Essentially, there are two types of sounds that one hears in a movie: diegetic and non-diegetic. The basic difference is easily explained. Diegetic sound is anything you hear that has a source within the story; non-diegetic is sound that has no source within the film itself. Usually this distinction is very easy for an audience to appreciate. When characters are dancing to a song playing on the jukebox, that is a diegetic sound. Those strings that accompany Norman Bates’ slashing of Marion Crane is non-diegetic. A conversation between characters is diegetic; narration from an unseen character is non-diegetic. So far, so good, but why is it important to know these terms?
Because if you are interested in becoming a filmmaker, it is wise to understand the larger possibilities inherent in effectively using sound. A diegetic sound can be manipulated to enlarge the spatial properties of your scene, for instance. This can be accomplished through a slow rise in volume, or through a stereophonic effect of having the sound bounce back and forth from right to left. Volume indicates distance. A louder sound gives the impression of close proximity, while a lower volume can suggest distance. How can you use this to your advantage in a low-budget film? Well, rather than indicating a vast geographical space when constrained by a location that is actually quite small, you can manipulate the audio to do what you can’t do visually. This isn’t limited merely to volume. Let’s say your scene is supposed to take place within a cavernous interior but you don’t actually have access to such a location. By skillfully adding an echo effect to your sound track you can quite easily create a believable sense of space. Remember, however, that a simple echo will sound like just that. By further adjusting the quality level of your echo you can further enhance the reality of distance between two speakers within your enormous visual space that isn’t really there.
Another fun way to mess around with sound is to combine diegetic and non-diegetic sound. Have you ever seen a movie or TV show where music that appears to be non-diegetic-that is, not from a source but as part of the soundtrack intended to manipulate emotion-suddenly is revealed to be diegetic after all? More often than not, this will be done for comedic effect. Let’s say your film is a parody of a suspense movie and as your characters are trying to remain inconspicuous as they follow their suspect, appropriately sneaky-sounding music begins to play. Then one character turns to the other and tells him to turn off the car stereo and it is revealed that this time the music was actually diegetic. You’ve seen something similar to that, no doubt. Another way to use this unexpected juxtaposition of diegetic and non-diegetic sound involves the use of a narrator who is not a character in the movie. Your narrator may exist-along the lines of the narrator of The Age of Innocence-only to provide information that the characters cannot, or to provide an informational overview of a particular time or place. The problem here is that you need the omniscient narrator because having a character narrate your story won’t work for some reason. But this also sets up a wonderful potential for using the two different kinds of sound to achieve an effect not otherwise possible.
One effect is to have your all-knowing narrator pose a question such as “If only Dr. Malarkey had known that saving Janet’s life would bring him such misery for years to come, what would he have done?” This would immediately be followed by a close-up of Dr. Malarkey looking directly into the camera and saying, “I’ll tell you what I would have done.” The natural inclination at this point is to believe that Malarkey is diegetically responding to the non-diegetic question even though that is impossible. Malarkey’s next line depends on how you want to juxtapose your diegetic and non-diegetic sounds. He could either say, “I would have bet the pot instead of folding” as you reveal that his first line was actually in response to an entirely different “what would he have done” query. Or, if you wanted to really play around in the sound space afforded by this opportunity you could have Malarkey’s next line be “I would have done exactly the same thing. My God, she was beautiful!” as the camera pulls back to reveal that he is responding to the same “what would he have done” question, but posed by an actual character. Orson Welles’ only on-screen contribution to The Magnificent Ambersons is as an unseen, all-knowing, God-like narrator who comments on the action. He takes advantage of the full potential offered by using diegetic and non-diegetic sounds under these circumstances.
Another way to use omniscient narrator and play around with both types of sounds is as an ironic counterpoint. The great French filmmaker Jean Luc-Godard literally took this possibility to the level of art in almost all his post-Weekend films. But the most famously successful use of combining diegetic sounds with a non-diegetic narrator is probably Stanley Kubrick’s Barry Lyndon, where the narration is almost entirely ironic. Barry Lyndon is a difficult movie for many, and has never been of Kubrick’s most popular films. It does have its pacing problems, but it is worth watching at least once to understand just how incredibly important understanding how manipulation of diegetic and non-diegetic sounds can be in a film.