Sunday, January 29, 2006

Death of the Protagonist

There has been a lot of talk about 'co-ownership' of characters, lately; and Vincent has proposed that perhaps we can let go of the idea that players play protagonists. Perhaps, he offers, we can "let the events of the game's fiction choose" whether a character is a protagonist or a supporting character.

Apart from the minor quibble that the fiction isn't really the kind of entity that chooses anything, this is a neat idea. Is this possible? Could it be fun? Could it, for instance, be fun to play a character that suddenly dies a deprotagonising death and is thus shown not to have been a protagonist?

Certainly. What's more, it could not only be fun, it could also be important. It opens the possibility of a new kind of narrative, a kind of narrative that is a critique of traditional kinds of narrative. The death of the protagonist (those who heard a resonance of Barthes in the title of this piece were absolutely right) is an important step towards the coming of age of roleplaying games as a form of art.

I will leave the abstractions now and speak about something very concrete: George R.R. Martin's beautiful A Song of Ice and Fire series. I recently read the fourth book, A Feast for Crows, and although it may not have been the best book in the series, it has not changed my judgement that Martin is writing the best fantasy epic ever.

But this fourth book made something clear to me that I might have seen earlier if I had been older when I read the previous books (Martin is not a fast writer), but which this fourth book made even more clear: A Song of Ice and Fire is not only the best fantasy epic, it is also the last fantasy epic. Not in the sense that it somehow ensures that people won't write epics anymore, but in the sense that it mercilessly exposes and destroys the ideology of the epic. It takes a traditional kind of narrative, seems to follows its rules to the letter, and then suddenly breaks them where it hurts most.

In A Song of Ice and Fire, being good and honourable, or even kind and innocent, doesn't mean that the author will protect you against evil - it generally means that you will die at the hands of those who are more ruthless. Here, a war doesn't end in victory, glory and things being set right again - it ends in Phyrric victories, death, a devasted countryside, plague, famine and horrors untold. Here, in a trial by combat the innocent person can die. As one of the characters was fond of saying: "Life is not a song", and "Knights have no honour".

But Martin's most powerful weapon is protagonist death. Or, in Vincent's words, letting the events in the world decide that someone was a supporting character after all. In A Song of Ice and Fire, focal characters that we have followed for many chapters sometimes die, out of the blue, suddenly, and utterly senselessly and in a deprotagonising way. More than once, I looked at the words in shock an horror as I stammered: "but, but... that wasn't supposed to happen!"

But happen it did. Life is no story. You can hear Martin laughing in the background and saying: "the ideology of the epic is false. You should learn to accept the reality of life, where fate intervenes suddenly and without human concerns. Life is not a song". After A Song of Ice and Fire, all other epics will be recognised by the reader as the lies they are.

(Whether Martin will be able to keep to his anti-ideological stance even when he writes the climax of the series is, of course, as yet unknown. But one can hope. I sincerly hope he'll levae us with both the ruins of Westeros and those of the epic.)

If I were to make a roleplaying game based on A Song of Ice and Fire, one of my main design goals would be to make sure that player characters could die in a sudden and deprotagonising fashion, without anyone at the table being able to prevent it. I would do exactly what Vincent proposes. These destructive blows against the very notion of protagonism are what make Martin's books the jewels they are. Sudden, unpredictable and senseless 'protagonist' death packs the punch that drives the message home.

It would be a game I'd love to play. (I suppose it is far too much to hope that the Game of Thrones RPG, which does exist and is called after the first book of Martin's series, is this game. But maybe I should check it out.)

No matter how innovative games like PrimeTime Adventures, My Life with Master, Polaris and Dogs in the Vineyard may be, the tales they can be used to tell are always safely embedded within a structure of human meaning. Taking away that safety will open entire new realms of possible tales - and dropping the idea of protagonism is one important (though by no means the only) way to make this happen.


  1. Wow, that was a really nice post!
    Got me thinking quite a bit about where I could head with a projetct of mine.
    Do you see other means that could do what you mention in the last paragraph?

    What structures would you use to give our tales continuity, if protagonists can die "randomly"? Families? Ideological currents? Geographical locales or history? Maybe even just ideas the players want to portray in game and would have talked about before (instead of chargen we'd have ideagen...)?

  2. Victor,

    ASOIAF happens to be my favorite fantasy series. I interpret the character deaths a bit differently than you do, however. Remember what Vincent wrote on character death:

    When a character dies in a novel or a movie, it's a) to establish what's at stake, b) to escalate the conflict, or c) to make a final statement.

    All of the protagonist deaths in ASOIAF escalate the stakes or make a statement. I won't go into details, as that would contain spoilers, but I don't think that any major protagonist dies a random or deprotagonizing death in the books. Minor characters, yes, though they are making a statement about the nature and effect of war. And sure, Martin's world is grittier and deadlier than the usual fantasy epic world. But I guarantee you that at least a few of the protagonists will remain until the end.

  3. It's interesting that people are bringing up A Song of Ice and Fire in this context. Because unlike you who is making a positive connection, I think about A Song of Ice and Fire, and how frustrated I am with the series. And then I think about the games I used to play in where my characters just died senselessley. And how it wasn't any fun (though perhaps the new game designs will consider that senseless character death needs to NOT result in player non-input to the game - yet in one discussion of how to handle character death on Monte Cook's ezBoard, folks who were praising A Song of Ice and Fire and using it as an example of why character death is good, and the integrity of the story is more important that the player not getting to play for several sessions).

    I really want to see what comes of Vincent's idea. But if it's "A Song of Ice and Fire the Real RPG" I probably won't be interested in actually playing (though I'll grant I am still reading the books, but I'm getting less and less enthused).

    On the other hand, obviously a lot of folks really like what Martin is doing, and they would probably really dig such a game (and some would probably even dig it with player non-participation being a cost of character death - though I wonder if that would only be so long as someone else's character died).

    Christian - I see your point about A Song of Ice and Fire, and perhaps that not quite as senselessness as it seems is what keeps me reading. But at this point, so many characters have died that while I hold out some secret hopes for who the enduring protagonists actually are, I'm starting to not care. Of course the issue is compounded by the length of the books, and the interval between releases.


  4. But Victor --

    Your reaction as audience to Martin's authoring character deaths is only half the picture when we consider roleplaying games. While it was certainly a surprise to you that Joe died, I'm pretty sure it was no surprise at all to Martin, and he in fact killed off that character for a very specific purpose (as Christian points out).

    So too a roleplaying game needs to afford us both author as well as audience roles. A game in which the system can decree character death or deprotagonization without any player intent towards that goal would result, I'd think, in very annoyed players. What input do I get as a player under such a system?

  5. An intriguing point here. Is it relevant, do you think, the George R.R. Martin is a role-player, and based early projects (that is, Wild Cards) on his role-playing games?

    You cite that it would be a new kind of narrative which critiques traditional narrative, but is it really new? Protagonist death isn't a new thing in role-playing. It is a frequent and often controversial issue within traditional role-playing. Might deprotagonizing death as it has been handled in earlier RPGs critique traditional narrative? Stretching farther, might Martin have learned from this?

  6. I did not know about Martin, but those are exactly the thoughts I had in reading Victor's post, John.

    "World-based" gaming, circa 1994, or what a friend and I earlier discussed as a "clockwork world" circa 1986, were already all about this kind of thing. Although for me, it ran into practical complications due to high levels of character-attachment among my college gaming friends. But I doubt I was onto anything new, even in the late 80's.

    I also doubt that this is really what Vincent has in mind for his project.

  7. Interesting how there are several different readings of Martin's work out here. I disagree with Christian that all protagonist deaths in ASoIaF can be understood as escalating the stakes or making a statement, but that is probably because I interpret the work as undermining the kinds of stakes and statements that these deaths seem to escalate and make. If anyone wants to discuss this further, I'll make another thread where we don't have to worry about spoilers.

    What may be an important point to make here, though, is that Vincent's three reasons for character death are neither the only reasons a novelist can have to kill off a character, nor are they all suitable for classical protagonists. Concerning the first point, my analysis of Martin (whether it is true of Martin or not) points to at least one other reason for killing a character: to undermine a form of narrative.

    Concerning the second point, I think Vincent claimed that only 'c' is suitable for protagonists. So even if Martin only killed, say, the character that so unexpectedly dies in book 1 because this escalates the stakes, then the fact that he uses a protagonist for this purpose is already breaking the traditional use of protagonists.

    It seems to me that games like Dogs in the Vineyard and The Mountain Witch have been specifically designed to ensure that only 'c' can happen to the protagonists. They never die to establish what's at stake, and they hever die to escalate the conflict. You can't, as a player, say "I'll have my character die so everybody has to triple their stakes!", but you can say "My master's honour must be avenged, so I ai-uchi the bastard traitor, killing us both!".

    But let it be very clear what 'c' is all about; it is about making a statement. But it is not about the character making a statement, and it is not about the player making a statement - it is about the narrative as a whole making a statement.

    Huh? An example. In Dogs, a player character attempts to save a little girl from being kidnapped by a group of bandits, escalates to gun-fighting, takes a lot of fallout, and dies. The character has made a statement, right? "Protecting innocence is worth dying for."

    But what if the tale afterwards plays out like this: the little girl grabs a huge cannon and destroys all the bandits, giggles evilly, and summons a lot of demons to destroy the poor bandit's families. Here, the narrative detsroys the meaning of the character's sacrifice. A player whom this happened to would rightly feel cheated, and would rightly think that his character was deprotagonised. (I think that some of Martin's protagonist deaths should be understood along these lines, if executed much more subtly.)

    But what this means is that according to our current sensibility, player character death is only acceptable if it allows the player to make a thematic statement and have it validated. (Which doesn't mean that the other players must agree with it, but does mean that they have to support the integrity of the statement.)

  8. I have to stress that adding 'deprotagonising' protagonist death to a game is not going to suddenly turn it into a critique of traditional narative. Of course not. Add "whenever you rolls only 1's, you character dies" to Breaking the Ice, and all you'll have is an annoying little game.

    And I guess that deprotagonising protagonist death used to be mainly annoying, in old-school roleplaying.

    To make it something more than an annoyance, you would need at least two things:

    1. A system that is coherently constructed with critique of traditional narrative (or some similar goal) in mind.
    2. Players who know what is going on and willingly participate in that, or who at least would be able to appreciate what is going on as it becomes clear to them.

    If in old-school D&D you had 99,9% of the game and the social focused on gamist goals and traditional, heroic narrative; and 0,1% - sudden character death - going against those goals, then obviously no critique was going to come off the ground. Those sudden deaths were not part of some greater movement; they were momentary bumps.

    So, Joshua, especially since RPGs are not really well-suited to do the "sudden surprise" thing that authored fiction does, I'd say that you, as a player, should be given tools to play out a certain kind of narrative from a critical point of view. You should not be given the tools to play out a certain kind of narrative and then suddenly see your tools falling away without anything replacing them.

    John - I must confess my total ignorance concerning the influence of RPGs on Martin's writing. :)

    Elliot - I'm pretty sure what I'm talking about is not the way Vincent wants to take his ideas on protagonists and supporting characters.

  9. I should add something to what I wrote above. I claimed that our current Nar RPGs safely embed the tales we tell within a structure of human meaning (I hope you can read my intent through the vague formulation), and that taking that structure away would open up new possibilities.

    Of course, sudden character death in old-school RPGs was also safely embedded in a structure of human meaning. It meant: you lose.

  10. Of course, sudden character death in old-school RPGs was also safely embedded in a structure of human meaning. It meant: you lose.

    Which of course directly conflicted with the text talking about how there aren't winners and losers...


  11. Of course, sudden character death in old-school RPGs was also safely embedded in a structure of human meaning. It meant: you lose.

    You the player or you the character?

    I think the player-loser idea was already being torn down in the 80's if not earlier, but as I mentioned above, I didn't personally have a chance to put the ideas we used to bat around on Usenet into practice--not much at least. I think other people did and do, though. I'll bet a lot of them are/were found in Runequest, Traveller, and Harn/Harnmaster groups. (Perhaps Paranoia, Pendragon, and CoC, too.) They also use(d) techniques like one-shots to isolate character death from player loss.

    As for character death = character loss, that starts to fall apart as soon as you take the fundamental step into roleplaying; characters have their own values and priorities, and survival may be secondary in some situations. But it's undoubtedly true that a character can lose. Basically, by letting characters lose while eliminating player win/loss, you open the way to a portrayal of subjectivities and destruction of unified game-world metaphysics.

    You may be onto something new, Victor--it'd be hard to say until we have an actual game in our hands--but right now (tongue in cheek here) you sound like a "Narrativist by habit" who's believes he's "discovered" Simulationism.

  12. Ugh, that may have come off as bitter or something.

    Whether or not the idea you're proposing is really new or just a rediscovery/reformulation of older concepts, what really matters is the execution in game form. (Originality is overrated.) What I should be saying is, "Great! You've hit on an approach to gaming that resonates with stuff I'm interested in, too."