Bryan and I were guests on the Nerdist Writer’s Panel podcast with Ben Blacker. We talked about the creation of the show, the move to digital, and even about the Movie That Shall Not Be Named. Ben does a great job with the podcast and has had some big names on in the past, like Matthew Weiner, Damon Lindelof, and Vince Gilligan, so I was honored to be one of his guests! It’s a great podcast, with in depth interviews with a lot of showrunners from some of the best series out there. A great resource for aspiring writers and people who just like learning more about how a TV show is made.
In the final part of this series on “Beginnings, Part 1”, we’re going to take a look at the script — essentially the blueprint for all the actors and artists once we move into the production phase.
Once the outline is written and we’ve received notes from Bryan and the network, I usually go over the notes with the writer before they head off to write the first draft. When I write a script, I also have all my other executive producer meetings to do, so often I’ll write for an hour here and there, sometimes at night, or whenever I have a spare moment. This isn’t my ideal way to write, but I make it work.
Because we’ve thoroughly worked through all the story beats in the premise and outline phases, the script writing is more focused on nailing down dialogue and describing the action in a clear, visual way. I find it’s really helpful to have to only focus on those parts, without simultaneously trying to figure out the plot as well. It’s like if you tried to build a house while also drawing the blueprints at the same time — there would be a lot of wasted effort going back and forth between the two.
One technique our writers Josh and Tim taught me is to take the outline and copy it into Final Draft (the scriptwriting program we use). Then you simply format the dialogue and action paragraphs and in no time at all, you have something that resembles a script. It’s much less daunting to start that way than with a blank page (and saves you from retyping things from the outline you may want to use). From there, I go through scene by scene, improving on and adding to the dialogue and action descriptions.
One tip when writing action descriptions, especially for animation (though this would apply to live-action as well): because an artist needs to draw what is written on the page, the action descriptions must be a specific as possible. Good writing always conjures images in the readers imagination, but there is little room (or need) to get flowery in a screenplay, so I make sure to use action verbs and keep the language very concrete. Very rarely do I use adjectives, unless it’s important to the way a character will be animated. Otherwise, adjectives don’t add to the clarity of the description. Here’s an example from the script.Wan shoots wild blasts of fire at the Chous and the guards, but their numbers overwhelm him. Little Chou closes the gap and jumps on Wan’s back. Wan spins around with his arm extended, trying to hit Little Chou, but instead he creates a ring of fire around them. Little Chou falls off Wan’s back, pulling off Wan’s mask in the process. Wan raises his fist. Little Chou cowers and closes his eyes.
There is one adjective in there — “wild” — to describe the way Wan is using fire. In this case I think it’s helpful for the storyboard artists to know how Wan is firebending. Since he’s still unskilled at this point, I went with “wild blasts”.
After about three weeks, the writer turns in the first draft. Usually, it comes in a few pages long, which is fine at this point. We’d rather have more to work with than less. Before we turn it into the network for notes, we once again gather in the writer’s room and do an internal punch up session. As a group, we project the script on a screen, and go through it page by page. We pitch alternate versions of lines and try to make the dialogue as good as we can. We also trim out any lines or action that aren’t strong or critical to the plot, with the aim of getting the page count shorter. Once we are all happy with the rewrite, we then turn it into the network for notes.
Once we get the notes back, we do one last punch up session (which is usually a lot quicker than the first), addressing anything that needs to be tweaked. We look for more cuts at this point as well. The goal is to finish with a 26-28 page script. Shows with a lot of action (like finale episodes) are usually even shorter, around 21 or 22 pages.
We call this script the Record Draft, and this is the version we use to record the actors and which the storyboard artists use to begin visualizing the story. You’ll see numbers next to each dialogue line — these are used to keep track of each line as we record.
Roughly eight weeks later, Bryan, the writers, the supervising producers, the director and assistant director, network executives, and I gather and watch the animatic. The animatic gives us an idea of how the story and action flows. We also get to see how long the episode is. We ship shows to Studio Mir at 23 minutes (for Books 1 & 2) and 22 minutes (for Books 2 & 3).*
Sometimes animatics run long. Like really long. Like 25 or 26 minutes long. Those are always the toughest because we have to find a way to cut them down without compromising the story. Usually it’s a combination of cutting (or speeding up) some of the action and cutting dialogue. I’m always surprised just how much we can trim out without breaking the story. This episode came in at around 24 minutes, about a minute too long. We make these cuts in yet another group writing session and turn in the Animatic Draft, which the directors use to finalize the storyboards before sending the episode to Studio Mir.
For this episode, the first sequence with Korra and the sage became shorter and shorter with each stage of the writing, until only the essential elements were left. Since this was Wan’s story, it made sense to get to him as quickly as possible. There’s also a scene on page 4 during the chase montage, where Wan is disguised as a woman. It was the least funny of the beats and I don’t miss it at all. It was a good cut. There’s also a short spirit wilds montage scene on page 15 that was cut out — again, it didn’t really add anything to the story, so we got rid of it. The Record draft is 28 pages, but the final version we sent to Studio Mir was 26 pages. And sometimes we have to trim out a little bit more when we edit the final episode.
By breaking down the story into the premise, outline, and script phases, it makes the story writing process much more manageable. It’s similar to the animation process. You can’t start with the final, cleaned up animation. You begin with very rough storyboards and build from there. Likewise with the script. It just wouldn’t work to sit down with a blank script page and start writing some dialogue, hoping the story will gel in the process. Working through the premise and outline helps you as the writer figure out what the story will be. And if you want to make any big changes along the way, it’s much easier (and less demoralizing) to rewrite a few paragraphs of description than an entire script.
Here’s the Record Draft of “Beginnings, Part 1”: K207_RECORD_DRAFT_1.16.12
*Due to Nickelodeon shortening our air times, the length of the shows decreased by a minute for Books 3 & 4. At first, I was concerned we wouldn’t be able to fit all the story we wanted into the shorter length, but it ended up working out fine.
If you missed part 1 or 2 of this series you can find them here:
The premise has been written, we’re received notes from the network, and now we’re ready for phase two — the outline!
Legend of Korra outlines are a detailed, beat for beat description and summary of the episode. Like with the initial story ideas, all the writers gather together in the “writing room” (AKA my office) to pitch out the story in further detail. We use the premise as our starting point, and over the course of two days, flesh out the emotional arcs and story beats as well as pitch out dialogue ideas and jokes.
On day one, we spend a good chunk of time working out the entire plot in detail. The premise provides us the story framework, but now is the time to nail down story specifics. In the case of “Beginnings, Part 1”, several of you pointed out how different Wan’s character felt in the premise compared with the final episode. His personality definitely evolved over the course of the writing process, due mainly to the fact that he was a brand new character. Bryan and I always imagined him as a classic trickster hero. In myth, the trickster hero often causes change in his or her world by messing with people and ignoring society’s rules. Our challenge was to keep Wan’s mischievousness and trickster tendencies, without making him too selfish or unsympathetic. We also wanted to give him a specific personality beyond the trickster archetype. With each stage of the writing, we honed in on his more generous side. He because a guy who always stands up for the downtrodden, whether they’re animals, humans, or spirits. In the outline, you’ll see how the beat of Wan giving up his stolen bread to feed the animals is missing. I added that in the script phase to show Wan’s generosity and also set up his connection with animals (which comes into play when he saves Mula from the trap).
A few logistics about how we structure the outline: The show is split up into three acts, and each act has seven to eight story beats. As we discuss the story in the writer’s room and pitch out the beats, the writer will write a sentence or two describing that beat on an index card and tack it to the wall. This is usually done in order, but sometimes we might have some beats figured out at the beginning and the end, but act 2 will be empty. Another way we approach the story structure is to look at the premise and decide what our act breaks will be. The goal of the act break is to have a dramatic moment that turns the story in a different direction (and keeps the audience watching after the commercial break!) We then fill in the missing beats around those tent-pole moments.
With this in mind, we knew that act 1 would have a lot of set-up involved. We had to show Korra in the beginning, establish Wan and his normal life, and see him enact the plan to steal the fire from the lion-turtle. For a while we thought the first act break would be Wan’s banishment, but that would’ve made the first third of the show too long. Therefore, we decided that the act 1 break would be Wan returning his tree house and showing his friends that he stole the fire, a bold action that causes the story to take a new direction in act 2.
Act 2 encompasses Wan using the fire to defend his friends from the Chous, being banished, and surviving in the wilds. The end of act 2 shows Wan saving Mula the cat-deer by standing up to the group of hunters. In act 3, having gained Aye-aye’s trust, Wan lives with the spirits and learns to master his firebending skills. Then he sets off into the world where he encounters Raava and Vaatu. He splits the fighting spirits and learns that his actions have great consequences, setting up “Beginnings, part 2”.
And after a day of discussions, we have all our index cards filled out. This is what the episode looked like, broken into beats.
- A weakened Korra is taken back to Bhanti Village. She wants to regain her memory.
- Shaman tells Korra her spirit is weak – if she doesn’t regain memory, she will grow weaker and die.
- Korra taken to isolation chamber. She must return “to the beginning.”
- Korra goes into sensory deprivation, transition into flashback.
- Wan chased by bullies through city streets – tricks them and gets away.
- Wan meets up with Jaya and Crazy Yao back at hideout. Wan has plan to change things.
- Goes out with hunters. Gets power of fire from lion-turtle, but ditches hunters and sneaks back to city.
- Wan shows Jaya he has firebending.
- Wan demands food from bullies. They attack – he firebends. They run scared but now town is on fire.
- Fire rages. Lion-turtle puts it out, Wan tackled and arrested.
- Wan banished. Lion-turtle takes mercy, lets him keep firebending.
- Wan’s first night in wilds – frightening, mystical. Can’t sleep, keeps getting attacked, ends up filthy and starving.
- Finds oasis, tries to trick spirit guardian but is driven away (pretends he’s a spirit).
- Comes across animal caught in trap. Tries to free it.
- Hunters show up and order Wan to turn over animal. Wan refuses.
- Wan fights hunters using fire and knowledge of the wilds, but he’s overtaken.
- Spirit from oasis saves Wan – tricks hunters into attacking each other – one gets away.
- Spirit brings Wan back to oasis to heal – “you’re different than the others.”
- Montage – intercut Wan’s skills growing with Wan’s legend growing. Wan lives in harmony with the spirits.
- Wan and cat-deer rest – spirits and creatures flee a giant rumbling/chaos. Wan checks it out.
- Dark and Light spirits fight and destroy everything. Wan splits them apart.
- Dark Spirit escapes. Light Spirit tells Wan he has thrown world out of balance.
- Korra twitches violently.
Now that we have all the beats, we spend the second day talking through the whole story, card by card. We pitch ideas for dialogue and jokes during this pass, while our writer’s assistant furiously takes notes of everything we discuss during the course of the day. At the end of day two, the writer takes the cards, all the written notes, and has about a week to write the outline. While it seems like all the heaving lifting was done in the pitch out, writing the outline still takes a lot of thought and skill. Many ideas are discussed over the two days, so it’s the writer’s job to hone all those ideas into the structure we discussed in an entertaining and clear way.
One thing to keep in mind as you’re outlining your own stories is to make sure the main character actively drives the story forward. Often a story will hit a wall or fall flat if the main character simply reacts to events around him or her. In Wan’s case, he steals the fire, saves the cat-deer, and splits Raava and Vaatu — all major story points which coincidentally (or not so coincidentally) correspond with the act breaks. Sure, events happen to Wan as well — he’s bullied, he’s banished, and he’s attacked by various spirits. But each of these events has an effect on Wan and results in him choosing to act in a new way moving forward.
In the outline, you’ll see there is a lot of sample dialogue. I’ll be the first to admit, most of it’s not that good. But that’s not its purpose here. Rather, the outline dialogue is used to get a sense of the kind of thing a character might say, but much of the time the dialogue is very “on the nose” and doesn’t always capture his or her voice, especially in an episode like this with many new characters. Lines and jokes from the outline often make it through to the final script, but they’re usually tweaked along the way. At this point, I don’t worry to much about getting the dialogue exactly right — I save that for the script. The goal of the outline is to give a clear picture of how the story will flow and how the characters will act.
Also in our outlines, there are headers indicating location and time of day. These are the same slug lines that appear in the final script. Their purpose is to indicate each time we are moving to a new location (necessary for the BG designers and storyboard artists to know what locations will be used). And later, the BG painters will reference the times of day in the script so they know whether a BG painting is supposed to be morning, day, night, etc.
As you read the outline, you’ll notice it much more closely resembles the finished episode, story-wise. But there is still some character-finessing to do as well as finalizing all the dialogue.
I find the story pitch out process and outline writing really fun. This is where the vision of what the story will be becomes much clearer and the possibilities (and problems) are more evident.
Click on the link to read the outline: K207_OUTLINE_11.10.11
Next time — the script!
This is the first of a 3-part post about the writing process for A:TLA and Legend of Korra. My goal is to describe the steps we take to bring a story from an initial idea to a finished script. Although this applies specifically to a half-hour animated TV show, I think the principals can be used when approaching any writing project. I also will provide examples of the premise, outline, and scripts from episode 207 “Beginnings, part 1” to show how the story evolved. I want to take some of the mystery out of the writing process. We’re used to only seeing the shiny final product, which seems to appeared fully-realized on TV. I always enjoy examining the process it takes to get to a finished book, movie, or show, so I hope you find this helpful.
Writing stories is a daunting process. My early attempts at creating short stories and scripts always fizzled out after several pages. I’d have what I thought was an intriguing premise or character, I’d set them in motion, expecting narrative fireworks to erupt, then… nothing. I’d put the pages away, always wondering how to go from an idea to a finished story. Jump forward to Avatar, and my first experience working with other writers in a writing room environment where I learned how to break a story down into beats. I’ve now done this process 113 times over the course of the two series. Combined with studying more about story structure over the years, I feel like I finally have somewhat of a handle on the story writing process. It makes it a little less daunting, but it’s still always challenging. I guess that’s what I really love about storytelling — it’s almost like a puzzle and creating new ones always brings up new problems to solve and pushes my creativity and imagination to new places.
Different shows have different ways of “breaking” a story. I honestly don’t know why it’s called this, since you’re actually “fixing” or “finding” a story, but that’s the industry-speak for what goes on in the writer’s room. For each season, the stories all move from very general to specific. In the case of Korra, for each season or “Book”, me, Bryan, and our writers brainstorm general ideas for the whole season – We ask ourselves big questions like: What’s the theme? What’s Korra’s emotional arc? Who’s our villain? We also come up with random ideas for a scene, or a cool action sequence, or even a vague notion of wanting to see a particular character more. So this first stage is about getting everything on the table and seeing what resonates with us. This process usually lasts a few weeks, but it’s an ongoing process through the season, since we don’t figure out every story beat up front. But once we have a sense of the beginning, middle, and end, we move on to the premise phase. Plus, the production moves so quickly, we usually have to start writing before we’ve figured out the entire season.
In the case of shows with continuous storylines like Lost or Breaking Bad, there seems to be a myth that the writers have (or should have) figured out everything from the start. TV just doesn’t work that way, due to the demands of production schedules (unless the show is True Detective, which was written all by one writer before production began). In an interview before the Breaking Bad finale had aired, creator Vince Gilligan was asked:
In interviews last summer you still weren’t sure how Breaking Bad was going to end. Was this just a matter of specifics? Or had you still not decided whether Walt was going to live, die, or go to prison?
“It was everything. We knew very little as of last summer. We knew we had an M60 machine gun in Walt’s trunk that we needed to pay off, and that was about it. We kept asking ourselves, ‘What would satisfy us? A happy ending? A sad ending? Or somewhere in between?’”
He’s talking about the final season and they had a machine gun in Walt’s trunk and weren’t even sure how to pay that off. That’s how these types of shows work. Every season is a new storyline with a new set of problems for the writing room to solve. So the best you can do is have a plan of which direction you’re headed, but realize that you’re going to have to figure out the path a bit on the fly.
As George R.R. Martin has said, regarding Game of Thrones:
I have a broad sense of where the story is going; I know the end, I know the end of the principal characters, and I know the major turning points and events from the books, the climaxes for each book, but I don’t necessarily know each twist and turn along the way. That’s something I discover in the course of writing and that’s what makes writing enjoyable.
Okay, back to the writer’s room.
Once we’ve decided on the major character arc and overall plot of the season, it’s time to tell that story episodically. Enter stage one:
For this part, we take all those ideas we had in the general meetings and make episodic stories out of them. Some episodes come easily, like season premieres — certain things are just going to have to happen, according to the story we want to tell. Other episodes are a little trickier to figure out. The premise discussions are basically a more focused version of the season overview discussions. For me, the episode starts to click when I know the character’s emotional arc for that episode. For example, in Book 1, Bryan and I had a big hole where episode 4 was supposed to go. We knew we needed something there, and had a bunch of plot ideas for it, but nothing that clicked. Then we hit upon the idea of Korra dealing with her fear, after learning of Amon’s power to take bending away, which was her biggest nightmare. Once we knew what Korra would be dealing with emotionally, it was much easier to tailor the plot around that.
When the writing team has a handle on what that story is about, one writer is assigned to an episode and has a few days to write up a 3-4 page premise, including a log line which is the quick pitch of what the episode is about. Our premises are pretty detailed, though as you’ll see in the example, there are still a lot of missing pieces to figure out. Our goal in the premise is to hit all the main plot points and emotional beats. It needs to be clear what this episode is going to be about and what the character is going to go through.
After a first pass of the premise is done, we usually all discuss it further as a group, figuring out any of the problem areas. The writer revises it, then it’s off to the network for notes!
Quick sidebar #2:
Premise is a word that is thrown around a lot in the writing world, and it can have different meanings depending on the context in which it is used. For our purposes here, I’m using it to refer to the actual document that we create, which includes what is traditionally is known as a premise — that one sentence blurb you see on the Netflix or Amazon descriptions. For this episode it was: “In order to regain her memory, Korra must delve deep into the Avatar’s past and learn the truth about the epic and mythic origins of the first Avatar.” What follows is a summary of the story that explains the premise in more detail. Some might consider it a treatment or outline, but as I’ll show next time, our process splits up the premise and outline phases. Every writer and every show will have their own way of getting the ideas on paper.
The link below will take you to the “Beginnings, part 1” premise, written by yours truly, though by now I hope it’s clear that this is hardly a one-man effort. It’s important to have one writer in charge of writing each episode so it has a consistent voice and tone, but as with all Avatar and Korra scripts (and most TV shows), it takes a group effort to create each story.
A few things you’ll notice if you’re familiar with the episode. Korra is much more talkative in the opening of the premise and many of the characters have yet to be included, like the Huntsman, the Chou Brothers, Aye Aye, and Mula (Wan’s trusty cat-deer). On page 3, there’s a mention of a man who has lost his mind from being out in the wilds, which later became Yao, the half-tree man. But even though there is a lot missing and the story still needs to be fleshed out, you can see the main beats of the final episode are all there: Wan being bullied, stealing the fire, being outcast, surviving in the wilds, and splitting Raava and Vaatu.
I realize this is a rough overview of the process, so if anything’s not clear, feel free to add questions in comments and I’ll do my best to answer them.
Next up… THE OUTLINE!