Inside Eric Chahi's Philosophy — Turning Lack into Margin

Introduction

Eric Chahi is a French game author. Self-taught from the 8-bit era, he became widely known for Another World (released in Japan as Outer World), which he built almost entirely alone in 1991. Drawn in polygons and almost wordless, this quiet science-fiction adventure told its escape-and-survival puzzles without dialogue and influenced many designers who followed. He is not a pure logic-puzzle maker, but on one point — how to stage the moment of solving — Chahi belongs deeply to the history of puzzle design.

I take him up now because rereading old GDC transcripts is a hobby of mine, and his 2011 postmortem has a density that never tires me however many refills of hot bancha I pour. This is not a tour of his games but a reading of the person: what he has said consistently, and where he struggled — relying only on his own public statements. I avoid assertion where the record is thin, and mark my own interpretation as interpretation.

Background

Chahi had been making games since around 1983 without a hit. In 1989 he set himself a constraint — to build a fully polygonal game — and took on an ambitious project. He recalls: “I hadn't coded for two years and my last experience with C language was a disaster... It was really depressing almost” (GDC 2011). He switched to assembly and a language of his own making, building the game's systems from scratch, from a modeling tool to a bespoke interpreted language.

Another World was thus completed by Chahi essentially alone over about two years, with music helped along by Jean-François Freitas. He went on to Heart of Darkness (1998), the god-game simulation From Dust (2011) made with Ubisoft, and the VR work Paper Beast (2020) from his own studio Pixel Reef. In Japan he is remembered as the author of Outer World, but his whole career reads as a long wrestling match with “the freedom to make things alone.”

Philosophy — Improvisation, and the player's memory

The most consistent thread across his statements is an embrace of improvisation as a method. Another World was built without a fixed script, “layer by layer without knowing where it was going” (GDC 2011). He likens this to theatrical improvisation: freedom lives precisely because the constraints are firm. His design outlook is condensed in one line — “Regarding the limits, I accepted them as improvisation constraints” (GDC 2011). Notably, this is not old news. In a 2017 interview he still said, “I did not want to define the gameplay in advance; things get clearer as the game making process advances,” and “As with Another World, the development of the scenario is done piece by piece, and we later set up the elements in chronological order” (bounthavy interview, 2017/pub. 2020) ↗ — the same method held for over two decades.

A second core is the belief that “The real media is the player's memory and imagination, so seeing it once is enough to create the game universe in his mind” (GDC 2011) ↗. Low-resolution polygons lose detail but gain a margin that invites the player's projection. What he returns to is not story itself but “the rhythm and narrative tension,” going so far as to say “It was not the story side that interested me the most” (GDC 2011).

And an obsession with immersion. Explaining the near-absence of UI: “I just wanted to immerse the player as much as possible,” because the score counters common at the time “pierces a game's overall immersion” (Nintendo World Report 2018) ↗. Constraint, improvisation, memory, immersion — these form, I read, a single circle in his thinking.

Obsessions — Suggestion over description, punctuating cuts

The foremost obsession running through his work is “suggestion over description.” He explains his visual style as one that “leaned more toward suggesting things than describing them in great detail” (NWR 2018) ↗. What began as a necessity born of the polygon choice he flipped into an aesthetic. Raising volume from few lines and shapes — a hand he admits is continuous with his training in black-and-white illustration.

The second is his use of cutscenes as punctuation. In contrast to Heart of Darkness, which tells its story in long scenes, “In Another World, it is the opposite. All cutscenes are very short, lasting just a few seconds... brief punctuations,” so that “gameplay and cutscenes are much more woven together this way” (NWR 2018) ↗. The black monster's appearance, taking the gun in jail — each stresses one point without stopping the play. His staging, I'd summarize, builds tension by cutting rather than adding.

Failures and how he moved through them

Taking only failures he has stated publicly. On Heart of Darkness, Virgin abandoned the project mid-development and he had to find a new publisher: “for six months nothing was certain and it was particularly hard to keep on developing the game.” Yet he calls it something that “turned out to be a good thing” (bounthavy 2017/pub. 2020) ↗ — by the time Infogrames took over it was nearly finished, making for a win-win partnership.

Nor does he hide the toll of Another World: “at the end of the development I was exhausted and this is the reason that Lester was almost dying in the end” (GDC 2011) ↗. The author's state bleeding into the work's feeling — his own words.

And difficulty: “Another World is a difficult game to play, but it's also improved a lot over the years; I did a lot of iteration since its first release on Amiga. People would be surprised how hard and unforgiving it was originally” (NWR 2018) ↗. He doesn't deny the early version's harshness; he kept sanding it down with each re-release. A stance, I read, of never erasing a failure from the record but examining it in public.

Design dilemmas

Restricting myself to dilemmas he has voiced. The largest is the “fork toward point-and-click” during Another World. He showed the game to Virgin France, whose producer “thought the game was too hard and told me to change it to a point and click adventure!” (NWR 2018) ↗ — the trend of the day. At GDC he called this “a major fork in the creation of the game” where it “could have become point and click,” admitting he was nearly convinced. But the effort to redo it “would have been too huge,” and friends who played it loved it, so he held firm. He “hates doing things twice” (GDC 2011) ↗.

The other is the tug-of-war between independence and stability. Granting that “being indie is a lot of stress,” he still says, “at least we have creative autonomy. We are totally free and that is priceless!” (bounthavy 2017/pub. 2020) ↗ He also notes team size: “Beyond 10 people, I think it would be harder for me to keep a fluid and flexible production process.” Staying small on purpose to maximize freedom — his choices share a consistent center of gravity.

Influences

Only influences he acknowledges. For story and world, first science fiction, and Star Wars in particular. Of Another World's laser gun he says he “was inspired by the dynamism of Star Wars' battles” (GDC 2011), and years later recalls: “For Another World, it was Star Wars and science fiction in general” (bounthavy 2017/pub. 2020) ↗.

Visually he names science-fiction and fantasy illustrators: “influenced by science fiction or fantasy work from artists like Michael Whelan, Frazetta, Bernie Wrightson and Richard Corben,” adding that “black-and-white illustrations were a major guide to understanding how to create objects and volume with simple lines or shapes” (NWR 2018) ↗. And for the VR work Paper Beast, “it's rather influenced by David Lynch” (bounthavy 2017/pub. 2020) ↗ — he speaks frankly of how his interests shifted over time.

Kizuki's reading

From here it is my reading, Kizuki's. I read Chahi as “someone who turns lack into staging.” Two years without code left him rusty; memory was short; polygons had to discard detail. Each of these he took up not as a defect to be filled but as a margin inviting the player's imagination. His line — accepting limits as “improvisation constraints” — reads, beyond technique, as a way of living. A publisher's abandonment, exhaustion, the early version's harshness: he narrates rather than erases them, as if he knew that what is missing is precisely what completes the work. That his method held unchanged across two decades is, I'd venture, because he keeps believing not in the answer but in the margin.

Closing

If you want to understand Chahi, the fastest way is still Another World (Outer World). Following Lester's journey for a few dozen minutes from that wordless opening will let the design ideas quoted here — suggestion, the cut as punctuation, betting on memory — land as experience. Recent re-releases are tuned to be more playable than the original, matching his own words that he “did a lot of iteration” (NWR 2018) ↗.

Drawing the line through quiet staging and cinematic tension, our study of Jeppe Carlsen (LIMBO / INSIDE / COCOON) sits comfortably alongside this one. Read side by side as a lineage that speaks through the rhythm of experience with fewer words, Chahi's position comes into sharper focus. He is, for me, the kind of author who makes me pour another cup of hot bancha and replay an old talk once more.

Sources

Primary sources referenced in this article:

・GDC Vault: Classic Game Postmortem — OUT OF THIS WORLD / ANOTHER WORLD (Chahi's talk, 2011) ↗

・Game Developer, “GDC 2011: Eric Chahi's Retro Postmortem: Another World” (talk report, 2011-03-03) ↗

・Nintendo World Report, “Interview with Eric Chahi on Another World for Nintendo Switch” (2018-07-16) ↗

・bounthavy.com, “Interview with Éric Chahi” (2017 interview, published 2020-07-21) ↗