I’ve mentioned visual novels a few times on my blog: a genre interesting first and foremost because it caters more to readers than to actual “gamers” as they are often defined by the press and industry itself.
Visual novels are not very often developed by main industry developers, even though there are several exception to the rule, like two mainstream series that have been going on for years, Ace Attorney and Steins;gate
Eliza, an indipendent game mostly developed by a single person, shows how far can visual novels reach thanks to a unique plot, smart game design and an array of meaningful choices for the player.
Developed by Zachtronics, usually known for pretty niche titles like TIS-100 or SHENZEN I/O, this is their first “adventure” game designed by newcomer Matt Burns. Inspired by his experience with “crunch culture” while working in several game studios and the whole Seattle tech industry, Eliza tells the story of the first AI counseling program and the people who developed it.
The game poses the player some very important and relevant problems: what is or could be artificial intelligence’s role in our lives? Can it actually benefit people to have an automated program to confess their psychological problems to? How far can said algorithm go when prying into people’s personal lives and details?
Thankfully, Eliza never spoonfeeds the player any straight answers, there is no unique preaching message here, but rather an invitation for the player to sit and reflect on those ever so important matters.
Of course, the stories of the people involved in developing and working with the AI are equally as important, the main protagonist, Evelyn, is struggling after leaving the project, deciding to find a job as a counselor for the Eliza app. She will have to be a “human face” for the people to talk to, while the AI will give her all appropriate answers, her role is limited to reciting the lines in an appropriate tone. But would that really help a troubled person?
Burns himself went on to explain his choice for developing a game, as opposed to a short movie or a novel: he wanted the “user” to really feel seated in the counseling room and not merely watching the events unfold.
It is true that, while the experience is probably too short to form real emotional attachments to the patients, at least in the way the designer hoped, the player does feel responsible for what the AI is making them say to someone who is troubled and in need of counseling and help. The amount of frustration that builds up in for not being able to say what is on one’s mind is the main foundation on which the narrative is slowly built and the reason the final twist works so well.
Another very interesting fact that Burns revealed while working on the visual novel is how he toyed with a cyberpunk setting, in the end deciding that “the most cyberpunk thing you can do today is accurately describe what’s happening in the world.” I totally and wholeheartedly agree with that decision.
Even though this is mostly a one man effort, Eliza doesn’t really look like a low budget affair. While characters are not animated during scenes, they are fully voiced, with a pretty good effort by the actors and the soundtrack is also the right kind of electronic and dub beats.
Probably a longer game would have been more effective in exploring the characters’ reactions and relationships, instead the visual novel doesn’t overstay its welcome, clocking in at 5-6 hours.
Some people have voiced complaints that many of the player’s choices ultimately make little difference. They’re not wrong, in that indeed Eliza seems to leave the player free to make many wildly and, apparently, conflicting different choices.
But all that seems to volatilize after one moment, just before the ending, where the player is allowed to choose from a list the desired outcome of the story. This basically means that all the choices made before matter little since, ultimately, it is possible to access every ending, even when not planning to.
On one hand, I understand the frustration, it is a common design feature that a visual novel has to be played again and again to unlock further endings, so as to change also the player’s answers and decision. Personally, I never could stand the necessity to replay whole chunks of games just to get the “real” ending.
My opinion is that Burns’ choice evens out in the end, I wouldn’t call it a design flaw per se, even if I would have liked it if at least the ending itself showed that some of the choices made (for example, sticking with the “corporation”) ultimately had at least some consequence instead of nothing at all. This means, though, that it is possible to act according to the player’s mood, which really feels very “human”. Perhaps, waking up today, our Evelyn decided that the corporation is not so evil so maybe that she should start treating her old boss better?
Like it or not, it surely is a design choice that allows the player more freedom in expressing emotions consequently allowing for more realism in the gameplay flow.
Eliza is not a game for everyone, it was already made clear from the moment it was classified as a “visual novel”. Naturally, if artificial intelligence and psychiatric counseling are among your interests, then definitely, do not waste any more time reading and see for yourself what it is all about.
Otherwise, should one like to pause a moment to reflect on current and probable future events, while not minding a slower narrative experience, I would also advise to give Eliza a look.
Burns’ interest and deep research in the topics really made the difference in writing the game, in my opinion, his efforts led to a pretty great example of game design.