have been a game designer for nearly 20 years, mostly in the adventure
game arena. However, at the moment, I'm doing something completely
different, working as Creative Content Director for the online gaming
site, WorldWinner.com, and for the first time in my life creating
games that are pretty much devoid of story and character, which
is pretty ironic given this articles topic: "Building Character:
An Analysis of Character Creation."
First, some terminology 101. This is real basic stuff, but just bear with me for a minute for the sake of those who aren't familiar with these terms that I'm going to be bandying about for the rest of the talk.
The distinction between player-characters, or PCs, which are characters under the player's direct control, and non-player-characters, or NPCs, which are all the computer-controlled characters in the game. The main thrust of this talk will be PCs.
with the category of PCs, is the distinction between first-person
and third-person point-of-view, or POV. With a first-person PC,
you're seeing the gameworld through the PCs eyes; examples would
be Myst, or Quake. With a third-person PC, you can
see the PC as you move him or her through the game world; examples
would be Kings Quest or Tomb Raider. First-person
PCs are often styled to "be" the player, as if the player
were injected into the gameworld, in which case the PC is left characterless
in order to preserve the fiction that you are the main character,
and first-person games don't deal with the visual appearance of
the PC, so its more rare in a first-person game to have a PC with
a strong, fleshed-out characterization. Some games switch back and
forth, such as the Tex Murphy games, Under a Killing Moon and so
forth, which use a first-person POV for the gameplay, but use a
third-person view for the cut-scenes, the movie-like non-interactive
When we talk about creating a character in a game, we're usually talking about characterization, which is everything observable about a character: what they look like, sound like, how they move, how they dress, intelligence, attitude, career, and so forth. I'll be focusing on characterization for the first and longest part of this talk.
on the other hand, refers to what's underneath — the human
heart, the essential nature. I'll be dealing with true character
during the second part of this talk.
of Character Development
The first question to deal with is why is good characterization important? Of course, this is dependent on the type of game; if it's a real-time strategy game, for example, with its relatively distant point-of-view, and its large quantities of interchangeable units, characterization isn't that important. So for today's purposes, let's stick to games where character is important, such as adventures, role-playing games, action-adventures, platform games.
Of all the aspects of such a game — the geography, the inanimate objects, the music, the action sequences, the interface, etc. — the element that is most likely to leave a positive lasting impression on players are the primary character or characters. Humans are hard-wired to respond to other humans (or human-like creatures).
This point was driven home for me a couple of years ago, when my son went through a period of extreme interest in The Three Stooges. He bought himself a life-size cardboard stand-up of the Stooges, and kept lugging it around the house and leaving it in different rooms. I kept walking past a room and spotting it for brief moment, out of the corner of my eye, and just that glance would often cause the most visceral, startled reaction. This continued even after the damn thing had been around the house for months.
you're going to expect players to spend dozens of hours with a character
you're creating, at the very least you want that character to be
interesting, easy to identify with, and hopefully very likeable
as well. The more a player can get into the skin of the character
or characters they're controlling, the more the experience becomes
something that's happening to you, rather than something you're
doing. Also, a strong central character serves as an almost iconic
representation of the game, which is damn useful as a shorthand
for facilitating word-of-mouth, and is useful for all sorts of marketing
hooks; furthermore, a successful character is a good, perhaps even
the best, way to build a franchise.
Let's look at a list of some of the most well-known and successful game characters:
- Sonic the Hedgehog
- Crash Bandicoot
- Leisure Suit Larry
- Putt Putt
- Lara Croft
- Fatty Bear
- Duke Nukem
- Carmen Sandiego
- Gabriel Knight
- Guybrush Threepwood
Certainly not a comprehensive list, and I'm sure just about everyone here could come up with a handful of good additions to this list with half-a-minutes thought. But it's a good, representative list, with characters from a number of different genres; a mixture of characters aimed at children, adults and at a crossover audience; a mixture of males, females, animals, and various fantasy creatures. But what they all have in common is that they're the focus of their respective games, and have spawned sequels, in some cases many sequels in addition to spin-offs into TV shows, movies, books, card games, action figures, and so forth. In other words... a franchise.
the way, I'm only going to be dealing with characters originally
created for games; I'm not going to be talking about cases where
you're transporting a James Bond or an Indiana Jones from another
medium. That process, of course, has its own set of issues.
first step in creating a successful character, especially one that
you're going to hang a game on, is to settle on what we in the development
biz call the "high concept" for that character. High concepts
for some of the characters on the previous slide would be "a
cute talking car" or "a marsupial who's been genetically
enhanced by a mad scientist" or "a female Indiana Jones
with mammaries the size of Volkswagens". Remember the two things
you're trying to do with this character: make an enjoyable and interesting
character that a player will want to adopt into his or her life
for the next few weeks or months, and create a character that will
be different and memorable enough to help you cut through the clutter
of the several thousand other games that you'll be competing with
for shelf, magazine, and player-awareness space. So at this point
try to think, what's interesting? What's cool? What hasn't been
Naming characters is a massively important step. A good name is a big part of what makes a character memorable; it is often what gives people their first impression of what the character is all about. Often, that character's name will be the game's name as well, or part of it, so this is a good opportunity to take that power into your own hands rather than letting the marketing weasels bungle it a year down the road, unless of course we have any marketing weasels in the audience, in which case, I meant to say, my very good friends who so skillfully and valiantly pilot our games through the intricate complexities of the marketplace.
As with the high concept, a character's name should be interesting and memorable. In addition, it should be euphonious, pleasing to the ear, and rolling off the tongue rather than twisting it. It should fit the character. Studs Steelpike probably wouldn't be a good name for the skinny accountant who solves crimes with his amazingly logical mind, and Milo Twigbody is probably a bad name for the professional wrestler who becomes an ace assassin for the CIA. I think Duke Nukem, for example, is an excellent name — easy to say and remember, and which instantly creates just the right mental image.
J. K. Rowling, the author of the Harry Potter series, is a master of naming characters is. There's no doubt when you meet a Severus Snape or a Draco Malfoy that you'll meeting an unappetizing character, that Hermione Granger will turn out to be a studious know-it-all, that Percy Weasley and Cornelius Fudge will be prissy and self-important, that Peeves will be one extremely annoying poltergeist.
a fun exercise to think of the names of successful characters from
various media, and notice how well their names conjure up the right
initial mental image — the solidly strong James Bond, the sinister
Darth Vader, the human-doormat Arthur Dent, the mischievous Bugs
Bunny and the everyman Homer Simpson.
Before you start developing a character, you need to know and thoroughly understand the character. The best way to do this is to write a background paper for each character. This can be just a paragraph or two for minor characters, and several pages, even 10 or 20 pages, for your main character. This is really important. It doesn't have to be in narrative form; lists are okay, and you should include stuff like:
- where was the character born?
- what was his or her family life like as a kid?
- what was his education?
- where does he live know?
- his job
- his finances
- his taste in clothes, books, movies, etc.
- favorite foods
- favorite activities
- personality traits, and how they manifest
- shy or outgoing? greedy or giving?
- what were the traumatic moments in the character's life?
- what were his biggest triumphs?
- important past romances
- current romantic involvement or involvements
- how does he treat friends? lovers? bosses? servants?
- political beliefs, past and present
- religious beliefs, past and present
- interesting or important possessions
- any pets?
- unusual talents
- what's the best thing that could happen to your character?
- the worst thing?
- tea or coffee?
- paper or plastic?
The list could go on and on. And, if your character isn't human, your background has to go a lot further, explaining exactly what, in your universe, it means to be a hobbit, or a Jedi knight, or an outcast half-orc half-troll, or what it means to be a robot warrior with a malfunctioning ethics chip.
You've got to know everything about the character, become the world's biggest expert on them, even if you end up creating 10 times as much background as you'll ever use. That way, once you start figuring out what your character will do in a given situation, you won't have to figure out, you'll know. And your players will know you know, even if it is just subconsciously, because they'll see your character acting and reacting in be real, natural ways; don't do this, and your character will be a shallow cliché.
Here's an exercise. Once you've written your background for a character, try to think of a dozen mundane or not-so-mundane situations, and say to yourself, "How would my character react in this spot?" How would he react when stuck in a traffic jam while late for an important date? When passing a panhandler? When confronted with a stray cat? When a comely co-worker makes a come-on? When pushed out of an airplane at 15,000 feet? If you've done your background development well, you should have the answers to these questions without even thinking.
I recently saw the movie Jaws for the first time since it was first released, and there's a terrific scene in it, when Roy Schieder, Richard Dreyfuss, and Robert Shaw are in the boat's galley, after a hard and futile day of chasing the shark. Dreyfuss' character, the naturalist shark expert, and Shaw's character, the hard-bitten shark-hunter, are trying to one-up each other, with near-shark experiences, rolling up their clothes to show off one scar after another. It's a wonderful scene, and the kind of writing that comes easily when you've done your homework, but probably not at all if you haven't.
Concept Art: Initial Design
Unless your character is a first-person POV character, or unless you've decided to eschew any hope of profitability and write a text adventure, a big part of characterization will be creating the character's physical appearance. And that means working with a concept artist.
Maybe you're thinking that your modeling artists are good enough to create without benefit of concept art. Maybe you're thinking that you don't have time in the schedule or dollars in the budget to afford concept art. Well if you're thinking that, think again. Concept art will save you time and money, big time. It's a lot faster, easier, and more painless to work out everything at the concept art stage. It doesn't mean that you'll never end up creating finished art that you have to throw away or redo, but it'll happen a lot less. And a good concept artist will bring a unique vision to the realization of the character that will, in many cases, be far more interesting and exciting than what you were picturing in your own mind and trying to articulate to the rest of the team.
Here's how this phase of concept art might go. This sequence is from a game that my development company, Boffo Games, was working on for Time-Warner Interactive about 5 years ago, a game that was never completed after TWI was reorganized out of existence. The working title was Reverse Alien, about a human who lands on a planet where the natural objects, the building materials, and the native race are all so fragile, that the human is to them as the monster in the Alien movies is to humans. The native race were called the Feebies, and I started by drawing this sketch:
As you can see, some of us need a concept artist more than others do. I'm like the ancient Egyptians — I never discovered perspective. This sketch, along with supporting text documents, went to the concept artist for the game, Les Nelken, who's currently an artist at Turbine Games.
Les produced a whole batch of sketches, variations on this theme. From those, I selected the head (right) and the body (left) that I thought worked the best:
Les took that and produced a new generation of sketches:
These then went to the art house that was doing the final art, a very talented company in San Jose called Dub Media. And here's a still from one of the few animations of a Feebie that was created before the project was consigned to that big hard disk in the sky:
Right after the Reverse Alien disappointment, I entered Concept Art heaven. The next project that we did at Boffo was an adventure game called The Space Bar, funded by Rocket Science. The concept art for the game was done by Ron Cobb, who was a founder of Rocket Science, and who had done art direction for numerous games and movies. Working with him still ranks as one of the big thrills of my career. The Space Bar was set in a spaceport bar filled with all kinds of pretty wild alien races, so we weren't just creating new characters every week, but entirely new races. One such race was a race of mobile plants, called the Vedj. Again, we started with one of my superb sketches:
Ron then produced this piece of concept art, which we decided wasn't "plant-like enough". A round or two later, he came back with a second image, which was perfect:
Ron's first piece of concept art of a Vedj (left), and the final draft (right).
As you can see, with each go-round, the sketches get more and more detailed as you realize you're getting closer to the final form of the character. And here's an image from the game itself of a Vedj, named Seedrot, seated at a table by the dance floor in the bar:
Now, sometimes the evolution of a character isn't dictated by creative reasons but marketing reasons. Here's an example from a platform game, for the Playstation 2, that we started working on for THQ when I worked at a THQ-owned developer called GameFX:
The game, with a working title of Smartacus, was about a 12-year old super-genius with an evil bent, his 6-year-old tomboyish sister and their adventures through space and time. The concept artists for Smartacus, was Richard Sullivan.
With each generation of concept art, we made her older, and turned her from Smartacus' little sister (left) into a classmate(right) of his.
THQ decided that the target users of PS2s, at least during its first year, wouldn't be interested in a 6-year-old girl, even if she was only a secondary character, and even if she could beat up grizzly bears without breaking a sweat.
By the final generation of concept art, she was even taller than Smartacus.
But to no avail; THQ killed the project at this point.
Now, just to show that the initial design phase of concept art is not necessarily a long process, here's one of many examples of getting it right the first time. Again, I turn to The Space Bar and the art of Ron Cobb. In the bar, serving as a fairly minor humorous diversion, was a race called the Fruufnids. They were less than a foot tall, stood on a table in the bar, and strongly resembled Polynesian drinks, when in fact they were high-level ambassadors from a very powerful planet. They were constantly being picked up by patrons, mistaking them for drinks, and were completely clueless why it was happening, and wildly indignant about it.
The first concept sketch from Ron (left), and an image of the two Fruufnids from the final game (right).
The second part of the concept art phase are the model sheets, a term that comes to us from traditional cel animation. I'm going to use some traditional cel animation model sheets as examples, partly because the project scope and overall tone of games is similar to classic cel animation, but mostly because model sheets from classic cartoon animation is much easier to come by than model sheets for contemporary game characters.
Here's someone I think we all recognize. Like the best model sheets, this one features many different poses and expressions, and allows those creating the final art to have a "bible" that they can refer to, so that original vision will be closely and uniformly followed. This model sheet of Bugs is from the late 1940s, nearly ten years after the character first appeared. It shows how, even after that many years and dozens of cartoons, the animators at Warner Brothers — in this case, director Robert McKimson's team — were still trying to improve and perfect Bugs, find new manifestations of his wise-guy attitude.
This model sheet (left), directed for MGM by Tex Avery, is from one of the great cartoons of all time, King Size Canary. Closeup of the model sheet (right).
This model sheet, directed for MGM by Tex Avery, is from one of the great cartoons of all time, King Size Canary. In it, a mouse, a cat, and a canary discover a bottle of "miracle grow" and use it to keep out sizing the others, with whoever's largest doing all the chasing. This model sheet, shows one of the most useful aspects of model sheets, as seen here in the close-up.
This model sheet (left), directed for MGM by Tex Avery, is from one of the great cartoons of all time, King Size Canary. Closeup of the model sheet (right).
Good model sheets will be filled with useful notes to the artists who'll be using it as reference material, laying down the rules that govern the character's appearance, expression, poses, movements, etc. A note for Mickey Mouse, for instance, might point out that his ears always appear as perfect black circles, no matter what angle you're looking at him from; a note for Marge Simpson might detail the proportion between Marge's height and the height of her hair, or what happens to her hair when she wears a hat, or goes swimming.
Here's an example from The Space Bar where a better model sheet would have saved us some grief. The native race of the planet where the bar was located were called the Marmali, a somewhat lizard-like race that, like kangaroos, use their large tail almost like a third leg. Once production art of various Marmali characters began to appear, Ron — who's something of a perfectionist — noticed that some had three toes on each foot, and some had four toes on each foot. And, sure enough, the number of toes was indiscernible in the concept art that the artists were working from:
So Ron hurriedly produced this additional model sheet which clarified a few details and which showed, beyond a doubt, that Marmali always have three toes:
had this model sheet been done at the time of the original concept
art, a lot of unnecessary work would have been avoided.
The third phase of concept art, and the only phase that doesn't necessarily come before the start of production art, is storyboarding. These include a sequential series of images of any type of movement.
A storyboard can be just a few images from the sequence, as in the case of a simple or not-too-important sequence. But in the case of a movement or action sequence which is very complex, or an important showpiece, like this animation from one of Tex Avery's Droopy/Wolf/Red cartoon.
A storyboard can be just a few images from the sequence, as in the case of a simple or not-too-important sequence.
want a highly detailed storyboard, which shows the position of the
characters and/or objects as frequently as every few frames. A storyboard
this detailed can actually be stacked and used like a flipbook.
I once heard a lecture by Chuck Jones, and he related a great story
about animation storybook flipbooks. The Warner Brothers cartoon
studio, Termite Terrace, was being visited by the producer
of the cartoon line, Leon Schlessinger. Like many managers of creative
and technical types, he was somewhat insecure and cowed by his own
lack of creativity expertise. Often, on these inspection tours,
he had seen animators pick up sheafs of storyboard sheets and flip
them on their arms to see the animation come to life. On this tour,
he wanted to show that he was on top of things, so he picked up
a shooting script for one of the cartoons, an all-text document,
put it on his arm, and flipped through it several times as if it
were a flipbook, nodding knowingly. Isn't it great how the spirit
of Leon Schlessinger is alive and well in many of our own industry's
In many of the kinds of games we're talking about, especially platform games and action-adventure games, one of the things that most defines a character is his or her moves, the actions he or she can perform. Just about every character can perform the basics — walking, running and jumping. What starts to separate characters is how they perform these basics — remember what Woody says to Buzz in Toy Story: that's not flying, that's falling, with style! Make sure that when your characters run and jump, they run and jump with style!
And then, you can build further by adding the more interesting moves — Lara's tuck-and-roll, Crash's hand-over-hand dangle from a mesh ceiling, Mario's butt-whomp. Unique moves can define a character's personality, and help them stand apart in a crowded field. One of my favorite sets of moves from the last few years is Banjo-Kazooie, a character-pair, where sometimes one, and sometimes the other performs the moves, including my favorite, the chicken run up steep hills.
Moves shouldn't just be limited to actions the player performs. Crash Bandicoot does a few things, which endeared him to me and, I thought, almost brought him to the level of some of the classic cartoon characters. Many of his death-moves were extremely well done and cute, such as his death by fire, where he becomes a pair of eyes in an outline of charred bandicoot, which almost immediately collapses into a small pile of ash where the two eyes fall onto the pile of ash, blinking in surprise. It's an animation worthy of Wile E. Coyote. Or when you go too long without giving Crash a command, he reaches behind and pulls out a wumpa fruit, and tosses it into the air, appears to lose it, and then when he looks away it appears and conks him on the head.
Dialogue: writing the words that go in the character's mouth. Your writing may be the final form of the dialogue, if it will be displayed as text, such as in an online game, where the size of the datastream is an issue, or a cart-based game, where the overall data size is an issue. Or, you might be writing a script that voice actors will use to produce recorded voices.
Try to find an interesting manner of speech for the character, which is consistent with who the character is. For example, I once heard a talk by Isaac Asimov, in which he was talking about the writing of his first "robot novel", The Caves of Steel, which is about a pair of detectives, one human and one a robot with human appearance. Asimov was trying to find a speaking style for the robot, R. Daneel Olivaw, and hit upon the idea of having him never use contractions: "I do not think we should go there." This was extremely effective in making him seem robot-like, since humans rarely avoid contractions. Asimov pointed out that this trick was also used for Spock's dialogue on Star Trek.
Another example is the character Claude Rains plays in my favorite movie, Casablanca, chief of police Louis Renault. He is constantly producing suave and witty lines even in the most pressure-filled situations; when Major Strasser asks him about the investigation of the murdered Nazi couriers, he says, "Realizing the importance of the case, my men are rounding up twice the usual number of suspects"; when Rick is holding a gun to Renault and points out that the gun is aimed at his heart, Renault says, "That is my least vulnerable spot."
About 15 years ago I wrote a game called Leather Goddesses of Phobos, in which your primary sidekick character was named either Trent or Tiffany, depending on whether you were playing the game as a man or as a woman. This character was a massively overenthusiastic, gung-ho, can-do personality. When writing Trent and Tiffany's dialogue, I always mentally imagined every sentence ending with half a dozen exclamation points, even though they rarely appeared in the actual text; this helped me maintain the right tone for their lines.
in his talk earlier in the conference on story elements in computer
games, Bob Bates pointed out an excellent way to use dialogue to
make characters memorable and underpin their personalities, which
is to give them a catchphrase. Schwarzenegger produces one of these
in just about every one of his movies: "I'll be back"
or "Hasta la vista, baby". Another good example is Robin
from the Batman TV show, with his "Holy (whatever), Batman!"
Of course, Bob rightly pointed out to be careful not to overuse
it, which is easy to do in a game environment.
Voice characterization is a fantastic way to get a lot of bang for your buck. Even if you line up some very talented professional voice talent, you'll be spending a fraction on vocals than you will be on graphics. And the human brain is equally attuned to audio and visual signals, so voice characterization is an excellent and not very expensive way to telegraph personality to the player.
Voice characterization is particularly important if your main character is a first-person character, and is rarely or never seen.
I think that Nintendo did a great job with Mario's voice in Super Mario 64, which is proof that you can get a lot of mileage from a little bit of audio. It's a cartridge game, of course, so there's not a lot of room for sound files. In fact, apart from his opening greeting, "it's me, Mario!", I'm not sure he actually speaks in the game. But the game is loaded with grunts and whoops that perfectly paint exactly the right audio tone for Mario who, in my book, is a "roly-poly guy who's had too much caffeine, too much sugar, or both".
another great Leon Schlessinger story that I heard at that Chuck
Jones lecture. They'd just created the first Daffy Duck cartoon,
and Mel Blanc, who did all the Warner Brothers cartoon voice characterizations,
was having difficulty coming up with the right voice for Daffy.
Well, as a joke, he did Daffy as a slight exaggeration of the lisping
Schlessinger, and everyone was in stitches, so they decided to go
with it. But then, as they were nearing completion of the cartoon,
they realized with horror that Leon would be screening the completed
film, as he did with all of the studio's cartoons. When the day
for the screening came, the team sat frozen with fear, sure that
they were all about to be fired. The film ended, the lights came
up, and Leon turned to them and said, "That wath great! But
where'd you guyth come up with that wacky voith for the duck?!"
Now we're going to move from characterization to true character. People have been telling stories in one form or another for thousands of years, and certain principles about the role of characters in a story are well documented. Here's what Robert McKee says about true character, in his book Story: Substance, Structure, Style and the Principles of Screenwriting:
"True character is revealed in the choices a human being makes under pressure — the greater the pressure, the deeper the revelation, the truer the choice to the character's essential nature."
I took McKee's three day course on Story Structure a couple of years ago, by the way, and I recommend it strongly to anyone interested in story-oriented games. Even though the course is oriented toward traditional linear screenwriting, and not game design, there were many rules that translate very well to our medium, and many thought-provoking moments that made me look at game design from some very fresh perspectives.
So, if true character is revealed by the choices a character makes, under increasing pressure, what does this mean for characters in games? In books or movies, a character makes decisions and we follow along with those decisions and learn about the character as layer upon layer of his facade is stripped away.
However, in a game, the character doesn't make his or her decisions — we, the player, make the decisions for that character. So, characters can never be observed to react to circumstances and through those reactions, reveal their inner selves. Sure, NPC characters, under the games control, can make decisions, and thus the potential is there for some classic character development. And you can have some limited story-telling by halting the game to play non-interactive cut scenes in which the main character's decision-making is taken out of the player's hands, allowing you to perform sporadic revelation of true character. However, it is pretty universally agreed that all but the most sparing use of such non-interactive sequences is terrible game design.
Furthermore, the spine of a story is often driven by a tension between the main character's conscious and unconscious desires. However, where a character's desires are the desires of the player-manipulator, how can such a tension exist?
So, are games hopeless as a medium for character development, and therefore destined to always be a weak medium for storytelling? Perhaps, in saying so, we're failing to acknowledge the power of interactivity, the power of putting decisions into the player's hands. Because, even if player-characters cannot make decisions under pressure to reveal their true character, players can make decisions under pressure, and perhaps by doing so reveal aspects of their own character to themselves. Alternatively, perhaps, even more excitingly, by making discoveries about themselves, players could even be a changed person by playing a game.
I don't think this has been done, I'm not sure anyone has ever even attempted this, but I think it's possible. I know it's something that I've never tried to do. I once did a game called Stationfall, which was a sequel to my first game, Planetfall. In both games, your sidekick is an affectionate robot named Floyd.
In Stationfall, you're on a space station, which has been taken over by an alien doomsday device.
In Stationfall, you're on a space station, which has been taken over by an alien doomsday device, a small machine that takes over all the machines in its vicinity and turns them into people-killers, and turns them into a factory for manufacturing new copies of itself, which are then sent off to spread like a mechanical plaque. After a few days on board the station, Floyd disappears, and you don't see him again until the last scene of the game, where you discover him, in complete thrall to the alien device. He is the only thing standing between you and the device, and you have only seconds before it launches its deadly copies and spells an end to all human civilization. You have a choice — kill your friend Floyd to get at the alien device, or condemn all of humanity to death. It sounds like a classic choice under pressure that would reveal true character — except that it was really no choice at all. You could kill Floyd, and win the game, or launch the alien copies and lose the game. I, as the game designer, made the choice, not the player.
However, I think there is one arena where we might be seeing a glimmer of what I'm describing. All the games I've been talking about so far today are single-player experiences. When you make decisions for your character, you know that you are not affecting anything besides your own private, personal game universe, and only computer-controlled NPCs will be affected by your choices. Therefore, as a player, you make choices, which, like in Stationfall, the dictator/author has mandated are the decisions you should make to succeed.
However, what about multi-player games such as the huge persistent worlds of Ultima Online or Everquest? Here, players know that what they do may impact another human being, perhaps profoundly. Here, for the first time, players have to weigh moral issues as well as gameplay issues. Just a week or so ago, I was dining with a friend of mine who is a lead designer at Turbine, and she was relating her experience with Asheron's Call. She quickly reached a point in the game where she became less interested in advancing her character, and more interested in helping newer players to succeed. She said that she was surprised to discover how nice she was as a character in the world of Asheron's Call...and, upon reflection, this made her realize that she was a nicer person than she'd realized.
It's nothing earth-shattering...yet. But it's a beginning. Characterization is good; it's important. Do it well. But think about character too, because that just might be the key to the future of interactive storytelling.