Games have a bit of an annoying conundrum that hampers writing for it. Too many cutscenes, and it's not a game, or a game that blue balls you with stuff that you never actually get to do when you play.
But gameplay, at least popular gameplay seems to pretty much involve action. Whether it's stomping on a psychedelic turtle or shooting the shit out of a fat cop's beer gut, or....uh playing Rez, whatever the hell goes on there.
As a narrative, this has obvious limits. I'm trying to think of a way games could have a narrative that doesnt revolve around variations on one basic concept (usually get out of here, or kill all these people, or get here and kill these people before the other guy does)
People are pretty eager to compare GTA 4 to The Sopranos. While you could argue that it's written so well compared to 3 (and most other games but thats a moot point) that it is closer to The Sopranos than other games....That doesn't give it an iota of the depth of feeling, the penetration of the subconcious, the layers of black humor and irony, the establishment of themes.
A game where you played as Tony Soprano living his life wouldn't be fun. It's hard for any game to be true to life, which inhibits artistic potential- not to the point where games aren't art, but it does seem like it's going to be hard to advance much past here, at least artistically.
Posts
i think GTA is written excellently as well but what it lacks in its episodic nature is the real cohesive feel of 'this is the story we're telling, this is how we're going to tell it.' what it amounts to is a vast swathe of entertaining, snappy bit-pieces which don't really combine in the way they should. don't get me wrong, i think Rockstar are trying hard and if anyone's going to get there through this route it's them, but at the moment i think there's still work to be done. if anything i think they could (and probably have) learned from the Warriors, which had a similar style but the real underlining narrative thrust of a set timeline - when you got to the end it felt like the whole thing was working up to the finale, rather than being stumbled upon accidentally, as sometimes happens with the GTA games.
It seems like you are complaining that modern gaming is too much about alternating between action segments and wildly different cut scenes, and there's a bit of a disconnect there.
Maybe you're saying you want to do the sort of things that you normally see in cut-scenes. So lets say you've just finished a big gunfight in a game. Instead of watching your character talk to other characters to forward the plot, you're saying you want to control what dialogue he says, possibly with branching options that affect the plot. Heard of Mass Effect?
You also give the example of Rez as part of your problem, a game which doesn't really have cut-scenes and I would not consider under any means a narrative work. It has a rudimental skeleton of an abstract story, yes, but it's first and foremost an exploration of synesthesia.
You also seem to be complaining that "popular gameplay" seems to include action. Are you aware that there are genres other than action games? Yes, the majority of games with narrative structures are action games. If you ignore RPGs, I guess. Obviously action games are going to have plots that revolve around action. For a counterexample, look at Portal. It's a puzzle game with a narrative, and a fantastic one at that. Simple and understated, but there is a lot you can analyze and get out of it.
I don't know what your beef is with the GTA4/Sopranos comparisons. Both are works of satire that explore themes about the seedier side of modern American culture. Of course they are going to be inherently different, because you cannot approach a medium such as television the same way you can an interactive medium like video games. No, GTA4 is not perfect, but I would not say that it is lacking in depth. For me, the artistic merits of GTA4 come from the little moments. Niko's looks of disgust and dissatisfaction after finishing up with a hooker. The drunk driving effect. As an observer it is not as effective as watching an episode of the Sopranos, but you are not meant to observe GTA4. You're meant to play it. And to me, the increased emotional involvement that comes from interacting with the world makes up the difference.
You should also check out Facade (http://www.interactivestory.net/). It's very rudimentary, even to a fault, but it sounds like what you're looking for.
Other games don't really take advantage of the fact that they're a game at all, and instead, you play for a while and then stop playing in order to watch part of a movie, which isn't connected to the gameplay at all other than both the movie and the game use the same characters. This leads to stuff people jokingly point out a lot, like how Leon Kennedy can do all sorts of crazy martial arts in the cutscene but can't even jump when the player has control of him.
The player tends to have the ability to undermine the character that the cutscenes are trying to establish. As a random example, Niko Bellic wants to start a new, mostly clean, life in America. In the first mission in which you have to kill someone, Niko dejectedly says "Oh no! I swore I wouldn't kill anyone in this country!" By this point, I had already killed like 20 people. Niko Bellic's character is inconsistent. When he's under my control, he's a bad-driving murderer who will kill someone just to see if they can manage to escape the police after doing so. When the game has him, he's a likable guy who's unfortunately been put into a tough circumstance. The writers forgot to accommodate for the fact that I get to decide what kind of person Niko Bellic is.
I'm a writer in that to some extent I get to decide what the character does. I'm a character in the sense that I am experiencing the game's narrative via the player character. And I am an audience member in that I'm observing the story from an outsider's perspective. A game that does not acknowledge and cater to all three of these aspects is most likely, if not always, going to fall flat as far as being a work of "art."
http://www.audioentropy.com/
I'm trying to get at how I wish that the extent to which games can aspire was as vast as other mediums. It's easy to say what's wrong with a game wanting to be a game. There's nothing wrong with that. But I don't think its prudent to say games will always be closer to hitman than le samourai. But something has to change, fundamentally, and games have fundamentally changed over the years. So far at least.
This was a real problem with CJ though. He was too nice and too much of a victim/pawn
Tommy Vercetti was an asshole who deserved all the shit that happened to him, and CJ was everyone's bitch. Niko is by far the most complex, likable and (most importantly) believable character I've seen in the series. I would argue that the GTA series has (so far) proven itself to be a good measure of where games have come as a legitimate form of narrative, and where they are going.
If you can control the 'cutscenes', then that neatly sidesteps the issue of the game turning into a movie.
Gaming is art.
....in DMC 4 you see Nero do something with his sword where he basically does the DMC3 gunslinger wild ride thing where he rides an enemy around the arena only he's using the chainsaw edge on his sword stuck through the enemy into the ground to accelerate and control his momentum.
You never ever get to do that move in the game. I was so fucking pissed when I found that out.
Design is Law.
really? see this is my biggest problem with game narratives at the moment - they can't decide whether they want to tell you a story or whether they want you to tell one yourself. if there's no authorial intent behind the decisions of the protagonists - if they are arbitrary, and based on only a button push - where can the meaning be? how can the writers be treating that decision with the necessity and importance that the actions of characters in any other traditional narrative hold?
if you want us to tell our own story, give us such wide worlds and choice of direction that the storytelling can't be summed up by polemic morality choices or simplified 'branches'. give us worlds where we can create our own drama in a way that we would want to show our own experience to others, and depict it in a way that is singularly powerful. don't fuckign give us a binary choice whose only reprecussions might be surmised in an indented paragraph in a walkthrough or FAQ.
if a game writer is telling a story, they should take full control of it, and give you control of the irrelevant (but fun) action. if they want you to tell it, they should blow the world open wide enough that they can't possibly guess the story you'll tell.
that may be, but there's a big difference in kinds of art. i guess here we're talking about narrative and literature - telling stories in a way that's meaningful and impacting. that's an 'art', and it's one that videogames debatably haven't quite reached yet. i don't think mario guiding himself through to some remotely familiar tunes in a rom hack has anything to do with it
There are multiple games out there that do this, and do it well. "Games can not be art" is nothing but snobbish wankery that ignores why art exists in our society in the first place.
NNID: Glenn565
Actually I'd say a few developers have already done things with narrative in videogames that you couldn't really pull off as well in movies, even before the days of 3D.
Sure thing.
Good artistic literary narratives are about evoking thoughts and feeling in the observer....
Wait...this argument feels similar. That's because it is. You yourself stated your conditions for a succesful literary narrative in a game: "meaningful and impacting".
They're not different at all by your own definition of what make a good literary narrative.
Now I get the feeling you want to analyse the literary devices that games use, which is fine, but I was responding to your comment about games not using literary narratives in ways that evoke emotion and reflection in their audience, something you initially brought up. Just because you think they are unrelated concepts does not mean they are unrelated...they all fall under the spectrum of the arts. Dismissing it dismisses your own comments.
In any case, under your meaningful and impactful definition, there are games out there that do this, such as SOTC or half life 2.
I'll admit half life 2 is not very high brow intellectually stimulating, but the way in which they have shaped the world to reflect the first person view they limit the player to is a study in and of itself, and it works extremely well across a broad spectrum of audiences to create the emotional effect they wished to evoke. It is not deeply impactful per se, but it is a very good example of clever adaption to the medium at hand. And in some cases people have found some of the parts of the story evoke powerful emotions, if not much thinking.
SoTC on the other hand, is a very clever game that evokes powerful and mixed receptions amongst the people that play it. Many people feel a sense of guilt at killing the giants. Others have reactions of extreme aggression or justification "It is for the girl." A discussion of SoTc will inevitably have a post wherein one player mentions while playing the game he/she began wondering about morality or ethics, extremely complex cognitive concepts that are uniquely human.
The game starts with quite possibly the most base story cliche in existant but then transforms this cliche into something that evokes powerful and complex cognitive and emotional reactions in its audience through it's extremely clever use of the medium.
If you want to argue wether or not games can use literary narrative and be "meaningful or impactful", it would probably be best to start by tearing down the strongest counter examples to your claim, and not a throw away game like mario which never attempted to use literary narratives in a meaningful way in the first place.
Incidentally stating that 99% of games are mindless drek is not very helpful, as you can say that about every branch of the arts. You didn't say that so this isn't addressed at something you said, but it's still something to keep in mind as the discussion progresses.
Like ghost recon. Story has to be about a generic tyrant and America saving the day. I can imagine Axe would have pulled their sponsorship had the game depicted American involvement in Latin America more realistically.
It's hard for me to express. It was such a powerful feeling that I hadn't had before.
The first Metroid Prime had almost zero traditional story telling. Rather, the story and what was going on was unveiled by scanning and examining your surroundings. If you wished, you didn't even need to bother with the plot and just play through. The act of discovery there was very powerful, and it was enhanced by the solitude the player had. It was just you (as Samus) thrown on to a strange alien planet and left to unlock it's mysteries.
Even though I love it for it's gameplay, I was very disappointed that Metroid Prime 3 told it's story with cut scenes and a cast of supporting characters. Yeah, I could get the flavor text, but it's impact was diminished greatly.
SOME kinds, yes. Post-impressionistic art, definitely. But some are isn't about feeling, or expression. Impressionism? Modernism? Abstract? Post-modernism? What about Duchamp's The Fountain, or John Cage's 4'33"? Art isn't always about emotion. The line between high art and low art has blended since the 1890's.
There's a fine line when doing this sort of thing. Mass Effect did this perfectly, but also stumbled for the same reason. You felt like you were in complete control of every conversation, and they all flowed perfectly. Until you realize that it never mattered what you picked. Shepard always said the exact same line regardless of whether you picked "Please tell me more!" or "Spit it out before I shoot you!" and that there were only a handful of decision that actually affected the story.
I don't really know where I'm going with this. It's a good system to use, but you have to be careful and not make it too obvious that your actions aren't having any effect whatsoever.
shadow of the colossus might be different, but i haven't played it, so i can't really comment. immersion is probably also a huge factor there, but also the stunning visual design , if it's anything like ico. what it may well do right is use these giants as a metaphor for something deeper brooding under the surface of the protagonists. that's artistic storytelling, and it's using literary devices to convey a specific meaning - rather than the other spokes of gaming art, such as, yeah, immersion, choice, attachment through control. i don't care about those things. i'm a writer, not a game designer - i'm interested in this debate because i'm interested in storytelling more than i'm interested in games themselves.
so again: be meaningful, be impactful, emotional, creative, resounding, whatever you want to call it. games are going about doing it in lots of ways, and you're right, a lot of the time they're succeeding. but i think at the moment there are problems with the way they're going about doing it through story, and that's the topic of discussion here and that's what i'm interested
also the mario thing was a throwaway comment and i never mentioned any gaming dreck ahh work is busy back in a sec
In theory, there's no reason why you couldn't have a plot where every decision that you make matters. It's just hard to design, since there's so much that such a game would need to account for.
Thinking about it, the Geneforge series might be an interesting study in how to go about something like that. The games give a good deal of control to the player, both in the strategy used for tackling the game (you can actually complete Geneforge 4 without a single true battle) and in the ability for the player to choose the plot, with multiple factions opposing factions that change their views of you based on how you help or harm them.
Now I have to fail my final tomorrow because OH MY GOD.
Well, I suppose that requires a look at different kinds of conflict, as conflict is the only way anything interesting happens in a narrative - without conflict, every story instantly becomes a complete snooze-fest. However, I get the feeling that other forms of conflict require something that cannot be perfectly simulated by a computer.
Let me explain. In any conflict, there are set conditions to succeed. Physical conflict makes those conditions obvious. You punch and an opponent tried to dodge - did the punch connect, yes or no? If yes then the punch succeeds. Simple. Physicality simply asks "does this coincide with that?". Did the timing of the jump coincide with the ledge falling? Did the trajectory of the bullet hit the target? Everything the player does physically has a straightforward measure of success, which can then be translated real-time into the appropriate reaction. The punch knocks out the opponent, the player jumped off of the ledge in time, and the bullet missed the target so the bad guy got away.
These interactions are so simple and straightforward, that generally they can be done with ease - jump with A, fire with B, punch with X, and so on. Such simplicity leads itself to more flexibility - Within a span of a few seconds, I can jump, punch three times, fire two bullets, punch someone else, fire a bullet at a third guy... Every action not only has obvious succeed/fail conditions for the computer, but for the player as well. My punch won't work if he's blocking, my bullet won't connect if he dodges, and my jump won't work if I don't time it right.
Other forms of conflict are not so easy, though. Take intellectual conflict, for example (which would be a lot easier to simulate than something as fuzzy as something like emotional conflict). You are arguing with someone about global warming. You say it exists, he says it doesn't. Go. Hmm... well, how do you know when you've won? How do you even know what winning is? Even if "winning" is something clear-cut like "get him to vote Green Party", player interaction is still limited by their options. A game can set some pre-selected factors like "show the opponent Al Gore's slide show" or "point to statistical data", but unlike our previous example, it won't really help us to assign each to a button, because in this argument, as with most, a "move" is something that realistically wouldn't really help more than once.
You show the enemy Al Gore's film, he isn't convinced, you show him the film again... Whaa? Unless the computer is simulating someone who's not listening or who makes circular arguments, re-using the same "move" in an intellectual conflict is pointless. Let's look at another example - you're trying to convince a customer to buy a big-screen TV. You tell him it's a good deal, he says he's not sure he wants to spend that much, and then you tell him it's a good deal again. Sure, that may be how some salesmen work, but it doesn't make for a good intellectual conflict.
Because of this, game designers are generally forced down two paths if they do this kind of thing - either they can make a rudimentary set of actions and then try and make those actions fit into every situation in the game (ask, beg, threaten, show item), or they can make pre-selected options for each conflict. The former is less common, and it does provide some variety, but it generally results in narrative that is extremely vague - as in The Sims, where you know if a character is hungry or wants to watch TV, but you can't gauge their opinions on Nietzsche.
The latter is far more common, and it can make for a very detailed plot - as detailed as the game's writers can think up. However, The gamer is limited to whatever options the game's designers set up beforehand. In the aforementioned scenario, maybe the game's designers give you three pieces of Earth-friendly material to show your "opponent" to change their mind. Let's say they give you another option - you can bribe him to vote green party. Let's add another - you can increase a "charm" meter to win him over with your personality. How about adding the option to sabotage his air conditioning so he thinks global warming is making his house hotter?
Depending on how much time and effort a programmer put into a game, more and more options can be added, giving the player more and more freedom to influence the narrative. However, you're still working within the constraints of these arbitrary choices selected by the game designer, and in the end, there's no possible way a game designer could give the player anywhere near the kind of freedom they have in a real conversation, and even if they could, they wouldn't really be able to design an AI that could interpret such a wide range of responses.
So, as a result, this kind of conflict ultimately suffers, becoming either vague, or limited by the game designer's capacity to add options to the game.
You can press the A button five times to punch your enemy until he dies, but it's far too complex to realistically simulate a truly interactive conversation where you convince him not to fight. Either you're just selecting the "convince him not to fight because it was all a misunderstanding" option from a list, or your picking the "heart" icon and hoping your likability stat is higher than his "I want to kill you" stat.
Er... at least, that's my thoughts on it, anyhow.
That's an alright article on the subject, using "Earthbound" as an example
Yes. That's cognition, or thinking. If you see, the veeeery first line of that quote seperates cognition from emotion. So I quite clearly mentioned cognition multiple times.
Meaningful can mean something cognitive too...you can think about something from a different angle without any emotion at all and that can still be meaningful. Meaning is an objective word.
bjezz:
See now you are much more clear. Before I kind of figured you wanted to talk about writing but were using metaphors and expression from wider art in an extremely general way. I kind of figured I knew what you were talking about but I can only go from what you actually post, not read your mind.
I agree that games need a lot of work, as do most mediums of expression. I do however, think that games are a singularly unique medium and that to simply analyse them from a single point of view, that of a writer, is doing them a bit of a disservice. The immersiveness you talk of from Half life 2 is something some writers strive to accomplish using all manner of literary devices, and yet in that game it's done so well virtually anyone can appreciate it without having to read a long or flowery book.
I think you'd have to step back and really study the visual aspect of art as a whole, as well as music, literature and the human psyche, because if you want to create rich and compelling artistic storytelling using this medium, you need to bring all these things together in a cohesive way.
CaspianX brings up this point very well....
Ultimately, games are limited by the design limitations of their human programmers. Until we can create an AI that can increase it's own complexity without our own interaction, this will stay that way.
We aren't getting smarter and smarter as time goes on, we simply discover/create easier design tools and methodologies. Information systems design is changing constantly, because this kind of complexity is something human beings have real trouble grasping, crafting and creating.
It's very easy to theorycraft a perfect virtual world. But who is going to design this? Who is going to craft it? Who will pay them? What happens if the budget runs out? What happens if someone dies? I know of a major game right now that cannot upgrade it's network software because the guy who coded it in the first place left the company and didn't put in any programmers notes so they have no idea what's going on in there. So just because a storytelling idea might have merit or be better than the ones currently used, this does not mean they are feasible or even possible. If they are possible, they may only be so if you construct an entire game around them, because there wont be the budget, time or personnel for any more.
Yet again, we find that storytelling in a gaming medium cannot exist and be analysed eperately because you are crafting the concepts into something someone else can use, not just writing them down or showing a series of images the directing of which you can precisely control.
It's no surprise that cutscenes are prelevant if clever story without them has so many elements to consider.
I appreciate what you are saying bjezz, I just think you are approaching it too narrowly. I don't think discussing art in general will take this topic "off topic" as it will as long as it's clear that it is being discussed in relation to gaming....because there's a lot of applications to the gaming medium from every branch of the arts.
In the perfect game, the storytelling IS the game, not a cutscene. This is incredibly hard to do in a complicated manner. SoTC is almost a baby steps version of this. It tells a very simple, base story, and then really melds this into the medium. And even it has cutscenes.
That is a horrible article.
That is the best thing ever.
As has been mentioned by a few people in the thread, the center of a narrative is conflict. To the best of my knowledge you cannot have a narrative without some kind of conflict. It doesn't matter what kind and it doesn't necessarily have to be resolved, but it has to be presented in some fashion. It just so happens that when you put a person in a digital world with a bunch of other digital people, the most obvious form of conflict is physical. It's easy, so it's where we started.
Can you simulate intellectual or emotional conflict? Absolutely. In theory. Now I'm going to assume, for the moment, that the only thing stopping us from presenting intellectual conflict in a game is how simplistic artificial intelligence is currently.
How do you create emotional conflict in a game? Hell, how do you evoke an emotional response of any kind from the player? How do you predict or control which emotional response a player has if you succeed? As I suck at explaining complex things, I'll throw out some examples.
The most intense emotional response I've ever had to a game was Star Fox in the SNES. "Star Fox?" you ask. Yes, Star Fox. It wasn't the narrative that got me. I didn't care about Andross. I didn't care about Corneria or any of the other planets. I didn't care about how the war was affecting people's lives. It was my squadron. They have names and faces. They ask you for help, and if you fail repeatedly to save them, they get shot down. There is no "going in for repairs" like the rest of the series. When the mission ends, you just see their screen show up as static. For the rest of the game.
Story time! The closest I ever made it to finishing the game was the surface of Venom. Midway through that playthrough I lost a wingmate. Now, this doesn't have any impact on how the game plays. It was just one screen of static reminding me of my failure. A couple planets later, another. Up till now, I'd just found comfort in the fact that I wasn't alone. I had a support group. We weren't a group anymore. There were now two of us against an armada, and Slippy finally snuffed it in Venom's orbit. The game never stopped. No one mourned any of them. They were soldiers in a war that wasn't over, so Fox just kept fighting. The game just presented it like it was a cold, hard fact and nothing more. It was just me on the surface of Venom staring down Andross by myself. Easily the bleakest, loneliest experience I've had in a game, and the interesting part was that I never connected it to Andross. I didn't pick up vengeance as a motive. I didn't consider whether destroying him would stop these things from happening to others. I just knew I was alone.
Then he killed me.
Now, was I the only person that went through that set of events? Probably not. Was I the only one that responded it to it that way? I don't know. Maybe, maybe not. I assure you there were plenty of people that didn't bat an eye at it. The only reason it hit me so hard was that I was a child when I played it. Children's entertainment doesn't do this. When someone dies, it consoles you. It goes out of its way to control how you look at death. In Star Fox, it just happened and left you to yourself. I actually think it may be why I love the game so much, twisted as it may sound. Had I been older, I might not have even noticed it. The point (yes, I had a point) is that the developers probably didn't intend for me to come away from the game with the experience that I did, and that it was one of several experiences that I could have had were I looking at what happened from a different perspective.
This is a problem with creating emotional conflict in the player, or between the player and a character. Anyone can be playing the game, and a designer has to be prepared for the fact that they can have any of an innumerable number of personalities and world views. In a movie or a novel or whatever, you have a protagonist that can be defined in whatever way will make the story work. All the audience has to do is acknowledge their motivations, even if they don't agree. You can't do that when the player is both the protagonist and the audience and that severely hampers your options for what you can present in a narrative. You have to play to a lot of base instincts to manipulate the player into experiencing what you intend them to.
Half-Life 2 tricked me into watching a crow get eaten by a barnacle by startling me. Episode 1 used the flashlight segment with Alyx to give you a codependent experience. SotC did a lot of subtle things with Agro to make you worry about and care for him. Then there's Fable 2's dog. These are baby steps towards actually creating a game reliant solely upon emotional investment.
Imagine you want to make a game out of No Country For Old Men. It's a story about realizing how futile it is to struggle against chaos. It's a story about giving up. How do you approach making a player lose the will to continue playing while continuing to play the game? People have different tolerances for how long they will continue to try. How do you ensure they lose their drive at the correct time? How do you ensure that they acknowledge that you meant for them to feel that way? Maybe you skirt the issue by having the player control the force of chaos (Anton), so that Ed Tom can assert these things for himself. Have you just robbed the player of the ability to empathize with it when they are identified with a different character? I don't have answers for any of those questions. We aren't there yet.
A few games just sidestep the issue. Half-Life has resorted to using Alyx as an external protagonist. If they can't be sure of what you're feeling as a 'character,' they'll give you a little push in the right direction elsewhere. This only works in Half-Life due to how much the series makes a point of control. Gordon is a tool, not meant to have control over where he's going. It just makes sense that you're lead through what to feel as well.
Maybe you pull a Morrowind and leave character motivations out of the game and completely up to the player. Then again, maybe the player won't think to.
Occasionally, you can try to give the player a limited ability to assert things for themselves. Mass Effect had Virmire. I'd argue that the only thing holding back it's impact was shallow character development. The Witcher had a lot of great choices, with great consequences, but how much it impacted you as a player was directly proportional to how much you identified with Geralt as a character. This is also limited to how much you can plan for, how much time you have to create content for that many paths, etc. You're going to miss something.
Man, someone needs to tell me to shut up.
tl;dr: It's possible and maybe coming eventually. Definitely not soon.
All right, people. It is not a gerbil. It is not a hamster. It is not a guinea pig. It is a death rabbit. Death. Rabbit. Say it with me, now.
I suppose that's a good target to aim for in terms of giving the player control over a character. The character doesn't need to be a blank slate completely defined by the player, the character just has to be able to make decisions in such a way that the player is satisfied with them, and would make those same decisions if they were in his place.
Keep talking.
Also, write out every possible branch that your train of thought could have taken.
Then distribute them via flyers from orbit.
It's the only way to be sure.
Oh, keep talking, smart guys. You think I won't? I'm totally game. And then you know who's gonna be sorry?
I'll give you a hint: you guys.
All right, people. It is not a gerbil. It is not a hamster. It is not a guinea pig. It is a death rabbit. Death. Rabbit. Say it with me, now.
you see, what interests me about this discussion is where this broiling meld of story and immersion fails. which i think it does, a lot.
for me the primary concern is the silent protagonist. half-life is a key offender, but the number of games out there with either silent or purposefully vague player characters is hardly marginal. i guess the theory goes, if your character is devoid of specific detail, you'll better empathise with them - they won't jerk you out of the "this is me, i'm in full control" feeling by saying something stupid. and you're right, it helps with immersion, it makes you feel more like you're within a world. but it's a deathblow for storytelling.
stories need protagonists. they need a defined character through which the outside events can be focalized and given meaning. it's one of the key things that makes story more than plot.
so yeah, give us immersion, it's all part of the rich tapestry of a fine game, etc etc. but don't do it at the expense of story - or at least don't get into the habit of it. i think a lot of gamers don't realise what's lost when the protagonist trotted out is either mute or featureless or customizable to the point that they're arbitrary. it means the story has lost one of its key tools in giving a specifc and meaningful take on the other events of the narrative, and it probably means a lot of the other storytelling tools have been treated similarly.