I was recently asked by Chetan, a young animator, about acting in animation so I thought I'd post my thoughts here. This was supposed to be just one post but it's got rather large so I've had to break it down into two. This first one is on subtext in dialogue.
I was introduced to this term while working on The Tale of Despereaux. Put simply, subtext is the underlying meaning of what someone is saying, not what the words actually are. What we interpret as viewers is ALWAYS the subtext and not the text. An example would be the word 'OK', it can mean many things based on context and the way it's delivered.
Text = "OK"
POSSIBLE SUBTEXT =
1. "I'm OK"
2. "I'm OK?"
3. "Are you OK?"
4. "stop Hassling me"
5. "I really don't want to do this but OK - I will do it for you"
6. "I am resolute"
7. "I am pumped let's go!"
8. "I've finished"
9. "I understand you"
Here's a piece of dialogue from the film Joe Versus the Volcano, Joe's boss is on the phone so we only hear one side of the conversation -
Joe Versus The Volcano
here's the transcript -
"I know he can get the job. But can he do the job? Harry. Yeah, Harry. But can he do the job? I know he can get the job. But can he do the job? I'm not arguing that with you. I'm not arguing that with you. I'm not arguing that with you. I'm not arguing that with you' Harry! Harry, Harry. Yeah, Harry, but can he do the job? I know he can get the job. But can he do the job? I'm not arguing that with you. Harry, I am not arguing that with you! Who said that? I didn't say that. If I said that, I would have been wrong. Maybe. Maybe. I'm not arguing that with you! Yeah, Harry, I know he can get the job. But can he do the job? I'm not arguing that with you! I am not arguing that with you! I am not arguing that with you! Who told you that? No! I told you that! Me! What? Maybe. Maybe, maybe. Maybe!"
The character basically repeats the same three or four things, yet it sounds like a real conversation doesn't? How can someone have a conversation when all they say is the same couple of sentences over and over again? Because of the subtext, what the character means is not what he's saying, take the phrase "I'm not arguing that with you", at various points in the conversation it means different things -
Text = "I'm not arguing that with you"
"That's beside the point"
"Stop repeating yourself"
"Listen to me"
"Please answer the question"
"I don't want to talk about this anymore"
Although, this is obviously open to interpretation, there is no doubt that it feels like he is trying to move the conversation forward by changing the subtext of what he is saying. The writer played around with this idea, and there are many scenes in the film where the characters would repeat the same line with different sub textual meaning.
Some Like It Hot is a famously good script and it is riddled with conversations with no textual meaning but with layers of great sub textual content. Here's a clip -
Watch at around 2.05.
Here's the text from part of the conversation between Marilyn Monroe and Tony Curtis' characters -
"So who are you?"
"Haven't I seen you somewhere before?"
"not very likely"
"You staying at the hotel?"
"Not at all"
"Your face is familiar"
The characters words aren't saying much but there is definitely a lot going on in these word. The first layer of subtext is that she is trying to talk to him and he is acting aloof. But under the surface, there is a further layer, as we also know that he is pretending to act this way as he knows exactly who she is and what he has to do to attract her.
On one level his subtext is -
"go away, leave me alone"
But underneath this he is really saying
"I'm intriguing, I'm famous and handsome, talk to me."
Very clever stuff. The reason this scene is so entertaining is because of the subtext, which is clear in the body language and delivery of the dialogue, and that's what we the audience read, not the spoken words.
So how does this effect animation? Well this is a very important, we should ALWAYS strive to seek and animate the subtext, and the subtext should ALWAYS be different to the text. This is so important that for non dialogue shots, Randy Cook would record the unspoken dialogue (subtext) of a shot for his animators to animate to in their scenes.
Here's an hypothetical scene - A girl appears from a fitting room and shows a new dress she's wearing to her boyfriend, she asks his opinion, he says "yeah, it's great".
If we have his body language mimicking that thought, (it's great) the animation could be fantastic, but the idea isn't very interesting. It's what we call on-the-nose acting, and is just that, acting the words without any deeper thought as to their meaning or what the character is thinking. If the boyfriend says "yeah, it's great", yet his body language says "I'm bored, can we go home now?" we'll have gotten inside the character and given him some depth and therefore created something much more interesting and entertaining.