[In this essay originally posted on the What Games Are blog, and reprinted in full with his permission, UK-based game designer and Gamasutra contributor Tadhg Kelly explains why he believes there's no such thing as a player character, and analyzes what's really going on when most people play games in context.]

"There is no such thing as a player character" is the kind of tagline that gets me into trouble in some places. So is "the emotional connection between player and character that many game makers believe exists in fact does not."

Both contain a powerful subtext, questioning everything from a player's sense of identity to the validity of their experiences. Read the wrong way, they can seem to say that all the emotion you feel in playing games is made up.

Of course that's not my intent. When I say "there is no such thing as a player character" I don't mean that there is nothing. When I say play occurs through "dolls," likewise. My intent is to reinterpret the emotional experience of play within a game-native context, and so derive useful insight that could apply to all games. In other words, the emotions are real but our way of talking about them is broken.

This is an essay to fully explain this concept, to set what's really going on when most players play games in context, about the importance of identity and self expression. 

A personal story of emotional experience

When I moved to London I didn't really know why I was moving. I had had a long background in games at a kind of pro-amateur level, creating role playing games, card games and live action role playing games for the Irish convention scene. I had also had some experience in the industries surrounding games, such as working in the retail sector and as a technical writer at Havok. However, I didn't really know where I fit in. So I emigrated.

Luckily I landed my first game design job soon afterward, and the following year was a wild ride. I learned and did so much, from level design to scriptwriting to action design, but -- as happened to many others in the UK at that time -- ultimately the studio collapsed through a lack of funds. I took it hard, became depressed and needed to find a job. I was willing to take the first thing that came along, which happened to be a contract tester position.

Where my first year had been amazing, my second was miserable. My employers seemed to have a culture of shipping software in whatever state it happened to be in to meet release dates. So they produced a lot of churn content, and it was the sort of place where issues like quality were a non-starter. I was paid little, lived in fear of redundancy (testers are often only hired on rolling contracts), and spent my days testing crapware. For a long time I wallowed.

To avoid feeling that funk, we testers played games. Our lab had a local network of PCs, so we might play Call of Duty at lunch. However, the lab also had glass walls, which bred a classroom mentality. Management sat outside and looked in on us. Producers wandered by and stared into see if their game was being tested. Various people came in at the drop of a hat and complained over what they saw being done (or not). It was like a real world version of the office in the movie Brazil. We were the students being made to do our homework and our managers were essentially invigilators.

The truth was that testing was not difficult. We could go through each new build in about an hour to verify fixes, play new content, button-bash the interface, and so on. So we had a lot of free time, but had to appear as good workers being productive. This meant no Call of Duty, but we could get away with smaller and more hideable games such as emulated GBA ROMs. In so doing, I got surprisingly hooked by a soccer game.

I'm not an avid fan of soccer. When national competitions like Euro 2012 roll around I will cheer for my beleaguered Irish team, but I have no interest in leagues or the soap opera of the transfer markets or which player insulted who. However, this soccer game caught me. It had simple controls. It was fun. I would always play as a particular team because I liked their color and knew that they were supposed to be good. At first I played just single matches lasting four minutes, but later realized that the emulator could save game states. So I could play leagues, and I did. Hundreds of them.

I imagined that the different players on my team had identities, and I started playing preferential tactics on the basis that I liked one player over another. I became emotionally connected to these little dudes, and when I developed a winning strategy (which usually meant I won a match by 5-0 or more) I kept playing anyway. It was a wonderful place to go and to feel success, to imagine cheering crowds and trophies and so on. I even imagined a sort of backstory to what was going on behind the game.

At a time when my career felt like it had stalled, that little game became the highlight of my day. In retrospect it also proved to be a personal example of everything that games are, and I often look back on it as an example of modality.

The ineluctable modality of the playable

Whether starting with a great tune, a basic three-act-and-two-plot-point script, or move-and jump in a simple game, the fundamentals matter. They teach us a great deal about what an art form is and what tends to function well versus what does not. So in a sense, all media can be interpreted as elaborations of the same forms over and over.

It could be called "modalism," and what it means is: The simplest form defines the rules by which the rest of the form operates because, while the work changes, the mode of use does not.

A statue from 5000 years ago and a statue from today are modally identical, and while the modern statue may be part of a more complicated conversation with its audience than the ancient one, there are principles of form that hold true. The same is true of ancient poems and modern poems, of the Theban plays and modern drama, of Charlie Chaplin films and The Descendants.

Game makers often feel that games defy this kind of rule. They think that one day, when the technology is good enough or the audience is educated enough, games will become something more. I disagree. Games as they will be in the future will be the same as games as they are today, and the same as games when they were first invented.

The play, imagination, fun, flow, thauma and so on of the simplest game (such as my soccer experience) is no different from that of the most advanced production that money can build. Modally, they are what they are, which is to say they are identical. And this is ineluctable (meaning inescapable, inevitable, or unable to be resisted) because it's a limit stemming from the players' brains. Modality arises because of how people perceive, interpret, and think.

If you are crowded into a darkened area with a lit stage, you'll expect a performance. If it's auditory you will expect musical pattern and rhythms, and perhaps singing. If the performance involves speech and masks, you'll expect a story. If it involves rules, you'll expect a sport and fair play. These are basic understandings which stretch all the way back to ancient civilizations in cultures all around the world. They apply to plays, dance, ballet, opera and cinema. And if you sit down at a computer or console (or tablet or smartphone) and play a game, you expect fun.

Modality does not mean that we all have the same tastes. You may love science fiction spectaculars or romantic comedies. You might love the mosh pit of a Metallica concert or the rarefied air of the Sydney Opera House. You might love frenetic shooters or intellectually puzzling adventure games. It's just about understanding how they are modally the same sorts of experiences -- and how modality rules what does and doesn't work.

Metallica and the opera are bound by the same constants of musical structure, for example, and sci-fi and romcom stories share many fundamental traits of plot and character development. Even shooters and adventures, seemingly wildly different, share common roots of clarity, feedback, and fun.

Modern games are simply more elaborate productions based on familiar modes, like Prometheus compared to Forbidden Planet. They look and sound far more sophisticated than their fore-bearers but are played for the same reasons as Combat, Donkey Kong, and Tetris. They evoke the same sorts of pleasures and the same sense of thauma, emergence, story, fun, flow, and so on. And the same rules of good functional game design (called creative constants) apply throughout regardless of audience, aesthetic, culture, market, simplicity, or complexity.

One of those creative constants is this: the player is always herself.

Performers and gamers

At the most meta level, this is what I think play is:

To perform is to act, sing, dance, juggle, perform acrobatics, and otherwise entertain an audience with practiced routines. Performers learn lines, steps, songs, chords, and so on, and they perform in the magic circle of the theatre. For the performers, this is a kind of play.

There are some similarities between performers and gamers. Perfecting skills is one example. A trapeze artist training for the big show and a gamer trying to get really good with railgun headshots are in some ways doing similar things. Performers rehearse; gamers practice. Performers love the feeling of achievement that comes with a great show, and gamers love the feeling of winning while mastering a dynamic. Both are also very creative in their play. Yet they are not the same.

The poet wrestling with verse, the guitarist mastering a solo, and the comedian weaving elaborate jokes on stage are all attempting to draw an emotional reaction from an audience. Whether the material is original or adapted, performers interpret and master for the entertainment of others and this often involves pretending to be someone else.

A performer adopts a character, which might be a scripted Blanche DuBois (from A Streetcar Named Desire) or a stage persona. She wears a mask (figuratively or literally) and becomes someone both archetypal and unreal. As Cary Grant (real name Archibald Leach) once said: "Everyone wants to be Cary Grant. Even I want to be Cary Grant." It is this wearing of masks and adopting of voice that brings something about performance to life. The experience is joyful both for artist and audience and it's why they do what they do.

On the other hand, Rock Band's lasting appeal lies in mastering the dexterity required to get great scores, not the entertainment of others. Though you might encounter social disapproval if playing in a group for not getting into the spirit of it, you can play it in a totally literal manner (follow pattern, earn points) if you wish. Even in public arenas like arcades or football stadiums, gamers often play functionally. Footballers often foul and cheat on the pitch to fool the referee into awarding penalties, and the crowd hates it. Yet they do it anyway because they want to win.

Showmanship exists of course, from the extra acrobatics that dance game masters perform to the NFL wide receiver doing a dance when scoring a touchdown. Yet while games sometimes incorporate aspects of performance, they are not really about it. The key difference between these two types of play is that performers do what they do to entertain others, but gamers do what they do to entertain themselves. Games are about mastering for the self, not for other people.

This is basically why the "player character" is a flawed idea.

Good dog

When Clint Eastwood picks up the gun in Unforgiven, it is meaningful in a story context. Modally it is the same sort of drama that goes back further than Shakespeare ("Is this a dagger I see before me?"), but Clint is a master performer. His performance interprets this moment anew, for this story, and is of characterful significance which resonates with our deepest archetypal selves.

Doing it yourself in Red Dead Redemption, on the other hand, is very different. It's a straight choice of fight or flight, which weapon will you choose and what's your strategy for killing these six bandits approaching on horseback. It's win or lose, fair or cheat, and all about you. You would probably refer to John Marsden (the protagonist of Red Dead Redemption) as your "player character", but you are not him. You are you.

Gameplay in all forms is always highly literal. It's about clear choices, the creative manipulation of resources, mastery, tactics, and strategy. Sometimes literalism or creativity don't matter, such as with prize games and gambling, or tolerating bad gameplay for other reasons like humor. Jumping through preordained hoops can work under certain conditions, but the game that fails to be literal and clear is usually a failure sooner or later because it's simply not fun.

While all the evidence (completion rates, metrics, player behavior) indicates that players are in it for themselves, designers commonly stick with the idea that the player is in part a performer, a "player character" and talk about eliciting emotional engagement on that level. That's because of the legacy of pen and paper roleplaying games.

In games like Dungeons & Dragons, the "player character" is an imagined entity that a player controls, and every other entity is a "non-player character" under the control of a game master. Fundamentally the difference between them is one of ownership, a way of saying mine versus yours. However "player character" has acquired more meaning than just classification.

Pen and paper roleplayers commonly want to play their part, customize their character, level up, and engage with the game's fantasy as well as its frame. They want to care, to have meaningful experiences, and do exciting things. Game masters also usually want their games to be about more than just refereeing combats and calculating spell damage. This is why systems like alignment have taken root.

Alignment is a shorthand to describe persona, and is most successful when welded into a game system. If you want to play a paladin then you have to play in a lawful good fashion. If you don't, then your character loses his special paladin abilities and becomes an ordinary fighter. Similar penalties apply to other classes like rangers and druids, and magic spells exist which can detect alignments. Dungeons & Dragons even includes a quadrant graph to track player alignments, and the game master is encouraged to judge the behavior of the player along this graph. Other roleplaying games have similar systems, such as the Beast/Frenzy system in Vampire: The Masquerade or the Sanity system in Call of Cthulhu.

Some video games have adapted this idea with morality systems. Actions are tagged as "good" or "evil" and if you commit them then the game adds or subtracts from variables that determine whether you are Jedi or Sith. The notion is to encourage the player to play in the spirit of the game, to get into the world and the story and the character and become a performer.

For the vast majority of players, however, this is not what they do. Doing good in Knights of the Old Republic is nothing to do with actually being good. It's about killing the right sort of enemy in order to earn points and unlock powers. Being lawful good in Dungeons & Dragons is not about actually feeling that way. It's about the benefits that come with being a paladin. It's a pretense, like a sinner saying the Rosary 10 times to stock up some forgiveness from God before going out to gamble.

Why? Because alignment, morality, and behavioral grading systems are treated by players as mechanistic levers, and gameplay is highly literal. They are just another set of rules to be mastered, and just another type of extrinsic reward. And if they behave, then the game (be it the game master or the rules or whatever) pats them on the head and gives them treats like experience points: good dog.

Sure, a player might put on a voice or choose not to kill the children in the village if they know that they'll lose or gain experience or alignment points for doing so, but not generally for reasons beyond that. It's just part of the attraction, of being able to level-up and get cool new powers and loot. And it's why system-focused games like Dungeons & Dragons and Pathfinder survived while the pure storytelling roleplaying games of the mid-'90s commercially died.

In video games it is even harder to make this sort of system work as intended, because there is no social pressure from friends to behave. Players grok the optimal actions required by the game to be a good dog and they batch them. This is why conversation dialogue in video games often feels like strip-mining for facts, for example. It's also why finding the perfect sequence of button presses for L.A. Noire sidesteps the whole point of the interview system. It's a literal system like any other, and the gamer acts to master for self, not to perform. So, good dog. Have a treat.

That's all that's usually going on with the "player character." Most of the rest of the time, play is driven more by self-expression.

Expression and projection

When alignment systems try to force players to play in certain ways, they often revolt.

They choose to play chaotic neutral characters in Dungeons & Dragons and then act disruptively in every scene. The game master presents a quest that they could go on, and they say no. The king asks the player to save the princess, and instead he kills the king. Whatever. They also mess with the morality system in Fable to see just how far they can push it. They take to Twitter to complain about the ending of Mass Effect, which tried to channel them in specific directions.

Pushing players to act in-character reveals an interesting disconnect between why the game maker thinks they play (for story, for personal performance, for narrative experience perhaps) and why they actually play. This is what I meant when I said "the emotional connection between player and character that many game makers believe exists in fact does not." It's not a lack of connection. It's a different connection.

The idea that there is just one type of player who only likes one type of game is, of course, nonsense. Some players play just for the functional. Others want to explore, relax, or experience wonder. Some want to live a fantasy, such as running a football team, being a Formula One driver, or commanding a kingdom. Some just like puzzles and don't care otherwise. Some just want to make cool stuff in the game. Some just want to win prize money.

All of these types have something in common, though: It's about them doing, achieving, being, and winning. And being told to play in one regimented way is simply the antithesis of all that, regardless of who you are and what kind of game you like. Impositions get in the way of self-expression.

In my roleplaying days, players almost always defaulted to playing the same kind of persona. Though their characters may have had different backstories or costumes, they invariably became the same type over and over. They always ended up playing the tough guy, the creepy guy, the smart magic-using guy, or the rogue. Even if they couldn't literally be that person (such as in a game without magic) they still effectively carried on that way. This led me to think that players were not adopting in-character behavior (like an actor might), but rather that their character altered to become a projection of self.

The projection of self is very important to understand: When I say that a player adopts a role in playing a narrativist or simulationist game, for example, I don't mean they are performing. I mean that they see the job description that the game promises and that either tallies with an opportunity to self-express or not. They get to be the inner star footballer, or knuckleheaded barbarian, or vampire that they secretly fantasize about being.

This is also why archetypal fantasy races like elves, dwarves, and orcs feature repeatedly in games from Dungeons & Dragons to Shadowrun to Warhammer. Orcs are always orcs. Elves are always elves. They are signifiers of the opportunity to ego-project and to express self.

Games enable parts of our personality to come out by creating a safe other world with little or no consequence in this one. They engage our imagination because the world is there for us, on our terms, and if we want to leap tall buildings or solve puzzles or mow down bad guys we can. It's always intermediated through a controller of some kind (another creative constant that I call lensing), but in a sense that lensing heightens certain kinds of internal responses that we already carry with us. (A bit like how some people become different when driving a car).

In video games, I think this is also reflected in franchise loyalties. Much is made of the inferred complexity of popular game characters like Cole Phelps from L.A. Noire or Ethan Mars from Heavy Rain, but many of the really popular character-led franchises are blank slates. Nobody tries to make Mario or the Master Chief play "in character" with sets of extrinsic motivators, and they are hugely popular. EA does not mandate that when playing FIFA you play in the Rooney style, or give you Rooney points (good dog!) for doing so. Rooney is yours to do with as you will. Rooney is your doll.

My Niko Bellic is different to yours. We are not interpreting Niko Bellic in two different ways to try and decode him for an audience, like performers. We are turning Niko Bellic into projections of ourselves, dressing him as we like (or not) driving the cars we like, and solving missions in the way that we choose. My Niko is a murderous psychopath who mows down pedestrians; yours is careful and likes to avoid the police. All we are doing is turning the template of Niko Bellic into projections of our selves. He's just a doll.

This, more or less, sums up all games. They are literal, self-focused, self-expressed, mastery-driven, and bring out existing parts of our personality, rather than imposing one. Unrestrained by extrinsic rewards and good-dog behaviorism, self-expression rules.

So you get Leeroy Jenkins. You get this Facade video showing how it's often actually played, rather than how its makers hope it was played. You get Soulja Boy's review of Braid as shown in Indie Game: The Movie, which Jonathan Blow admitted he found very depressing post-launch, as almost nobody understood the deeper points he was trying to make. You get profane 14 year olds goading you on Xbox Live.

Sounds bad, doesn't it? Here's the good: Self-expression is also at the heart of creativity, discovery, exploration, delight, and emotional connection.

Emotional connections

In response to the piece I wrote about Tomb Raider, many commenters on Gamasutra responded that they have experienced emotional connection to their avatar in a game. My post was perhaps overly blunt when I said:

Perhaps how it should have been phrased was:

In childhood, most of us own dolls and we invest a great deal of time in them. Even as we grow up, we often hold onto one or two teddy bears or favored G.I. Joes, and for adults there is a considerable industry in action figures and figurines.

In fact, we develop emotional connections with objects all the time. It's normal, even healthy, to do so. We connect with cars, computer brands, favored cups for coffee, buildings, clothes, and so on. We name things, sometimes even talk to them (remember Wilson from Cast Away?) Sometimes we identify with special items for social reasons, such as treasuring a pair of Manolo Blahniks. Other times it's because of personal significance, like your dead grandfather's pocket watch. Perhaps it is because the object represents an investment of creativity, time or identity.

Taken in combination with projection and expression, object connection is (I believe) the correct way to understand what "player characters" really are. It's also why I call them dolls -- meaning a treasured possession.

The emotional connection to a Shepard or a Link comes from the way that they become us, we manipulate them, we dress them, and grow them, and we invest identity in them. We customize them, make them our own, and interface with a whole world through them. And we grow fond of them. Yet we are always aware they are not actually alive, and we are not them. We are their makers; their parents, in a sense.

Emotional connection becomes even more complicated when story is involved.

There are two Niko Bellics. There is the Niko of the cutscene, the war-weary criminal who feels that he must obtain revenge for a past wrong, help his cousin Roman and get involved in the happenings of Liberty City. This Niko is taciturn, wise beyond his years, and wry about the world around him.

The other Niko is the little psychopath doll that I control, the one I described above. The one who is my conduit to Liberty City and whom I attire as I choose. There are also two Laras, two Marios, two Shepards, two Drakes, and two Clouds.

There are times when they are characters and others when they are dolls, and the disconnect between the two can be quite odd. When a game reveals that the character version of my doll is not who I thought he was, for example, that can either be very clever or totally inappropriate.

When the game over-characterizes my doll (perhaps with ambient audio) to make him or her unlikable, or unlike the self-expression that I project into it, then that can feel strange. Sometimes in a good way, but often not. (Which is where my post on Tomb Raider came from.)

The duality of character and doll is perhaps most starkly illustrated by Heavy Rain. There are two Ethans. Character-Ethan is the grieving father having already lost one son, now tasked to find the other. His marriage is broken down, his life is a mess, everything he says or does is affected by a deep and painful sadness.

Doll-Ethan, on the other hand, is an android. He (you) wanders around his own house opening drawers to find out what's in them and talking to people (such as his wife) to find out who they are. He plays swings with his children, but it's a dislocated experience because he has no idea of his relationship to them. He talks to his remaining son in a playground like a machine, polling him with questions for answers. He even walks like an android, perfectly straight and turning clockwise or counter-clockwise on a dime.

Game makers like David Cage believe that the interplay between dramatic scenes and control strengthens the connection, in a kind of movies-plus-doing model, but my contention is that this is not so. Though lavish, Heavy Rain is modally no different to Jet Set Willy, and the same creative constant of self applies. Interposing duality mostly weakens the parental connection with the self-expressed doll and relegates it to play-time/story-time. "It's okay," says the game. "You just press buttons when you're told. I'll handle the emotional part."

And so you get interminable cut scenes which just don't seem to matter to the literal game. That's why (no matter how well written) a cinematic-story led approach to games always feels oddly cold. It's also why storysense works.

The storysense way

"Storysense" is an approach to narrative which relies on the creation of an interesting world, a discoverable set of threads and bits of story, a minimalist approach to goal direction, but dispenses with dramatic plot and character development. It treats story as a backing track to the play of the game, and so the player can participate or not as he likes. There is no time given over to extrinsically rewarding the player for being in-character, and the only rewards are literal -- just as the game is. There is no elaborate characterization, no attempt to insert unnecessary meaning, and no emoting at the player to try and make him or her feel.

Storysense is at the heart of successful pen and paper roleplaying games, where a good game master understands how to change up if the game is getting boring. Storysense is at the heart of virtual promenades like Dear Esther, where the relatively simple addition of a disconnected monologue elevates the experience of wandering around a desolate island without needing to pay exact attention to its sequence. The objective of storysense is to enhance the sensation of a world in motion. All expression and judgement is left to the player.

So there is only one Gordon Freeman, only one Executor from StarCraft, and only one "you" in Journey (although there are cutscenes, your nameless adventurer simply watches them). They are all just you.

In Half-Life you can lark about while conversations go on around you, or you can pay attention. It's up to you. In StarCraft you can get with the spirit of the mission briefings and elements in the world like the alongside dialogue of the units, or just play it as a sport. Again, it's up to you. In Journey you can participate with a fellow traveller generously, run through the game by yourself, or go about collecting every little power-up you can find. It's your business. In Deus Ex, you can stop and read every book or email and listen to every conversation -- or be an oblivious psychopath. It's your call.

The mistake is to then try and turn that into a movie. Half-Life 2, for example, traps the player in rooms with over-long conversations with key characters that go past the point of dull. StarCraft II employs many meant-to-be-stirring cutscenes and dialogue sections in between missions which add little to the experience other than time. Red Dead Redemption is plenty of fun, but then there are some scenes where it's all about characters reflecting on the hard life of the Old West, and they're all a bit random, really.

Though cutscenes can be informative or visually arresting when used well, using them for character drama is almost never interesting for very long, regardless of the quality of writing (The one significant exception to this is if the scenes are hilarious). Again, it's because of the modality of playing games. That's tough to hear for designers who want games to be like movies, but it's just the reality: create a big saga game with lots of emoting and storytelling, and most players will not bother to finish it.

Here's another tagline that annoys some people, but is true:

I believe that the evidence shows that there will never be a great told story in games, because they modally do not work. Ultimately, they are paradoxical. They only ever produce the need to either tolerate the game because you like the story, or vice versa. Either way we will always see actual storytelling arts like books or films streak very far ahead of games because telling stories is what they are all about, modally speaking. No amount of production budget or technology is ever Ever, EVER, EVER going to change that.

Rather than being dramatic, games demand to be something else: thaumatic. Storysense works because it treats your doll as a doll, as an interface or conduit into the world. It works because it allows the players to fully self-express. It works because it keeps the goals literal. Accepting that involves a big sacrifice: It means giving up on formal storytelling, and reducing it to the role of setting objectives.

The short cutscene that says "go here, do this" and nothing more is one example. The quest which is one line on a message board that you accept is another. The inherent goal that arises from the pressures of the game, such as a boss moment. The urge to run away from a scary noise down a dark corridor. The experience of playing a small GBA soccer game and going into that world to feel success. From the smallest to the largest games, this is the approach that works because it is game-native.

Great storysense requires a great doll and a great world, great opportunities for self-expression, and clear literal tasks. Then the game can earn legitimacy to create poignant moments. The moment in Journey when you are walking up the side of the bleak snowy mountain is one example: It's a journey you have made with your robed doll; you see it slowing down and the shrinking of its ribbon. You see the ice, foresee the end, and that thaumatic feeling grows: You are there. You feel cold. It is perhaps the end.

How much worse would it have been if it had transitioned into cutscene mode and started talking at you?

Children and the self-selected

Two counter-examples are often raised to the above points that should be addressed. The first is how children play.

Let's Pretend is a common example used to illustrate a storymaking experience. Kids will, through dolls or other games, construct stories. Even something as simple as a deck of cards can be a vehicle for meaning, such as playing Snap with my 3-year-old niece, wherein she would not let me play the Jack of Spades on the Jack of Hearts because they apparently weren't friends. And a 7 and a 3 could snap because apparently they had been married. Children do this sort of thing all at the time. While the play brain learns to focus on narrow and specific problems, the art brain learns to see the world as story and situation through fantasy.

Even into adulthood we continue to sort events into patterns that we understand, editing out the unimportant parts and so render factual memories into narratives (this is why witnesses in court cases are often unreliable). However this is different from a child's experience. Children do not necessarily understand the difference between reality and unreality, but (sane) adults do.

Although adults attribute character and attachment to objects (see above), they also know that a thing is just a thing and they are themselves. They do not actually think their favorite car has a personality, nor that their grandfather's watch is capable of magical connections. Children might really believe in Santa Claus, but adults know better. Unlike a child, an adult art brain is developed.

You might play a game of Space Hulk and imagine some vague back story going to the events of the game but it's unlikely that you find everything in the game numinous. Game makers have to work hard to make a "The cake is a lie" moment have its intended effect precisely because adults will not invest as much imagination as children do, and are not as easily inspired as children are. Because the adult gamer is expressing himself rather than becoming someone else, Portal has to work with that. So the cake is a lie to you. It signals danger to you. It creates the impression of a larger world around you.

The second example is the gamer who tries to play in character.

In every roleplaying game there are a small contingent of players who try to do this. In World of Warcraft, for example, there are 57 'RP' and 'RP-PVP' servers where the house rules maintain that players should act in-character. However, that compares to 465 'PVE' and 'PVP' servers where no such requirement applies. (source: Wowwiki.com)

Anecdotally, this tallies with my experiences. About one in every ten players wants to get into the persona side of the game. The other nine play literally. So, without dismissing the 10 percent, there is an element of outlier behavior to in-character play, and this is reflected in many arenas. The old MUSH scene, for example, was always an activity for tiny minorities and usually little more than a cybersex forum. Ditto Second Life.

Broadly speaking, in-character play falls into two camps. There are those for whom it is about mimicry. So when playing City of Heroes they might create their own Incredible Hulk, or their own Picard in Star Trek Online. The more adventurous types in this category might even be into activities like cosplay. They are the people whose in-game characters often take names and identities from other sources like Tolkien, anime, superheroes, or Star Trek. They self-express through hero worship, saying that Legolas or Naruto or Commander Data is them.

The other kind of in-character players are those who maintain that they do in fact try to act as the character. These are the kind of players that game makers who believe games will become something modally different look for. It is the makers' hope that those players will one day go from being the tiny minority to the large majority, and so validate games and storytelling.

This kind of player (let's call them "dramatic players") tries to play Max Payne while trying to be Max. He often says something along the lines of how he identifies with the character, feels for his struggle, and that he is guiding the character through the quest. He likely plays BioShock and finds the moral choices of the game affecting, or wanted to play Deus Ex without having to kill anyone at all. Or so he says.

I'm not sure he really does. In pen and paper and live action roleplaying games, dramatic players tend to act in character during moments where skillful play doesn't matter (such as having an enjoyable conversation). However when the levers of the game need to be pushed, they are become just as win-motivated as everyone else. In a sense, it's all well and good to speak like a noble in front of a dragon, but when the dragon is dead it's usually all about loot drop.

The theatric player wants constant in-character interaction to take place, wants to find it meaningful, and wants for their experience to be as deep as Shakespeare. However being in-character is often only a momentary thing, and even at that defaults to the same self-expressed archetypes again and again -- of always playing the elf, the dwarf, the chaotic good guy, and so on. Only rarely have I ever seen a player who fully gets into dramatic mode (such as making choices which disadvantage them because they're in-character). Even in the relatively social setting of roleplaying, this is uncommon. In video games I'd be amazed if more than 1 in every 1000 players regularly plays that way.

As a result, I think theatric players (who exist in all kinds of game) self-select. As the people who are supposed to find this kind of thing meaningful, they rationalize away the duality of their game experience. They'll write articles about how meaningful a game is supposed to be, calling it the Citizen Kane of video games, or expounding upon what the emotional connection between the player and the character should be. They are often very vocal, but behind them is a gaping void of reality.

No, Mass Effect is not the most important science fiction work of the new millennium. No, L.A. Noire does not represent a coming of age for games-vs-movies. In the cold light of day they're ham-handed games trying to be something other than their modality allows. (And secretly I think even the most ardent dramatic gamer knows this.)

So I'm wary of over-interpreting the play of children or the self-selected dramatic gamer. Instead I look to the experience of the median adult who plays games. What does he or she find special? What does he or she come back to and play again and again? Where does she find her 100 hours of gameplay?

The first time your mind is expanded by a game is a powerfully imaginative sensation. It's like the first time you see Star Wars, the first time you really understand the Mona Lisa, or the first time you hear Pink Floyd play. However, as with all forms of novelty and revelation, that is not the common experience of play.

The cinephile whose mind was blown by Star Wars eventually becomes conversant in the language of film and so finds them interesting in a mature way. He doesn't expect all films to be as inventive as Star Wars, and realizes that there is more depth to the form even if it is modally always the same. The art fan does likewise, as does the music fan. Cinema, art, music and so on have more complicated conversations with their matured fans than with their neophytes, again along those same modal lines.

So, too, with games. The indie game scene is full of games which have a complex conversation with the kind of PAX-attending fans who know what platformers and roguelikes are. They are the sort of people who see through the veils of Facebook games quickly, who form niches around specific types of game, and occasionally even become tribes. They are the matured played-100-games-already players.

For these audiences, all of the games they enjoy tend to play well. They are not reaching for explanations of how children play, nor extreme novelty, nor rationales on the basis of what games should or should not be. They are not always reaching out to the pages of Wired to find players who know little of games and so are likely to be enchanted by their novelty. They are as they are, and they have complicated conversations with their existing audience. The are examples a maturing culture, for whom there is a steady stream of interest beyond the first example and for whom the modality is just fine.

The dramatic player wants that modality to change, for the conversation to be different, and for the ratio of roleplayers to literal-players to change from 1:9 to 9:1. Sadly, it won't. For the rest, games are already an art and a culture as they are with their own internal values, such as play rather than show.

Play, don't show

One of the great lessons in film, which took a generation to learn, is "show, don't tell."

The earliest silent movies often used prompt cards and staged sets to tell stories. When talkies came along, the cards went, yet the sense of staging did not. In fact, "staged film" still survives in sitcoms and Bollywood. Staged films were choreographed and dialogue-heavy, and the transition into cinematic entertainment didn't happen all at once.

Starting with Citizen Kane and some others, and then evolving into the New Wave, filmmakers started to realize that they weren't making recorded theater. With little changes in technology, critical thinking, and some banner examples like Welles' work, they gradually came into their own. Staged films like sword-and-sandal epics or Westerns continued to be made right up to the '70s, but by then were dead. Directors like Kubrick, Scorsese, Coppolla, and Lucas showed what cinema could really do, and everything was different after that.

I see a similar situation in games, except where cinema used many of the conventions of theater, games use many of the conventions of cinema. We're passing through an era of "filmed games," just as film passed through its era of "staged films." And just as the lesson to learn in film was "show, don't tell," the lesson in games is "play, don't show."

What's the best-designed game of the last five years? Everyone will have a different answer, but for me it's Left 4 Dead. Left 4 Dead understands urgency and agency, clear goals and simple tasks, and variety. More than that, it has fantastic storysense, tapping right into that Dawn of the Dead apocalyptic zombie meme that is so popular these days. Even better is the way that it ties its gameplay into its storysense through limiting simple things like your ability to stand up when knocked over. And after playing a session or two you soon realize that you can't leave a man behind. You will be overwhelmed.

It's play, don't show. Paint the world and let me play in it. Let me self-express. Let me ignore your story if I want to. Let me experience the synergy of frame and fantasy all in one place. Let me believe. Your specific answer to that best-designed question will vary (another recent candidate for me is Journey), but I'll bet the quality of the experience you're looking for doesn't. That magic of games, the one you first encountered all those years ago is what you want to escape to, where you want to be and where you want your players to be.

Meaning is made rather than given. Belief happens when your game gets out of the way. Context is irrelevant. Characterization is irrelevant. Authored emotion is irrelevant. Regardless of genre or fiction, rule or role, emergence or experience, strategy or skill, enabling the self to play and be inspired is what games are. On their own terms, modally, for the player.

I would love to see BioWare build a role playing game in the Mass Effect universe which understands modality as well as Elite and Frontier did. I would love to see Quantic Dream take the android-idea of Kara and realize that that kind of blank slate would lend itself to a much more powerful game. And I would love to see both studios trying to do that without resorting to cinema or miring themselves in unneeded "Show." Staged films like Ben Hur seem quaint and bombastic to the modern cinematic audience, and filmed games are going that way too.

Conclusion

Dolls (avatars, player characters or whatever you prefer) are conduits of experience, of self and of self-expression. They are you, you act through them, and sometimes they become your children. They are precious. They are not characters.

As a maker of games, you get to pull a player into a world in which they can self-express. You get to sling ideas at them while you do, inspire meaning, and help them discover. You don't get to tell them what to think and feel, though, because the relationship is not really like that. You don't get to creep them out and then expect that to be regarded as art either.

The player is not a performer. He's not interpreting your tale or making up a story in his head at your direction. He is instead a seeker, a parent, and an artist. The world is there for his play brain to master, his lizard brain to fear, and his art brain to find meaningful. In providing him a doll, you transport him to this other place and time where the rules are different, the outcomes are more certain than life, and within that magical space things can happen.

From the simplest soccer game to the most massively complex world, he can win. He can create. He can self-express.

He can believe.