Sponsored By

Designing Mechanical Narratives (Part 2)

On the virtues, and difficulties, of approaching rule design to tell stories with mechanics

Taekwan Kim

February 6, 2013

9 Min Read

My last post ended up being rather academic, so I wanted to start off Part 2 by explaining the practical purpose of this whole discussion.

The goal in Part 1 was to debunk the assumption that “gameplay narratives” are always and necessarily player-driven, that the designer should have little to no say in the formation of these types of experiences.

But this assumption—this ideological stranglehold, really—hinders us from deliberately designing rules and mechanics to produce specific experiential narratives. The implicit, foregone conclusion is that crafted narratives can only be told through words and cutscenes, and the end result is that we associate narratives that are told through gameplay with “magical” properties, as if they somehow arise of their own accord. That is to say, the possibility of “gameplay narratives” is often left up to luck, instead of up to design.

(The term “mechanical narrative”, therefore, was used instead in an effort to break from the conceptual and ideological baggage that the terms “gameplay” and “narrative” always seem to bring when placed next to each other—to emphasize the idea that mechanics can deliver crafted stories, too.)

Why should that even matter, though; isn’t it better that these types of narratives are player-driven?

Let me propose a seemingly unrelated example. Minecraft creations that get widespread recognition are often reproductions of existing creations (World of Warcraft, King’s Landing, etc.). But why? Typically, they are noteworthy because of the man hours and dedication required to produce them.

From a motivational viewpoint, having to recreate an already existing pattern gives a firm ruleset which needs to be followed—a significantly more restrictive set than the freeform blank slate that the base game presents. Such restriction, however, provides structure, logic, and tangible goals, and an instant means of gauging progress. In other words, it creates the high level of motivation required to keep players engaged long enough to make these endeavors noteworthy.

Likewise, Skyrim and similar games are fun as long as there is a set of rules to struggle with, to beat. But once you’ve done the quests (sets of narrative impositions), once you’ve become so powerful that the rules are meaningless, once you’ve run out of self-imposed rules because the base rules don’t restrict you enough, the motivation disappears and not much game remains. This is perhaps surprising in that the purported value of an open world game is the “freedom” it provides, and it suggests that even player-driven gameplay is ultimately about following restrictive rules, self-imposed or no.

These are really just examples of an already common observation: unrestricted creative freedom can be more stifling than having to work within set restrictions. And that should really be the purpose of rules in a game—to produce a set of limitations which immediately motivate the player precisely through the tension between limitation and agency. And wherever such tension exists, so too does narrative.

Generally speaking, I would say that developers are getting better at thinking about rules in this manner. But the fact that every other game suddenly has zombies in it says a lot about how much we struggle to make rules that connect, that “make sense”. Everyone knows (or thinks they know) what it means to have to survive a zombie apocalypse; we automatically have a sense of what we need to do in the game, and pre-conceived notions on how to play the rules. Which is to say, we are already invested in the rules when the motivational logic is clear.

I'm not really saying that designer-driven is superior to player-driven or whatever. The argument here is that the more we think about rule design from a motivational, narrational, or experiential standpoint—the more we think about designing rules around and about specific experiences and narratives—the better we get at designing immediately engaging rules with clear motivational logic, the less we rely on reusing the same rules and settings time and again.

My hope is that I will be able to contribute a bit to this goal by talking about some of my own struggles with this type of design thinking.


Let’s pick up where I left off my last post, then, where I started to talk about how I botched the mechanical narrative in Sola Scriptura, the third of my entries in the last Indie Speed Run.

Part of the difficulty of designing mechanical narratives is that games require win conditions or some sort of “winning” state. On this I am always reminded of Nietzsche’s “What is happiness? The feeling that power is growing, that resistance is overcome.” This concept is actually deeply engrained in most games. Indeed, players expect this, to such an extent that it is practically part of the ludic contract.

The difficulty results from two things: 1) Not all experiences are meaningful when they can be “won”. Sometimes loss itself is the point, and these experiences are difficult to communicate with games because 2) It is assumed that player actions that follow the apparent rules will eventually result in “winning”. It is assumed that unless a struggle is made against an opposing objective, that objective will not be overcome. It is further assumed that not struggling is the same as not exercising one’s agency.

The trouble with Sola was that the experience I was trying to communicate was essentially at odds with this implicit contract. The mechanical narrative of Sola was supposed to demonstrate that unforgiving ideological witch hunts are more dangerous than any external or physical threat. What this meant, though, was that the actions the players took in the game would actively harm their chances to win. I needed a set of verbs that would solve this problem.

In implementation, this translated into four verbs: Lecture, Observe Crew, Recruit Spy, and Interrogate.

Forgive me, but in order to discuss the goal of these verbs—before continuing with the “resistance is overcome” problem—I need to take a detour about an important conjunction of narrative and mechanics.

I mentioned in the comments of my last post that every verb has both a narrative and a mechanical definition. The former tells us what a verb does to the game world, the latter tells us what a verb does to the game’s mechanical variables and values. For instance, let’s talk about “assaulting” in Skyrim.

In Skyrim, assaulting is considered a crime. You commit the crime of assault when you attack a “peaceful” non-player character. The actual mechanical definition, though, is less intuitive. You “assault” someone any time you attack pretty much anything that is not in active hostile state towards you. This has more to do with hidden “faction” assignments than it does to do with actual assault.

Those occasional random dragon spawns that just fly around without attacking? Those guys are “neutral”, and attacks on them (dragonshouts included) count as assaults. But nothing about these dragons distinguishes them from the others (that are hostile by default) except this hidden value. Similarly, the bandits around the Lord’s Stone look exactly like every other bandit in the game. And again, attacking them counts as assault unless you give them time to go hostile first. Other examples abound.

The problem is that “assault” has a pre-existing, narrative definition (unprovoked, but more importantly, unjust attack) that doesn’t quite fit with the way factions have been assigned in the game. It’s not so much that the verb’s narrative goals and its mechanical goals are at odds with each other (as in ludonarrative dissonance), it’s simply that the mechanical interpretation of a verb’s narrative definition isn’t obvious to the player. The same thing happens in games where players can take narrative actions that actually have no mechanical consequence (the root of the protest against Mass Effect 3’s ending). [Perhaps this is still just a rehashing of ludonarrative dissonance, only the emphasis is placed differently.]

At any rate, the point of this detour is that the relationship between a verb’s mechanical definition and its narrative one can be bungled or exploited.

Moving back to Sola, it was this exploitable that I tried to employ to bypass the “resistance [needs to be] overcome” assumption.

The primary action cycle is as follows: to uncover the saboteur, the player must place all crew members under some sort of scrutiny (observation, spying, interrogation), each form of which decreases their resolve to various degrees, which directly impacts the operation of their assigned compartments. A compartment run by someone with no resolve is inoperative.

Resolve is recovered by the distance the submarine travels (towards safety) per turn, modified by the efficacy of the control room. The player must pressure the crew enough to uncover the saboteur before the submarine is either immobilized or life support systems are disabled by external actions and sabotage.

So far, so good. But the goal was to criticize interrogation, not to empower it, and that’s what had me stumped for some time. Eventually, once I realized that I could make doing nothing an active verb, a verb that has no mechanical result in itself—a verb that allows the crew to naturally recover resolve—the overall design seemed to click.

The hope was that the stark difference between the narrative presentation of this verb (“Lecture at the Shrine to the Articles of Self-Government”) and the actual mechanical result ("Do Nothing")—and the power of using this verb—would restore some of the criticism against ideological absolutism that was the original objective of the design. This was further underlined by making the “Shrine to the Articles of Self-Government” the only compartment that did nothing, where all the other compartments are necessary for survival.

Ultimately, I was OK with the amount of message I was able to get across with the mechanics. I say botched, though, in that I was unable to come up with a satisfactory way in which there is no saboteur, which would have delivered the narrative most clearly. Such a hidden rule would have violated the formerly stated “resistance is overcome” assumption, since literally doing nothing for the entire game would pretty much automatically result in a win. The only game would have been to figure out this hidden rule, and I had no confidence that this would constitute “enough game”.

A game’s win conditions, then, say a lot about the game’s philosophy, and this was unintentionally made explicit in my second ISR entry, Exodus, which, due to time constraints, ended up having no win conditions at all—and becoming more meaningful for it.

Once again, this has carried on much, much longer than originally intended—there will be a Part 3 to this discussion :( , though that should definitely be the last of this long-winded series. In Part 3, I’ll finally finish up the assessment of my ISR entries, and touch upon any other observations on mechanical narratives that may remain.

Read more about:


About the Author(s)

Daily news, dev blogs, and stories from Game Developer straight to your inbox

You May Also Like