Gamasutra: The Art & Business of Making Gamesspacer
View All     RSS
October 30, 2014
arrowPress Releases
October 30, 2014
PR Newswire
View All
View All     Submit Event

If you enjoy reading this site, you might also want to check out these UBM Tech sites:

Systems vs. Stories: A Response
by Taekwan Kim on 06/28/13 01:25:00 pm   Expert Blogs   Featured Blogs

The following blog post, unless otherwise noted, was written by a member of Gamasutra’s community.
The thoughts and opinions expressed are those of the writer and not Gamasutra or its parent company.


Mr. Dan Whitehead over at Eurogamer published last week another entry in the long running battle of “Systems vs. Stories”.

While I agreed with many of the sentiments expressed in that article, it seemed to me that the core of Mr. Whitehead’s arguments were based on the premise that “events that don’t require player input” = “stories”, and “events that do” = “systems”. This is somewhat problematic for many reasons, not least of which is that it’s entirely possible to have largely pre-written—or “scripted”, as Mr. Whitehead phrased it—stories that are systems driven.

Fallout 1 is the game I always point to as an example that does this exceptionally well. FO1 works primarily because of the strictness of its systems—how hard it “pushes back”, to once again borrow from Mr. Whitehead. In FO1, significant restrictions mean that developing one set of skills is always done at the real expense of all the other skills.

At the same time, many of the game’s essential narrative choices (challenges, really) are skill based (that is, they require skill checks). Coupled with the relatively punishing difficulty of combat survival, these conditions create serious, system-wide competing demands for every single skill and attribute point.

A low level cap (20) and extremely limited opportunities to increase base character attributes compound the tension. So, unlike Planescape: Torment, for a comparative instance—or even FO1’s sequels for that matter—the player can’t simply grind herself into omnipotence. (PST has attribute checks for plot crucial dialogs, too. It’s just that they basically don’t matter.)

As such, narrative (or, indeed, any sort of) progress in FO1 is always a matter of juggling mechanical priorities. Systemic, mechanical, exclusive choices about character build directly become systemic, mechanical, exclusive choices about the story. By causing dialog demands to genuinely impact all the other build considerations (and vice versa), and by giving skills mechanical potential both inside and outside of dialogs, FO1, for the most part, is able make story-mechanics integration transcend mere disconnected, standalone events or one-time pass/fail conditions.

I’m trying to avoid sophistry or pedantry here, but really, what actually does a script do? It takes an input and returns an output. What we perceive to be “systemic”, then, is really just a bunch of co-dependent scripts—layer in enough scripts and it begins to appear systemic.

Let me repeat that. The number and randomness of the scripts involved are pretty much the only things that change an event from being “scripted” to being “systemic”.

Obviously, it’s true that you can have “stories” that unfold in games without any input from the player. But it’s also true that the most compelling events in “systems” equally happen completely independently of the player. This independence is precisely what makes them compelling—what causes them to be perceived as stories. And again, none of this means that you can’t also have “stories” that require player input in tangible ways.

In the modern sense, though, narration requiring player input has fundamentally been simplified to a state which makes Mr. Whitehead’s arguments against “story” entirely valid. They are often no more than forced dichotomies isolated from the rest of gameplay, or non-simulative “choice and consequence” chains. These divergences exist either purely for their own sake or to artificially inflate replayability, as opposed to existing in order to systemically simulate believable reactions and outcomes—existing because they are part of the systems of the game.

As much as I dislike employing an overused scapegoat, my opinion in this case is that this type of divergence is almost certainly a result of prioritizing “accessibility”. The thinking, it seems, is that we don’t want to punish players in their narrative options because of character build choices that they’ve made. (Honestly, it’s more or less come to the point that we don’t even want to punish the player in combat, or “actual gameplay”, for these decisions either.)

And that, for all intents and purposes, is the “modern” divergent narrative—excised of all mechanical consideration.

But that’s not all. Taking out mechanical wrangling from these decisions makes them eminently boring—unchallenging. The “modern” response? Just double down. Manufacture difficulty by removing the player’s ability to make informed decisions. Force them to pick between two unknowns. Then call it “moral ambiguity” to disguise the fact that there’s about as much player decision-making involved as resorting to flipping a coin.

I said that as if I believe this entire line of thinking to be mistaken, which isn’t actually true. It doesn’t really make sense to hold players accountable for failing mechanical requirements they had no way of anticipating. And having to try to pre-empt these requirements by dumping points in attributes you don’t really need feels wasteful and unfair. It’s precisely this type of thing that causes players to just hoard unspent character points. Or, when the player does decide to gamble by investing in attributes she thinks might eventually come into play, well, gambling is precisely what it is—just another flip of the coin.

I think at this point in the discussion, it would help to touch upon a hypothetical solution that attempts to address these issues. How can we maintain “accessibility”—that is, lower the psychological barrier to mechanical commitments—without reducing mechanical complexity or tension in systems based scripted narratives?

Something I’ve discovered in the thousands of hours I’ve spent playing RPGs is that, generally speaking, I don’t actually care what narrative results the game gives me as long as the course of action I’ve taken embodies the identity I am trying to build in the game. Negative or positive, give me that chance again and I’d probably just do the same thing, because the principle of action and my sense of self (either as the player or the player character) are stronger than the judgment made by the game upon that decision. If the game “rewards” me for it, so much the better.

Let’s extrapolate from this. Say that we base fundamental aspects of the build mechanics straightforwardly on worldview and personality. For instance, consider the classic offensive/defensive, attacking/turtling, expanding/entrenching behavioral split. These are fairly deep, consistent preferences—usually one style is strongly favored over the other for each individual player.

For the sake of argument, let’s assume that these preferences stem from the familiar extrovert/introvert personality orientation. Applying this to a game with path-finding and survival mechanics, say that you can invest character points in extroversion or introversion, which would then translate into differing approaches for affecting self-rescue from a hostile environment.

An extroverted approach seeks to gain as much ground as possible, taking serious risks if necessary, in order to achieve extraction from said environment. Morale/willpower is gained when risks are vindicated, and lost when the player behaves in too stationary a fashion. On the other hand, an introverted approach is going to excel at creating shelters, extracting every last usable object from each area, while receiving penalties to movement. Take too many risks, and the player’s morale will plummet. Specializing in one orientation or the other unlocks access to extreme abilities, like being able to hunt bears or alligators with a sharpened stick (for the extrovert) or being able to create blazing fires out of a handful of wet hay (for the introvert).

Taking this to dialog integration, then, instead of basing dialog checks on specific proficiencies, what if we based them on the strength and orientation of the player character’s personality or approach? An introvert with a liberal leaning is probably going to be more successful at persuading an NPC with a similar outlook than an NPC that’s stridently conservative and extroverted. Similarly, an introverted NPC might be more open to process-oriented solutions coming from an introverted PC than to goal-oriented proposals coming from an extroverted PC. Etc.

It’s tricky though; once you start going into behavioral generalizations, you risk introducing implicit judgments and cultural stereotypes that can alienate the player (already, the idea that introverts can’t be goal-oriented or open to risk-taking is pretty absurd). As well, it’s hard to say what kind of reaction the player will have when a character build so strongly based on personality fails.

Anyway, the actual point of this hypothetical proposal is that simply by changing a few descriptors, we can make what are ultimately still just skill checks seem far more deeply integrated into the overall systems of gameplay, as well as make the systems seem more firmly integral to the scripted narration.

In the end, I can’t say with complete certainty that I’m not intentionally missing Mr. Whitehead’s point here. I mean, I get what he’s saying. That the most interesting narratives in games happen through the interplay of rules, not through dictation. It’s just I can’t shake the feeling that the idea that “storytelling” and “interaction” are mutually exclusive is misguided—or more accurately, reactionary.

Is it possible that the desire to differentiate games from films as “unique” is just another manifestation of insecurity in our favorite medium? A thought process still operating under the shadow of the accusation that games don’t matter? Why actually do we need this distinction, except to be able to say (as if we can’t already) that games are culturally significant in their own right, too? And aren’t we doing games a disservice if we stigmatize certain forms of expression within them?

It’s a difficult situation, and it’s complicated by the fact that cultural significance does indeed matter. But I think that a real danger exists in trying to escape the perceived clutches of the influence of other media, leading games to be limited to not-films and not-books—that is, still defined by other media—instead of being whatever form of expression that best suits the situation and experience the game is trying to communicate. And that downward spiral would be truly counterproductive.

Related Jobs

DeNA Studios Canada
DeNA Studios Canada — Vancouver, British Columbia, Canada

Analytical Game Designer
Blizzard Entertainment
Blizzard Entertainment — Irvine, California, United States

Senior User Experience Designer, Irvine
Blizzard Entertainment
Blizzard Entertainment — Irvine, California, United States

Senior User Experience Designer, San Francisco
Avalanche Studios
Avalanche Studios — New York, New York, United States

UI Artist/Designer


Luis Guimaraes
profile image
Disclosure: I haven't thought of this whole subject for a while, so maybe it all doesn't make much sense, but might be food for thought anyway, so here it goes... (and by the way it's not entirely answering to the article).

First Video games are awesome to tell stories. Period. This whole idea of games trying to be "as good as movies" is ridiculous because games are actually better already in many ways, yet different enough for both to remain interesting media in each's own way.

The medium just needs to be used correctly. Checking the basics of storytelling techniques is a good start for developers not wanting to fall into the wannabe-movie-director trap by accident.

Of course some genres naturally fit best into the medium, and also different people has different tastes. But as I seemingly only care for story in two specific genres of of the medium, I'll try to shortly analyse these genres only:


Whatever the medium it's in, the genre uses both present tense events (plot) and past tense events (backstory). So it makes easy to separate player and designer roles and have both to contribute a desired amount to the story, while having the mood/atmosphere working as a tool to define how much the desired amount for each involved part is.

Being plot a mix of player actions and developers authorship, considered involvement percentages can vary, and backstory mostly the job of the developers authorship even thou part of the discovery can be delegated to the player as desired and, most importantly, is the narrative-time which leaves the most room for negative space, it's the designer job to, well, design, the correct combination of weights of the input of all involved parts, paint close attention and exploring the mood/atmosphere in order to achieve the ratio desired (by the designer) and expected (by the user).

The bulk of the story is divided between present and past tense while involving the player into an active discovery process (exploration versus exposition), therefore allowing for better and acceptable use of authorship in levels otherwise unattractive for an interactive medium.

Adventure Games

Many might disagree here, but the genre basically revolves around figuring out how to move the story forward (gameplay-McGuffin).

The puzzle is usually about:
a) figuring out something unrelated to the story: solving puzzle machines/panels;
b) using knowledge previously gained from the story: a clue given in a certain conversation/document about something hidden somewhere;
c) guessing what happens next in the story and pushing for it: character steals a car to pursue the vilain, tricks someone into doing something, etc.

The entire story of the game can be done using present tense and still make for a good game because the story itself is the toy the game with played with, making the active discovery of the story be the game itself (exploration plus exposition).

Taekwan Kim
profile image
I think there are basically 2 tests for determining how systemic a game's events are:

1. Does it feel like a dialogue, or just two monologues (or worse, soliloquies) taking turns?
2. How many solutions are there?

#2 requires a bit more clarification: Does the player's playstyle actually matter? Will two different players with completely different strategies and tendencies get exactly the same result?

I don't think both of these tests necessarily have to be passed for a sequence of events to feel systemic, but failing both is probably a good indication that it's not.

But again, it's kind of tricky, because the first test is about as subjective as the second test is objective. For me personally, the genres you've mentioned, particularly adventure games, feel fairly "monological" (I don't think that's actually a word!), which probably comes down to the fact that ludic obstacles in most adventure games are devices that exist for their own sake, not devices that support dynamic systems.

I said most, though, because it doesn't have to be that way. Blade Runner is a pretty good example of a more robustly systemic adventure game. (Ah, Blade Runner. That game brought me to tears as a teenager, heh.)

Ramin Shokrizade
profile image
I like the idea that a game can allow the player to create the story. This is the promise in most MMOs, though it is rarely realized. If every choice a player makes slowly moves the story in one direction, while closing the story in other directions, then the player is really crafting their own story within the existing game systems and can feel some ownership in the result. What often fails in MMOs in this approach is that the games are not balanced, such that there are only a few "good" paths and the other ones become undesirable. Those that went down the "wrong" path want to start over or respec and in allowing this you make the story somewhat meaningless since they can just rewrite their story at will.

I'm trying to translate your discussion to multiplayer games, since to me that is the future of gaming. If you do truly allow every player to craft their own story, and in doing so become a unique (or almost unique) member of the community with skill sets no one else has, then the result can be a game world (and economy) where every member feels useful and productive. This is where the promise of a virtual economy is realized, when every member is able to contribute to a whole and is not "obsolete" before they realize it.

Then you get situations like "Wow Jane decided to go to the Faerie Islands first and learn Dust Magic while the rest of us were just trying to get to L60 first. Now Jane is the only Dust Sorcerer in the world that can make Eldritch Dragon Sand for our Spectral Siege Cannon. Let's make sure she joins our faction!" If you let everyone do everything in time, instead of making meaningful choices, this scenario can never happen.

Taekwan Kim
profile image
"What often fails in MMOs in this approach is that the games are not balanced, such that there are only a few "good" paths and the other ones become undesirable."

That's precisely the thing, right? How many MMOs ever reach that level of balance? But I think there's more to this problem. The more balanced a game is, the flatter the differences between players become, the less distinctions between different builds matter. It's a bit like the 1984 scenario: the three competing entities are so completely balanced that they essentially become the same. The only difference is the labels applied to them.

That is to say, when choices are limited, the illusion of uniqueness requires imbalance. And I would say that the Dust Sorcerer scenario is actually an example of such. Which brings us back to the same problem of everyone rushing to the next imbalanced build--everyone rushing to be Dust Sorcerers in that scenario (the "flavor of the month" phenomenon). So uniqueness disappears once again. Whatever personal story a player tries to craft out of their decisions ends up taking a far, far back seat to mechanical demands--their story becomes overridden by group dynamics.

I think the difficulty is that competition makes identity building (besides displays of expertise or conspicuous consumption--appearing competitively successful, basically) a secondary goal at best. And all of this is absolutely fine for purely systemic stories. Maybe not as great for hybrids, though--you'll probably need to skew the design towards completely segregated and non-mechanical story branches in order to remove the pressure of mechanically "correct" choices.

That's a really interesting design problem, though. I will enjoy mulling over it further.

Ramin Shokrizade
profile image
A well constructed economy can actually support a "fairly" imbalanced competition. For instance in World of Tanks the system tries really hard to balance both sides such that there is not a lot of variation, and those on the losing end of the allocation end up on the winning end eventually. Using a complex economy as a balance, you can actually have an unbalanced encounter occur with both sides winning. It is a bit complex and I don't really want to explain how.

If you allow social systems to form organically in multiplayer games, then even though people might be designing their avatars or builds to fit a certain role, the peer-to-peer interactions can tell a very complex story that could never be predicted by your game design. This is what makes such games so compelling. I talk a bit about this in my Group Monetization paper if you are interested.

Taekwan Kim
profile image
Right, I understand the Dust Sorcerer example was given as an illustration of incentivizing mechanical interaction between players.

I was just trying to express, given the goal of this post (which admittedly seems to have been somewhat unclear), that it would be significantly difficult to incorporate pre-written story elements that have mechanical impact in a competitive multiplayer setting. Mechanical concerns would override any narrative motivations, particularly since competition incentivizes the leveraging of any advantages that can be had. Which is what I meant when I said that the design would probably be skewed towards removing mechanical considerations from pre-written narrative choices.

[User Banned]
profile image
This user violated Gamasutra’s Comment Guidelines and has been banned.

Michael DeFazio
profile image
I can recount the best game narratives I've experienced and none of them are "scripted", were the result of some arbitrary decision, or came as a revelation from game dialog.

Think about the "narratives" in some of the most popular events that occurred in games in history:

The "Immaculate Reception"
The “Miracle on Ice"
The "Tony Hawk's 900"

And I can think of parallels that happened to me in video games:
Beating the flamelurker in Demons souls (Experienced a triumph like Tony Hawk with his 900)
Defeating a grand master backgammon player (was like participating in the "Miracle on Ice")
Single-Handedly taking down a Tank in Left4Dead and reviving my team (was a little like experiencing the "immaculate reception")

... and each of these above carried more weight in my memory than any "Would you kindly" moment.

These narratives were born from the game's systems/rules/constraints...The narrative IS the game event, and what decisions I made and my skill in the game using the systems, not just the context or pre-scripted decision making.

If we were to analyse the narrative of the "Thrilla in Manilla" and only take into account Ali's antics, that only accounts for setting up the context of the event (rivalry)...what happened in the ring was far more interesting and what makes the event memorable.

IMHO that is something for video games to aspire to... create great systems and allow the narrative context (David verses Goliath, Snatching Victory from the Jaws of Defeat, Heated Rivalries) to be created around these systems without having to "talk at the player" exposition style.

...I suppose you lost me with these paragraphs (it seems to narrowly define game narrative as only "scripted things"):
"I’m trying to avoid sophistry or pedantry here, but really, what actually does a script do? It takes an input and returns an output. What we perceive to be “systemic”, then, is really just a bunch of co-dependent scripts—layer in enough scripts and it begins to appear systemic."


But it’s also true that the most compelling events in “systems” equally happen completely independently of the player. This independence is precisely what makes them compelling—what causes them to be perceived as stories. And again, none of this means that you can’t also have “stories” that require player input in tangible ways.

--but perhaps I am not interpreting them correctly.

Here's the power of games creating narratives IMHO:


Taekwan Kim
profile image
Ah, it seems there’s some confusion as to what I’m trying to communicate with this post. Once again, I am failing at clarity, and for that I apologize.

I’m not trying to deny the power of systems or anything like that. I mean, gameplay _is_ systems, in my understanding. The only thing is how complex those systems are.

But my point is that you can create systems out of scripted components—indeed, if you go down to the programming level, that’s actually what systems are: scripted components designed to interact with each other. There’s no real reason that these components, then, can’t also include pre-written narrative elements, is all I am trying to say.

At the programming level, many systems are independent of player input. Or, to put it another way, they follow rules that are not of the player’s making. That’s what makes them feel emergent, because the player has no immediate or direct control over what output they will receive from the game.

For (a rather long-winded) instance, let’s talk about Crusader Kings II. The precise details are a bit hazy for me now, but I recall one playthrough in which, as the King of Ireland, I married off a second generation daughter to a son of the King of Scotland in a matrilineal marriage. This daughter had a child, which, due to the rules of inheritance, became the King of Ireland once my first generation character died—that is, became my replacement character—an event for which I had absolutely not been planning.

Eventually, once again due to rules I had no control over, that second generation daughter ended up being Queen of Scotland. Entirely unanticipated—systemic. And, since the marriage was matrilineal, her son that was now King of Ireland (me) was positioned to inherit an entire other kingdom.

At this point my character was already getting quite old—inheriting that kingdom before my claims became diluted was a pressing concern. Moreover, the Kingdom of Scotland itself was struggling with multiple rebellious vassals; it was in danger of being split.

It occurred to me, then, that if I simply assassinated the Queen, I could significantly expedite the matter. Even as I was planning the assassination of my own mother, though, it happened that England declared war. And here’s the real kicker: my mother, being the dutiful ally that she was, took up arms in defense of Ireland despite her own significant internal problems. Again, _while I was plotting her death_.

Such unanticipated loyalty oblivious to the insidious duplicity that was about to occur was rather mind blowing—seriously gripping stuff. And once more, I do agree, systems are just that amazing.

But the point of this (purely systemic) example is that the bulk of that “narrative”, outside of the initial decision to marry off a daughter, was completely independent of the player, just as “scripted narrative” sections can be. That is to say, the rules that were in operation here were just as pre-determined and out of the player's hands.

Anyway, I wrote this post because I wanted to critically examine the idea that you can’t genuinely incorporate pre-written elements into systems in a robust fashion, and to explore some methods through which that can be done—not to say that games should be “scripted”, or to define games as this or that or anything else.

I hope I've been able to clarify the aim and thinking of this post?

Michael Wenk
profile image
You know, reading this brings back memories. I'd actually submit that Fallout 1 (and 2) are best enjoyed with cheats setting various stats to levels that are not normally achievable. It is very entertaining to play when you're as dumb as a rock(no offense rock!) or smart as a genius.

Bart Stewart
profile image
On a side note, I found it very interesting to contrast Dan Whitehead's generally pro-system argument with Tynan Sylvester's recent anti-system "Simulation Dream" article.

"Anti-system" sounds like it puts that argument too strongly, but I don't think it does. Calling the desire to build dynamic systems a "dream," and asserting that all that's really necessary are facades of systems... that is a direct shot at emergence as a valid source of play experiences.

I read Dan Whitehead's Eurogamer piece as a proper and necessary response to the "faking it is good enough for telling the developer's intended story" claim made by the "Simulation Dream" piece. My only beef with the "Systems vs Stories" article is that it didn't go after that claim nearly hard enough.

Is the appearance of a dynamic world (BioShock: Infinite) to tell the developer's story necessarily better than actually simulating some dynamics (System Shock) to allow player stories to emerge? I don't think we're near the end of that debate... happily for the gamers who get more enjoyment from creating their own stories than from being pushed through the developer's.

[EDIT: Also, read the last section of Rock Paper Shotgun's Jim Rossignol's interview with Syndicate developer (and now Satellite Reign Kickstarterer) Mike Diskett.
ndicates-simulation-legacy/#more-158898 They understand why systems and simulation, not scripting, are what distinguish computer games from interactive movies. I can give it some more thought, but right now I don't think seeing this distinction "stigmatizes" the scripted films-as-games design model, so much as observes that it just functionally does not realize as much of the power of computers to create new realities for us to experience as the systemic worlds-as-games model does.]