Eric Lindstrom, creative director of Tomb Raider: Underworld at Crystal Dynamics, delivered an impassioned and even frustrated presentation about the need for more nuanced storytelling techniques and roles within the development team, to create satisfying next-gen game stories.
Setting up his talk, Lindstrom says, "I am a creative director which means I don't actually do anything -- what that means I get a lot of other people to do things. I am a designer, and have been for many years."
This is the understanding and leveraging of how to use all of the techniques available to make games more meaningful -- from game design to cinematic techniques. And Lindstrom's argument is that the understanding and effective us of the latter must improve.
The Challenges Facing Tomb Raider
He framed his talk around the challenges he and the team are facing telling the story of this fall's upcoming Tomb Raider: Underworld (being careful to disclaim that they might not solve all of these problems in that release -- but that they are working their hardest to do so.)
"The first storytelling challenge [of Tomb Raider] is an obvious one: the original characterizations of Lara Croft and tomb raiding were not deep enough to sustain many games without change. It was good enough for a couple of games, but after that it got pretty thin."
Lindstrom contrasts this against James Bond, which does not have this problem because he can battle a new criminal mastermind in each film -- giving the franchise a strong character impetus. "Lara is different -- she was going out and finding an artifact." Though even the first few games had stories with human antagonists, "after a couple of games it became obvious what that template was."
When Crystal Dynamics inherited the franchise from prior series developer Core, the developers chose to change her back story, by having Lara accidentally kill her mother -- which affected the story of Lara's father and gave Lara a motivation: to find and deactivate these dangerous ancient devices.
It turns out that fans disagree with whether Lara is a hero or an anti-hero. Lindstrom polled the room -- half the room thinks she's a hero -- but several also went for anti-hero, which actually surprised Lindstrom -- he chalked it up to game professionals being more aware of the anti-hero. "It's kind of a trick question, because Hollywood has trained us into [confusing] what hero and anti-hero mean."
Hollywood, Lindstrom thinks, complicates the definition of hero and anti-hero by presenting anti-heroic actions in a heroic context -- like James Bond killing essentially defenseless soldiers. "What we have are these rules which advocate violence if it's in service of a noble cause."
Lindstrom presented the description of Lara from the original game's design document -- version 1.3, dated March 3, 1995. "Lara Cruz [not Croft] is a modern day adventurer and procure of rare artifacts which she relieves from Johnny Foreigner with the gay abandon of a five-year-old stealing Mars Bars from the local cornershop."
The point? "She was originally designed as a psychopath... in the original design she was in some ways a criminal, thief, and murderer. But the truth is irrelevant -- because so many people know and love Lara Croft." The Hollywood movies have made her even more positive -- which has complicated things for the design team.
There are thus two audiences for the character, one which believes her to be a hero and the other which believes her to be an ant-hero, says Lindstrom, "which need to be reconciled. So what we do is help people think what they want to think. We depict Lara doing harsh things, what the anti-hero crowd wants, but we depict it in Hollywood ways that make it palatable to the hero faction."
The back story introduced in Legend -- where her mother may not have actually died, and decides to investigate this and potentially rescue her mom -- was appealing to the hero crowd, not the anti-hero crowd, it's worth noting.
The "Inflammatory" Storytelling Technique
The term Lindstrom used -- "inflammatory" -- is a response to his obvious frustration with the lack of forward motion in storytelling quality in games. "Storytelling methods really still have a long way to go."
Lindstrom doesn't mean to tackle the problem of innovative interactive storytelling -- because others are working on that. He's talking about using the basic tools that have been used in other media, which "are not being used at all, or not being used effectively, and there's no reason why they can't."
Here's his mini-manifesto:
Stop saying that storytelling is less important than game mechanics. "There are lots of people who say this, but they don't really mean it."
Start putting storytelling on par with other pillars of game creation. "There are plenty of people out there who say this is true, but when push comes to shove, it's not true."
Stop hiding behind the word "interacrtive". "If there's really one thing to take away today -- it's that 'oh, but it's interactive' is used as an excuse for bad storytelling all the time, and it just doesn't wash."
Start training and employing storytelling experts. "Hollywood knows how to write dialogue more than anybody in the industry on average. The last 10 movies I saw, seven of them had pretty crappy dialogue -- so it's not going to be perfect on average. But you're going to find more people who understand storytelling." It's worth noting that Lindstrom is pro-training on all fronts, not just hiring from outside the industry.
Advanced Storytelling Techniques: Tomb Raider's Choices
As a creative director, Lindstrom clarifies, "My role is to make sure [all of the following techniques] are understood and can be included in the final product. Really, it all comes down to details. Because details matter -- all of them matter. It doesn't mean that every game has to be a big story game... but they all matter."
Content. "To emphasize drama and avoid stereotypes." A major issue right now? "First, don't think of stories as screenplays -- that's why most people get locked up by the interactive problem. If you stop thinking of stories as screenplays, you will stop asking writers to give you screenplays, and will stop looking at them and saying they're unusable. Stop asking for screenplays -- ask for treatments."
Compelling characters. "Imagine if we created Lara Croft using stereotypes -- she would have a generic Hollywood starlet voice, be a sex kitten, and speak in innuendo." However, Lara speaks in a witty, smart, and meaningful manner, is attractive but not overtly sexual, and speaks in an upper-class British accent.
If you build your character out of clichés, Lindstrom says, "It's very easy to stop seeing the top line and start seeing these little bits. Don't fool yourself. Look at the top line. What are people going to see? Don't talk yourself in to what you really have." The component elements of the character are not the character, and one mitigating positive quality does not rescue you from cliché.
Memorable settings. "Lara's [settings]... are memorable and consistent. An experiment that was made that was less successful was Tomb Raider: Angel of Darkness. It was in Paris, a large city... yet the city was unpopulated, so the credibility was largely zero. It didn't function credibly as a setting for Lara Croft, or even as Paris. Don't fool yourself."
Unusual Circumstances. Here he discusses the concept of the log line -- the basic summation of the story. He started with Tomb Raider: Underworld's, which isn't predictable. Then he delivered a log line for a sci-fi shooter -- "A soldier finishes basic training and fights a series of pitched battles to stop the waves of invading aliens." He wouldn't divulge which actual game it was from, but notes, "I didn't make this log line generic just so I could talk about 30 games."
Structure Matters. Lindstrom again takes cues from Hollywood, and illustrates a salient point. "Games are about progressive growth and success, but stories are about sudden change and failure. I've had a lot of argument about this over the years -- that people don't want failure in their stories because game players don't want to fail. But look at any action movie, they're about failure. Look at John McClane in Die Hard. He's running away the whole time." How do you make that palatable to the player? "The player can succeed even when the character fails -- it's about how you frame the action."
The three-act heroic structure -- an exceedingly common framing structure for Hollywood films and other fiction -- was discussed briefly -- and almost surprisingly, both Portal and BioShock fit this structure extremely well.
In contrast, Lindstrom notes, "If your story is that a nuclear bomb is going to go off and it's fight, fight, fight, fight, fight, and then at the end you cut the red wire, that's not a story, it's a framework."
Size matters. "Especially in video games. Just because your game is 20 hours long doesn't mean you have a 20-hour story. Imagine what Lord of the Rings would have been if you multiplied it by two. People just don't remember that much -- especially in the context of interactive, as they play a game."
Lindstrom brings up an excellent point about storytelling in games: "People don't play in one session -- they play over two days, two weeks, two months. They will forget. So it's important to build in reminders. You don't want a lot of exposition, but when you have dialogue where the characters don't use each other's names, you don't know who anybody's talking about. It takes a good writer to thread in a lot of information that, in a movie, people would not have forgotten -- because it was 15 minutes ago."
Interactivity matters. "It does matter, obviously -- but I did want to talk about how it matters with respect to what I'm talking about. We sympathize with characters we see, we associate with characters we control." A story about a weak character who becomes strong by the end -- that's "classic Hollywood" but it won't work in games, says Lindstrom.
"Movie characters can behave in ways game characters would not. What matters is whether you believe people are controlling their character like a puppet master or if they are inhabiting their character, in their shoes -- it changes the way you write."
Don't injure your game progression for story. For example, a movie might bring a character down to nothing -- and have him rebuild. But in a game, that would translate to taking away the main character's full equipment and you have to build everything up again -- which is no fun in a medium that's built around consistent progression.
Mood matters. "Mood matters a lot -- every moment has tone, mood, and emotional goals and you need to know what they are all the time, and support them all the time. Score, audio design, lighting, camera angle, color correction and other full screen effects, and weather... all of these affect the mood. You need to know how to use them," Lindstrom says, and clearly believes strongly. "As a storyteller, if I know you're not going to save the girl at the end of the scene, the music needs to foreshadow that."
To explain more, Lindstrom illustrated how Underground starts with Lara on her yacht, on a nice sunny day. "It has to be nice, to get bad later." As the plot turns more serious, all of these techniques are used to change the mood of the game: the game darkens, color correction changes from bright to dim, and the soundtrack gets more foreboding, among other things.
"Even later, things are getting really bad -- the score becomes creepy and dangerous. The lighting becomes drab and washed out, and we get more extreme and claustrophobic camera angles." Lindstrom believes you must employ a cinematographer -- either hire a Hollywood guy or train one of your people, an intensive process either way -- "but it doesn't just happen" typically.
Quality matters. "Too often story design and written dialogue get lumped into the same discipline and they're not. That's why you get story credits and writing credits in Hollywood. Story design is about compelling interaction, it's about beginning and middle and end." In fact, Lindstrom believes that many roles should be more specialized. "I originally had vocal performance broken out into talent and direction -- but every [facet of production] should be broken into talent and direction."
"It looks simple but it's really not -- understanding what your audience wants. It's a very important part of not only your basic story construction, but when you are dealing with the interactivity of your story." And the bottom line is, Lindstrom says -- as you would for any other discipline -- "Employ experts or make experts -- because that really matters."
A question was raised about whether back stories are really deeply necessary when creating characters for games. Lindstrom is not so sure, "I don't think that back story is a very important piece because it's hard to deliver in games... it leads to exposition, which is bad. What happened with Lara is that I don't think people thought we would be making nine games."
"When she started off, she was adequate... but when you start to get past Tomb Raider 3 and 4... back story was an imperfect solution, among many imperfect solutions. We could have completely redesigned her character, but we chose not to do that," which Lindstrom notes is the route Prince of Persia recently took.