Simlish, Explained: How EA Built a Language With No Real Grammar

Simlish is the spoken language of The Sims, and it has no grammar. No dictionary. No consistent word-to-meaning mapping. No phonology. It was built deliberately to be untranslatable, so that players would project their own stories onto the characters without being contradicted by anything the Sims actually said.

The Sims franchise has sold over 200 million copies since its launch in 2000. Simlish has appeared in every entry in the series, has been adopted into TikTok trends, and has been sung by real musicians who recorded Simlish versions of their hit songs. The Brazilian pop artist Anitta counts Simlish as the sixth language she sings in.

If you want to experiment with the sounds, our Simlish Translator converts English text into Simlish-style gibberish using the phonetic patterns the Maxis audio team established. Like Minionese, Simlish is not a real language, so any converter is an approximation of a system that was designed never to be systematic.

The Sims was created by Will Wright at Maxis, with the first game released by Electronic Arts on February 4, 2000. Wright's design philosophy was that players should build their own stories. The game would provide the simulation, the characters, and the objects, but the narrative would come from the player's imagination.

Language was a problem. If Sims spoke English, players who spoke English would hear specific words and interpret them literally. A Sim saying "I am sad" would fix the emotional interpretation for the player. Wright did not want that. He wanted a player to see a Sim crying and decide for themselves what the Sim was feeling and why. Real language would mess with the player's storytelling by imposing fixed meaning.

But silence was worse. The audio team, led by audio director Robi Kauker and sound designer Kent Jolly, with composer Jerry Martin, knew that human vocal sounds carry emotional weight that no other audio can replicate. Kauker has said that "when a baby cries, everyone in the room reacts." That reaction is automatic and cross-cultural. The team needed vocal performance without semantic content. They needed a language that conveyed emotion but said nothing.

Voice director Claire Curtin was brought in to solve this. The team's first attempts did not work, and the path to the final solution went through several failed experiments before they found an approach that stuck.

The Maxis audio team tried three approaches before landing on the one that shipped.

The first attempt was to use real languages that most players would not recognize. The team recorded voice actors speaking Estonian and Ukrainian. The idea was that these languages would sound like language to the ear but would be meaningless to the vast majority of the English-speaking player base. This failed immediately. Players who spoke those languages heard real words and reported that it was distracting. Players who did not speak those languages still picked up on the fact that it was a real language with real grammar, and their brains tried to parse it. The whole point was to prevent parsing.

The second attempt was to use musical instruments to create vocal-like sounds. Kauker has described this as inspired by the Charlie Brown cartoons, where the adult teacher's voice is represented by a trombone making "womp womp" sounds. The team tried building Simlish from instrument-generated vocal approximations. Kauker said this approach "failed spectacularly." The sounds did not trigger the emotional response that real human voices do. A Sim crying sounded like a saxophone, and nobody cared. The human brain is tuned to respond to human vocal cords, and instruments could not fake it.

Both attempts taught the team the same lesson: the voice had to be human, and it had to be meaningless. Those two requirements seem contradictory, and the solution was not obvious.

Claire Curtin solved the problem with an improv exercise. She brought two voice actors into the studio. One actor would perform a poem or monologue in gibberish on the spot, making up sounds that felt like language but had no meaning. The other actor would "translate" the gibberish into English for the audience, improvising a meaning based on the performance.

The exercise produced exactly what the team needed: human vocal performance with real emotional content and zero semantic content. The gibberish sounded like language because the actors were performing it as if it were language. They used real intonation patterns, real pauses, real emotional coloring. But the words were invented on the spot and would never be repeated.

The recording process that emerged from this exercise was intensive. Voice actors would watch an animation of a Sim performing an action, then voice their own interpretation of what the Sim was saying in gibberish. Sometimes the team recorded up to 20 takes per line. The takes were not variations on a fixed script. Each take was a new improvisation. The team selected the take that best matched the animation and the emotional beat they wanted.

The result was a language with no structure that would not repeat. Will Wright got what he wanted: emotions that resonated without Sims saying anything meaningful. Players could not parse the words, so they projected their own interpretations onto the scenes. The Polygon feature on Simlish history describes this process in detail, and The Verge's 20th anniversary coverage traces how the improv method evolved across the series.

The recording process has stayed close to its improv roots across 20-plus years of The Sims games. Voice actors perform in gibberish while watching animations. They do not read from a script because there is no script. The voice director shapes the performance by asking for different emotional tones, different speeds, different levels of intensity.

Jackie Perez Gratz is the current Voice Director for The Sims. Robi Kauker has been the audio director since the game's inception in 2000. The continuity matters. Kauker was there for the original improv experiments, and his presence ensures that the recording philosophy has not drifted toward scripting or fixed vocabulary across the decades.

The constraint that Simlish must not repeat is taken seriously. If a voice line sounds too similar to a previously recorded line, it gets re-recorded. The goal is that a player who spends hundreds of hours in the game never hears the exact same phrase twice in a way that makes them think "I know what that means now." The moment a player recognizes a repeated phrase, the illusion of meaninglessness breaks.

This is a different design goal from most game audio. Most games want recognizable callouts. "Enemy spotted," "reload," "behind you." These are functional, repeatable, and meant to be learned. Simlish is the opposite. It is designed to be unlearnable. You can compare this to the Minionese Translator, which handles a similarly unstructured fictional language, or the Aurebesh Translator, which is a proper one-to-one substitution cipher where every symbol maps to a fixed English letter. Simlish and Minionese share the no-fixed-mapping philosophy. Aurebesh does not.

Despite the no-fixed-vocabulary design, a few Simlish words have become recognizable through sheer repetition and fan adoption. "Sul sul" is the Simlish greeting, roughly meaning "hello." "Dag dag" is the farewell, roughly meaning "goodbye." These are the closest things Simlish has to a dictionary, and they emerged from the early recordings rather than from a deliberate naming process.

Fans have compiled lists of other recurring sounds and their inferred meanings based on context. These lists are community-maintained and not official. EA and Maxis have never published a Simlish dictionary, because publishing one would contradict the design philosophy. The Standard Galactic Alphabet from the Commander Keen series is a useful contrast here: that system is a proper substitution cipher with a fixed alphabet, so a dictionary is not just possible but trivial. Simlish resists dictionary-making by design.

The famous words matter for fan culture more than for gameplay. "Sul sul" appears on merchandise, in fan art, and in social media posts. It functions as a shibboleth for the Sims community. If you know "sul sul," you are in the club.

One of the most successful features of The Sims franchise is the Simlish music program. EA has had real musicians record Simlish versions of their songs for inclusion in the game's radio stations. The artists perform their own songs with the lyrics replaced by improvised Simlish gibberish that matches the melody and rhythm of the original.

The EA official Sims Sessions page describes the program as a way for musicians to "rediscover their songs in Simlish." The Brazilian pop artist Anitta has said that she counts Simlish as the sixth language she sings in, alongside Portuguese, Spanish, English, and others. Other artists who have recorded Simlish versions include Katy Perry, Lily Allen, Paramore, and many others across the series' history.

The Simlish music recordings follow the same improv philosophy as the voice acting. The artist listens to the original track, then performs a gibberish version that preserves the melodic and rhythmic structure. The result sounds like a real song in a language you do not speak. Your brain tries to parse it and fails, but the emotional content of the music still lands. This is the same effect the voice team built for Sim speech, applied to a different medium.

Simlish music has become a marketing tool for both EA and the artists. A Simlish version of a song introduces it to the Sims player base, which numbers in the tens of millions. For the artist, it is a novelty that generates press coverage. For EA, it is licensed music that does not require lyric localization for every market, because there are no lyrics to translate.

The Sims franchise is in a transitional period as of 2026. Project Rene, which EA has been careful not to call "Sims 5," has been in development for several years. In 2025 and 2026, EA confirmed that Project Rene has been repositioned as a mobile-only social spin-off. This is a significant change from the original expectation that Project Rene would be the next mainline entry in the series.

A separate "next evolution" of single-player PC and console Sims is being teased by Maxis. Details are limited. What is known is that Maxis is treating the mobile social experience and the traditional single-player experience as two different products rather than one unified platform. This split suggests that EA learned from the mixed reception of The Sims 4's online features and wants to keep the single-player sandbox separate from social multiplayer.

Simlish will appear in both products. The audio team's philosophy has not changed, and the voice recording infrastructure is established. Whether the mobile social spin-off will introduce new constraints on Simlish, such as the need for more repeatable callouts in a multiplayer context, is an open question. The single-player product will almost certainly maintain the traditional improv-based approach that has defined the franchise since 2000.

The linguist David Crystal, in The Cambridge Encyclopedia of Language (Cambridge University Press, 2010), notes that invented languages in media serve functions beyond communication: they build world identity and signal membership in a community. Simlish has done this for 26 years. Whatever form The Sims takes next, the gibberish is the brand.