Jet black poetry looms overhead in the blinding white. A shrouded leviathan sweeps across the horizon while the void offers nothing but sweet torment, inviting you to give up the search for firewood and succumb to the arctic frost gnawing at every fiber of your being.
This is the harsh reality of Imagined Leviathans, a scavenger tale set in Britain’s arctic future that deftly treads the line between beauty and existential horror by combining an experimental diegetic narrative with wonderfully abstract visuals.
The result of that heady cocktail is an experience that lives long in the memory. Although still teasingly pitched as “coming soon,” we recently went hands on with The Imagined Leviathan: Prologue — a demo that offers a glimpse into the haunting world developer Far Few Giants’ is currently piecing together.
Intrigued by what we saw during our brief time in the British tundra, we caught up with two of the project’s key players, 3D artist Richard Tongeman and writer Antony de Fault, to better understand how Imagined Leviathans‘ philosophical and poetic take on survival horror came to be.
Gamasutra: Imagined Leviathans is unashamedly abstract. Words hang above players as they wander the English tundra and voices on the wind share stories and philosophical musings. What was the inspiration behind that unconventional approach to storytelling?
Richard Tongeman: As a studio we’re interested in approaching narrative design from a new angle each time — in this case, can we deliver a narrative entirely via diegetic text? Like you’d see in What Remains of Edith Finch.
We wanted to do away with the typical ‘find a diary extract pinned to a tree’ trope you often see in first person games and make the narrative more all-encompassing, like you’re stepping through the pages of a novel. We landed in territory that is similar to The Witness’ audio logs, with philosophical musings grounded in personal anecdotes.
When we design a game we often start imagining a single image we can capture in 3 to 6 seconds via a gif on Twitter. If we can capture the game’s whole pitch in one moving image, then we know we can sell the game. In this case it began with the words on trees, we wanted to evoke a claustrophobic feeling of being haunted by something ethereal and bigger than yourself — and that visual seemed to communicate it well.
The game began more abstract than it’s ended up, we thought you’d be hopping between islands of tree clusters and forging out into the void to find the next breadcrumb, but much like playing a spooky game with the lights out, people felt uncomfortable alone in the white, so it effectively acted like an invisible wall and hindered progress.
People were even more uncomfortable trying to orient themselves within the stark environment and have a visceral reaction — it’s a very love-it-or-hate-it aesthetic. To counter this we’ve been scattering bursts of detail throughout the map to bring it closer in density to your typical open world game. But we have left in a few important moments where you have to make a leap of faith into the whiteout in order to progress which provides a controlled moment of sensory deprivation which can be really exhilarating.
Gamasutra: Once you’d decided that’s how you wanted to connect with players, how did you actually implement those audio-sensory narrative systems? For instance, how did you choose which words would confront players on their travels?
Richard Tongeman: The game started out purely as a reflection on climate change but we quickly realized that we had the opportunity to discuss the larger issue behind climate change and the other huge problems we face today: we don’t make space in our lives to really confront them. These problems are like glaring suns that can blind an individual who looks directly into them. In an increasingly individualist world, we lack the framework for confronting global issues. After realizing that, Antony wrote the first iteration of the script in one day, opening himself up to every existential crisis he’d ever had, and spilling it out onto the page. I wouldn’t recommend that technique to everyone but, uh, it’s that kind of game.
We specifically didn’t want to have your typical ‘sad dad’ voice over, exploring your traumatic past like you’d see in Dear Esther. We’re not entirely sure who the narrator in Leviathans is, how they relate to the players’ character and who the stories are about, we’ve kept it more universal so that the player can insert themselves into the stories instead. I suppose they’re intended as a sort of personified zeitgeist.
To trigger the words displaying we use Koreographer in Unity, a VJ’ing plugin which acts as a timeline to synchronize game events with audio. In it, we’ve tagged each of the important words in the audio timeline at the point at which they’re spoken. We then have text boxes scattered on the trees like branches, and we pick which ones to render the words to based on whether they are in the player’s view.
Gamasutra: We played the Steam demo, and that captivating drip-feed approach to narrative is accentuated by the world itself: a surreal arctic void that feels both impossibly vast and blindingly claustrophobic. How did you expand the experience using environmental storytelling and world building?
Richard Tongeman: I grew up in The New Forest in England, where we have these postcard-perfect thatched cottages buried amidst ancient cathedral-like redwood trees. We’ll only see maybe one day of snow a year but nonetheless I wanted to bring the winter apocalypse to Brockenhurst.
The world building is essentially a flash tour of English heritage countryside. That’s the big thing that we have over somewhere like the United States, the simplest stone wall in a farmer’s field probably has more history than most of their post-colonial architecture, and we wanted to embrace that. It was really important to us during development to isolate quintessentially British features to ground it with a nostalgic, recognizable sense of place.
Environmental storytelling hasn’t changed much since Half-Life, I don’t think. It’s about getting in a character’s head, thinking how they would behave in the situation and then going around with a sledgehammer destroying things. Gamers will miss most of what you’re putting out when they’re focusing on the business of playing, so it’s important not to be too subtle!
Gamasutra: Were you ever concerned about becoming too obtuse in your pursuit of narrative subtlety? It feels like you’re trying to walk a fine line between giving players room to chew the fat and think freely, while also dropping enough breadcrumbs to keep them engaged. How did you approach that balancing act?
Richard Tongeman: We learn a hell of a lot by releasing prototypes and demos as we go and observing people play on YouTube. We’ll get a huge range of responses, from some people just enjoying the survival, to others diving deep and creating thirty minute video essays. In the later months it’s been difficult to get in-person playtesting during the pandemic so we’ve been running playtests through Discord streaming instead, asking players to voice their thoughts out loud as they go.
Playing the game without the voice-over can get lonely very quickly, so we definitely wanted to give you a vocal companion like you’d see in Firewatch. We also wanted to provide a deeper thematic layer to the story than just what was immediately in front of you, like in David O’Reilly’s EVERYTHING, but it doubles up as an approachable narrative layer for players who don’t enjoy reading to muse on while wandering around.
We’ve never been able to afford voice actors before as we tend to lean into big stories with comparatively small budgets and we don’t want to compromise on performance quality, so Antony has been voicing the lines himself and Richard Campbell has been processing them to sound ghostly. Antony originally provided the placeholder performance and didn’t really want his voice in the game, but everyone around us insisted it’s great and he sounds right for the role, so it stays.
When deciding whether to include voice as part of the game I think that often players will decide in the first few seconds of a trailer whether the narrative delivery method is for them, and if there’s a strong vocal performance then it tends to elevate the game in perceived quality up to the next level. If we can nail that, it means that hopefully more people will be willing to take a risk on it and engage with some of the more complex and wordy narrative delivery methods hidden inside the game.
Gamasutra: Why did you choose to implement survival mechanics instead of making the title a pure narrative experience? Were they part of the game from day one, or something you worked in later?
Richard Tongeman: There’s a definite trend emerging of walking sims with “more to do.” I’d put ourselves in the same bucket as games like Adiós or The Magnificent Truffle Pigs, the latter of which I had a hand in the early development of. As a genre we’re not very techy, so it’s baby steps but we want to offer the player something relaxing and manageable for their thumbs to do while we bombard them with quite heady story.
The game was ideated as The Long Dark through the lense of Jack London’s To Build A Fire, so the survival mechanics were the first thing we built. We see survival games coming out with insanely detailed inventory systems but a general disregard for any sense of storytelling. We wanted to topple Maslow’s hierarchy of needs right down to just building a fire and trying to stay warm to give room for the narrative to shine. Which means no inventory, no food, no crafting – just picking up sticks and deciding what to keep and what to drop.
Of course it wouldn’t be a Far Few Giants game if it were not dripping with metaphor, so in picking up sticks you’re effectively keeping your head down and focusing on your immediate survival whilst ignoring the larger problems affecting the wider world.
One of the big problems still facing narrative games is negative perception due to short run-times, you can see this recently with Before Your Eyes. Steam gamers are still caught up on price / time, so having the survival mechanics will keep the player busy letting us pace out an hour of story delivery over a more palatable length.
You can imagine a version of The Witness, for example, where you just listen to the audiologs one after the other, and the game’s over within two or three hours. Instead, in that game you contemplate those audio logs while meandering around the island solving puzzles. The logs themselves are enriched by this breathing room, and to the average player the runtime is more attractive. That’s the kind of experience we’re aiming for.
We’re not ideologically against short games, in-fact we originally released this game as a short, but people wanted to see more. I’d personally like to see more games dare to be short. It’s just that this particular game’s mechanics lend themselves to a slower, more contemplative experience.
Gamasutra: What was your approach to actually writing the story? Did you have everything down on paper beforehand or iterate as you went along? Perhaps your process evolved during production?
Antony de Fault: Iteration, iteration, iteration. As we said before, I began by spilling all of the existential crises I’ve ever had onto the page in a day. Then I came up with ideas for little stories that, in a few lines, would express each crisis to the listener. I had them wrapped up within another day or so. These went through a few versions before recording, and I’ll likely rewrite and record all of them at least once more before launch.
Later on, we decided to add the text-based encounters to the game to give it a ‘proper’ narrative in addition to the figurative one provided by the poetic readings you discover. In it, we planned a rough story idea where you’d be building a community to help you confront the game’s Leviathans, in response to a small YouTuber’s video essay where he correctly pointed out a metaphorical flaw with the original prototype’s ending.
I then mostly disappeared into the ether for a few weeks. When I returned, I had five documents: each one a comprehensive summary of one ‘branch’ of the game’s open world, which would be laid out like the legs of a starfish around a central gathering location, the Steel Henge. Each one had its own theme, character and associated arc, distinct locations, and conclusion.
I wrote the first branch out in full in-game myself, but then I got super busy with programming. Luckily, we were able to work with Faith Brownlee, who came onboard to turn all of my planning documents into a first draft pass for the rest of the game. She’s working on that literally as I type, using a tool called Fungus in Unity, which allows the setup of narrative flowcharts for authoring branching writing. Once she finishes, I’ll take a few redrafts, and the game’ll ship!
Richard Tongeman: The reason we choose to use a node-based narrative engine over something more procedural like INK is because we like to treat our narrative similar to a non-linear editing suite – triggering events, visual effects and audio from the narrative flow. Every storyline is hand-authored and each mission unique, this allows us to build out the impression of a much grander system than what we actually have under the hood.
One of the things that allows us to build narrative games as quickly as we do is that we spend the time we would have spent programming a complex system on writing more content instead, and in the process we fake them!
Gamasutra: Based on what you’ve learned on the project so far, do you have any tips for other developers looking to try their hand at abstract storytelling?
Richard Tongeman: Be really careful around making your visuals or story so abstract that it’s no longer palatable. I’ve seen a few developers, ourselves included, plant their stake in the ground with a completely out-there visual or narrative approach and in the process alienate their players, publishers, and reviewers. Be acutely aware of who you’re aiming your game at and adjust the weirdness accordingly.
I think players can be open to new ideas, and I think it’s important for both the industry and your own creative fulfillment that we have breathing room to explore those, but there needs to be something which grounds it.
The most obvious way of achieving that is drawing heavily upon your inspirations, by recycling visual motifs, art styles, archetypes, themes, settings, voices or characters from the media that influenced you while making it, so that your players have something recognizable to view your work through a lense of. Alternatively, heavy playtesting early on can expose those unintentionally off-putting moments, allowing you to build more structure to gently guide the player into it.
If you were to look at a creative partnership like David Lynch and Mark Frost on Twin Peaks, you can see that the wacky world and characters of Lynch are balanced out against the more mainstream friendly soap opera format brought about by seasoned showrunner Frost. You can see in the third season without Frost as a grounding force the show becomes a very isolating experience. It’s important for commercial indies to identify who in your partnership plays each role and attempt to balance each other out.
One thing we’ve built into our creative process is to attempt to stand out visually and build a consistent visual identity, but I think we can only get away with making a game this visually unique because we had support from a local funding body to make it. I’m glad we stuck to our guns and kept it weird, but without the safety of that funding, we would have been forced to make the game more ‘mass-appeal.’
I’ve seen a few people on Twitter remarking with regards to our game that they’re glad to see black-and-white games are becoming more accepted or mainstream. And I’d caution against jumping at that too readily, I certainly don’t see it as part of an emerging trend — though I’m excited to see the games coming out and announced this year generally exploring more experimental art styles.
For me the black and white style came from a very specific creative place and unique set of limitations. I wanted to create an open world with a lot of visually-intriguing ‘hero’ assets extremely quickly. So eliminating the variable of color enabled that — most of my scenes take less than an hour to build from start to finish. The commercially unfriendly nature of black and white game art was offset by the fact it wasn’t feasible to make in any other way.