I feel like this is a topic that’s been explored a lot on this website, and I certainly haven’t read all of the metacommentary. It has, however, come up a few times in my immediate orbit and with regards to String Zero directly, so I wanted to touch on it briefly.

First, a brief update: Chapter 3A will be releasing publicly within the next week or two.

Now, let’s get into it.

Choices and Cybertexts

To summarize the questions put to me most often: Is a visual novel “missing the point” by not including choices? And what kinds of choices are meaningful enough to justify their inclusion?

To answer these questions, I think we need to define some terms. Without diving too deep into literary theory, I think it’s important to outline a basic concept critical to understanding the nature of choices in digital literature:

Cybertext: A medium where a user or player obtains a different outcome based on the choices they make. A subcategory of ergodic literature.

To quote Espen Aarseth directly, who coined the term in his seminal 1997 “Cybertext—Perspectives on Ergodic Literature”:

“In ergodic literature, nontrivial effort is required to allow the reader to traverse the text. If ergodic literature is to make sense as a concept, there must also be nonergodic literature, where the effort to traverse the text is trivial, with no extranoematic responsibilities placed on the reader except (for example) eye movement and the periodic or arbitrary turning of pages.”

Here, “extranoematic” simply means “active or bodily engagement” and is used to make a clear distinction between texts that simply do not function without “extranoemetic responsibility” or user agency/input (cybertexts), and those that exist just fine without it (traditional novels, for example).

Aarseth illustrates the concept with a number of examples, notably “choose your own adventure” novels and the divinatory “I-Ching,” because both only result in a complete story once they receive input.

Why is this important? Because in cybertexts, the players are literal co-authors, integral to the process of constructing the narrative through their input. Yes, the author might construct all of the choices, but what the author has actually done is create a web of possible outcomes. The story the player experiences is the story that they choose.

Reader-response and other poststructuralist theory might seem to challenge the idea by suggesting readers are always co-creators simply by interacting with a text (we all bring our own unique perspectives to everything we read, and interpret things differently, so no two readings will ever be the same, etc.), but I think this actually strengthens the cybertextual argument by making a clear distinction between what the reader creates on their own vs. a cybernetic feedback loop where the narrative is fundamentally and illustratively altered/affected/created by player action.

i.e. Video Games writ large.

Choices and Visual Novels

We make a distinction in visual novels between linear or kinetic visual novels that do not contain choices, and nonlinear visual novels that do. We do not, however, make any sort of clear distinction regarding what level of agency is required to make a visual novel “nonlinear.” Is a single minor choice all it takes to be labeled “nonlinear?”

Some might say a visual novel having any choice at all invites player agency, and so by definition is “nonlinear” because there is some variable outcome. For example, many visual novels allow player reactions to specific situations, which results in different NPC reactions and variable dialogue, but otherwise have no impact on narrative progression.

Is this a cybertext?

I think the better question is: Would the story still be the same if the choices were removed? I’m not going to argue these sorts of choices have no impact, because I think they can showcase and illustrate different facets of character personality. They can still do work.

But do they rise to the level of “different outcomes,” “co-creation,” or “non-trivial effort”?

This is where some of the critiques of “ergodic literature” come from—and by extension, the confusion of how much choice is enough choice—because choices simply existing at all satisfies Aarseth’s “non-trivial” definition, even if their output does not fundamentally change the textual and narrative experience.

So is any choice enough choice?

Well, no.

One can look at AAA games like Horizon: Zero Dawn to see how choices are still doing work, but fail to meaningfully impact the narrative. In H:ZD, the player is allowed to choose Aloy’s reactions to certain situations for roleplaying purposes—choices which notably have no impact on the story, but allow for players to direct the tenor of conversation in a way that will provide them the greatest emotional resonance. That emotional resonance invites players to invest themselves in Aloy, but isn’t accomplishing much else.

This is part of why I think the argument for “roleplaying choices” is stronger for self-insert main characters, since the process of defining that character is (or at least should be) integral to the game’s design and narrative experience.

But for games with an established main character, who will go on to make their own decisions regardless of player input, I begin to wonder what purpose these sorts of choices serve. Even in the above example, choices for emotional resonance feels like sleight of hand, because again: Would the story still be the same if the choices were removed?

Yes.

But would the experience be the same?

That will ultimately depend on the player. But arguably? Also yes.

Answering the Questions

So back to my initial questions.

Is a visual novel “missing the point” by not including choices?

If the goal is simply to tell a story, no. Incorporation of choice is a fundamental game design decision, and the visual novel medium is on its face a “visual novel.”

Just because the medium allows for choices to be made does not mean allowing for choices is fundamentally meaningful. The “agency for agency’s sake” argument is weak and often frustrating when that agency leads to nothing. For choices to really matter, they need to be an elemental part of the story structure. A true input-output cybertext.

Otherwise, there’s plenty of room for linear novels enriched by all of the other things a visual novel can accomplish: visual storytelling in sprites and backgrounds, auditory storytelling in music and sounds, etc.

Then what kinds of choices are meaningful enough to justify their inclusion?

Choices can provide characterization and emotional resonance, but their success at doing so is a subjective exercise that is going to vary greatly by reader. I don’t think stories are necessarily harmed by including them, but they can be, and removing them fundamentally changes very little, so their inclusion begs an answer to the question: “Why?”

It’s important to remember that choices are a promise for consequence. If the output does not match the input, then player expectation is frustrated.

For example, if I can make choices to approach a date in different ways, then the outcome of that date should be impacted by those choices. If the options simply serve as roleplaying “color” but do not affect the date’s outcome, then what was the point? If the story is best served by the date having a certain outcome, then the story is also best served by writing the most interesting scenario that will effect that outcome.

How This Impacts String Zero

In short: I thought about choices for a long time, and ultimately decided that the String Zero story was so involved, that I’d essentially have to write a different novel for every possible outcome. That might work for a AAA game, but I don’t have the time and don’t want a decades-long development cycle.

I also thought about including “roleplay” choices that immediately affect conversations but little else. I’ve seen this succeed in some games, but mostly for comedic moments, because anything provoking lasting tension would require lasting consequences. In general, it seems like players pick whatever choice strikes their fancy in the moment, but then largely forget about that choice in the next scene because by their nature, the impact of these choices must be ephemeral.

As mentioned earlier, I think these sorts of choices work best for self-inserts to help roleplay and define the character being developed “in motion,” but SZ is not a self-insert game, so the argument for them here is pretty weak, I feel.

Final Statement

This is not a judgment on other games. I know some developers feel very strongly that any agency, even if it serves no lasting purpose, helps players feel invested. I simply don’t find the argument very persuasive, but I do think if it motivates creativity, then by all means.

Create!!