Chris Avellone On Writing & Making Game Stories Engaging

Please introduce yourself. What companies have you worked for? What projects have you contributed to?

I'm Chris Avellone, and I've been a junior designer, animation producer, area designer, systems designer, script doctor, script surgeon, narrative designer, writer, lead writer, creative director, project director, and business owner. I've had the good fortune to work on a number of projects – Planescape: Torment, FTL: Faster Than Light (for free, because I loved it), Divinity Original Sin 2, Dying Light 2, Star Wars Knights of the Old Republic II: The Sith Lords (sensing a pattern?), Baldur's Gate: Dark Alliance (but not the main Baldur's Gate series, or 3), Prey, Fallout 2, Fallout: New Vegas and most of the DLCs, Alpha Protocol, Into the Breach, Judas, Weird West, Tomb Raider, Horizon: Forbidden West (script consultant, minor), and Star Wars Jedi: Fallen Order, among others.

I'm usually known more for narrative design than other disciplines, but I enjoy area design and system design very much – the area design allows me to use some elements of my Architecture minor, but for the purposes of creating fantasy locales.

I'm currently working on four projects, but the main project is a fantasy RPG, working with Adam Williams at Republic Games. I'm enjoying it more than I think I've ever enjoyed working on a project.

I've also been a comic book writer (mostly Star Wars), an (awful) actor in Fallout: Nuka-Break, written some bad short stories, and a movie script writer (Legend of Grimrock). I also foster kittens and cats whenever I can.

How and when did you start working on stories for games? What was your first project?

I started as a pen-and-paper game master for D&D, Superworld, and Champions, and then began taking the adventures I designed and submitting them to various companies for consideration. This resulted in a lot of rejections (even from Monte Cook, who helped design the Planescape setting Torment was eventually made from), but eventually, a company called Hero Games gave me a chance.

After I'd gotten published once, I found it was easier and easier to write articles and modules, but it didn't pay very well. Realizing I was in a financial crunch, one of my editors at Hero Games recommended me to Interplay Entertainment's Dragonplay division as a junior designer. 

That said, my first video game project wasn't at Interplay, it was a "random encounter supplement" for Hero Games' Dark Champions called Hudson City Blues. The first project at Interplay I got a chance to work on was Conquest of the New World with Scott Bennie. Scott was one of the designers on Fallout 1, and he has since passed – he was a great guy and gave me my first place to live when I came out to California. On Conquest, we were writing historical snippets to flesh out the flavor of the game, and it was one of a few projects I contributed to in the first few years (Star Trek: Starfleet Academy, Descent to Undermountain, Die by the Sword) until I moved on to Fallout 2 and Planescape: Torment full-time.

I definitely learned a lot on each project, but it wasn’t until Fallout 2 and Torment that I got to finally do in-depth narrative design, dialogues, and reactivity.

What is the first step when writing a game story? Do you work with game leads or concept artists to build a common understanding? What helps you find the first brick? Perhaps you could share an example or two.

Whenever doing game writing, I fall back on two things at the outset before writing:

1. What's the "vision" and "emotional vision" for the game (e.g., Resident Evil is survival horror with an oppressive/anxious feel, Fallen Order is a third-person action title capturing the space opera feel of Star Wars) and…
2. The basic game mechanics (Fallen Order had a certain camera and power set for the main character, while Resident Evil used to have fixed camera angles, heavy use of sound design, a constrained form of combat that contributed to tension, a certain layout to fodder enemies, and so on). Narrative is intended to support game mechanics, but you need to know what the mechanics are in order to support them.

After this, you evaluate the mechanisms for how the story will be conveyed: is it cinematic, is it menu-driven talking heads, or is it devoid of all conversations and solely relies on environmental storytelling? You also want to be conscious of the budget: If you only have the budget for X lines of voice dialogue, or 10 minutes of cinematics, those are very important "design bookends" to know before deciding how you're going to flesh out the story and what kind of story you can tell with those limits.

You also need to start from the fact that it isn't just your story. It's also the player's, and it's also the team's: You want the team to have ownership of the game elements they're creating, and the chance to create their own visual and written narratives as long as they fall within the bookends above. I've found that giving designers their own space (whether key characters, certain areas of the game, or a faction) gives them more freedom and more investment in what they're doing vs. continually shifting those same elements across different designers every few months.

Once you have the basic game story premise down, you ideally present it to all the leads before anyone else, get their take on it, and make sure that at the end of the meeting, they can "speak" to the story if asked by someone in their department. As an example, if an animator comes to the lead animator and asks for information on the Think Tank in Old World Blues, the lead animator should be able to give a brief explanation of what they are all about, and also point that same developer to key people involved in that story aspect – say, Brian Menze for the foundation concepts and inspirations, then the level designers or I for any quirks each individual Think Tank member may have, and so on.

Also, one thing that I've found helpful – decide what's important to you and what's not, and if it's not key to the narrative, consider letting other team members explore an idea on their own. For example, in Knights of the Old Republic II, there were some companions and antagonist concepts I was very invested in, but there were others where the concept artist had more freedom to explore the characters' visuals the way they wanted, as long as it met the basic criteria of what the character was supposed to do.

So after you've run the narrative premise by the leads, got their approval/included their suggestions/did a revision, etc. then you present it to the whole team, using as many assets people have already created (for example, if Brian Menze had done some concepts of the Ghost People for Dead Money, I would include those concept art pieces in the story/game presentation for the DLC). People like to see how their work is being used to support the game's theme and story, and the more concrete examples you can use of people's work, the better.

After this story presentation (which you should record so you can play it for others coming on to the project later), then ideally, each team member should have a grasp of the vision for the DLC, what mechanics we're focusing on, what gear is to be found, what the story is, and who the key characters and antagonists are. We did story presentations like these as standard practice for most of the Fallout New Vegas DLCs, and it seemed to work well.

Let's there's a huge open-world game – Dying Light 2, for example – that needs lore. How do you structure this task? Do you start with the main story and then develop the world elements as branches? What are the main challenges?

Dying Light 2's probably both a good and bad example because the lore felt like it kept getting revised monthly. It's difficult to build characters and lore for a world in flux, but it's still possible as long as you can rely on the system mechanics being solid enough to build a foundation on. So, using the list I mentioned before, you can apply that to the lore as well.

Start with the emotional feel of the game: Survival horror, zombie-killing, a sense of humanity's survival teetering on the edge. This means the lore should emphasize a lack of safety, questions about "at what cost survival," and designing a game space to emphasize this.

Start with what key mechanics DL2 had: Zombie-killing, parkour, and "reactivity." Zombie killing is relatively easy, but you still need to make sure they are present as a threat and some flexibility in that threat, so there's always a new source of enemies and tactics. Parkour also informs the lore and game space – a city works far better as a game space than an open plain for parkour.

Now, on the reactivity element: I love designing reactivity in characters, factions, companions, locations, and anything that allows the player to make a mark on the world. So the design of the lore was to create factions that could change over time, had different agendas, had different approaches to the "survival" theme, and the setting was designed to allow for the introduction of new factions and enemies at points in the game based on clear, player-inspired actions.

That said, none of these designs made it into the game. I recall doing several treatments for a branching, reactive story, but none of them seemed to land, and it became apparent that unless the idea came from particular leads, it wasn't going to gain any traction. Ironically, some of those same leads expressed similar frustrations with trying to build lore and the story, to the point that at least one of them quit the lead role in the later stages (this was after I left). The lore process was also compounded by the fact that there was a rotating queue of writers that seemed to come onto the project and vanish. While I lasted the longest, it made keeping a consistent lore and tone approach difficult.

Techland and the Dying Light 2 team had great developers in the trenches, some of whom I've worked with on later projects and others who are friends to this day. That said, due to the other difficulties mentioned, I wouldn't work with Techland again.

Do you prefer working on large game worlds with tons of characters or on more character-focused experiences where side characters are kind of helpers that develop main characters? What are the main differences?

In a perfect world, either one is fine. In a budget and constraint-driven world, however (like any business), I tend to find a smaller, more reactive cast is better, if only because those characters are forced to have considerable depth. And the more reactivity a limited cast can respond to, in detail, can result in story moments that shine.

In a world with thousands of characters, that's still achievable (example: Fallout 2, New Vegas), but the cost is considerable, as evidenced by having four voice-over studios working at once to record VO for New Vegas (and while I did write for New Vegas, it was nowhere near as much as others). This likely resulted in Bethesda telling us, "Because of that, you only have 10K voiced lines in all the DLCs."

Now, the good thing with that is it meant we were forced to have a small cast for the DLCs, which caters to how I prefer to design narrative anyway, so that was fine (I also liked it in Prey).

You've worked on so many games and stories. How difficult is it to find something fresh each time? What are the tricks to creating something that doesn't feel too familiar?

It helps that the writing is in different franchises, as each franchise has its own set of narrative bookends that influence the writing. There might be challenges put forth by the antagonists that can only work in certain settings, for example: The Transcendent One's attempt to sever itself from you in Planescape: Torment and the resultant damage to you as the protagonist wouldn't have worked in Knights of the Old Republic II, and Kreia's view on the Force as her agenda would have been similarly out of place in Planescape – the franchise helps dictate the stories you tell.

Also, if your themes aren't fresh, then that can be an issue. I usually find each project has a theme they're hitting either on purpose or not, and usually, there's a new "song" in each theme that's interesting to explore, and that helps as well. For example, in Prey, representing Morgan's isolation being reflected back at him by the chief engineers struggle was intentional, and seeing the "damage" that your relationship with your brother had caused to various NPCs, in audio logs, also was intended to showcase that even Morgan's minor actions in the past had serious consequences on others... and showcasing that one may believe their actions are isolated, serious ripple effects can occur.

A lot of games have stories that feel like fillers – they give a starting point and some milestones, but don't keep players engaged. Why do you think that happens? What can be done to solve the issue?

I feel your pain. This can happen in a variety of ways. If you don't make areas and characters integral to the location and the plot, if the quests are killing forward momentum vs. supporting it, then it's hard to keep the players engaged beyond leveling mechanics. As an example, if you're given a task to save the world from a tidal wave of spirit armies beyond the Wall of Bells that encircles the north, but the very next town you go to they are having troubles with the rats in the granary, that definitely feels like filler and a disconnect in the importance of your role in the game as well.

Narrative designers, area designers, system designers, area designers, character modelers... hell, the whole section of the team involved in the design should all be working together and be on the same page. Audio engineers can do a ton of story support with background sound effects, ambient noise, and music stingers – often far more effectively than a writer can by themselves. An area designer's prop placement, which tends to shine in Bethesda games, can also tell stories all by itself. Enlist everyone to help with the story when you can.

So not only should everyone be involved, there should be a reality check on the content and a reality check on the player's approaches to the game. As a developer, if the content sounds like filler to you, if it sounds like a detour, if it sounds like a diversion, the player will probably see it the same way. Put yourself in their boots, and ask yourself, "Why should the player care?" (I usually say, "Why should the player give a sh*t," but the intent of the question is the same) If you don't have a good answer to that question, do another pass of the original idea and see if there's a way to involve the lore, the plot, or (best case) the player themselves in the area and the quest.

In the initial drafts of Tyranny, we tried to solve the "filler" problem by having everything you do advance the game's plot, since a lot of it was reputation-driven. Once the player got "noticed" by the antagonists through actions they were doing, they would attract the next plot point... but in the interim, the players were free to do any activity in the world they wanted to do. Players in open-world games tend to gravitate to exploring the world on their own vs. following the critical path linear plot, so this design catered to that desire. The goal was to create an "open-world story" to match the open world we were striving for, and an open-world story should have as much freedom to explore it as the world you've designed.

Finally, could you share some tips for beginner writers who are willing to create their own stories for games? Where should they start? What skills should they develop?

Since this could be a very long answer, I'll try to hit the highlights. For those who want a longer answer, possibly tailored to the particular challenge they're having, feel free to contact me on LinkedIn or whatever social platform you prefer, and I'll do my best to give what advice I can.

One, play a lot of games, consume as much media as you can, and consume as many personal stories as you can. Interact with people. Travel. Analyze, store, and remember moments that resonated with you. If a villain in a book did something that literally infuriated you while reading it, analyze that: break down the steps in the design that caused that emotional reaction in you, and use that as part of your "how to make an a**hole villain the best a**hole he can be" design.

Do not solely focus on games for your research; you want to cast your net wide. Try studying cinematography (very important for games now that more and more "movie" quality cinematics are made), acting, new languages... you'll find that these pieces of knowledge will surface while writing and make it stronger. As Scott Bennie, whom I mentioned before, used to say to budding designers: "Read interesting history." He's right. Real history is far more insane and fantastical than some fantasy settings, and there are tons of fascinating stories echoing throughout the past and into the present.

Lastly, respect writing and take the craft seriously. Learn the ins and outs of English grammar before you start breaking rules. Learn to edit. Learn to do cinematic or cut scene scripting if you can. Learn how to maximize your time in the voice-over booth with an actor and prevent any delays or pick-ups. I've worked with game writers in the past who have said things like, "Oh, that's the editor's job." It's not. It's yours, too. All of it makes for a successful story.

Chris Avellone, Narrative Designer

Interview conducted by Arti Sergeev