker draws from Sobchack and Maurice Merleau-Ponty’s phenomenology,⁸ calling for closer attention to the place of tactility in film spectatorship. Tactility, Barker argues, is “a material mode of perception and expression.”⁹ Barker does not limit “cinematic tactility” to haptics alone, but conceives it as contact “enacted and felt throughout the body”,¹⁰ shaping a relationship with cinema that transgresses binaries of subject and object, spectator and screen: 

To say that we are touched by cinema indicates that it has significance for us, that it comes close to us, and that it literally occupies our sphere. We share things with it: texture, spatial orientation, comportment, rhythm, and vitality.¹¹

Barker argues that the spectator does not “lose” themselves in a film, but rather emerges through this sense of touch and being touched.¹² In this bodily relationship between spectator and film, the “film’s body” manifests as a “concrete but distinctly cinematic lived body, neither equated to nor encompassing the viewer’s or filmmaker’s body, but engaged with both of these.”¹³ The spectator can see the “perception and expression” that makes the film’s body possible, but this body is typically invisible.¹⁴ Thus, the film’s body is expressed through cinematic devices – editing, framing, camera angles, lighting, soundtrack – and felt as a body in the “space of contact” between spectator and film.¹⁵ A mutual bodily relationship emerges, as both the spectator and the film’s body have a tactile relation to the world.¹⁶ We can better understand this by approaching the film’s body through something it actively performs: it can puncture, pierce, scrape, stumble, shudder and stutter. 

Barker conceptualises the film’s body in three ways: the skin, the musculature, and the viscera. The film’s skin makes itself felt through our haptic perception, touching our skin in the process.¹⁷ Grainy images can feel grimy, and “grappling bodies” onscreen can feel overpoweringly tactile.¹⁸ The film’s musculature is its body language, evoking gestures from the spectator’s body.¹⁹ For example, a jump scare in a horror movie can “reach out” of the screen, prompting the spectator to flinch without physically touching them.²⁰ Barker’s concept of the film’s viscera offers a particularly effective framework to conceive the crying child and their vocalised crying as a cinematically constructed presence. This section is, admittedly, more abstract than her other two, and Catherine Clepper is wary of these latter sections falling into speculation.²¹ Conversely, Mark Cousins highlights this section as a rich framework for filmic analysis,²² and I concur: this calls for a sharpening and broadening of Barker’s idea of the film’s viscera. 

Barker describes the viscera as organs that constitute the internal body – an area that can give rise to intense affective responses or “gut feelings.”²³ Films draw attention to the viscera by highlighting rhythms that we are not directly conscious of in our day-to-day life, like our heartbeat or our breathing. They sustain the illusion that our being in the world is continuous, rather than composed of multiple intermittent movements.²⁴ Disruptions like heart palpitations and hiccups are ways of bringing the rhythms of the internal body into consciousness.²⁵ The viscera are not usually accessible but can be perceived through parts of the body that we notice and can access, such as muscle or skin.²⁶ Barker draws a compelling link between the film’s body and the spectator’s living body: both are made up of discontinuous movements and rhythms in “space and time.”²⁷ Katarzyna Weichert emphasises this connection, reminding us that while film has its own rhythms, it can concurrently simulate the living body and multisensory experiences.²⁸ Therefore, the discontinuities within the film’s body – as much as its continuities – can resonate with the spectator’s bodily experience of space and time. Barker argues that when films shift the spectator’s attention to the discontinuous rhythms of the viscera, they can evoke a feeling that is simultaneously “pleasurable and discomfiting” in the spectator.²⁹ In this interruption, we are awakened to the viscera’s hidden, internal labour. We feel the frailty of our living body and the processes that keep it going.³⁰ 

While the tactile qualities of the voice may, at first, seem to locate the child’s vocalised crying closer to the film’s skin in Barker’s model, her framework of the viscera is particularly productive to understand the affective force of crying child scenes in film. Vocalised crying asks to be heard and interpreted as a site of authenticity, invoking the possibility of the “real” suffering child and their suffering body in the profilmic. Steven Connor argues that “suffering mutes and muffles articulation, which produces a further intensification of suffering … that impeded voice then becomes the authentic voice of suffering.”³¹ Connor’s term “impeded voice” remains most appropriate for my discussion of vocalised crying and its parameters.³² When we hear the impeded voice, we hear noises intervening, overcoming or disrupting language and breath. I suggest that the impeded voice not only includes vocalisations, but also incredibly subtle sounds omitted or transmitted by the voice. This visceral corporeality, encompassing the workings of the lungs, larynx and throat, highlights what Roland Barthes calls the “grain” of the voice.³³

The visceral qualities of the child’s impeded voice secure it as a site of acute affectivity. Barker explicitly highlights the relationship between the film’s viscera and spectator as an affective one: “… the rhythmic filling and emptying of the lungs can enact and express attitudes and emotions such as desire, fear, anticipation, and relief.”³⁴ The central rhythmic force that impedes the crying voice is the breath, which Davina Quinlivan describes as characteristically tactile.³⁵ She argues that breathing destabilises the body’s boundaries and the distinction between visibility and invisibility.³⁶ Barbara Creed considers the rupture of bodily boundaries as chiefly visceral, but Quinlivan adds that we can think about breathing as simultaneously inside and outside, with the skin intact.³⁷ Quinlivan’s approach resonates with the impeded voice as a visceral force that undermines bodily boundaries of internal and external. When the child cries, we do not witness their organs being physically regurgitated, but the impeded voice can still generate an uncanny copresence of the viscera inside and outside the body. We hear the impeded voice as the viscera, not only through the disrupted flow of air and mucus, but as the violent fleshy mechanics of the organs that produce and sustain them. 

The crying child can induce the spectator to feel for and with the child by summoning the viscera as a mode of bodily contact. Thus, the spectator’s body experiences the child as a tactile, suffering cinematic body. This cinematic body is not equivalent to the real, living child actor in the profilmic, yet it consists of the child’s performance, the technologies that represent it, and the film’s production. Heightened attention to the continuity and intermittence of the child’s disrupted breath and body secures their cinematic body as affective, immediate, and most importantly, material, producing the effect of the “real.” It is fair to caution this phenomenological approach as the cinematic spectator’s experience cannot be universalised as one experience, or one visceral feeling. Katharina Lindner notes that Barker generalises about the spectator’s body without accounting for embodied differences such as gender, class, race and sexuality.³⁸ However, she suggests this is a point to develop research on embodiment, rather than criticism of Barker’s work. I acknowledge that my discussion of corresponding structures between the film’s body and spectator’s body limits perspectives of embodied difference. My body is undoubtedly implicated in my discussion of the viscera. Yet there is merit to embracing our own feeling responses in analysing filmic texts: it allows us to understand how we orient ourselves in relation to films, and the social and cultural contexts of their production.³⁹ In turn, we can better understand how films orient particular spectators. 

I am not suggesting the spectator’s affective response is predetermined or fixed, and I recognise that what counts as a realistic, effective, childlike performance historically, culturally, socially and generically changes. Rather, I am emphasising the viscera as an underdeveloped avenue to explore how certain affects are solicited, why the child’s impeded voice has affective value, and the ways narrative cinema represents this. From here, we can approach the film’s body as a product, representation and expression of being in the world, and address one of the most significant visceral aspects of the crying child: their voice. To understand the film’s body, we must look at its inner workings, and the sound technologies that render the crying child’s voice as palpably real. 

Up Close and Personal

How do film sound technologies enable the adult spectator to feel for and with the crying child onscreen? How do they generate bodily contact through the viscera? Jacob Smith argues that a major issue in film scholarship is that it struggles to balance discussion of the human voice with attention to the technologies that render them.⁴⁰ Film scholarship often privileges the voice in cinema as an index of identity while ignoring the social and cultural context of the media apparatus. Conversely, psychoanalytic scholarship privileges the media apparatus and excludes the human performances behind them.⁴¹ Smith calls for an awareness of both, aiming to address this gap by developing discourses on voice acting and sound recording.⁴² Lisa Coulthard observes that sound scholars characterise post-Dolby sound as uncannily intimate.⁴³ Almost noiseless sounds – of which she includes voices, whispers and breathing – have an amplified presence, creating the equivalent of an audio close-up.⁴⁴ In a post-Dolby era, cinema has had the opportunity to render a crying child’s voice in more detail, which means their presence can feel, quite literally, closer than ever. It is useful to return to Smith here, as he reminds us that vocal performance and microphone placement is contextualised within a broader history, beginning with the development of the radio.⁴⁵ The child’s vocal performance of crying should not be seen as something that Dolby definitively changed, as performance has always been shaped by the technologies that record it, and vice versa. We can instead assess how sound recording technologies have facilitated and sustained culturally significant modes of performance. 

Smith cautions against assuming there is one uniform microphonic style, and highlights that sound technologies do not reproduce sound, but rather represent it.⁴⁶ Close miking, for example, enabled certain performance styles to flourish and is generally recognised to signify complicity and “immediate bodily presence”.⁴⁷ Through the sonic close-up, radio supported a cultural shift in the modern era from “character” – representations of moral values, such as work, citizenship and duty – to “personality,” the presence of the vivid, magnetic star.⁴⁸ Now “organic vocal sounds” were made audible, “the liquidity of the saliva, the hissings and tiny shudders of the breath.”⁴⁹ Adjusting relations between sound and space, the microphone cultivated new modes of performance that emerged in film. Closely miked vocal performances could be more individualised, nuanced and expressive, since the performer no longer had to rely on projection as they would in a theatre space.⁵⁰ The culture of personality, associated with the visual and audible close-up, was invested in the spectator’s proximity to the presence of the individual – the effects of which resonate in popular films today. 

The child actor’s performance is valued when they manifest as an authentic, unmediated child. Karen Lury has argued that the reality effect of the child’s onscreen performance is especially potent when the boundaries between acting and not acting are uncertain – when they merge, overlap, shift and toggle between one another.⁵¹ Spectatorial fascination with the crying child onscreen circulates around the child as a source and promise of this uncertainty: the possibility that the spectator is witnessing “real” emotion. This is especially evident in films that build up to scenes of the crying child as a key affective moment, and when the child actor is celebrated as moving. While not all children vocally cry in the same way, their crying is affective when it establishes legible childlikeness – or, more accurately, what is socially and culturally perceived as childlike. When the child cries onscreen, they are open to the scrutiny of the adult spectator. Naturally, sound technologies can render distance and delay as effectively as intimacy and immediacy, but it is no trivial thing that popular films so often key us to listen to the child’s impeded voice so closely. Therefore, I am consciously limiting my discussion to audible proximity regarding the crying child protagonist.

Cinematic sound does not have to be analysed as purely engineered or purely organic. Like Smith, Pamela Wojcik aims to bridge the divide between cinematic sound and sound technologies, advocating for an approach that does not pit acting and sound recording against each other.⁵² Another major problem with analysing cinematic sound is the lack of public information around sound production.⁵³ Wojcik suggests that rather than making assumptions about a sound’s source, scholars can enrich their analysis by focusing on specific aspects of performance, such as vocal qualities and editing: “We might consider not only whether a performance was recorded through ADR, but also how pitch and timbre were or were not massaged, how the sound was mixed and with what, and more.”⁵⁴ I intend to take a similar approach, but I argue that research on the voice cannot be so clinical as to discard assumptions altogether. Cinematic sound relies on spectatorial imagination of its origin. For example, Smith argues that a term seemingly straightforward as “timbre” is especially difficult to describe and notate. Our vocabulary for timbre, characterised by adjectives, is not wholly adequate and heavily reliant on the listener’s perception of timbre, rather than their objective assessment.⁵⁵ This is an opportunity to think about how cinematic sound shapes and mediates the spectator’s assumptions, instincts, gut feelings and fantasies. As a result, I am aware that my analysis cannot be – and should not be – entirely objective. My subjective experience will pervade my descriptions of the child’s impeded voice, how I interpret its source, and where I locate it.   

Though the voice in cinema is never in its organic, unmediated state, cinematic representation certainly invests in the idea of the unmediated. Wojcik amends Michel Chion’s concept of the acousmatic voice, highlighting that all cinematic sound is, in some way, acousmatic: it has been parted from the body through recording, and must return to the body through technology.⁵⁶ The relation between “original” and “recorded” sound has been a central debate in scholarship on sound technology, and one James Lastra decisively puts to rest. Similarly to Smith, he elects for the term “sound representation” over “sound reproduction,” which frames recorded cinematic sound as an “interlocking set of diverse practices and not simply as an act of pure inscription.”⁵⁷ As to the concept of the “original” recording and the “real,” Lastra argues that this is tied to effects of immediacy and authenticity, which are historical, social and culturally defined concepts and conditions.⁵⁸ Lastra does not exclude the original, authentic, or the real from scholarship on sound studies, but like Lury and Smith, he highlights that we should be asking in whose interests are these credited or denied.⁵⁹ As sound representation, the child’s impeded voice produces an effect of the real.

Talking Different: Milo Machado-Graner in Anatomy of a Fall

Anatomy of a Fall follows celebrated novelist Sandra Voyter (Sandra Hüller) and her visually impaired son, Daniel (Milo Machado-Graner). Sandra’s husband, Samuel (Samuel Theis), is an oppressive unseen presence in the opening, renovating the attic of the family’s chalet to a blaring steel pan cover of 50 Cent’s P.I.M.P. Downstairs, Sandra struggles to conduct an interview over the music, clearly frustrated. The interview is cut short. Soon after, Daniel comes across his father’s body (apparently having fallen from their attic balcony) and Sandra is accused as the lead suspect in her husband’s murder. The film alternates between the highly publicised court case, and private, emotionally charged moments in the Voyter home. Sandra’s desire to protect Daniel’s innocence conflicts with her desperation to maintain his trust, especially when darker details of her marriage – infidelity, jealousy and violence – are exposed. Director Justine Triet intentionally withholds painting Sandra as unequivocally guilty; she is more interested in how truth and testimony are constructed. The fragile boundaries between fact and fiction are all the more evident when Daniel is onscreen. He is the person Sandra cares about most, with arguably the most to lose. With his father dead, the outcome of the trial threatens to jeopardise what trust he has left in his family. His close relationship with his mother is fracturing. He faces condemning her as a murderer, complying with the story that his father was suicidal, or acknowledging him as manipulative – all of which he can’t be sure.  

While Anatomy of a Fall has been critically celebrated for Milo Machado-Graner’s performance, he wasn’t who Triet first had in mind for Daniel. Initially, she insisted on casting a visually impaired or blind actor and intended the boy to be blonde to match his onscreen mother, Sandra Hüller.⁶⁰ When Triet found herself struggling to find a match, she decided to include sighted children. Her partner and co-writer Arthur Harai brought her attention to Machado-Graner, who she had dismissed at first because of his dark hair: “Arthur came to me and he said, ‘Look at this kid’s tape, but actually don’t look. Just listen to the sound of his voice’ … The way that Milo was talking was different. It didn’t sound like a little kid, and I really liked that about him.”⁶¹

The content of Machado-Graner’s first audition is unclear. Machado-Graner remembers “there was some improvisation.”⁶² His second audition, however, secured his role. Triet called him back to perform a crying scene and directed him to “really open up emotionally”: 

in 10 seconds [h]e had tears all over his face, really. And in that moment I asked him to say the lines of his character at the end of the movie, which no one had really done perfectly for me before. And he did the whole scene, right there, right in front of us […] It was so complex, all the things that were going on with him at that moment. But he made it look so easy.⁶³

Machado-Graner’s account suggests a more mutual effort: “she put me to the test and worked on a scene that needed a lot of emotion and she pushed me and worked with me and brought me to the point where I got emotional and started crying right during the audition.”⁶⁴ Triet was taken with this performance, comparing it to child actor Henry Thomas’ famous audition for Elliott.⁶⁵ She describes having the same conviction in Machado-Graner as Spielberg had in Thomas: “I sent the tape to the producer: Here he is, we found the boy.”⁶⁶ The similarities between the two auditions, forty years apart, outline that the child’s crying performance continues to have affective value when they negotiate tensions in ideologies of childhood. Interestingly, Henry Thomas is the child actor with affective value who persists in cultural memory. A host of other equally accomplished, equally affective children could easily replace him, but the privileged status of the accomplished boy, implicitly white, resonates in popular film. Machado-Graner has inherited this specific legacy, Henry Thomas’ legacy, by presenting as a kid “who doesn’t sound like a little kid” and credibly embodying maturity and childlikeness. 

But how did Triet and Harai predict that Machado-Graner was affectively capable before the second audition? Machado-Graner’s appeal lies in sounding like an unmediated child; a hybrid voice and body, where character and actor are compounded. Machado-Graner’s impeded voice is Daniel’s impeded voice, and vice versa. In the initial stages of the script, Triet was advised to change Daniel’s dialogue as it was considered not childlike enough, but she was reluctant to change it: “I said, ‘Huh? My 8-year-old daughter speaks like this. It is definitely possible.’”⁶⁷ 13 years old at the time of filming, Machado-Graner’s voice has not broken yet, which makes him credible as Daniel, a child two years younger. Yet his voice is also on the cusp of change and has a slightly husky, rasping texture that surfaces in emotionally intense moments. Today, it is unlikely he’d be chosen for the same role, not because he lacks the emotive capacity, but because he has physically and audibly changed. In a Gold Derby post-film interview, he has rapidly grown into adolescence: his voice has broken, far deeper and no longer legibly childlike.⁶⁸ In Anatomy of a Fall, Machado-Graner’s 13-year-old voice ambiguously lingers on the threshold of a child. This must be, at least in part, the strange maturity that Triet noticed and was drawn to. Machado-Graner’s voice also ensures he does not sound like a scripted child. His scenes are typically dialogue-heavy, but managed effectively through his emotive voice, particularly striking when he cries onscreen. In moments when we doubt who we are watching onscreen – when Machado-Graner’s performance of crying doesn’t sound performative or planned – we cannot refer to actor and character separately, or paired as Daniel/Milo. We witness a fleeting, impossible fusion where character and actor seemingly become one. Potentially exclusive to scenes of distress, we can call this body onscreen the hyphenated “Daniel-Milo.” The possibility of the real infiltrating the fiction emerges in these virtuosic, rare moments where Daniel-Milo breaks out.

Universal acclaim for Machado-Graner’s performance proves Triet and Harai’s instincts were well-founded. He won Best Youth Performance at the 8th Seattle Film Critics Society Awards⁶⁹ and Best Supporting Actor at Les Capucines du Cinéma Français,⁷⁰ an award for French short and feature films. Machado-Graner was also nominated for several others, including Best Young Actor for the 29th Critics’ Choice Awards⁷¹ and Best Male Revelation at the 49th César Awards.⁷² Critical praise for Machado-Graner is strikingly similar in tone to acclaim of Henry Thomas: he is a “talented” performer who offers a “terrific, painfully conflicted turn”⁷³ as Daniel and takes on a “monumentally mature task.”⁷⁴ Katarina Docalovich from Paste Magazine praises how “Machado-Graner’s tears push past ‘generic sad kid’ and plumb the depths of distress.”⁷⁵ Several critics identify Machado-Graner as the emotional core of the film, its “secret weapon,”⁷⁶ adding “a much-needed element of heart to the film”⁷⁷ and deeply influencing the film’s success.⁷⁸ Machado-Graner’s maturity and performative prowess undoubtedly constitutes his affectivity, alleviating Triet’s early concerns about him not looking like Hüller’s son. Onscreen, he is described to embody facets of Hüller’s personality, sharing her “quickness and confidence.”⁷⁹ For Hüller, Machado-Graner was “a true partner, like every grown-up partner I could wish for.”⁸⁰ Yet Machado-Graner’s affective power and presence is not limited to his performance alone. The dynamic between Triet and Machado-Graner during the audition process highlights his affectivity as a collaborative effort. Accordingly, we must account for how the film represents and supplements Machado-Graner’s affective presence.  

Crafting the Child’s Visceral Presence in Sobs, Sniffs and Snuffles

In Anatomy of a Fall, Daniel-Milo’s impeded voice is represented as intensely proximate, producing a viscerally tactile encounter between spectator and the child’s suffering cinematic body. Two consecutive scenes foreground the child’s impeded voice as a visceral presence. In the third act of the film, the date of the trial verdict is close and new details have emerged. Sandra claims she once found Samuel unconscious with white spots in his vomit. She suggests these were pills, traces of a failed suicide attempt, which could explain Samuel’s fall as deliberate. Frustrated with his marriage and disillusioned after his failed attempts to write, he leapt off the balcony. It seems unlikely this argument will turn the tide as there is no-one else to corroborate Sandra’s story. The weekend before the closing arguments, Daniel coolly requests for Sandra to be removed from the house and is left with only his court monitor, Marge (Jehnny Beth). This is a calculated decision. Stealing aspirin, he feeds it to his guide dog, Snoop (masterfully played by Messi, winner of the Palm Dog Award at the 2023 Cannes Film Festival). The following morning, Snoop is unresponsive, and Daniel panics.

Anatomy of a Fall

How can we empathise with a child who deliberately poisoned his own dog? Snoop is integral to Machado-Graner’s affectivity onscreen, staking a claim on the spectator’s affective investment. Snoop and Daniel are introduced as close companions in the film, and in Daniel’s fractured family unit, Snoop is the only figure he can completely rely on. Kathyrn Bond Stockton highlights that dogs in cinema are often closely linked to children and that “the dog is a living, growing metaphor for the child itself […] engaged in a growing quite aside from growing up.”⁸¹ By “growing sideways,” Stockton means the width and depth of a person’s experience and understanding of the world, which can bring adults and children “into lateral contact of surprising sorts.”⁸² Borbála László makes a case for the dog in film as “threshold creature” that moves between realms of adulthood and childhood.⁸³ Like Stockton, she argues that the dog does not so much enable the child’s linear maturation, but instead reintroduces childhood into the adult world and becomes an identifiable figure of childishness.⁸⁴ Snoop’s poisoning is a catalyst for Daniel’s lateral growth in the film. Daniel develops from a casualty of his parents’ volatile marriage to instigating his own crime. He later testifies that Snoop had been in this lethargic state before, and suggests Snoop overdosed on the aspirin in his father’s vomit: unequivocal evidence of a suicide attempt. Sandra is acquitted. Snoop, the childhood companion, is conspicuously separate from Daniel for the remainder of the film, and Daniel’s priorities have shifted: he now takes responsibility for Sandra. When his mother returns home, he strokes her head comfortingly, knowingly, as a parent does.

Daniel’s distancing from childhood is rationalised through his distress. His crying is only moving and meaningful if the spectator is assured that he truly cares for the dog and fully feels the consequences of his betrayal. To encourage the spectator to identify with Daniel’s distress, the film focuses on Snoop as a key affective figure. In an interview with The Hollywood Reporter, Triet recounts that Snoop was incredibly affective at the time of shooting: “It’s the only scene in the movie where all the team was crying at the same time because it was too … I don’t know why … the team was really, really attached to this dog.”⁸⁵ Hüller jumps in, “Yeah, I had the same feeling … when I saw the film for the first time, that was the scene that touched me the most.” László proposes that the dog in film is especially affective and poignant in cinema when attention is drawn to its phenomenal presence, the dog as dog.⁸⁶ Animals, like children, often occupy the same space of ambivalence in relation to their performance. Dogs are ideal nonhuman actors to double human actors as they are considered highly trainable, intelligent, and psychologically complex.⁸⁷ Like the child, Messi’s performance onscreen invokes questions of the dog’s agency. His lifelessness is convincing enough to cause spectatorial concern, with some even speculating as to whether he had been drugged on set.⁸⁸ Snoop’s trainer, Laura Martin Contini, reveals that Messi’s lifelessness is the result of two months of training, beginning with “lying down, playing dead” and not reacting to being picked up.⁸⁹ Contini was inspired to teach Messi to stick his tongue out after noticing it lolled out when he was tired. Doubt as to whether Messi is acting or drugged provokes the spectator to be uncomfortably conscious of the dog’s phenomenal presence, the real in the fiction. Just like Daniel-Milo, Messi’s breathing body becomes a hybrid of actor and character: “Snoop-Messi.” 

Machado-Graner’s performance is just as critical in upholding these narrative stakes as real. If his crying came across as performative or insincere, the scene would lose its affectivity, but to be completely inconsolable would paint him as too childish. Instead, he performs a frantic state of distress that is just below the threshold of hysteria. He does not compete with Snoop-Messi for the audience’s compassion; he carries the scene’s burden of distress and becomes a surrogate crier for the audience, a conduit for their concern. Reduced back to helplessness, distraught and seeking adult intervention, Daniel’s innocence is reinstated. He is a legible child again. Shukin and O’Brien observe how an animal’s pain can create an emotional blow that rouses the spectator to feeling, which is even more powerful when the spectator is reassured by “another’s humanity in their affectability”: “it isn’t necessary to be a member of homo sapiens to qualify as human, but it is necessary that one demonstrate a capacity for feeling.”⁹⁰ Daniel is rehumanised when he cries for the nonhuman. Even if the spectator does not forgive Daniel, the film asks us to believe him.  

Anatomy of a Fall intensifies the child’s visceral distress by sonically highlighting Snoop-Messi and Daniel-Milo as equally proximate, suffering bodies. The first scene, where Daniel discovers Snoop, centres around the dichotomy of asphyxiation and breathing. Diegetic breathing haphazardly clutters the soundtrack, and the lack of extradiegetic music keys the spectator to every breath, gulp, sob and shudder. The unsteadiness of the handheld camera enhances the immediacy of the scene and underpins it with a fraught, nervous energy. Daniel-Milo’s impeded voice is audibly, overwhelmingly, viscerally proximate, and especially salient alongside Snoop-Messi. His agitated gasps and sobs indicate his living, hurting body, drawing parallels with (likely dubbed) low, laboured groans and snorts from Snoop-Messi – sounds which suggest restricted airflow through his nasal cavity. Oxygen is unevenly distributed between the two; Daniel-Milo fights with a surplus of air, while Snoop-Messi fights for a fraction of it. The irregular rhythms of their lungs establish the visceral presence of a pained – and painful – breathing cinematic body. This cinematic body still evokes the child’s body as it emphasises Daniel-Milo’s impeded voice as the central affective force. Snoop-Messi is characterised by sonic sparseness and lapses of breathing, while Daniel-Milo’s impeded voice is continually, relentlessly present. 

As Daniel calls out for help – Marge is heard washing dishes offscreen, downstairs – the shot zooms out, framing him and Snoop in the doorway. Daniel-Milo’s heavy, ragged breathing, however, remains audibly close, punctuated by another shallow snort from Snoop-Messi. These discontinuities with sound and space should not be considered a threat to the spectator’s sense of the real and their immersion in the film. James Lastra explains that in classical Hollywood cinema, we can discern two modes of sound representation in relation to spatiality: the first is the phonographic model, where all aspects of sound are considered significant, and the goal is to faithfully reproduce its material qualities.⁹¹ The second is the telephonic, which frames sound through a hierarchy, often prioritising intelligibility over “material specificity”.⁹² Dialogue in film tends to adhere to the telephonic approach, which is evident when we can still hear a character speaking clearly even if they are facing away. Lastra notes that this close frontal miking became the norm for dialogue in Hollywood cinema.⁹³ He rightly rejects arguments that “real” space is only upheld by POA sound (point of audition), where spatial characteristics manifest. Scholars championing “sound perspective,” claiming that close miking violated this perspective, struggled to hold up under scrutiny; films don’t strictly operate on the basis of “precise perception” and consider narrative emphasis.⁹⁴ Evidently, privileging a specific mode of sound representation as more immersive than another is oversimplistic, idealising one at the detriment of another. 

Telephonic emphasis on breath in Anatomy of a Fall places narrative emphasis on the child’s affective body. While dialogue and breathing are closely miked, breathing sounds closer. Marge and Daniel’s dialogue subtly adheres to POA sound (or hearing within the diegesis). Their voices have a slight reverberation that situates them in the room onscreen, and their dialogue is louder in close-up shots. Conversely, Daniel-Milo’s sobs and Marge/Beth’s breathing is heard as close frontal miking with no spatial signature, breaking out of POA conventions. This noticeably reoccurs when Daniel-Milo’s face is buried in the dog’s fur. His breathing has a nearness, crispness and clarity that should be muffled. With the spectator intensely aware of the anxious rhythms of Daniel-Milo’s body, the child becomes a tactile, visceral force that temporally and spatially breaks out of the body onscreen. It is still the child’s voice, and of the child’s body, but materially closer than the camera, bringing the spectator audibly closer as a result. I suspect that this too-close breathing is ADR (automatic dialogue replacement), as it often doesn’t visibly sync to Machado-Graner’s body onscreen. His chest and body don’t heave with the audible force of his sobs, and his lips don’t move. However, shots change restlessly, creating a sufficient distraction to mask this lapse in simultaneity. Daniel-Milo’s overly proximate impeded voice, which seems separately recorded from his dialogue, could also be a result of Triet toeing the line between keeping Messi relaxed during shooting and retaining the integrity of Beth and Machado-Graner’s performances on set. She recounts that it was difficult to keep the dog calm alongside Machado-Graner’s “nervous” energy.⁹⁵ Here, Beth and Machado-Graner’s dialogue in the profilmic appears largely intact. Shrewd editing retains the excruciating, distressed presence of Daniel-Milo through his breathing, which becomes physically oppressive and suffocating in its closeness. 

When Snoop vomits and gains consciousness, the audible and visual rhythms of the film’s body slacken, and the child calms down. Daniel-Milo’s impeded voice becomes attuned to the voices around him. Beth’s voice is closely miked as Marge comforts Snoop and Daniel in the scene, shushing and speaking steadily, “Good dog … He’s ok, he’s breathing. Looking at us.” Marge/Beth’s voice dictates the scene’s slowing tempo and Daniel-Milo’s suffering body within it. After vomiting, Snoop-Messi’s panting highlights airflow back through his open mouth and throat. This shot of Snoop, eyes open in Marge’s lap with Daniel’s anxious hands reaching for him, is the longest of the scene, lasting 18 seconds. Daniel-Milo’s breath distinctly slows down to match Marge/Beth and Snoop-Messi – we can identify it through the rattle of mucus in his throat. His voice is coaxed away from sobs, carrying residual traces of his tears. But there are moments where, overlapped, these three characters become a single respiratory unit, and it is difficult to differentiate between them. The intervals between each breath become longer, more measured, and their rhythms merge. Daniel-Milo’s impeded voice ebbs away, allowing the listening spectator a fractional, physical reprieve. The film’s body retains the presence of the child’s body and the motions of its viscera through these commingled voices. Ensuring the spectator is conscious of Daniel-Milo as the primary distressed body in this scene, the film produces a powerful sense of his budding relief.

The following scene draws the spectator close to the jagged rhythms of the child’s viscera by coordinating the sounds of Daniel-Milo and Snoop-Messi. It begins with an audible shock: Snoop-Messi is loudly, sloppily drinking from a water bowl. We can hear Daniel-Milo crying, his voice still within telephonic conventions of audible proximity, and we hear Marge ask, “Can you explain?” Now that the spectator is reassured of Snoop’s recovery, the focus of the narrative rests squarely on Daniel. Machado-Graner’s responsibility as performer is arguably even more difficult here. The conversation is a series of shot-reverse shots between Beth and Machado-Graner, caught on a stabilised camera, which means his impeded voice and body are under direct scrutiny. In fact, the camera insists on it, slowly pushing in on Machado-Graner whenever he comes into frame. Temporally, spatially, audibly and visibly, he must appear to cry during the profilmic, in the profilmic. Consequently, Machado-Graner’s voice and body are snugly synchronised, and these realistic conventions secure the immediacy of his presence onscreen. 

This soundtrack drips with a tangible wetness, cued by Snoop and the water bowl. Machado-Graner’s performance is explicitly teary; his throat and nose are clogged, rendering his voice snotty and thick. The moistness of Daniel-Milo’s impeded voice authenticates his crying as “real” bodily fluid, but it also has potent tactile qualities that invoke and provoke the spectator’s own internal body. Lisa Coulthard argues that moistness and mouth sounds in particular – spittle, chewing, breathing – generate acoustic disgust, especially when they sound too close.⁹⁶ Coulthard notes that the “over-amplification and unwanted intimacy of something so tied to life” creates “uncanny sound objects, both familiar and unfamiliar.”⁹⁷ Up close and moist, Daniel-Milo’s voice forces the spectator to become intimately acquainted with his mouth, but even more uncomfortably, his internal organs – the wet workings of his lungs, larynx and nasal cavity turned inside-out. 

Anatomy of a Fall

Machado-Graner incorporates the soggy, audible invasiveness of the mouth in his performance repertoire. He avoids wading through dense streams of dialogue in the same emotional register and instead calibrates his delivery to fluctuate and build to an emotional crescendo. Spontaneous hiccups and swallows penetrate Daniel’s dialogue as Daniel-Milo’s voice grows more emphatic and stressed. One gulpy hiccup trips him up mid-sentence and he wipes his mouth self-consciously after, a movement that evokes Snoop’s mouth at the water bowl: “I figured he’d … I don’t know, caught a virus or something, because (Daniel-Milo’s voice snags, he repeats himself) … he acted strange for days.” The hiccup is a spontaneous intrusion, a legible vocal marker of his authentic suffering. Over-proximate mouth-sounds continue as Daniel-Milo becomes increasingly distraught and breathless: 

Now I’m thinking maybe he ate my dad’s vomit, and that made him sick, and Mom told the truth! So my experiment with the aspirin was to see how he’d react. He slept for 14 hours. Now he’s drinking all the time. He smells exactly the same! Everything! Everything’s exactly the same! 

Daniel-Milo’s taut voice (“and Mom told the truth!”) has an inherent brittleness that threatens to snap, slipping into an unvoiced whistling where the vocal chords don’t vibrate. These high frequencies are ordinarily difficult to hear, reinforcing our nearness to the child’s bodily presence and the organs within it. Snoop-Messi’s concurrent body is reinstated when we hear him loudly lapping at the water bowl as Daniel remarks that “he’s drinking all the time.” This sound is slyly timed to substantiate Daniel’s story. We shift to a shot of Snoop, head raised and attentive to Daniel’s pain. Water audibly drips from the dog’s mouth before he starts drinking again. Through the tears and mucus that dribble out of Daniel-Milo’s voice, and the water that Snoop-Messi gulps and trickles, wetness endures as a disruptive, visceral force that invokes the crying child’s body.

While Triet prompts acoustic disgust through wet sounds, this does not mean there is a lack of bodily empathy between adult spectator and child. If the spectator’s bodily discomfort is aroused, then they also come into contact with Daniel-Milo’s own bodily discomfort. Amplifying this discomfort and making it tactile, the film unifies spectator and child. All the while listening to Daniel’s story, we have been conscious of the inner workings of the throat, the arrhythmic persistence of its labour to ensure the body can speak, breathe and swallow. The film has nudged us into trusting Daniel again, as we feel the child’s suffering with the child. The camera, too, insistently moves closer to Daniel-Milo as he becomes more distressed, and the scene ends with a close-up. For almost ten seconds, Daniel-Milo thickly breathes in and out, his words spent. Snoop-Messi and Marge/Beth’s breathing bodies are audibly absent in this final pause, while the residue of Daniel-Milo’s tears and mucus begins to settle. The scene ensures his visceral presence does not overstay its welcome while inducing us to wait and listen a little longer. We return to the child’s sole breathing body when Daniel’s defences have, at last, crumbled. 

What can we learn from the child’s sobs, wails and whimpers? To be in visceral contact with the intermittent rhythms of the child’s vocalised distress is to encounter them as a felt and feeling body. Anatomy of a Fall not only intensifies the child’s suffering by drawing the adult spectator close to Daniel-Milo’s impeded voice, but highlights Snoop-Messi as a significant corresponding distressed body. And why are they so affective together? Triet carefully intertwines their laboured breathing to mould an excruciatingly tactile presence. Their visceral affectivity, however, also relies on qualities unique to the child and the nonhuman; Daniel-Milo and Snoop-Messi share the same space of ambivalence that offers the adult spectator a troubling, provocative promise of “real” suffering onscreen. We might ask ourselves how films capitalise on the child and the nonhuman as sources of affectivity to make them so moving. A burden of affectivity is carried in dribbling mucus, hoarse sobs, sniffles and hiccups, especially when they writhe their way into us and make their contagious rhythms felt. 

I would like to thank the School of Arts and Media at the University of New South Wales, and my Honours supervisor, Dr. Jodi Brooks, whose support and expertise was invaluable in writing this work.

Endnotes