Third Helix is now Kickbomb Entertainment! Please visit kickbomb.us for all future updates.


18 May 2009

[ Continuing with my new weekly series, "Monday Musings", this week's post explores why we fail at designing failure, and posits a new way forward. ]

Failure Is Conflict

Traditionally, the protagonist of a story must endure a series of failures before ultimately achieving success. These failures mark key plot points and are evidence of conflict, the critical element in any story. Without failure, there is no conflict, and without conflict, there is no story. Therefore, stories require failure.

*Death *seems to be the most common form of failure in contemporary narrative games. You're given a challenge to overcome, and when you fail you're kicked back to the beginning to do it again. The problem with death is that it completely shatters narrative flow. It means the story has ended, because there is no longer a protagonist to carry it forward. Reloading a saved game or recent checkpoint is akin to turning back a few pages in the novel, or rewinding a few minutes of the film, and hoping things will work out differently this time around. From the protagonist's perspective, this is obviously absurd.

From the player's perspective, however, it works out conveniently. While death is a terminal story-breaker in the narrative of the protagonist, it's actually just a plot point in the narrative of the player. The story of the player struggling to complete Tsavo Highway on Legendary is quite distinct from the story of Master Chief leading a convoy of marines through enemy resistance to the city of Voi. The former is ludic narrative (the narrative of the play of the game), while the latter is internal narrative (the narrative contained within the game), and their interplay is critical to the design of narrative games.

Narrative Disharmony

Narrative disharmony *occurs when the *internal narrative and the ludic narrative are sending different messages. As mentioned above, death is one such example: here, the internal narrative indicates the end of the story, while the ludic narrative merely indicates a plot point.

Sometimes, games use forms of failure that don't *break the narrative flow. One such example occurs in the original *Half-Life, roughly halfway through the game, where you're subjected to a difficult battle against a group of assassins in a darkened loading dock. Immediately after completing this sequence, you exit the area through a door that was previously inaccessible, whereupon the lights go out and you hear sounds of a scuffle: you've just been ambushed by grunts and taken captive. You awaken in a trash compactor, stripped of all your weapons, and must escape before being crushed.

Despite the unbroken narrative flow, there are still a few problems here, chief among them the fact that the internal narrative has just described a significant failure (the ambush) in response to the ludic narrative's success (defeating the assassins). It feels as though, for completing that difficult sequence, you've been "rewarded" by having all your weapons stripped away by means of an ambush you had no chance to prevent. And after this event, the game carries on as normal, almost as if the ambush never happened... this is narrative disharmony at work.

In this case, we can frame the interplay between the internal and ludic narratives as a challenge and response. The ludic narrative presented the challenge of the assassin encounter, and the internal narrative gave the response of your ambush and capture.

Some games use forms of failure in which these roles are reversed: the internal narrative issues the challenge, while the ludic narrative provides the response. Fallout and Fallout 2, Morrowind, and The Witcher all allow you to make decisions whose consequences may not be known for quite some time; often, this involves the death or disenfranchisement of a key NPC as a direct result of the player's actions. In these cases, the internal narrative -- seeded with a given player decision -- presents an unexpected situation, such as an NPC with a necessary bit of knowledge having been killed earlier in the game. Here, a narrative point is established, not in response to the player's completion of an immediately-preceding challenge, but in response to a seemingly-disconnected decision some time in the past. The internal narrative issued the challenge to acquire information from the NPC, but the ludic narrative describes the player's struggle to deal with the unexpected, and now-inconvenient, consequence of his former decision. (This scenario naturally assumes that the requested information is also available via other means; if not, then the game is technically broken.)

Although these games avoid the problem of "rewarding" ludic narrative success *with internal narrative *failure, they have their own issues. Most commonly, the failures in question don't actually raise the stakes (in *either *narrative thread). And frequently, the player simply thinks he's broken the simulation, so reloads an earlier save from before he made the "mistake" instead of continuing on and dealing with the consequences in play.

It may seem that internal and ludic narratives cannot be reconciled, but there is a way forward.

Rules of Failure

I propose three simple rules for designing failure:

  • Do not break narrative flow
  • Provide a clear way forward
  • Raise the stakes

The first rule is pretty much a given: if you break the flow of the internal narrative, the only way to reconcile the ludic narrative is to end it also. Completely. If the death of the protagonist is, narratively-speaking, complete (that is, he's not going to be resurrected), then the cessation of the play experience must also be complete. The only game I know of that was this daring was Steel Battallion, which at its hardest difficulty level wiped your save game at death. Steel Battallion *was awesome -- largely because of its custom mech cockpit-simulating controller -- but way too hardcore for just about everybody on the planet. Players generally want to keep playing, even when they fail. Let them, by *not breaking narrative flow.

The second rule has to do with motivation; specifically, keeping the player immersed in the narrative instead of reaching for the "reset" button. If the failure provides a clear way forward -- something that can clearly be done, right now, to keep playing and proceeding -- players will think of the failure as part of the game experience, rather than their own personal screw-up to be erased from the annals of history by reloading a saved game. This integrates failure as a plot point in *both *the internal and ludic narratives.

A good example of providing a clear way forward is found in Mass Effect, at the climax of the Virmire mission. Allies Kaidan and Ashley are both in dire trouble at separate locations, and the player must choose which to aid; only one can be saved. The death of a crew mate is inevitable, and that inevitably is communicated at the time of the decision. The game provides a clear objective -- the rescue of the chosen crew-mate -- and carries that objective through the remainder of the mission, even as the narrative "failure" takes place. The obvious inevitability, and the constant presence of a clear way forward, reassures the player that this failure is part of the game's internal narrative, and not the result of his or her own inability to play the game correctly. The ludic narrative failure, on the other hand, is embodied by uncertainty as to the rightness of the player's decision; he or she may experience regret that the wrong person was saved, or that both could not be saved.

The final rule also deals with motivation, but here it's not about keeping the player in the game so much as it's about not letting him or her dwell too long on the failure itself. While failure is necessary to manifest conflict, nobody likes to fail. We want to ensure players are looking forward to positive experiences as quickly as possible, so we *raise the stakes *to give them an immediate sense of purpose: avenging a fallen ally, perhaps, or escaping a perilous situation.

Two excellent examples of raising the stakes occur in two of my favorite games of all time: Chrono Trigger and Skies of Arcadia. In Chrono Trigger you eventually acquire a time machine called the Epoch. Soon thereafter, you find yourself in the Kingdom of Zeal circa 12,000 B.C., accosted by an arrogant prick named Dalton. At a certain point, Dalton tricks your party and steals the Epoch, effectively trapping you in this era. The combination of your sudden inability to travel through time and your burning hatred for Dalton -- and he is an eminently hatable character; a perfect character study for a villain, in fact -- results in an incredibly compelling motivation to hunt him down.

Skies of Arcadia establishes a similar scenario: late in the game you spend a lot of time building up your own personal air pirate headquarters. Lots of memorable events take place here, establishing cameraderie amongst allies and making the place feel like home. Near the end of the game, you return from a mission to find your headquarters ablaze, under assault by an airship fleet led by the rotten prince of Valua. The sense of losing your home, and the clear assignment of blame to the prince, makes your desire to exact revenge upon him shockingly intense.

By maintaining narrative flow, providing a clear way forward, and raising the stakes, failure can be effectively woven into narrative games without confusing, frustrating, or boring your players. Furthermore, integrating the **internal and**** ludic narratives** can create a greater sense of *harmony *in the game experience, reducing or eliminating improbable or nonsensical narrative situations and greatly increasing immersion and engagement.

Posted In:

design-essays game-design monday-musings video-games