If you've ever lain awake at night pondering the fascinating question of how one approaches subtitling films, then this is the essay for you. It was submitted as part of my film degree, and left my course teacher bored to death. I sympathise. But I still feel the opening three scenes of Once Upon A Time In The West are among the most beautifully crafted in film history. (Mercifully, the footnotes have been lost in time. The plan of the soundtrack can be found at the bottom of the page.)
The enclosed plan is an inventory of all the sound that can be heard in the opening scene of Once Upon A Time In The West directed by Sergio Leone. The plan has been organised from the point of view of a subtitler working on a version of the film to be presented to a deaf or hard of hearing audience. What is immediately noticeable is that the bold sections, which represent the actual captions that will appear on screen, only begin after a considerable number of sounds have already been heard. These represent almost a minute and a half of screen time. So before the viewer is given any aural information, the ‘soundscape’ of the film has already begun. Why is this the case? Why do certain parts of the soundtrack seem to be privileged over others? The reasons that subtitling convention throws up are revealing of its attitude towards the importance of film sound and its subordination to the visual narrative. By examining how a subtitler treats Once Upon A Time In The West, and by discussing that treatment beside examples of complex soundtracks in other films, the whole relationship of the soundtrack to the rest of the film can be fully explored and it can be shown that contemporary subtitling may not be adequately conveying the richness and complexity of that relationship.
There are four reasons why subtitling chooses not to recognise every element of a film’s soundscape and all four reasons are based around what the viewer needs. In other words, subtitling is not based around what the filmmaker may deem important or of aesthetic value; it is concerned with what the viewer needs to understand the film and not be isolated from it. It therefore privileges the visual information given by the film: its image track. Given that the deaf viewer has a lack on one sensory level – the aural level – it seems sensible to allow the visual level as much expression as possible. Thus the philosophy is that captions should be kept to a minimum and be as unobtrusive as possible where the images can do the talking. In the given example, many of the sounds recorded need not be captioned because the actions that cause them are plainly visible. Indeed, to subtitle the sound of boots clumping or a windmill creaking when they are in full frame might well be deemed patronising and this is a second reason why subtitlers are wary of captioning everything. It must not be presumed that just because a viewer cannot hear the soundtrack that they then cannot understand the tone, texture or meaning of the film. There are several factors in the visual construction of a scene – mise-en-scene, lighting, framing, editing pace – that contribute to its meaning; from this perspective, sound is only one more container of information in a cornucopia of others. The viewer must be credited with the ability to discern from these other signifiers, not only the tone of the overall film, but also the nature of the soundtrack itself. In this opening scene, the exaggerated slowness of movement, the emphasis on long takes and minimal cutting is mirrored by the pauses between captions on screen (there are many seconds of silence between each noise on the soundtrack). So, purely from visual information, the viewer has learnt a lot about the filmmaker’s intended style and the tone of the scene in general.
But, perhaps most importantly, subtitling privileges narrative. In other words, only parts of the soundtrack that give the viewer necessary information regarding the plot or characterisation are deemed worthy of recognition – and that is whether the image track gives that information or not. Dialogue, therefore, assumes paramount importance and, to all intents and purposes, is the raison d’etre of subtitling. It is the main provider of meaning within the soundtrack and is the one area on which the image track can provide no information or clue whatsoever (save for the small percentage of lip readers). Furthermore, characters’ motivations and intentions are predominantly signalled through what they say and so dialogue is a large part of the overall structure of the narrative. Sounds are only deemed important where they enter the cause-and-effect dynamic of the narrative, ie. if characters respond to a sound heard off-screen like a telephone ringing or if a sound signals a shift of action in a particular scene like the cheering and applause at a rock concert. In Once Upon A Time…, the initial sounds do not further the narrative in any way and are merely background noise. The buzzing of the fly and the drops of water are captioned, however, as the gunmen in each case react to these sounds and the physical objects that cause them.
So where does that leave music? Well, conventional subtitling does not recognise scored music at all. The reason, following logically on from the approach to dialogue and sound, is that scored music does not contribute to the narrative action. Furthermore, it is not part of the diegetic world. The sound of the wind from the opening shots may not be subtitled but it is part of the environment in which the characters are situated. Ennio Morricone’s music heard towards the end of the scene is not; there is no orchestra playing by the side of the train station! As David Bordwell points out, ‘non-diegetic sound has no relevant temporal relationship to the story’; you may say it acts as a complement to the narrative but it is not part of it. Subtitling convention deems that to caption the music would be to confuse the two spaces in which a film operates: the diegetic space and the non- or
extra-diegetic space1. Subtitling, being focused on the narrative, is concerned only with the internal diegetic space. Thus, the fourth reason why subtitling does not deal with all of the soundtrack is that it sees music, the third sonic element, as ‘outside’ of the film proper. By saying that, it implies that a film’s music purely acts as a commentary on the action but is not necessarily a part of it.
Where this argument falls down is in subtitling’s treatment of found music in film. If the filmmakers choose to use a recognised song or piece of music in place of a specially-written score, this fact is noted in the subtitles. So, for example, the repeated use of Gershwin’s Rhapsody in Blue by Woody Allen in Manhattan would be accompanied by a caption like this: # GEORGE GERSHWIN: Rhapsody In Blue. In this case, the subtitles would be recognising a part of the soundtrack outside the diegetic space. The reason given for this exception is that, as the music may be well-known to the viewer, they might appreciate knowing of its usage. Furthermore, knowing the quality of that particular music, they might be able to see how it works in relation to that scene. However, it still acts as an inconsistency in the logic of subtitling and throws up questions about the relationship between sound and narrative surmised so far. In the case of Manhattan, Allen, like other users of found music, has used Gershwin for two reasons: firstly as a point of recognition for the audience who may be familiar with Gershwin and that era of music but also as a way of complementing the image much as an original score does. The movement from romantic saxophone to full-blown orchestral bombast in Rhapsody In Blue is synchronised with images of Manhattan and underscores the narrator’s feelings about the city. Similarly, in Once Upon A Time…, the ‘As A Judgement’ motif underscores the menace of the ensuing gunfight and the apprehension of the gunmen and yet no mention of it is made on screen.
Allen’s use of found music in Manhattan is also particularly instructive in showing how conventional subtitling’s dismissal of music may be detrimental to appreciating the film. In the final scenes, Allen is recording a monologue into a tape recorder. He remembers Tracey, a former girlfriend, and how much he still misses her face. His voice dies away and He Loves And She Loves by Gershwin fades into the scene. Its romantic melody is operating not so much as an accompaniment to the image but as an interior monologue. It is the continuation of the words Allen was speaking about Tracey before he cut off and the music broke in. Thus, the music here has a narrative purpose; it replaces dialogue’s function of telling us what a character is thinking. He then leaves the apartment to run to Tracey’s hotel; the more upbeat music of Strike Up The Band is heard on the soundtrack. This fast-tempo instrumental not only gives the scene momentum but raises the whole course of the film to its crescendo and is indicative of the inner spirit of Allen’s character. When he finally reaches the street where the hotel is situated, the music abruptly changes to But Not For Me, a bittersweet romantic number. This sudden change of mood cannot possibly be conveyed by subtitles and yet it is vital to the understanding of the turbulent state of mind of the protagonist. Throughout the above example, Allen has conveyed the tone and the narrative import of the scene through music alone. Is it enough then to simply name the tunes employed or is it more important to convey the kind of music being used? And if the latter is true, then should not all music in film have its character conveyed to the deaf viewer?
For Morricone’s music in Once Upon A Time… also has a sophisticated relationship to the narrative. Each character in the film is introduced along with their own musical motif, in Bronson’s case, the harmonica riff. These motifs contribute to the way we respond to and read the character. Christopher Frayling sees Cheyenne’s theme as a ‘straightforward musical cliché from the Western: the plodding horse, the campfire whistle, the lazy rhythm.’ In other words, from its use of honkytonk piano and relaxed rhythm, he has built onto it a series of associations that colour his interpretation of the character. In the case of Bronson’s harmonica, Morricone and Leone had different ideas themselves on the consequences of using that instrument. Morricone thought it recalled the ‘solitude of the country’ while Leone saw it as ‘more sinister.’ In the event, the listener can build both associations onto the harmonica riff and both are pertinent to Bronson’s character – the lone gunman who refuses to settle down in the new railroad town. But these motifs not only colour in the various characters but speak for them, too, in much the same way Gershwin’s standards spoke for Woody Allen. The rumbling piano theme for Morton, the businessman in Once Upon A Time…, is overlaid with sounds of crashing ocean waves. These are extra-diegetic and it’s debatable whether they would be subtitled or not. But they are vital in our understanding of his absolute determination to make the railroad reach the sea. The ocean waves are a signifier of his ‘unspoken’ motivation. Similarly, Bronson’s first words in the film are not spoken as such but iterated through his harmonica. It operates as his call to the gunmen but, as Frayling points out, the strains of the riff can also come across as a ‘lament’.iv The reason for this may be that the harmonica motif itself has a complex relationship to the narrative proper. In the final scenes of the film, we learn that, as a child, Bronson’s character was forced to play the harmonica as he tried to support his dying brother. When his strength sapped, he collapsed, and the notes of the harmonica riff are a representation of the final gasps of breath through the harmonica organ. So the motif is actually a concrete part of the diegesis; its strange three-tone structure is explained by the extraordinary circumstances in which it was played. This point is reinforced when the riff represents the dying breaths of Henry Fonda’s gunman, Frank. Thus the interpretation of it as a ‘lament’ is pertinent, being linked, as it is, with death, and significantly, Morricone dubbed it ‘Death rattle’. So, in this instance, a detailed description of the music may not only be desirable but vital, as it constitutes a complex series of signifiers both within and without the diegetic space.
Furthermore, the motifs do not operate independently of one another. In the scene following Bronson’s first appearance, the McBain family are massacred by Frank and his henchman. As Frank comes onto screen, the ‘As A Judgement’ motif is played, this time on a strident electric guitar. Thus this same motif operates as a musical crossover, linking Bronson and Frank thematically across the film. Although the listening audience does not yet know the link between them, they associate the two with each other, and this comes about through the repetition of extra-diegetic music on the soundtrack. Royal S Brown sees this ‘theme sharing’ as a common trait in Morricone’s scores and one which helps the film ‘avoid literalising the film/music associations.’v In other words, it prevents the
audience from making pat connections between one motif and one character and allows the score to be understood on a more complex level. More significantly, Leone saw the technique as vital in making the viewer ‘understand what dialogue couldn’t explain.’ There is no need to ‘literalise’ the relationship between Frank and Bronson’s character in dialogue if the music can do it more subtly and unobtrusively.
Leone also saw that the score could replace dialogue on a more basic level. At one point in Once Upon A Time…, the music actually constitutes a moment in the action. In the final scene, Bronson’s character is riding into the sunset while Cheyenne follows behind accompanied by the Cheyenne musical motif. Only Bronson can be seen on screen as suddenly the music breaks off. The implication is that Cheyenne has died – with him, his musical motif must vanish, too. This fact is not signalled by any visual information or dialogue- it is wholly carried through music. But what is even more significant about this passage is that Bronson turns as if he himself ‘hears’ the break in the music. It’s a cue for him as well as the audience. So the music in the scene seems to be operating both for the on-screen character and the audience, just as the harmonica in the opening scene is a cue for the gunmen as well as ourselves. Once again, the film seems to be dissolving the barriers between the diegetic and extra-diegetic worlds.
So scored music can have a more complex relationship with narrative than subtitling convention allows and can supply more character information than the dialogue alone. But the soundtrack does not need to be explored solely in its relationship to narrative. Although hard of hearing viewers may be able to tell a great deal about the tone and texture of a film from visual stimuli, certain directors use sound or music in a way that doesn’t just enhance individual scenes but adds to the overall theme and ideology of the film itself. In his controversial adaptation of the Marquis de Sade, Salo, Pier Paolo Pasolini periodically punctuates the narrative, in which a group of fascists torture young teenagers in a country house, with the drone of military aircraft overhead. At first, it seems this noise is part of the diegesis and there will be some consequence arising from it. But nothing occurs and it becomes clear that Pasolini is using this sound in the same way other directors may use music – to supply texture and mood, in this case the ominous sense of doom overhanging the young victims of the villains. And just as the aircraft might seem to be a source of coming relief if they bomb the house and kill the fascists, so the dispersal of the noise represents the lack of hope for the young people and the recurrent nature of their torment. In this instance, it is noise, not music, that is operating as a source of arguably extra-diegetic texture and thematic colour. Examples abound in film of sound being used to add to the meaning of a film instead of just being an accessory to the image track. Andrea Truppin has shown how Andrei Tarkovsky uses sound creatively in films such as Stalker, where sounds, whose sources are either on-screen or off, are recorded so that they deliberately appear not to be coming from the physical space within the frame but from an extra-diegetic space. So, for example, in a series of long shots depicting the outside of a ruined house and a nearby waterfall, the audience can quite clearly hear drops of water that sound like they are falling in an inner space such as a cavern or cellar. That it would be impossible to hear such sounds in the outside space allows them to take on a mysterious quality and enhances the sense of enigma surrounding the ruined house in the film. Truppin even suggests that they give ‘a sense of some transcendent, unlocatable space’, lending the film a mystical quality.
In these two cases, both the nature of the sound and the way it is recorded for inclusion on the final soundtrack are important carriers of meaning. What is significant about the series of sounds at the start of Once Upon A Time… is that they are so clearly defined and distinct. The unnatural loudness and clarity of the sounds is a deliberate ploy on Leone’s part to draw attention to them. The sounds are objects to be appreciated in themselves, not merely the accompaniment of physical actions on screen. Furthermore, they have an aesthetic foundation. Morricone has told how he and Leone discussed a concert where a musician simply made a stepladder ‘creak and squeak’. This concept of using everyday, natural sounds interested both artists and so the first scene of the film, previously intended to carry a score, became ‘a symphony of exaggerated sounds.’ Notice the musical term ‘symphony’; for the filmmakers, this sequence of noises was actually a surrogate score. Indeed, Morricone quipped that it was ‘the best music I’ve ever composed’ and Royal S Brown refers to it as an ‘intriguing musique concrete.’ Thus, the captions that appear on a subtitled version of the film are not just carrying sounds that are meant as narrative information, but sounds that are meant as mood signifiers. Morricone goes on to point out that not only the sounds but the pauses between them are significant:
‘any sound at all from everyday life, isolated from its context and isolated by silence, becomes something different that is not part of its nature.’[ My italics]
So the extreme silence between each individual sound adds to the effect of that sound and its impact on the audience. This silence is difficult to convey in written captions and, without being signalled, may be lost on a hard of hearing audience. The opening scene’s soundtrack is then heavily stylised and constructed almost as a mini-overture to Morricone’s score. The bland, economical captions of the subtitled film do not capture the strangeness of its effect and that is detrimental to one’s appreciation of the scene precisely because Leone is overtly calling attention to the sound. We are being asked to listen, to savour almost each individual noise. Thus, when the bizarre harmonica motif is finally heard, the audience has been guided to ask ‘What is that sound? What does it mean?’
What is also distinctive about the ‘soundscape’ created by Morricone is that sounds actually merge into the music and become part of its rhythm. In the final shoot-out, the full version of the ‘As A Judgement’ theme is played out. As it reaches its final notes, a funereal bell tolls and then the gunshot is heard. This bell is not part of the diegetic universe; it acts as the final note in Morricone’s music. But it also acts as a semi-humorous comment on the action; this is, after all, Frank’s funeral. This merging of sound and music is part of Morricone’s aesthetic and the way he builds the harmonica riff into the main score is exemplary of it, too. For Morricone sees this motif not purely in musical terms. The three notes, he said, were ‘articulated… as a physical force like a heartbeat.’ In other words, a musical construction was made to represent an actual physical object or effect like a heartbeat or a man’s breathing. This physical presence is then articulated into the score as a whole. Conversely, Morricone takes an existing physical phenomenon, like the human voice, and makes it into music. The soprano solos in the main theme, linked to Claudia Cardinale’s character, are an example of ‘the human voice as musical instrument, vocalising without a text.’ And so, once again, the concept of music ‘speaking’ what dialogue cannot becomes a part of the film’s structure.
Michel Chion has put forward a theory of three types of listening: causal, semantic and reduced listening. The first ‘consists of listening to a sound in order to gain information about its cause.’ It is the listening of identification and, in cinematic and subtitling terms, is primarily to do with sound effects – what they are and where they’re coming from. Semantic listening is listening in order to obtain meaning and can be most closely associated with dialogue, information carried through a character’s voice. Reduced listening, however, is listening purely for the quality of the sound itself. Now, as subtitling regards music as supplemental to narrative as opposed to being part of it, and as music is usually composed to be listened to anyway, it can be placed in the reduced listening category. But the previous examples have shown how the three listening modes are not necessarily distinguishable. Indeed, in the case of Morricone’s music for Once Upon A Time…, we can see how film sound can engage all three listening modes simultaneously. The opening overture of ‘amplified natural sounds’, which acts as a kind of prologue within the score, is linked to definite physical objects, sets the audience up for the strange harmonica sound that will follow and allows the sounds themselves to be appreciated for their own intrinsic quality. The score proper, meanwhile, has associations built into each individual character motif and, through musical crossovers,
can make links between these individuals. Thus, the Once Upon A Time… score collapses any boundaries between the three modes of listening. Therefore, to tackle the film with the same subtitling conventions as one would use for a film that is less creative in its use of sound would be to deprive the viewer of its richness and the essential thematic colour it provides. Ultimately, then, it can be seen that some films require more sensitivity to their use of sound than others and that no blanket conventions can hold. In the case of Leone’s film, there is a strong argument to suggest that captions should be fuller and be more descriptive of both the musical and sound effect passages. After all, subtitling is the last process in production before the film reaches its audience; seen in those terms, shouldn’t it be allowed the same margin of creativity as all the other stages in the production process? In that way, the complexity and depth of an intelligently-structured soundtrack has a concrete presence in the captions on-screen.
Creak of door
Unidentified creaking off camera >
As the intruders enter the station, the stationmaster scrapes his chalk on the blackboard.
There is a minute-long sound mix of footsteps, creaking doors, the wind outside and the twittering of birds
Footsteps on wooden boards as the gunmen enter
The Indian lady lets out a gasp of fear
Stationmaster: If you want any tickets,
you’ll have to go round to the front of…
Oh, well, I suppose it‘ll be all right.
Third gunman: Chitter, chitter! (at the bird in the cage)
Stationmaster: Three?
That’ll be seven dollars…
..and 50 cents.
Cock crows.
Jack Elam: Sh! (as he pushes the stationmaster inside a cupboard)
Screech of door as it closes, slams shut. LOCK CLICKS.
(First credit title.)
Indian lady babbles incoherently as she runs away.
Rattle of spurs and the sound of boots on boards as the gunmen take positions.
Elam pulls a chair back. The creaking of windmill becomes clearer and identified.
In shot and in close-up, it becomes a whirring.
Third gunmen swirls water in trough. (This may be subtitled as TRICKLING OF WATER)
Creaking of Elam’s chair as it rocks.
CLATTER OF TELEGRAPH MACHINE
CLATTER STOPS (As Elam destroys it)
SPLASH OF WATER (On Woody Strode’s head)
THUD OF WATER ON HAT
CRACK OF KNUCKLES
BUZZING
KNUCKLES CRACK
THUD OF WATER
Creaking of the windmill
BUZZING STOPS
Slam of Elam’s gun into the side of the bench
BUZZING INSIDE THE GUN BARREL
BUZZING STOPS
Elam looks inside
BUZZING STARTS AGAIN
TRAIN WHISTLE
This is amplified and accompanied by the screech of the rails when the train is shown on screen.
TRAIN APPROACHES
Strode swallows the water in his hat
Click of gun. The chamber is loaded.
Train comes to a halt. Whistle. BELL RINGS
As the train is at a halt, steam is let out in a rhythmic beat. A door opens and a package is thrown out.
TRAIN WHISTLE
Puff of engine as the train breaks away
SINISTER HARMONICA MUSIC
(The music is made up of three notes sliding between pitches played in a slow, lazy manner. It does not make up a tune or melody as such but the soundtrack music cuts in and builds on it with strings and percussion. This part of the scored music to the film is a citation from a larger theme heard later called “As A Judgement” by Morricone.)
Bronson: Frank?
Elam: Frank sent us.
Bronson: You bring a horse for me?
Elam: Well, looks like we’re…
(LAUGHS) Looks like we’re shy of one horse.
(ALL LAUGH)
Bronson: You brought two too many.
Gunshots. Horse whinnies.
All crash onto the floor.
Windmill creaking.
Bronson’s hat and boots make dull thuds on the boards as he gets up.
NOTE:
The captions written in bold are the actual subtitles that would appear on screen in a version prepared for the deaf or hard of hearing. Note that the actors’ names would not appear in the final captions.