Joe Keiser at Next Generation has written an article statying why he agrees with Roger Ebert with regards to computer games being an inferior mode of storytelling. Ebert, in response to a letter from a reader, expounded on his comment, saying that:

Yours is the most civil of countless messages I have received after writing that I did indeed consider video games inherently inferior to film and literature. There is a structural reason for that: Video games by their nature require player choices, which is the opposite of the strategy of serious film and literature, which requires authorial control.

I am prepared to believe that video games can be elegant, subtle, sophisticated, challenging and visually wonderful. But I believe the nature of the medium prevents it from moving beyond craftsmanship to the stature of art. To my knowledge, no one in or out of the field has ever been able to cite a game worthy of comparison with the great dramatists, poets, filmmakers, novelists and composers. That a game can aspire to artistic importance as a visual experience, I accept. But for most gamers, video games represent a loss of those precious hours we have available to make ourselves more cultured, civilized and empathetic.

Keiser rightly refutes one of Ebert's central points by saying that authorial control in most cases is never really relinquished at all, because the portions of the game in which the player has control does not normally impact the story. Think about Halo 2's story: how you play it has no impact whatsoever. If you survive, you advance to the next level, see another cutscene, and keep playing. If you die, you respawn and try again. Halo 2's story, for the most part, can be experienced in its entirety by watching the cutscenes that bookend each level, with only a few quotes from key characters missing here or there. The major difference between such a sequence of cutscenes and a computer-generated feature such as Toy Story is the quality of the graphics, as well as the quality of the performances by voice actors. But these are specifics relating to individual works, not an inferiority inherent to a medium.

However, this refutation does not go far enough. Keiser asserts that games can preserve their ability to tell a story well by creating the illusion of choice: "A storytelling game should give the illusion of control, the idea that you can do anything, while at the same time putting the idea in the player’s mind that they want to do a specific thing."

I think this is a cop-out. This is, in fact, what most current games do. It creates the illusion that you're participating in a story while in reality you're just along for the ride, and the only choice you really have is to proceed as expected, stop playing, or die. When you play the game as the developers expected, you unfold the story as the developers wrote it. Hurray for us.

What Ebert presents as an inherent inferiority in games as a storytelling medium is the only real advantage it has over other forms, one almost completely unrealized to date; the ability of players to form and shape the story they experience as a collaborative effort with the games' designers, rather than experiencing it exactly as the auteur envisioned.

Storytelling in games has taken only baby steps in the past dozen years; the difference between Doom's hen-scratchings, Marathon's terminals and Halo's cutscenes are ones of quality, not kind. Where gaming has a chance to emerge as a wholly unique entertainment medium is being played out in the Massive Online Multiplayer arena, typically among role-playing games, where sincere efforts are made (sometimes, but not always) to create a fictional alter ego for oneself and then act as that person within a fictional context, both with others performing similar actions, and interacting with an environment created by the developers.

Even those efforts are in their infancy. The most popular game so far of this type, World of Warcraft, is still at best only a little removed from Blizzard's own other hack-and-slash dungeon crawls. The degree to which any one player character can star in a heroic story is somewhat limited. The strongest foes, once fallen, rise again in the next instance created for the next party, diminishing the value of the dramatic experience from within the game world.

Look at the struggles that have surrounded other MMO worlds, such as Star Wars Galaxies. When you're playing a single-player experience, like one of the Jedi Knight games, you're the star-- you're the aspiring young Jedi who chooses between the Light and Dark sides of the force. Once that experience is moved into an online realm, not everybody can play that role. It was understandable that SWG would severely restrict the number of players who could attain the status of Jedi, and understandable that paying customers (at least some of them) would object to this restriction. The hue and cry from taking Jedi status from being an accomplishment to a simple user choice has been similarly defeaning. How this kind of collaborative gaming/storytelling is supposed to function has not been worked out yet. Stories need minor characters as well as major ones, but it's unclear whether customers paying the same monthly fee are going to be willing to accept minor roles just because their stats are lower than other characters, or because they don't play as well or as often.

Some believe that the solution to that is just to instance all the core gameplay-- combat, treasure hunting, stat-building and exploring-- and retain the massively multiplayer features for the social aspects. Richard "Lord British" Garriott of Ultima fame said as much in an interview with GamesFirst back in 2002, when he put forth a kind of instanced, theme-park approach to MMO-building:

In Disney World, if you think of the main area as the massively multiplayer space, where it’s very easy to find each other or get from one fun activity, called a ride, to another fun activity, and even if you’re on opposite sides of the park, you can get there quite expediently either by walking, or using the train, or in our case teleporters even to make it faster. But when you go on a ride at Disney World, like Pirates of the Caribbean, when you get on a boat, you become blissfully unaware of the other people on the other boats. You can still see them, and you occasionally bump into them, but if it were an instantiated activity, you wouldn’t, and if the Pirates of the Caribbean were a pirate battle instead of a passive boat ride, you could imagine that here we have a quite contained hub world where you go from one fun activity, you come back and say, "Haha, we had a great time on that ride. Let’s re-equip ourselves and see what else we want to do."

I honestly don't think that's the answer. Garriott's approach takes away the problems of encountering delays in combat areas, and worlds being too small for large player bases. It doesn't solve the story problem. Every player in the world then gets to be a hero, and you can't have a story where everybody is a Chief and nobody wants to be an Indian.

There are a lot of other problems to solve, and I happen to think that this one will be the last and toughest nut to crack. MMOs already have the capability to take over the storytelling duties for their worlds that game developers mostly keep to themselves and film auteurs take for granted. What needs to happen is for game developers to keep coming up with problems-- quests, challenges, things that need doing-- within the context of a framework system flexible enough to allow multiple possible solutions without anticipating each possible outcome, and then allow the players to evolve the story on their own. Some stories will be of ignominous defeat, some will be of cowardice, perhaps others of betrayal, and perhaps some player or group of players will come up with a formula for triumphant victory. The story of the game world will be all those stories. MMOs currently are making the mistake of trying to coax each and every one of its players towards the triumphant victory ending, and letting each player take a shot at having it in their turn, conveniently resurrecting the terror at the end of each instance, making the storytelling value of the encounter nil.

Garriott advances exactly this scenario without mentioning the huge problem it creates; what I think is the real base of the problem Ebert points out when he mentions authorial control. The real authorial control within films and literature is not really so different from that in traditional single-player games. The portions in which the player has control are so divorced from the story being told that the loss of control is completely illusory, and thus has no impact on the game as a storytelling medium.

Such a lack of control could be a problem in a MMO environment; but to date, developers have skirted that problem by trying to recreate a single-player story experience within a persistent, online, massively-multiplayer framework. In this context, a game is an inferior storytelling medium:

Garriott: If I understand your question correctly, in the massively multiplayer setting, everyone is there together, and so if you could remove the word avatar and replaced it with the word Jedi, then there can be multiple Jedi. The fact that you went on a Jedi quest and became a Jedi doesn’t mean that somebody else can’t also become a Jedi. So even though the players have private experiences, it’s just like the real world where if you and I both played Ultima 9, we can meet and not be offended that we’ve both finished it. Our scenarios will be designed to keep that in mind. You’re not the main hero of the hub world, but you have defeated the bugs of Arachus, and even though others have also defeated the bugs of Arachus, it’s cool that you’ve both saved princess Lea.

What Garriott ignores here is that it isn't really "cool" that all the players in the game have rescued the Princess. In fact, it makes having saved the Princess worthless. At best, it's become a shared dream the players have-- a fiction within a fiction, much like two people sharing notes after they've watched the same movie. The fact that they both saw the same story and participated in it vicariously is not diminished by the fact that they've both seen it. But their involvement in it was at a much lower level. To really involve players in a story within an MMO context, it cannot be just a shared dream-- it has to be as real as any other event within that world. When someone rescues the princess, she has to stay rescued-- and only one player or group of players should be able to lay claim to the feat.

Serious roleplayers bemoan the amount of out-of-character, or "OOC" chat that goes on in MMOs like World of Warcraft, but given the repetitive nature of many of the game's encounters one can hardly be surprised. From the perspective of those players, how real is the gameworld where everyone can be the hero, nothing ever stays dead, and the act of selling an enchanted pauldron to a merchant has more verisimilitude and permanence than any exploit on the field of battle? Is it any wonder that the social hubs of these online worlds become virtual office water-coolers, where the events of last night's raid and last weekend's keg party are given equal validity?

It's as if, after one player killed Lord British in Ultima Online, they decided to make LB the game's ultimate boss and let the players queue up and kill him themselves in turn. That event lives on in the lore of MMO games because it is one of the only unique events that has ever occurred-- everything else is just a theme park ride.

Moving away from that model puts a lot more stress on game developers. Making up a few dozen or even hundreds of encounters that thousands of players can go through multiple times is one thing. But if each quest was a theme park ride that could only be ridden once, by a single party, the demand for creative content from the developers goes up exponentially. Even if such increased output were possible, some players would just have to live with the fact that a red dragon attacked the town while they were making a snack, and someone else killed it, and its like may never be seen again in these parts. But that's what makes the whole thing worthwhile for the guy who did kill it.

Think that idea is crazy? Maybe so. But the creators of Eve Online, if this Monday's column by Mark Wallace at Walkerings can be believed, are making this happen now-- and have been doing so for several years already:

In fact, EVE's uber items are too expensive for any one player to afford alone. So trust and cooperation becomes a resource in a way that it is not in most other game worlds. You could solo all of Azeroth and the only thing you would miss would be killing a few boss dragons that require a raid group. Even then, you could probably get into a pickup raid, even if you probably wouldn't come out a winner. A solo career in EVE can be just as rewarding a thing, but there is also a vast stretch of the game (most of it?) that is not accessible unless you're part of a player corporation and even an alliance of corps. The benefit of social interaction in EVE is much greater than in WoW. (In a sense, what it comes down to is that CCP isn't afraid to make a good bit of its game off limits to most of its players. In fact, this can be seen even at the early stages, given that the game's learning curve is so much steeper than almost any other MMO.)

The focus here is on economic rather than combat achievements, but the principle is the same: in the real world, scarcity equals value. That is just as true of experiences as it is of currency and durable goods. Eve has also found a solution for the problem of exclusion: those experiences and goods are blocked from even the most powerful single players, but are available to social groups. This gives more meaning to those groups than just raid participation.

If there's a technical magical bullet for solving such problems, I have to admit, I've no idea what it is. But this is not an inherent inferiority of gaming as a storytelling medium. It is unrealized potential for what could be the single best storytelling medium ever devised.

Ebert's parting shot, that time spent playing games is wasted, time better used improving ourselves, presumably with great films and great books, attracts the predictable retort that not all the films and literature being produced today are all that and a bag of chips, either. To say that games are inherently inferior is to say that somehow film and the printed word are somehow superior; and as long as people allow Uwe Boll to pick up a camera and Danielle Steele near a typewriter, there are going to be examples of film and literature that make Serious Sam look like an Oscar contender.

Games are a unique medium that present creators with unique challenges. To shrink from them and blame the medium is the cowardly way out. Where Ebert errs is not, as some have suggested, being too old and set in his ways as to fail to see what gaming really is. It is in underestimating the ability of human creativity to turn what he sees as limitations into opportunities.

UPDATE: Kotaku pointed out a well-written entry on Mitchell's blog Soo-Mahn that covers some of the same ground.