On Friday I attended a Pixar Story Masterclass with Pixar's head of story, Matthew Luhn, who's been with the company since the original Toy Story. There are only three of these seminars scheduled for North America for the whole of 2011: Montreal, New York City, and Austin. And as if the odds of the class coming to my city weren't already low enough, it just so happened that the venue -- the amazing Long Center for the Performing Arts -- is literally just a few blocks from my apartment.
I think this is the kind of thing storytellers tend to refer to as "fate". :P
This is a writeup of that all-day event, in the same vein as my GDC writeups from the past couple of years.
The "controlling idea" is central to the story. Also referred to as a logline, elevator pitch, or high concept, this is a one-sentence description of the story. You should come up with this first, and nail it. Everything else about the story emerges from this controlling idea.
Here are some examples. Note how easily you can identify the following movies:
- A journey of self-discovery by a brilliant mathematician once he was diagnosed with schizophrenia. He eventually triumphs over tragedy and receives the Nobel Prize.
- An Epic tale of a 1940s New York Mafia family and their struggle to protect their empire, as the leadership switches from the father to his youngest son.
- A meek and alienated little boy finds a stranded extraterrestrial and has find the courage to defy authorities to help the alien return to its home planet.
Once you've nailed your controlling idea, you can start looking at the high-level structure of the story. Classical story structure, on which the vast majority of stories are built, looks like this:
- Inciting incident
You've probably seen this before. But one thing that jumped out at me is the crisis/climax pairing. Previously I had always thought of structure in simpler terms: exposition, rising action, climax, resolution. But crisis/climax is important: the crisis is the moment at which the hero is compelled to make the choice that will change him forever, and the climax is the action that occurs as the result of that choice. Typically, one of these is an "upper" moment while the other is a "downer".
Our brains are hard-wired to think about stories in this structure, because this structure is how we give meaning to our life experiences. It makes more sense when you think of it in terms of the "story spline", which is essentially a structured plot prompt:
- Once upon a time...
- And every day...
- Until one day...
- And because of that...
- And because of that...
- And because of that...
- Until finally...
- And since that day...
- The moral of the story is...
Luhn illustrated the power of the story spline with the help of an audience volunteer and a quick improv segment. The rules were simple: each player fills the answer to the next point on the story spline, with the first thing that comes to mind. Here's what they came up with, in about 60 seconds:
- Once upon a time... there was an armadillo.
- And every day... he rolled across the desert to get to work.
- Until one day... they built a highway intersecting his commute.
- And because of that... he had to find an alternate route to work.
- And because of that... he ended up rolling through a cactus field and meeting a bunch of hippie roadrunners.
- And because of that... he joined the roadrunners and became a drifter.
- Until finally... he realized the value of having a steady job, and returned to work.
- And since that day... he's livened up his job with the experiences he had with the roadrunners.
- The moral of the story is... don't take life too seriously.
This is far from a perfect story, of course, but the point of the exercise is to show the power of structure. Even though the details of this story don't make a lot of sense, we understand it as a story because it fits the structure. Understanding structure, and working within its constraints, frees your creative mind to come up with the details without worrying about inadvertently ruining the story.
Characters must have identifiable human traits. The story is not just a stack of plot points that the character climbs, but rather the sequence of answers the characters has to the challenges in his path. A well-defined character is much easier to build the story around.
When creating a character, consider:
- What does he fear?
- What are his strengths? (Traits, not talents.)
- What are his weaknesses? (What do others say about him behind his back?)
- What is his dark side, the very worst thing he's capable of doing?
- What traits get him in trouble?
- What traits does he admire in others?
Fear is arguably the most important thing to know about a character, because fears (or deeply-rooted passions) drive his choices. One of the best ways you can learn what makes someone tick is to understand what they fear.
Strengths and weaknesses tell you how the character will react to any situation. Characters play to their strengths and their opponents exploit their weaknesses. The "dark side" hints at the protagonist's behavior at either the crisis or climax moment in your story: these are the moments when the protagonist is pushed to his very limit.
Traits that get your character in trouble are great jumping-off points for coming up with the successive complications of the story. Traits he admires in others tell you what other characters the story needs.
Most characters go through an arc, meaning they undergo some kind of change as they move through the story. You can start building an arc by asking these questions:
- Who is the character?
- What does he/she want?
- How does he/she take that want to an unhealthy level?
- How does he/she realize it's unhealthy?
- How does he/she change?
The answers to these questions are driven by the character's fears, strengths, weaknesses, dark side, trouble traits, and admired traits. A character who fears loneliness wants friends; one who fears obscurity wants fame and recongition; one who fears poverty wants money. Weaknesses and trouble traits show you how the character will take his want to an unhealthy level: a character who is obsessive pursues his want at the cost of those around him, and one who is short-tempered may drive away the friendship and acceptance he seeks. The character's "dark side" can inform the moment or means by which he realizes that his pursuit has become unhealthy, and the change - the culmination of the arc - is the lesson learned, or (in tragic stories) the failure to learn.
This approach to character development helps you develop characters that seem human, and that audiences can relate to. One of the most common mistakes writers make is to build characters by simply copying or compositing characters from other stories. Draw from personal experience when developing your characters; otherwise they're likely to seem cliche.
To pull it all together, Luhn had the audience think about their childhood bully (everybody had one) and describe him/her in terms of the above questions. Here's mine:
- Fears: Not being the best.
- Strengths: Confident, sociable.
- Weaknesses: Poor sport, short temper.
- Dark side: Violence and aggression.
- Trouble traits: Short-tempered, can't cope with defeat.
- Admires: Intelligence.
That only took about a minute to work out: the point of focusing on a childhood bully, rather than inventing a character from scratch, is that everybody vividly remembers that real person and thus can draw from life very quickly. Note that you don't know who my bully is: you don't have a name, a physical description, even an age at which I knew this person. But what you do have already is a pretty good idea of who this character might be, and how he might act in different situations.
Which, incidentally, is exactly where Luhn took it next, asking the audience to consider how their bully would react to four different situations. Here's mine:
- What if his power is threatened? He responds with aggression, an "alpha-male" display.
- What if his bike is stolen? He rounds up his friends and goes to find and beat up the thief.
- What if he gets in trouble with his teacher? He jokes around, plays it off like it's no big deal.
- What if he's forgiven for wrongdoing? He withdraws, acts uncomfortable and irritated.
Those four situations represent a pretty broad spread, but I was able to put my bully in each of them and very easily imagine how he'd react. Once you really understand your character's fears, his strengths and weaknesses, his dark side, his trouble traits, and the things he admires in others, his reaction to any situation becomes all but obvious.
This part of the class was the most directly related to animation and film, as opposed to general storytelling, but the concepts presented are easily applied to the development of scenes in any medium.
Sequences are distinct story points, containing a point of conflict and a progression of intensity. In a film, they're usually a few minutes long. Monsters, Inc. had 31 sequences, ranging from about a minute to just over five. A sequence may consist of numerous beats, but every one is driving toward the sequence's main story point.
Developing a sequence is 80% thinking and 20% drawing storyboards. When working out the beats, ask lots of questions. What should the audience feel during this sequence? What are the characters' intentions? Get as much information from the director as possible!
Break sequences down into beats: individual, brief moments that generally last just a few seconds each. When brainstorming beats, ask yourself "What is this sequence about?" Answer in broad terms: it's about "giraffes dancing", or "cowboys doing laundry". Don't get too specific. "Wild Bill Hickock doing laundry" limits you to only those ideas which make sense for that specific cowboy. If your sequence is about "cowboys doing laundry", jot down everything that comes to mind that might happen in that context. The specific beats you choose aren't nearly as important as supporting the sequence's overall plot point and delivering on its premise.
Once you've selected your beats, identify which are the most intense and which are the least intense. Intense beats don't have to just be physical action: they can also be the divulgence of critical information, moments of intense emotion (either positive or negative), and so on. Sequences should have variation and progression of intensity. You can increase the intensity of a beat by using value contrast (dark-on-light or vice-versa), diagonal lines, and closer shots. You can reduce intensity with flatter shading, horizontal lines, and wider shots.
Horizontal lines are passive, stable, low-intensity. This is called "flat staging". Comedy (especially TV sitcoms) works best with flat staging, but otherwise you'll want to avoid it unless you have a very good reason for using it, because it's not very interesting. Vertical lines are medial and aware. These are good setups for action. Diagonal lines are active and off-balance, best for your most intense beats. You can get these with upshots (looking up from a low position) or downshots (looking down from a high position), as well as with dutch (tilting the camera left or right), and with clever use of perspective and motion in the frame.
Lower the horizon for more dynamic shots. A common mistake is to put the horizon right through the vertical center of the frame, which looks boring and flat. The lower the horizon is in the frame, the more dynamic the shot will feel. You can even push the horizon off the bottom of the frame entirely, which typically results in an upshot and strong, active, diagonal lines.
Symmetry generally works only when you're showing something regal or important: a king on his throne, the Cross, etc. Most of the time you'll want to keep characters and focal points to one side of the frame or the other. This is related to the "Rule of Thirds" in composition, where you divide the frame like a Tic-Tac-Toe board and put points of interest on one of the axes or at intersection points, rather than in the center of the shot.
Consider shot economy. Don't go to a new shot unless you must, and have a very clear motivation (such as cutting to a different speaker). Zooming in and out do not count as new shots. What you're really looking at is staging: where is the camera relative to the subject? If the camera isn't moving or rotating, you're still in the same shot.
If you're changing shots too frequently, consider combining shots. For example, can you move the action in shot B into the background of shot A, or line up A and B and pan across instead of cutting to a new shot?
With moving shots (and moving characters), be sure to preserve screen direction. If a character is moving from left to right in one shot, and then you cut to a new shot, he'd better still be moving generally left to right unless we saw the character explicitly stop and change direction. If you break this rule, the audience gets confused and doesn't understand the layout of the scene. Similarly, if a character is facing left, keep that character facing generally to the left from throughout the sequence; and if he's on the right side of the screen, keep him generally on the right.
Breaking the above rule is called "breaking the line". Imagine a line that bisects your scene: it doesn't matter where, though if you have multiple actors onscreen it will generally run between them. The camera must stay on one side of the line throughout the entire sequence in order for the audience to understand what's going on. If the camera crosses the line, we immediately lose our sense of spatial awareness. You can move and rotate the camera all you want, just as long as it stays on one side of the line.
While there were a lot of things in this class that I already knew, the real value of classes like these - for me, at least - is that they provide a different perspective on that knowledge. It's like I have all the pieces of the puzzle on the table in front of me but I don't necessarily know what order they go in, and then I go to a class like this and it's like, "Oh! You put these over there and this up here and then it makes sense!"
I've been vaguely aware of a weakness in my approach to story development for a while now, which is that I tend to try to build plot first and then add characters to it. I had read that the best stories are character-driven but for some reason that never fully clicked, until this class. What did it for me was seeing, and thinking through, the specific applied examples I outlined above, and realizing how many ideas just "fell out" once I had answered the right questions, like "What does this person fear?" I'm not thinking of story development as brute-forcing my way through the structure any more. Now, I'm thinking about it as a series of simple questions, designed such that their answers lead naturally to the right next questions.
As a game designer, it's only natural that I think about ways to apply Pixar's story development approach to games. It seems obvious when you think about applying it to game stories, but what about applying it to game design in general? What kinds of questions can we design, that lead naturally to the right next questions? I feel like we often start new game designs by asking: "What's the core mechanic?" Does that answer lead us in the right direction? How often does it lead us to the game's central aesthetic, to the identity of the player/avatar, or to the mood and tone of the game? Would we be better off if we started somewhere else, somewhere like "What life experience is this game about?"
Miyamoto has often said he takes game design inspiration from simple life experiences like gardening (which led to Pikmin) or managing his health (which led to Wii Fit). In this 2008 USA Today article (thanks, Google!) he said: "When I design games, what I am trying to do is find a way to take something that is fun or entertaining from something I have experienced and to bring that to other people so they can experience that same degree of joy." Is that perhaps a more Pixar-like approach to game design? It certainly doesn't seem to track with conventional wisdom - where does the ubiquitous first-person shooter come from, if we all design games based on our life experiences? - but this is Miyamoto we're talking about, and also Pixar, and I can't easily dismiss that.
The class left me with a lot of unresolved thoughts. Maybe those will turn into an article or two sometime in the future, but in the meantime, what do you think?Posted In: