Christopher Doll on The Inception

SMT-V is the Videocast Journal of the Society for Music Theory; it has been publishing since 2015 using the Vimeo platform. The most recent title is Christopher Doll's “Was it Diegetic, or Just a Dream? Music’s Paradoxical Place in the Film Inception”: link. [Link checked on 3-01-2024.] 

Here is the abstract:

Between "diegetic" film music (heard by the characters) and "nondiegetic" film music (heard only by the audience) is a paradoxical space called the "fantastical gap." A film such as Inception (2010) makes traversal of this gap into an overt theme, obscuring our sense of place to such a degree that even the literal plot of the movie is open to interpretation, and thus also illustrating the extent to which filmmakers can manipulate an audience's understanding of the filmic world through the blurring of the diegetic/nondiegetic divide.

The video opens with succinct example clips for the categories of diegetic, nondiegetic, and "fantastical gap," then traces the connection between the latter, musical themes, and narrative. Although posted to a professional music theory site, “Was it Diegetic, or Just a Dream?" is very accessible and would be an excellent accessory to the opening pages of Chapter 3.

Hollywood Musicals Quiz

A Hollywood musicals quiz is available on the Oxford University Press blog: link.  [Link checked on 3-01-2024.]    The quiz is based on a chapter in a book by Todd Berliner. The chapter "traces the history of the convention that characters in Hollywood musicals burst into song without realistic motivation."

This idea is related to—but is distinct from—the audio dissolve that we discuss in Hearing the Movies, chapter 3. Although we do say that the audio dissolve "serves as a transition to song and dance," we use the term as Rick Altman defines it, for the transition from diegetic song and accompaniment to song with a non-diegetic accompaniment. In such cases, the initial status is clearly diegetic ("The character seems to know very well that he or she is singing"). In fact, "one very common device in classical Hollywood musicals is to show one or more characters gathered around a piano." Once this is established, and "as the song progresses, nondiegetic orchestral accompaniment enters, replacing the piano."

Berliner is concerned, then, with what happens before the singing starts and with the transition into it. This notion that musicals permit "bursting into song without realistic motivation" is consistent with our generalization that film "musicals do not maintain the clear separation of diegetic and nondiegetic registers and so cannot rigorously enforce the boundaries of the diegetic world, which seems to constantly dissolve under the force of song." (Quotes are from Hearing the Movies, second edition, p. 69).

Reba Wissner on Hearing the Unseen

On The Avid Listener (W. W. Norton's in-house blog), Reba A. Wissner writes about "Hearing with Your Eyes: Science Fiction Television and Hearing the Unseen": link. [link updated on 3-08-2024]  Her video examples include a spooky scene from 19th century composer Carl Maria von Weber's opera Der Freischütz and three scenes from TV sci-fi shows (The Invaders, The Outer Limits, and The Twilight Zone).

Although she doesn't use the term, Wissner is writing about what Michel Chion calls the acousmêtre, an unseen agent "embodied" by sound (Hearing the Movies, second edition, pages 74-75), but she also discusses actions (physical attacks) carried out by these agents and also not presented onscreen (Hearing the Movies, chapter 3, describes and illustrates a number of types of offscreen sound).

As with Jonathan Godsall's Avid Listener post that we recommended a few days ago (link)  [link checked on 3-08-2024], Wissner's post has discussion questions at the end. These, however, assume fairly substantial student knowledge of television repertoires. One might complement or substitute for those questions an exercise that compares/contrasts techniques and effects in the TV clips with those in a feature film scene, perhaps from one of the films mentioned in Chapter 3, one of the versions of The Invisible Man, or even a teen horror flick (some films in that genre have achieved a kind of cult status and might well be familiar to many undergraduates; your current writer has found that students may seem a little embarrassed about their own knowledge of this repertoire, but one has only to ask and they will quickly come up with titles and also be enthusiastic about the chance to make presentations to the class).

Week 4 and Week 5 Lectures

I'm a little late getting the week 4 and week 5 lectures up due to a big conference on music, sound and film I attended last week.

I opened the week with a discussion of the acousmêtre (acoustical being). I developed the concept using the work of Chion, Bonitzer and Doane. Here are some quotes:

Acousmatic. Dictionary definition: “a sound that is heard without its cause or source being seen.” (quoted Chion, The Voice in the Cinema, p. 18)

Acousmatic—“Pertaining to sound one hears without seeing its source. Radio and telephone are acousmatic media. In a film, an offscreen sound is acousmatic” (Chion, Audio-Vision, p. 221)

Definition—“A kind of voice-character specific to cinema that in most instance of cinematic narratives derives mysterious powers from being heard and not seen.” (Chion, Audio-Vision, p. 221).

Between one (visualized) situation and the other (acousmatic) one, it’s not the sound that changes its nature, presence, distance, color. What changes is the relationship between what we see and what we hear. (Chion, The Voice in the Cinema, p. 19)

The use of the voice-off always entails a risk—that of exposing the material heterogeneity of the cinema. Synchronous sound masks the problem and this at least partially explains its dominance. But the more interesting question, perhaps, is: how can the classical film allow the representation of a voice whose source is not simultaneously represented? As soon as sound is detached from its source, no longer anchored by a represented body, its potential as a signifier is revealed. There is always something uncanny about a voice which emanates from outside the frame. (Doane, “The Voice in the Cinema,” p. 40)

[The consistent use of voice-off for the villain in Kiss Me Deadly] gives to his sententious voice, swollen by mythological comparisons, a greater power of disturbing, the scope of an oracle—dark prophet at the end of the world. And, in spite of that, his voice is submitted to the destiny of the body . . . a shot, he falls—and with him in ridicule, his discourse with its prophetic accents. (Bonitzer, quoted Doane, “The Voice in the Cinema,” p. 41)

The acousmêtre . . . must haunt the borderlands that are neither the interior of the filmic stage nor the proscenium—a place that has no name, but which the cinema brings into play. (Chion, The Voice in the Cinema, p. 24)

De-acousmatization of a character generally goes hand in hand with his descent into a human, ordinary, and vulnerable fate” (Bonitzer, quoted Chion, Audiovision, p. 131).

For examples I used Wizard of Oz and Psycho. In the past, I've also had good luck with 2001: A Space Odyssey, though I've found that film takes a bit longer to get the concept across. The process of de-acousmatization is particularly clear (and amusing) in Wizard of Oz. Psycho, on the contrary, resists showing the final de-acousmatization, which helps produce the uncanny effect of its ending, a gesture that the slasher film has taken over and made its own.

* * * * *
We then moved on to chapter 4 proper and examined the various transitions covered there:

• Sound Advance. I showed a long sound advance in When Harry Met Sally. We hear Harry and Sally talking on the phone about Casablanca as we see a montage of Harry's and Sally's days both together and apart. Only after about 90 seconds of this do we get synchronization, with each in their own house, in their separate bed, but watching the ending of Casablanca together as they discuss it on the phone. This example is discussed on p. 94 of the text.


• Sound Lag. For the lag, which is much less common than the advance, I use Goodfellas, which we discuss on p. 95 of the text. Here, the transition from the night club act to the airport caper are linked through the continuation of Henny Youngman's act well into the beginning of the airport sequence.


• Sound Link. For the link, I showed two examples. I began with a brief sequence from Emma, where Emma must tell Harriet that Mr. Elton is getting married (HtM, p. 96). We then watched the more elaborate "Would You" sequence from Singin' in the Rain. This sequence is discussed on pp. 96-97 and 100 of the text.


• Sound Match. A Match is generally less elaborate than a Link, and the Match from De-Lovely discussed in example 4-2 (p. 97) could easily be analyzed as a link by analogy with the example from Emma discussed above since in both cases the connecting sound is continuous across the cut and more or less identical in both scenes. A Match may also be based on resemblance rather than identity, however, in which case the resemblance may suggest a metaphorical connection between the two shots. A good example of such a link occurs in The Big Broadcast (1932), where pounding on the table turns into pounding on a car horn. This example is discussed on p. 97 of the text.


• Hard Cut. For the hard cut, we watched a short sequence from Blue mentioned in the text (p.98). Note the abrupt shift in sound mirrors the abrupt shift in image, and together they help mark a sharp temporal break despite the lack of formal transition.

* * * * *

The distinction between synchronization and counterpoint (playing with or against the film), covered on pp. 98-110, was the next topic. I used a couple of sequences from Singin' in the Rain to talk about general issues of synchronization. We started with the failed premiere, which is useful for emphasizing that it takes a lot of work to get synch sound right.

I then showed the bit where Cosmo illustrates how they might substitute Kathy's voice for Lina's, followed by the short scene of Kathy looping in Lina's dialogue (which was actually done by Jean Hagen, the actress who played Lina). If I had time, I would have screened the "Would You" sequence again, as it brings together many of the concerns of synchronization.

We next turned to the general issue of playing with the image. We watched the sequence where Rick and Ilsa meet from Casablanca, discussed on pp. 102-06. This is a particularly good sequence for illustrating dialogue underscoring, which in this case follows the scene very empathetically. The real effectiveness of Steiner's practice here can be illustrated by comparing it to the French language dub, which on my DVD uses very little of Steiner's score, substituting different music, also based on "As Time Goes By."

For counter- point or playing against the image I showed the sequence from Catch Me If You Can discussed in the text, p. 107. Nat King Cole's rendition of "The Christmas Song" plays, apparently diegetically, while Frank approaches the house and sees his mother happily a part of a new family. Chion notes that anempathetic music, especially when nondiegetic, often signifies the world's indifference to the characters. There is certainly a degree of indifference here, but the music serves as much to emphasize Frank's exclusion from the scene, as though we are hearing music that insists on a point of view that is not Frank's. Choosing to withdraw from the scene rather than force himself back into his mother's life, Frank seems experience empathetic recognition, which is underscored by having the music lag into the next scene as we hear his sentence handed down. The empathetic recognition also helps convince us that Frank is in fact redeemable, which aligns us with Carl.

Music that seems at odds with the image often produces an effect of an intruding—or at least shift in—point of view. Near the end of the opening battle of Gladiator, a solemn hymn suddenly drifts in over the images, which also shift into slow motion with distorted sound. This kind of gesture is occasionally used to underscore battles from the perspective of a "lost cause," but here the film remains very much on the side of the victorious Roman army. The shift in music does seem to signify a new point of view, however, as the cut to the Emperor watching from afar indicates.

* * * * *

The above material took a little less than two class periods to present (NB: my class meets twice a week). In addition, I also prepared the class to write a screening report. For that, we looked at the opening ten minutes or so of Catch Me If You Can. As we did not hold class for the first meeting of week 5, the students worked on their screening reports instead. For the second class meeting of the week, we spent about an hour on the commutation test (pp. 110-13). First I showed the "Non Nobis Domine" sequence from Branaugh's Henry V, which basically consists of a single virtuosic tracking shot across the body strewn battle field accompanied by Patrick Doyle's version of the hymn (in fact that's the composer shown singing at the beginning of the clip below). I have prepared an alternative sound track to this segment using "Vergangenes," the second of Schoenberg's Five Pieces for Orchestra, op. 16, and some foley effects. The contrast could hardly be more marked; I've found that with the change in music students will actually see a different preponderance of colors: where with Doyle's score students will generally say they see blues, whites and greens, with Schoenberg's music they will generally say they see, reds, oranges and browns. Schoenberg's music also lays the emphasis on the dead bodies that are strewn across the image, whereas Doyle's score encourages us to see the growing numbers of people marching across the scene. Of course, I play the Schoenberg first, and I don't let the students know that I have changed the sound track. I then play the Doyle under the pretense of listening to the sound track again. If you like, you can easily turn the discussion to the rhetorical force of music, since in this case the sound track alone seems to turn the scene from expressing anti-war sentiment (Schoenberg) to war as something quite a lot more positive, if not exactly warmongering (Doyle).

We also did an actual commutation test, using a scene from Taxi Driver and a playlist put on shuffle. The scene consists of Travis cruising through a red light district to the sound of a soft, jazzy nondiegetic saxophone.

* * * * *

Bibliography on Acousmêtre
Chion, Michel. Audio-Vision (1994).
_________. The Voice in the Cinema (1998).
Doane, Mary Ann. "The Voice in the Cinema: The Articulation of Body and Space." Yale French Studies 60 (1980): 33-50.

Chapter 3 Examples

Like Chapter 2, Chapter 3 covers a lot of terms. I began my lecture by identifying three basic distinctions: onscreen/offscreen, diegetic/nondiegetic and foreground/background. We then watched the "Tears Flowed Like Wine" performance sequence from The Big Sleep. The diegetic performance begins offscreen and in the background but quickly moves onscreen and to the foreground.

We then watched the party scene from Glory (HtM, 67-71) as a second example; as mentioned in the text, this is a very rich example, featuring extensive intermingling of diegetic and nondiegetic music as well as distortion of diegetic sound.

I then lectured a bit on diegetic and nondiegetic as a distinction that applies to the level of narrative. I find the following quotes to be useful:

Diegetic—"Pertaining to or part of a given diegesis and, more particularly, that diegesis represented by the (primary) narrative.”

Diegesis— “1. The (fictional) world in which the situations and events narrated occur. 2. Telling, recounting, as opposed to showing, enacting.”

—Gerald Prince, Dictionary of Narratology

And this succinct definition from Claudia Gorbman:

"Diegetic music: music that (apparently) issues from a source within the narrative" (Unheard Melodies, 22).

The third example I used was the de Lesseps' dance sequence from Shakespeare in Love (HtM, 71). As mentioned in the text, this example has subtle and narratively effective blurring of diegetic and nondiegetic music. This example is also a good one to initiate a discussion of the audio dissolve.

I find the title-song sequence from "Me and My Gal," which we discuss in the text (HtM, 83-84), to be particularly good for illustrating the audio dissolve proper, as the process is slow and deliberate. I also showed "The Way You Look Tonight" from Swingtime (at right) and "Cheek to Cheek" from Top Hat. In both the transitions back to diegetic sound after the song involve collapses that very much emphasize the way the nondiegetic sound creates an idealized space. "Cheek to Cheek" has the added complication of beginning with a dance orchestra so the dissolve itself is more conceptual than real, involving the disappearance of other dancers from the screen and the orchestration seeming to react empathetically to Astaire and Rogers dance rather than a shift from music anchored diegetically to nondiegetic orchestral support.

The best example of an audio dissolve used dramatically probably occurs in Casablanca, the transtion to the flashback sequence. Because I reserved this example for a quiz, I chose to illustrate the dramatic use of the audio dissolve with Pippin's song from the "Sacrifice of Faramir" sequence in Return of the King, which is discussed in the text (HtM, 83, 85).

* * * * *

The other big topic of chapter 3 concerns issues of onscreen and offscreen sound. In fact because we often infer nondiegetic status from its systematic offscreen placement (since it does not belong to the scene depicted), it is not always easy, nor wise to separate the two distinctions and I use this slide to relate the two. To illustrate offscreen sound, I showed the scene from Bleu mentioned in the text (HtM, 74) and the first "Que Sera, Sera" sequence from The Man Who Knew Too Much, which uses offscreen singing to unite mother and son conceptually; the idea of this curious unity returns near the end of the film when the mother sings the song again in an effort to locate her kidnapped son. We also watched the end of The Apartment, with the misrecongized champagne cork (HtM, 78) to illustrate how filmmakers can play on the inherent ambiguity of offscreen sound. In the past, I've also used the approach to Skull Island sequence from King Kong, which features a misrecognition of drums and also seems to blur the distinction between diegetic and nondiegetic, and the end of the Indian attack from Stagecoach, where Mrs. Mallory recognizes the diegetic cavalry bugle amid the din of nondiegetic music.

Voiceover was the next topic. I used the opening narrations of Laura, Clueless and The Apartment. The openings of American Beauty and Sunset Boulevard (HtM, 81) are worth considering as well because they feature the "impossible" condition of the dead narrator. Letter from Siberia (HtM, 80) illustrates how effective narration can be in telling us how, and even what to see in an image; this holds important consequences for thinking about how nondiegetic music functions.

The opening narration of The Apartment also uses musical sweetening to underscore the returns of accounting machine (HtM, 86), which makes this example effective for making a transition to the concept of mickeymousing and sweetening. I showed two examples from Casablanca to illustrate this idea. I began with the final scene, after Strasser has been shot and Renault has to decide whether to accuse Rick. After Renault decides to protect Rick, he throws the Vichy water into the trash bin, which Max Steiner's score underscores with a sharp stinger. We then watched an earlier scene where Rick first sees Ilsa (image at right), a moment Steiner also underscores with a poignant stinger that serves to mark Rick's painful recognition (HtM, 87-88). Here, the stinger serves not as a sound effects sweetener marking a significant narrative action, but as a nondiegetic representation of emotional shock.

I concluded this week's lectures by returning to the "Mapping the Possibilities" slide and filling in the very counterintuitive fourth quadrant, onscreen nondiegetic. My favorite example of this is the band in There's Something About Mary, which functions something like a Greek chorus and goes completely unnoticed by diegetic characters despite being onscreen. As an example, I use the final scene because it is relatively short and the singer ends up being inadvertently shot by one of the diegetic characters.

I also played a clip from Ferris Bueller's Day Off; the title character's asides to the camera seem similarly nondiegetic, intended solely for the film's audience and externalized in a way that makes them difficult to interpret as interior monologue. The end of the Chez Quis restaurant sequence illustrates the difference by interpolating a clear interior monologue by Ferris' sister (rendered as voice-over, here understood as a voice of interiority rather than nondiegetic) and then shifting to Ferris speaking to the camera in the bathroom with the attendant completely oblivious to his words. (The mirror is perhaps a complicating factor in this particular scene, though most of Ferris' asides do not involve a mirror.)

* * * * *

Because I had a quiz scheduled, I deferred discussion of the acousmêtre. I plan to use Wizard of Oz and Psycho to illustrate the concept. In the past, I have also used 2001 and Friday, the 13th.