Dear Esther Review

Dear Esther is not a game.

There are no obstacles to overcome, no other players, no score, no goals except to reach the ending. Why then, you may ask, are you reading about it on a website that’s ostensibly about video games? Because Dear Esther is a computer-generated interactive experience. That is to say, in the common sense of the term, it is indeed a video game, or at the very least, it intersects in interesting ways with video games as we know them. But it’s not a game.

To be more specific, Dear Esther is a single-player Half-Life 2 mod, created as part of The Chinese Room, a sort of research project into first-person gaming, and is described by its creators as an interactive ghost story. It was built by a team of six, including a musician and a voice actor. It places the player on a deserted island in a grey sea on a dreary, overcast day. As they explore the island at a meandering pace, a narrator reads from a series of letters he has written to the titular Esther. When they finally arrive at the end of the mostly linear path, which won’t take more than about an hour, the game ends. As far as mechanics go, that’s it.

Whereas mechanics are central to a typical video game, the narration is the real heart and soul of Dear Esther. The quality of the voice acting and script are far beyond what you’d expect from an experimental mod made by a handful of people, and the voice acting in particular puts even most triple-A titles to shame. The island is brought to life by the narration, each place making an impression on the player as the narrator explains his connection to it or elaborates on its history, with different possible readings for some locations selected randomly by the game. When the player sees a pair of gouges cut into a cliff face, the narrator may describe the custom of the island’s people to carve such gouges, and entering a broken-down shack elicits reminiscences about the time he spent there in isolation. The music also is of startlingly high quality for such a small project, and complements the narration in building the atmosphere and sense of place for each of the island’s locations. It closely follows the highs and lows of the script, moving from simple melancholy piano or strings to more complex pieces, culminating in strange, distorted choral crescendos.

Gradually a story intimately tied to the island begins to emerge from the narrator’s disconnected monologues. He tells alternately of a car accident, a historian’s chronicle of the life of a hermit, and his own exile on the island. As the game progresses the various threads draw more closely together, but their real connection, like much else, is only suggested and never made explicit. The identity of the player is also left a mystery, as is the exact nature of the island: does it really exist, or is the player’s traversal representative of some inner journey? Mysteries abound in the environment as well. What is to be made of the technical diagrams scrawled on rocks throughout? Some have obvious connections to the narrative, others not. Possible answers are subtly hinted at in the narration, but ultimately the player is left to draw their own conclusions. Dear Esther’s isn’t about telling a story so much as it’s about drawing the player into its world, forcing them to invest in it; in the island by teaching them about it as they explore, and in the story by leaving tantalizing blanks and vague patterns for their imagination to fill.

While the game succeeds impressively at this, it is sometimes hampered by poor level design in the more open first two levels. In the first level especially, it’s possible to reach several spots which seem like they’re not meant to be accessible, which in turn makes other parts that aren’t accessible seem like they ought to be. What’s worse, several such locations bring the player to the edge of the level geometry, leaving them on the edge of perfectly straight cliffs, looking out into empty skybox where ocean should be, or seeing through the paper-thin ground to the rootless bases of the trees. This kind of glimpse behind the curtain subtracts considerably from the immersion and punishes exploration of the open environment. It’s the game’s biggest weakness, and it stands out next to its remarkably high quality in other areas. Mercifully the latter two levels are more strictly linear, making such accidental detours impossible towards the climax where they would damage the experience the most.

Dear Esther’s interactivity is barely more than nominal. At most, the player decides in what order to experience things, and in the more linear levels not even that choice is present. Would it not have been better off in another medium, then? I don’t believe so. By putting the player in the position of the protagonist and allowing them to make their way about the island, the experience is owned by them in a way that watching someone else traverse the island isn’t. The player is in this place, not someone else. It’s something happening to them, not something being told to them. As limited as the freedom to explore may be, the illusion of agency it creates is crucial to the work, a major part of how it makes the player invest in the narrative. For Dear Esther, the video game is the ideal medium, and any other would have made it less than it is.

Dear Esther is not a game. It can be called a video game, but it’s still radically different from most of those. What it is an eerie, sometimes poignant experiment in interactive storytelling. One suspects that it or something much like it would eventually have been made even if the video game as we know it had never been invented. Like Tale of Tales’ recent The Path, it demonstrates how far the term “video game” has now diverged from its root. It is an experiment – and anyone with an interest in the kind of stories video games can tell will find it a very interesting one.