Constraint, Coherence, and the Limited Parser

I’ve spoken before about matching interactive stories to narrative platforms, and briefly mentioned experiments with limited parsers. That talk was about how to use a system’s built-in strengths to make your narrative stronger: Twine’s sense of containment, Texture’s high-ask tactile affordances, Inform’s exploration via cardinal directions and world modeling. But limited parser pieces do the opposite: they press against the constraints of parser games, and by doing so, allow creators to tell stories about the frustrations of agency.

In traditional parser games, the player has a set of verbs to choose from. Developers can implement new verbs, nouns, and directions, but a stable of standard verbs is assumed. Many games play with and subvert this feature: Sam Barlow’s one-move game Aisle draws upon the entire set of possible verbs to create a palimpsest narrative, graspable only in fragments. Limited parser games, on the other hand, strip the set of usual verbs to only a handful–in some cases, even just one or two. Tools like Alice Blue’s Parser Parer have recently become available, making the disabling of verb sets less tedious and leaving creators more time to focus on implementing descriptions and custom parser responses that will prune the parser’s output even more closely to their vision.

Limited parser games tend to offer a sort of narrative momentum that trad parser games often lack. Velocity–Sam Kabo Ashwell’s term for the correlation between plot movement and play pacing–mounts as the game progresses. Limited parsers are better able to manage velocity so as to avoid slack moments, as every verb is part of a specifically chosen stable rather than a default set. These games, by cutting down on what one can do, build in a coherence of action: there are regular ways in which to engage with the world model, and the verb set chosen communicates to players very clearly what this story’s preoccupations are. By stripping out extraneous verbs, it’s possible for every turn to keep the story moving along the path of its plot arc.


Lime Ergot, made by Caleb Wilson for a jam in 2014, uses the parser’s limited scope to offer a hallucinatory vision of colonialism and decay. The rot here is physical but also moral, implicit and pervading the protagonist’s perceptions; implicated in the colonial enterprise, they are contaminated by the same specter of imperialism which rots the town. The game’s verb set is small: outside of the final verb, it consists of ‘examine’ (or ‘look at’; both produce the same effect), ‘put’ and ‘take’. The parser-staple cardinal directions are essentially disabled: when you try to move, you are only given description.



That way is an impassable tangle of stems and leaves, too flimsy to support your weight anyway.


The landscape itself resists your presence, and only by a hallucinatory examination are you able to act. This colonial outpost has collapsed in on itself under the weight of imperialist offenses; in its last days, the mapping of St. Stellio imposed by the colonial governor is no longer functional. The restriction of verbs makes it impossible for the character to act outside of what the general and the game command. It is impossible to have clean hands in this crime.

Lime Ergot’s pacing contributes to this impossibility of escape: after each lime is deposited, a new task, a new lime is demanded, and there is only one description that the game allows players to examine, one “direction” in which to move. A traditional parser game might ask players to roam the landscape, collecting 3 St. Stellio limes, and to return when they have discovered them all. But Lime Ergot structures each lime as a discrete sequence, building one on the other, so that atrocities are revealed in taut, stark description and cannot be returned to. Once they’ve surfaced, they’re out in the world; the player cannot leave the place they occupy without accomplishing their task. The architecture of the city that betrays the violence opens up only at the very end, after the player has a bit more understanding of the island and its relationship to the colonizing power, even if that understanding is only grasped in hallucinatory vision.


>x palace

It was beautiful while it lasted (which was two and a half centuries). Now it’s slag. Beyond the Palace is a cemetery.

>x cemetery

These days the cemetery is no longer unique in being littered with large chunks of white marble, or being full of dead bodies. Just uphill of the cemetery is the arbor at the summit of the hill.


Only here, at the very end of the game, the sheer scope of the venture is revealed: two and a half centuries of exploitation, with all that entails. Lime Ergot is paced out in such a way that this description cannot be stumbled on by simply exploring in a cardinal direction; the protagonist is bound within these larger structures that they are powerless to subvert or escape. The structural limitation of a narrow parser suits a game about observing the ravages of colonialism; a player cannot > go east and then > solve structural inequality. By deliberately removing agency from a medium that usually promises the illusion of it, Wilson highlights the structural forces that have shaped St. Stellio rather than focusing on individual acts.


Katherine Morayati’s TAKE, written pseudonymously as Amelia Pinnolla, is even more pared down than Lime Ergot, and even more tightly paced. Where Lime Ergot relies on a confined set of verbs, TAKE requires only the titular verb to complete, although “examine” proves illuminating. The game plays off the original function of “take” in solving traditional parsers’ medium-size-dry-goods puzzles to turn nearly every object the protagonist can observe into a source of immediate consumable (and discardable) opinion. Many of the significant nouns can be taken, which set me scouring the landscape of the game with a critical eye, sizing everything up for possible use. Which is how I play traditional parser games as well–what part of this room description will yield the answer? But where looking in a traditional parser game (and in fact most modern video games) is a free action, assumed not to change the state of the world to the extent that picking up or using an object would, TAKE collapses the distance between observing and affecting. There are no neutral actions in this game’s world.

The PC is a gladiator who must file takes on her surroundings–her preparations, her handler (or rather, her editor), the agora as meat market–as she prepares for single combat. Upon examining those takes closer for clues about her world, however, the game’s writing–simultaneously polished and slightly elliptical–suggests that the combat is the sort that takes place in bars and coffee shops and across the chasm of a white tablecloth with wait staff hovering at your elbow. The sort of messy spectacle pieces like The Cut’s weekly sex diaries offer readers as, presumably, a complicated sort of escapism. I have never felt anything but a deep welling of despair after reading any article ever published by New York Magazine, and the feeling I got from directing the character to carve up digestible portions of her own life was similar. But I kept playing because I wanted to see, which situates me both as player and voyeur, as part of her audience hungry to consume the takes she is pressured to produce.


>take Karin

You expound upon the dynamic of your relationship with Karin, once again. You’ve done so many takes on this, but there are always more variations. One variation reads “power dynamics” and “autonomy.” The other, “slavery” and “servitude.” The other, “bullshit” and “garbage.” Perhaps if you get a break on the battlefield you’ll find out which version your audience got.


The monitor pulsates. Your brands ache.


Unlike Lime Ergot, TAKE is turn-limited; after a certain number of takes, the game’s plot will progress, pushing the player character inexorably towards her appointment with her fellow gladiator. There is no escape. It isn’t possible to win. The game indicates the state of your monitor–which reports to you how well individual takes perform with your audience–but how hot or cold your monitor is does not influence your final outcome. The limited parser here suggests a bleak and brutal reality where dating is oppositional, where human connection is surveilled and passed back to your opponent’s jeering friends and your curious audience. No affection can bloom here. There are no verbs for it.

But the second play of Morayati’s game is possible to win. It relies on knowing the standard parser verb set, being familiar with the tools that are not available to you as the protagonist preparing for battle. The lateral thinking of a traditional parser game is present; here, at last, modern romance is a problem that can be solved. But not for everyone. In the second play, you see, you play as the opponent. The winning move is > USE GIRL.

TAKE, then, is about who has access–to standards of beauty, to privilege, to attraction and desirability, to community knowledge that will allow them to navigate a space in a way that will allow them to ultimately be triumphant. On its own and on only a single playthrough, it stands as a guttural howl of frustration at embodied heteronormative cis femininity. In its larger context, though, TAKE reads as a critique of systems and the assumptions around them that go uninterrogated, including the very parser system that it uses to construct its critique.

One thought on “Constraint, Coherence, and the Limited Parser

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s