The Game Design Forum

The Architecture of Dreams

by Patrick Holleman

Although this site does not usually operate like a blog, I think it is worth responding as such to two recent articles about stories in videogames.  The more important of those two pieces is the set of published letters exchanged by critic Simon Ferrari and author Tom Bissell.  While their letters covered a number of topics pertinent to video games, one of the most striking things I read was that Ferrari felt he was unable to go back and play two of his favorite games from youth because his adult perspective on the low-quality stories in those games would destroy his fond memories.

One of the two games Ferrari cites as his favorite is the SNES classic Chrono TriggerChrono Trigger isn't in the same esteemed place as Hamlet or Anna Karenina by any stretch of the imagination.  As an adult, however, I can look back through 15 years and have a deeper appreciation for how Chrono Trigger's designers manipulate the player's expectations, both in terms of what they expect from the story and what they expect from the game as a series of levels.  (There’s actually a graph included with this post that illustrates exactly how they accomplished what I’m talking about.)  In doing so they created one of the most effective mid-story surprises in all of videogames without resorting to cheesy plot twists.  What’s more, they did it in a way that could only be done in a videogame.

But rather than start with the response and fill the first page with spoilers, let's start with the critical position Ferrari might be coming from when he says this, and go from there.

There are Games, and then there are Games

Will Wright recently said that games are not the right medium for telling stories. Now to be fair, this was probably something influenced by his media people as a way to freely publicize his new TV show in major venues like CNN.com, that ran the story. Still though, he’s Will Wright and I’m certain he had some say in which of his words were printed. Wright offered an alternative: “Video games are more about story possibilities.” I’m not sure how something can be about a possibility, unless you’re talking about an insurance policy. Maybe this is just a concession on his part so that he doesn’t sound too harsh.

If he's saying that games should not tell stories, my immediate reaction is “or what?” What will happen if games tell stories? Of course that question is moot because games already do tell stories. Many games even tell stories in the much-maligned cutscene. And yet many of those games do perfectly well in the marketplace, which seems to justify their existence. Is Wright saying the buying public ignorant of where the “real” games that don’t tell stories are sold? Or does the buying public simply have poor taste? Those points are moot too because, as I said, games that tell stories exist, and they’re popular. And I can’t imagine that even Will Wright can stop the people who make story games from earning their living that way.

I think there is an underlying issue here that motivates Wright and other people in the gaming industry to say that games should not tell stories, or at least not in the way they do now. The issue is that Wright sees videogames as an extension of traditional games, which don’t have stories in any conventional sense. (By traditional games I mean chess, poker, Monopoly, etc.) And I agree that saddling Stratego with a story would be a bad idea. There are "tabletop" games that have stories, like Dungeons and Dragons, but tabletop role-playing games are very different. Right?

As I mentioned at the beginning, another well-known figure in the field of gaming commented on this recently. Critic Simon Ferrari said,

I hold that there are few formal differences between the design of analog games and digital games, while nevertheless recognizing that there are financial differences, cultural differences, differences in taste, and differences in apparatus—for instance, the encyclopedic and procedural properties of the computer allow for the storage and retrieval of complex game states; dynamic or realtime play; and the material, rather than conventional, execution of rules.
Ferrari seems to be saying that although the apparatus of the game has changed, the fundamental properties of games have not changed; they just look a bit different.

I think this is perhaps where the break occurs between people who are interested in the storytelling potential of games and those who aren't.

The way that Ferrari connects historical games to the present (via writing and scholarship about games) is through ancient China and Greece, and their “critiques and celebrations of play, sport and games.” I think that right there he has incidentally highlighted an important distinction. Sports are games, too. When Mohammed Ali taunted George Foreman in their 1974 fight—with right hand leads and mocking encouragement shouted at him through his mouthpiece—he was employing a classic game strategy: baiting your opponent. The same thing can be done in chess or poker with equal success. And yet if you look at the best sports writing, it emphasizes things that are very different than the best game writing. I can’t think of any scholar of games who proposes that we eliminate the physicality of sports or the use of instant replay by large refereeing crews because they’re not like traditional games.

In the same way that sports and traditional/tabletop (for lack of a better term) games are both different kinds of games, so too are videogames different. So when Ferrari says that he sees no formal differences in videogames and traditional games, I'm not sure how he accounts for artificial intelligence and the discrete simulated world. In a traditional game, the action of the game is driven either by the players or by chance. True, some videogames use chance in traditional ways, but many videogames do not. If a player’s ability to score a headshot in a first-person shooter were dependent mostly on chance, the game would lose much of its audience. Even when computer opponents haphazardly avoid the headshot, they’re not acting in a random way; they’re being controlled by a complex artificial intelligence.

Beyond that, when artificial intelligence drives the action of the game, the computer opponents and obstacles controlled by artificial intelligence are not always bound by all the same rules that the player is, even though they exist in the same playing field. That might have some similarities to the "dealer" in many games, but then again the AI isn't usually taking a "turn". Moreover, the players do not always initiate the AI programming that drives a game as they would if the AI were the equivalent of dice, a spinner or an events deck. They simply arrive within the AI's purview in a section of the simulated world. At least in that the mechanics that drive the videogame are fairly different from traditional models.

(I realize that people do turn the game on and press the start button, which activates the AI. But activating the game is not a play action any more than putting the chess board on the table is a play action.)

Ferrari explains the simulated world as merely a “material, rather than conventional, execution of rules.” Certainly the bounding walls of a level and its physics and other conditions can be an execution of rules. But then again, there are levels in some games that don’t have boundaries. The player character can run forever (or at least for a very long time) away from the action of the game into a shapeless brown blur or endless repeating background without being stopped or penalized. This is almost always an oversight on the part of the designers, but it does highlight a difference in design. The world or level in which gameplay takes places is not merely a set of reified rules, it is a virtual space in which play, according to rules that don’t create the space, is possible. Thus we have terms like “sandbox games” and “open world.” That world is not enough like a chessboard for the comparison to be meaningful. And of course is to say nothing of bonus stages, cheat codes, extra lives, varying victory conditions, high scores, secret areas, warps, difficulty settings, loot drops, etc which are all unique ways that videogames address their world and their rules.

Does any of this constitute a difference in “form?” I don’t know. But if you generalize broadly enough, almost anything can be said to be the same. The differences in the form of my body and that of a chimpanzee are not so great that a third party couldn’t say we have no formal differences, if they really wanted to. The differences in our respective behaviors would reveal a more telling distinction.

I do not mean to make light of what Ferrari said or take it out of context only to dismantle it for the sake of doing so. Nor do I mean to make the final distinction between all the various kinds of games that exist. The purpose of writing this is to show that there are different kinds of games— indeed there are even different kinds of videogames that differ at a fundamental level. I’d also like to raise some amount of doubt about the notion that the traits of the traditional game should govern the traits of videogames, story-related or otherwise. Sometimes they do, but sometimes they don't. Whatever the case, I do not claim to have solved the problem and cleared everything up. I think that, if anything, I have only muddied the waters.

Stories that Belong Nowhere Else

It seems that Ferrari and Will Wright are of a common mind that videogame stories are either superfluous or simply bad. Wright says of games with stories, “That’s not the kind of game I like playing.” Ferrari expresses similar sentiments numerous times in his exchange with Tom Bissell, although saying so was not the sole purpose of his writing. (The article had other topics, and was generally thorough and an interesting read.) Ferrari does, however, bring up a very interesting case in videogame stories. He says of the SNES classic, Chrono Trigger:

I’ll admit that I still count Chrono Trigger and Earthbound, two obviously story-heavy JRPGs from the SNES era, among my favorite videogames. But I haven’t played those two videogames in a long time, because I know that returning to them now would only replace my nostalgic idealization of their stories with a number of depressing realities about their narrative shortcomings.

Chrono Trigger is actually a great example to discuss because its narrative could not be achieved in any other medium with the same success. To be fair, it was made for a younger audience—indeed it was made for the young people who grew into the mainstream audience of today. So some of what Chrono Trigger has in it looks a bit saccharine and trite from a distance of fifteen years. No game is perfect, but I’d contend that Chrono Trigger has one moment that really defines what it means to tell a great story in a videogame. That is, it tells a story that really couldn’t be told as effectively any other way.

The player’s first arrival in the magical Kingdom of Zeal is one of the most effective surprises in all of videogames—narrative or otherwise. And to describe how the game pulled this surprise off, I’ll try to recapture the player’s experience of the game as best I can in a brief space, both from a narrative and gameplay point of view.

First it is necessary to establish some context about the events leading up to said arrival. The five or six hours of play leading up to Zeal have been primarily structured as a quest to find and defeat Magus, the archetypal sorcery-villain who is the creator of an eldritch abomination that will eventually destroy the world. His cartoonishly foreboding castle is a classic dungeon-crawl, and by far the most challenging portion of the game yet.

(The dungeons up until this point averaged about 33 enemies and one boss total, and the median number of enemies was 32. By contrast Magus' castle contains 118 enemies and 4 bosses, while the Tyranno lair contains about 65 encounters and 2 bosses, still much larger than before. And the T. Lair appends to the end of the previous dungeon almost immediately. See the chart below for a visualization of this)

The player characters start gaining experience much faster from the masses of powerful enemies. Accordingly this is usually the point where those characters first earn their higher-level combat abilities. The flashy new attacks—and the fact that these attacks are also essential for the difficult boss fights—lend a sense of gravity to the dungeon.

Chrono Trigger Dungeon & Story Map (Small)

(Click the image or here for full size. If you're looking for the raw data that went into this, make a request on the forums.)

The climactic battle with Magus himself is a fairly clever one, at least for a mid-nineties console game. Magus uses a shield that absorbs damage and has an attack that can wipe the player’s party out in one hit if the player is not careful. The first one or two times the player fights Magus, they stand a good chance of dying and having to reset. Moreover the attacks Magus uses are (for the time) quite graphically impressive, so the gameplay really reinforces the fact that Magus is as tough an enemy as the story says he is. Composer Yasunori Mitsuda’s soundtrack is at its absolute best right during this fight. It’s a classic showdown.

Very shortly after Magus' castle, the player is thrown into another huge dungeon crawl. Magus, as it turns out, is not the real villain; the game is only about 10 hours long at that point so it’s not too much of a shock. The player winds up in the distant past next, faced with another crawl through the prehistoric Tyranno Lair, a dungeon about as challenging as than Magus’ castle. After battling through that dungeon, the player is confronted with another iconic villain: the fire-breathing dragon (in this case as a fire breathing tyrannosaurus). This fight, while not necessarily as tricky as the Magus fight, is much longer. In fact looking back on it, with fifteen years perspective, it is clear that this boss offered a test of endurance rather than a fight for survival. This test of endurance feels especially long because for the past three or four hours the player has been doing nothing but crawling through dungeons filled with enemies and fighting ever-more-difficult bosses.

After the boss is defeated, the eldritch abomination around which the story is centered introduces itself by falling out of the sky, blowing the dungeon to hell, and leaving a giant crater with a portal that sends the player onward to time unknown. The characters briefly wonder aloud where they are.

And then silence.

What happens next is the absolute money-moment of the entire game. For the next twenty minutes there are no mandatory battles, there are no mandatory cut scenes, and most importantly: there’s no quest. The player has no directions to follow. This is very disorienting because the player has just been through dungeon after dungeon, battle after battle, each one proving more climactic and more revealing of the plot than the last. Surely there is another dungeon close at hand? Some objective to fulfil? The game has been so clear about these things up until this point.

Empty Ice PlainsThat first minute is as bewildering as it is thrilling, as the party emerges from a small cave to find themselves in a completely unfamiliar world. They enter into an ice age, with no civilization—and no dungeons for that matter—visible anywhere on the first screen. The only fully accessible location on the map is the mysterious building labled “Skyway.” There are no instructions on what it’s for or where it leads. Of course with nowhere else to go, the players step on the glowing platform that beams them up into the Kingdom of Zeal.

Describing Zeal is somewhat pointless since the words coined to describe places like it are largely reductive with overuse: floating continent, utopia, magic kingdom. The rapid progression of videogame technology has, of course, made Zeal graphically ancient. (I think it still looks good, because it was cleverly crafted, not just technologically robust.) That’s inevitable, if somewhat lamentable. But the genius of the place was not its graphics; the genius of Zeal is that never once is there an inescapable, sweeping camera shot that reveals the entire kingdom. Never once does an unavoidable NPC rush up to you and start reciting a lengthy exposition piece. Never once does a mission objective pop up and tell you where to go and what to do. The player is merely left to wander, uninhibited, through a marvelously persuasive dreamworld whose existence they never could have imagined—even half an hour ago.Kingdom of Zeal

It is possible to talk to NPCs and they will talk about Zeal, its history, its culture, its people—this indirectly ties the kingdom to the main plot. You can buy equipment, find secret chambers in the elaborate cities, and find a number of rare items in obscure locations. There are even books to read, if you look for them. The mysterious gurus—characters already encountered in the game are introduced in their proper context. The royal crest of Zeal appears everywhere, and after a few minutes the player realizes where they’ve seen it before: it’s present on every door that has been “Sealed by a mysterious force.” A large number of previously disconnected ideas click together in the player’s mind revealing the shape of a much larger plot than the player imagined they were a part of before this point.

This is an amazing moment in videogame storytelling, at both its most narrative and game-like. That which is most characteristic to videogames—the discrete simulated world—has provided a place where the player can play their way through a story. Some of this is art direction, clever writing and great music. But it’s difficult to overstate how well Zeal works as a level in a game. Upon first arrival there is disorientation because first, players never expected Zeal to happen and second, because players just crawled through a huge amount of dungeon. Zeal acts as a “breather level” that breaks up the monotony of constant dungeons, allowing the player to explore for both game purposes (secret items) and plot purposes. In that regard Zeal is the center point of the game. Magus’ Castle and the Tyranno Lair were merely the final tests of the first half, and Zeal is the intermission before the climax of the game begins.

Architecture and the World of Stories

I liken the presence of Zeal in Chrono Trigger to the art of architecture. Ultimately, architecture exists because people like to move around in beautiful spaces. The job of an architect isn’t to make a building safe or functional in a basic way—that’s more the job of the structural engineer and the contractors who build it. Architects do their work adapting the space to reflect a mood, an atmosphere, or even a philosophy that pertains to what the space is for.

It would be easy to say that players don’t need the breather level that Zeal affords them; why don’t they just turn the console off if they want a break? It would be easy to say that the time and money spent on embellishing and populating Zeal could have been spent on more complicated gameplay. But that’s the same as saying we don’t need architecture when what we’re really after is shelter. People like their houses to look nice, they like their offices to look impressive, they like their restaurants to look classy, they like their gardens to look cozy. In a videogame, designers have the ability to be the architects of kingdoms, of civilizations, of dreams. Why wouldn’t we want to fill those games with great stories, interesting characters, fantastic cities and surprising twists and turns? These are things that people like—these are things that people have demonstrated they will buy a game for.

Bad videogame stories exist, but they are not the only kind of story, and they don’t own the future of gaming. I am repulsed by the thoughtless brutality and machismo in God of War and Call of Duty: Black Ops as much as anyone else. I do not, however, take that as an indictment of all videogame writing. It occurs to me that science fiction once (and perhaps still) labored under the constant withering attacks of critics of “real” literature. The author Theodore Sturgeon, after decades of apologizing for bad science fiction said that he believed, if you apply the same standards to everything, that “ninety percent of film, literature, consumer goods, etc. are crap.” So if good AAA stories seem to be few, perhaps it is because a sufficient body of work has not been established. Especially if we are to judge what “should” and “should not” be done, we need more stories to examine. Videogame stories with high production values are young yet; give them time to show their value.

Want to leave a comment? Hit the forums.

Site questions? webmaster@thegamedesignforum.com

All material copyright by The Game Design Forum 2011