Six Authors In Search of a Character: Part 8, Voice Acting

While characters moved to the beat of the music in early sound cartoons, the use of the human voice was not as important. Dialogue was relatively sparse as animators were inexperienced with lip synch.

Early voices were done by cartoon studio employees rather than trained actors. Walt Disney provided the voice of Mickey Mouse and Marcellite Garner, who worked at the Disney studio tracing animator drawings onto celluloid, provided the voice of Minnie Mouse (Barrier, Speaks 20). Carman Maxwell, the production manager for the Harman-Ising studio, provided the voice for their cartoon star Bosko (Barrier, Speaks 17).

Had an animator been responsible for all of a character’s behaviour at this point in time, animation may have developed like puppetry with the animator supplying his or her character’s voice. For example, Jim Henson not only provided the voice for Kermit the Frog, he was also the manipulator of the Kermit puppet. While Henson acted with something external to his own body, making him closer to an animator in that regard than a live actor, he was still in complete control of Kermit’s behaviour. Because animators had given up exclusive control of a character’s behaviour in the silent era, some animators on a character would be working with another person’s voice. From there, it was a short step to all the animators working with another person’s voice, and that opened up the job for professional actors.

As the 1930’s progressed, animation studios became more sensitive to the possibilities of the voice track. Character voices became more distinctive. Mickey Mouse’s voice is simply a falsetto, but Goofy (who debuted in 1932) and Donald Duck (who debuted in 1934) both have voices whose sound contributes to communicating their personalities to an audience. The sound of Goofy’s voice emphasizes his lack of intelligence; the sound of Donald’s squawk communicates his out-of-control rage.

Like animation, radio was also maturing as a medium in the 1930’s. Lacking visuals, the radio actors who succeeded were those who could create full-blown characterizations with just their voices. Many actors who worked in animation first worked in radio and continued their radio careers concurrently. Mel Blanc, who started in radio in Portland, Oregon, provided voices for the majority of the Warner Bros. cartoon characters (Barrier, Speaks 28). While acting in cartoons, he also performed on The Jack Benny Program, The Great Gildersleeve, Baby Snooks and Blondie (Blanc 132-33). Arthur Q. Bryan, who provided the voice of Elmer Fudd, was first a regular on the radio show The Grouch Club (Barrier, Speaks 29). June Foray worked on radio with Jimmy Durante, Danny Thomas, Bob Hope and Red Skelton before doing voice work for Disney in the 1950’s. She was later the voice of Rocky the Flying Squirrel on the Bullwinkle TV show (Barrier, Speaks 31).

In the first few years of sound, voice tracks were recorded simultaneously with music and sound effects (Barrier, Speaks 17). This meant that the dialogue for lip synch would be timed out before it was recorded, forcing the voice artists to match the lips of the animated characters, much the way any live action film might be dubbed. With the difficulty of getting music, sound effects and voice all working together in a single take, there was an incentive for dialogue to be simple and brief.

West coast animation studios moved to pre-recording dialogue while east coast studios continued to post-record throughout the ‘30’s (Barrier, Speaks 17-18). These two approaches led to very different results.

In pre-recording dialogue, studios guaranteed better synchronization since the soundtrack would be phonetically analyzed and plotted on the exposure sheet. It was relatively simple for animators to match the lips of their characters to the existing soundtrack, so dialogue became more prominent in cartoons. More importantly, however, animators no longer went first in the collaboration with the voice talent.

When the animators went first, they had more control over how a character would emote while speaking. If a line of dialogue was “Look out!” the animator would be the one to determine how broadly the character would move while saying it and the voice actor would have to scale the reading of the line to match the visuals. With pre-recording, the actor went first. In reading “Look out!” the actor would decide the pace and how much emotion to put into the line and the animator was forced to match the voice track. In effect, the person who went first got to shape the character’s behaviour. Once pre-recording became standard, the actor got to interpret the script and the animator was forced to interpret the actor.

Pre-recorded voice tracks are the raw material of animated behaviour, as can be seen in this advice from Disney animators Frank Thomas and Ollie Johnston: “Listen to [the voice track] carefully; listen to the thoughts and ideas – they are your character’s. He is thinking them, and you must capture them” (467; emphasis in the original). Animator Dick Lundy stated, “It was the dialogue that had the personality that you tried to portray” (Barrier, Hollywood 118).

Andreas Deja was the lead animator on the character of Scar in The Lion King, voiced by Jeremy Irons.

“As a voice talent and actor, he was able to do so much with the dialogue and was a great springboard for the character,’ recalls Deja. ‘He had a way of playing with the words and twisting them so that they would come out very sarcastic and always a bit unexpected. I would watch him at the recording sessions and then run back to my desk because I couldn’t wait to get started with the animation” (Ghez 263).

It’s not just the emotional tone of the behaviour that comes from the voice track. It has an impact on the timing of the character and determines what part of the motion the animator will emphasize to accompany the soundtrack. Animator Chris Buck says that “once the scenes are handed out, I’ll listen to the voice. That will dictate a lot about the character, whether his movements are sharp or slow” (Cawley 203). Animator Mark Kausler says that,

“If I’m starting on a new scene I listen to the [voice] track, if there’s a track on cassette of the character’s voice, 30, 40 times. Then I look at the exposure sheet I’m given and I circle words that seem to be the most important words to hit. That’s your key. Usually the soundtrack will guide you, if there is one, to what’s the most important points in the scene to hit. When I circle those words, I know that’s were I’ve got to make my emphasis” (Cawley 111).

At the Fleischer studio, located in New York City, post-recording continued to be the norm. However, something interesting happened on the Popeye series. William Costello was the original voice of Popeye starting in 1933 (Cabarga 87). His voice was gruff and Popeye was portrayed as a dour, roughneck character. Costello was replaced by Jack Mercer, who was also a talented writer (Cabarga 88). Mercer created ad libs in the recording session, even though there were no animated mouth movements to match his dialogue. Because the animation was already done and would not be changed to reflect Mercer’s additions, he was limited to commenting on the on-screen action. In The Paneless Window Washer, Bluto closes a window on Popeye’s neck and Popeye says, “Hey, you give me a pane in the neck!” In I Never Changes My Altitude, the covering of an airplane wing rips off and Popeye comments, “I think I need a little more mucilage on that fuselage!” Instead of Mercer having to take second place to the animators, restricting himself to accompanying their actions, he used his freedom to add dialogue to build an additional dimension to Popeye’s character, giving out a constant stream of self-conscious humour.

When you compare the Costello and Mercer Popeye cartoons, Mercer’s contribution to the character is clear. Mercer softened Popeye, counterbalancing the violence in the cartoons with good-natured comedy. Mercer was able to have as much impact on Popeye as other voice actors who got to pre-record their dialogue because the Fleischers ignored the necessity of lip synchronization.

Pre- and post-recording had another effect on how voices were recorded. In a post-recording situation, the actors all had to be in the studio at the same time, as music, sound effects and voices were recorded together. It gave actors the same opportunities they had in theatre, film or radio, to work in an ensemble situation where they could react to other actors’ performances in their own. In a pre-recording situation, there was the possibility of recording actors separately. If an actor was recording a scene with two or more characters, he or she might record without the other actors being present, having only the director and the storyboard to guide the interpretation. The DVD release of Open Season includes a documentary on the voice actors that reveals that Martin Lawrence and Ashton Kutcher, the two vocal leads, did not meet during the three years that they each worked on the film.

Some of this is determined by the medium. Based on my own experience, you can record an animated half hour in 3-4 hours. The rush to get it done means that it’s often more efficient to put more than one performer in the booth at a time and let them play a scene out. On a feature, performers may be called back several times, months apart, to record additional dialogue as story ideas develop during production. Re-assembling the entire cast for a few new lines is not efficient, so features default to actors being recorded separately.

Studios and voice actors have their own preferences in this manner. Jay Ward, the producer of Bullwinkle, preferred to have the performers in an ensemble situation (Neuwirth 215-216). Hanna-Barbera worked in an ensemble fashion and Don Messick (the voice of Scooby Doo, among others) liked this approach due to his experience working in radio and on stage. “I like to feel the relationship of the other people I’m working with” (Dobbs, Messick 28). By contrast, Billy West (Ren and Stimpy) prefers to work alone in a recording booth (Dobbs, West 42).

It’s interesting that animation is the only medium where actors are prevented from interacting. One of the purposes of rehearsal is to allow the actors to gauge each other’s performances so that they can blend the performances into a consistent approach. Because animation people take fragmentation for granted, they actually impose it on actors, forcing them to act in a vacuum. The performer has to rely solely on the director’s feedback to judge how he or she is doing. Animation has no equivalent of a table read, where the cast sits around the table reading the script and discovers what chemistry might exist between performers or characters.

Just as a separate character designer and animator split a character’s appearance and behaviour, it has a similar effect on voice actors. Often before creating a voice, an actor will be shown a drawing of the character so that he or she can create a voice that’s appropriate. In the 1940’s, Sid Raymond was shown a drawing of Baby Huey and he created several voices that he thought were appropriate to the design (Dobbs, Raymond 38). John Leguizamo was supplied with a statue of Sid the sloth for Ice Age before creating the voice and “knew the voice couldn’t be myself” (Neuwirth 230).

On the other hand, there has been an increasing reliance on star voices in animated features in recent years, and this has become a major influence on the creation of animated personalities. Disney used the voices of actors who were known to the movie-going public at least as early as Snow White and the Seven Dwarfs in 1937. Billy Gilbert, a supporting actor who appeared in many two-reel comedies and in features like Captains Courageous, was the voice of the dwarf Sneezy. Other character actors lent their voices to Pinocchio (Walter Catlett, Cliff Edwards and Frankie Darro) and Dumbo (Ed Brophy, Cliff Edwards and Herman Bing).

For the character of Geppetto in Pinocchio, Disney originally cast Spencer Charters and a one-minute animation test with a design based on Charters’ physical appearance was done. When Charters was replaced by Christian Rub, the design was altered to more closely resemble Rub (Frank Thomas 203-204).

While Rub was known to audiences from his live action roles in films like Little Man, What Now? and You Can’t Take It With You, he was not a major star or box office draw. In recent animated films, voices of stars have been used for their name value and characterizations have been built more on a star’s persona. Eddie Murphy was cast for the role of Mushu in Disney’s Mulan. Tom Bancroft, the lead animator on the character, acknowledged Murphy’s influence.

“There’s a lot of Eddie Murphy in the character. The point in the production when the story really began clicking was when they started writing for Eddie. At first, they were writing for the character and what they felt the needs of the story were. Eddie recorded that, but he was struggling with it. So, he would read it the way we had it and then he would ad lib. And it would always come out so much funnier, whether it was a funny line or not, because he’d put these inflections in it and make these straight lines sound pretty funny. So they went back and rewrote almost all of Mushu’s lines and added some gags and made it more “Eddie-esque.” From there, it really helped the character click a lot, So, early on, we really did start gearing Mushu toward Eddie Murphy” (Lyons 96).

Murphy was later cast as Donkey in Shrek. This character is also “Eddie-esque” and so Donkey’s behaviour and personality very much resemble Mushu. Murphy was cast specifically to bring his own persona to both roles, in the same way he might be cast in live action films and for the same reasons. The audience pays to see Eddie Murphy, so the animators also need to be “Eddie-esque.” They are not just visualizing the emotions and attitudes in the voice track, they are visualizing Eddie Murphy’s persona as much as they can within the confines of the character design.