Who’s In Charge?

It’s a handy coincidence that Victor’s brought roleplaying games into the discussion at this point — because just recently I’ve begun playing them for the first time. In this post I’m going to try and start where Victor’s left off, talking about my initial D&D experiences and the issues of DM control, and then try and relate this back to IF and the dialogue we’ve been having there.

The Dungeons and Dragons campaign I’m playing at the moment is with a group of housemates and close friends who simply fancies giving RPGs a shot, very few of whom had any experience of the game. One of the more experienced took on the role of DM and starting plotting out a campaign and world for us to take part in. A we began, a few things became very clear: first, that our style was to play fast and loose with the rules and aims laid out both by the book and by our DM; second, that the DM’s style was to lay out a huge, open-ended world space for us to play in. The DM has drawn up a rough map of the archipelago we’re campaigning in, along with power alliances, racial groupings and dynamics, and so forth, and our job is to try and explore it in whatever way we like (we very quickly became pirates). The result of all this is that the business of narrative a world-description has become very much a co-authorial experience. The DM rarely knows what kind of things we’re going to try and do, and so is constantly forced in invent new locations, NPCs and events to accommodate our eccentricities. We have in effect wrested a great deal of power from him — or, perhaps, been given power by him — in terms of authorial control.

I’ve been having a second thread of thought as well. Because most of the players, and the DM, are relatively inexperienced at playing the game, we’re fumbling around an awful lot — making mistakes in judging difficulty,  forgetting rules, storming down dead end, and generally struggling in completing the quests we set ourselves. Observing the struggle our DM is having in setting appropriate cues and clues, as well as judging the difficulty of puzzles and battles, I’m seeing a lot of the same issues come up as do in IF authorship: How can I guide players’ actions without making them feel railroaded, and without puzzles seeming immediately obvious? How can I preempt what players are going to attempt, so that I can have appropriate responses developed in advance? And, in line with the direction of the current dialogue: How can I take the players seriously?

When these issues are misjudged by the DM, or the players don’t meet his expectations, there’s invariably an extradiagetic social conflict between the players and the DM: an argument about the appropriate difficulty level, or, more crucially, an argument about the plausibility of the narration. Occasionally players protest about the way the DM has narrated an event — when the DM doesn’t meet their expectations. In essence, players are claiming for themselves rights and powers over the way the story should be told.

Much of this is well-trod territory in IF theory, criticisim and authorship. Graham Nelson’s bill of players’ rights is well known, and it’s an elementary account of taking the player seriously by simply allowing for a players’ reasonable expectations about a game to be me through depth of simulation and appropriateness of difficulty. But at its fringes, on the radical lines Victor and I have been discussing, the ideas bring more pressing accounts of different ways IF can be written. I would argue that much of this comes down to the question of who’s in charge of IF narratives — questions of authorship (see my previous notes for an essay).

When the player is frustrated by not being able to use reasonable synonyms for verbs, by not being able to examine described items, by not being able to speak preferred dialogue, by not being able to progress in the expected way, they are effectively making an authorial demand on the IF work: they claim their right to author their own story, and are being frustrated in this by the limited and totalising authorship of the work’s implementor.  These are limited examples of claiming authorship, but throughout this dialogue Victor and I have been tackling the way more adventurous IF functions in terms of authorship, exploring what can happen if more authorial power is relinquished to the player.

I understand Poster’s fear that “you don’t have a unified story” if authorial control is relinquished, but I woudl argue that allowing for co-authorship — the subject of this thread — brings different types of narrative, and of narrative experience. A story does not have to be the complete conception of a single mind (just as the world and experience are not the product of a single mind, with a single epistemology); co-authored narratives — in IF, works where the interactor is given a greater degree of control over the narrative’s direction by the implemtor — can represent the intersubjectivity of ideas, or the chaotic nature of story-telling, or can yield results entirely beyond any author’s expectations.

This leaves us with the question of “how you can get away from a deterministic model for IF and still have it be IF”. Well, I would argue that this dialogue  has produced some answers: extternal to the game-world, through giving interactors opportunities to work with open-source codes (or through new interfaces), through player-responsive systems of narrative generation. I suppose the next challenge is to bring these theories to fruition — though there’s still a lot of ideas to be explored and theories to be talked through! I’m interested in what anyone has to offer.

Taking the player seriously II

In this post, I will continue my musings on taking the player seriously. One thing which I wish to say immediately is that my emphasis in the previous post on giving the player several choices and having ’serious’ immediate responses to those choices should only be seen as that, an emphasis for the sake of theorising. Harry’s points in his post below serve as an important balance for this attitude. First, he discusses the possibility of what we might call “background effects”: changes in the game that are not immediately obvious to the player, like a slowly evolving playing style. Background effects are alternative, and non-obvious, ways that the player can influence the game. It would be interesting to think about the play experience of a game in which you are being taken seriously, but almost exclusively through background effects.

Second, I read in Harry’s prison cell example in his discussion with Jason (hello, Jason!) the possibility of putting the player in a situation where it is not at all obvious which possible actions the player can take. This serves to make my notion of the game “offering the player a choice” slightly problematic; and it too could be an interesting topic for discussion. Two related entries in my Gaming Philosopher Blog are Veiled and unveiled spaces of possibilities and Veiling and Unveiling I.

But in this post I would like to continue with my emphasis on obvious choices and their direct effects; and I would like to make a comparison between interactive fiction on the one hand, and contemporary “indie” roleplaying games on the other.

Taking the player seriously was perhaps the great challenge that faced the authors of roleplaying games. The problem is basically this: the traditional division of a roleplaying group into a Game Master (GM) whose decisions are law and players who are bound by the rules introduces a serious imbalance of power. In practice, this often leads to minor and major problems, the most common symptom of which is that the players do not feel themselves to be taken seriously. (Because of a huge number of assumptions that were in place, they might not conceptualise it in this way themselves; but a strong case can be made that this was the root of most problems.) Apart from such non-solutions as “choose the right GM”, traditional roleplaying games offered only three ways out. One, by using pre-made adventures, the burden of power was removed from the GM and put onto the shoulders of the game designer. This reduced social tension, but only at the expense of destroying the possibility of a dialogue between GM and players. Two, by focusing on those aspects of gameplay that can easily be put into formal rules which both the GM and the player must abide by. The best example of this is Dungeons & Dragons 3rd Edition, which has an ingenious system of “challenge ratings” that ensure that the GM plays fair with the players, and has an intricate set of rules that allow the players to make serious tactical and strategic decisions. Three, by dropping the idea of dialogue altogether, and postulating that the players are there to listen to the GM’s story, sit back, and enjoy the ride.

Thus, thus who wanted to tell a story with each other in dialogue were left out in the cold. If you play without rules, you have chaos; if you play with rules, you have a power imbalance that makes dialogue very hard to achieve.

Over the last couple of years, this problem has been satisfactorily solved in a number of interesting ways. I wish to talk about a few of these ways in this post, deferring the question of whether and how these could be ported to the medium of interactive fiction for a later post.

• Universalis. In the game Universalis, there is no Game Master. The right to narrate (that is, to tell what is happening in the fictional world) is bought by players through coins, of which every player receives a certain amount at the beginning of the game. You can use coins to establish facts, but if other players disagree with your proposals, they get to use their own coins to challenge you. Such challenges are resolved through dice, where players who stake more coins on the outcome of the conflict have a better chance of winning.
• Inspectres. This game superficially looks like a traditional roleplaying game, with the Game Master having the power to narrate, call for skill checks, and set the difficulty of these checks. However, whenever a player wins a skill check, she is allowed to narrate what happens, which narration can include new facts about the game world or other characters. There is also a nifty thing called “the Confessional”, where players can sit in a special chair and reflect on what is going to happen - like in a TV documentary, where people look back on the events that the documentary reconstructs. Thus, as the characters are entering a building, a player can take a Confessional and say: “Little did we suspect that we would all come out covered in layers of green slime”, and everyone will have to make sure that this is actually going to happen.
• My Life with Master. This game tells the story of an evil Master (basically, the GM) and his unwilling minions (the players), who really only want to be loved but are forced to do horrific things. The game divides the story into scenes, each of which is of a particular type: basically “being given a quest by the Master”, “carrying out the quest in a villainous or violent way” and “trying to find love”. The general outcome of these scenes is decided by the dice; thus, the narrative structure is enforced by the rules, not the GM or the players, who are there to fill in the content. In addition, the Master and the setting are created by the whole group together, before the game starts; and whenever the GM wants to assert his own vision during the game, he must do so through the voice of the Master, and the character gets a fair chance to resist this - hence, out-of-play power struggles are converted into in-game events that fuel the story. (And the player will finally get her revenge!)
• Shades. My own game, which can be found here. There is no Game Master. Whenever it is your turn to narrate, you have unlimited power over what you narrate. However, since what you are narrating are the memories of your character, and since memory is unreliable (and the characters in fact want to believe that things were different than they actually were), saying something does not make it true. The players must contradict each other in order for the game to progress, and must reconcile these differences before the end. At the end of the game, each player is asked to evaluate whether or not he or she has exerted an non-proportional amount of power over the final story.

How, in these four instances, is dialogue achieved? Sometimes, the traditional power structure is changed into an in theory egalitarian power structure. In Universalis, competing visions battle each other through a formal betting system. You can enter this battle every time you wish to assert your own vision, and the winner gets to have power over the narrative. In Shades, something like the reverse happens: competing visions must be reconciled through an entirely non-formal, ’soft’ system where you always have only the power to give, never to take. In Inspectres, everybody is allowed to add to the narrative more or less when they wish, and the game is zany enough not to make this a problem. In My Life with Master, finally, most of the power to influence the large scale structure of the story is put into the hands of the completely impartial dice, and any remaining power struggles are encouraged, turned into fuel for the story, and lead to a cathartic ending.

Can any of this be useful for interactive fiction? That is a topic for another time.

Talking Across the Divide

I’m very interested in dualisms — or, at least, in their deconstruction — and so it’s always with a smile that I find myself caught in one. Victor is entirely right to point out that I’d rooted my criticism — somewhat unnecessarily — in a strong player/author dualism. He throws this observation out almost as an aside, but it’s actually quite essential: this IFdialogue is now tending towards the discussion of what I originally termed the world-internal approach to co-authorship in interactive fiction, but the world-external should not be forgotten: Victor’s proposal to overhaul the methods of IF authorship can stay at the back of our minds as we discuss styles of writing as well as methods. A dissolution of the external/internal divide of IF worlds would be truly exciting — indeed, we found this intimated to an extent only recently, in IFComp07’s DeadlineEnchanter, which radically undermined and ambiguated the roles of author and player.

***MINIMAL SPOILERS FOR DeadlineEnchanter FOLLOW***

Interestingly, DeadlineEnchanter is in most senses highly uninteractive. Player freedom is minimal, as is puzzle-solving: one is instructed throughout how to behave in the game-world. The use of interactivity in DeadlineEnchanter is for emotional content only — allowing the reader to participate in the unfolding of the story helps pace the argument, complicity in the actions of the story helps give its arguments emotional impact. These uses of IF should now be well understood — they have been features of the best IF for many years.

In short, the game gives us an excellent example of what Victor terms monological IF: there is one voice present, which is the voice of the author, and the role of the player is simply to assist that voice in speaking. The author’s skill in this IF lies in making the player desire to assist — usually through emotional or intellectual engagement, or simple through the desire to solve puzzles. We know, then, what dialogical IF is not. As we risk defining here purely by exclusion, I’d like to add to Victor’s ideas about what we would expect to see in dialogical IF some of the results of a previous dialogue I had with Emily Short on rec.arts.interactive-fiction.

One of the ideas which emerged was the importance of “that quality that allows the game to respond to the will of the player and engage the player in creating meaning”. This bears a resemblance to what Victor has in his post fleshed out as taking the player seriously — to “offer the player a choice about which she can form reasonable expectations as to the outcome [...] and to which the game then responds in a way that honours those expectations (and thus the reasons for choosing it).” (As an aside, I found that regarding Emily’s own Galatea my satisfaction levels for many of the endings I found was fairly directly correlated to how much I felt they followed naturally from the conversation — the more arbitrary held a breif surprise but little lasting effect, while those I felt I was responsible for stay with me even now.) The idea of responding to the will of the player is fairly basic, but what is key is that this helps the play “create meaning”. By honouring the reasons the player had for making a choice, the game takes not only the player seriously, but also the player’s interpretation of the game-world seriously.

Many ambiguous works of IF have been written — some authors (Andrew Plotkin for example) write mainly works which are an enigma, which allow for multiple aesthetic interpretations. This allows some form of player-empowerment to take place — the player is given some authorial control. But when the player is asked to interpret the game and then respond to it, and the game is able to take that interpretation seriously — this is the point at which IF becomes truly dialogical.

In my dialogue with Emily a few potential (and largely well-trodden) techniques emerged which could allow us to do this. Plot-branching, “moral choice” IF, interpretative amiguity — these can all aid dialogical aims. Also mentioned was the interesting idea of “responsiveness to play-style”. This has mainly been explored so far through adjusting difficulty levels and providing in-game hints — but I’m excited by the prospect of trying to write IF which can take account of the way the player is playing it. Does the player like to attack or take everything in sight? If so, descriptions could become terser and violence more enjoyable. Does the player like to investigate everything in the room? This could lead to longer descriptions and uncovering of new mysteries. Does the player rush through rooms or take eir time? Does the player go straight to the hints when solving puzzles? Detecting these things would require new explorations into programming techniques, and a willingness to do a lot of “unseen” writing — but the effects could be exciting.

The effects are what we must hold in mind. Why would we want to write and play dialogically? What does it allow us to experience? To my mind it enriches the gaming experience by allowing me to contribute to the way it is experienced. Perhaps a lesson intended by DeadlineEnchanter — a game I actually found highly dissatisfying to play, and I wondered if this was not in fact intended, if the game did not hold a deep ironism — was that excessive authorial control, and excessive desire to impose an authorial view on the player, leads to dissatisfying play and unconvincing argument. But something else I find exciting about dialogue is the way it effects all participants — new ideas emerge which neither player nor author was expecting. I would not want, as a player, total authorial control of a game, sandbox-style — then I would have nothing to learn from the game. But what I would find most exciting as an author would be to find players putting new spins on my work, taking it in new directions. (This was also a feature of Victor’s original essay.) How we can develop new techniques towards these ends is something worth exploring. It would require unique co-operation of programming and literary skill — which is, of course, IF’s singular domain.

HG

Taking the player seriously

Let me start by thanking Harry Giles for the invitation to join him in a dialogue on interactive fiction, and for the excellent first post with which he set the wheels of this discussion in motion. I agree with almost everything he wrote below; except perhaps the firm use he makes of the author/player distinction in his criticism of my original proposals for radical co-authorship.

But I have no wish to defend my former proposals, for they are too radical, too ideal, too far in the future for them to have any real impact on how we write interactive fiction right now. And that is what I would most like to talk about: ways of getting significant co-authorship in our current interactive fiction, with the tools we process now. What we write here should be useful to all the authors out there who are trying to come up with a good idea for the next IF Comp. Let’s see if we can write theory which engenders action.

What do we mean when we speak about co-authorship, about playing as dialogue, about the player producing manifold narratives? We can not simply be speaking about giving the player the power to influence the text which appears on her screen, for the player has that power in all interactive fiction, including that which is least co-authored and least dialogical. Let us look at an extreme example of monological IF:

“Please, don’t kill me!” , the nymph implores. “But kiss me, and I will turn into a beautiful princess.”

> kiss nymph

You should have known better! You die a horrible death.

> undo

The previous turn has been undone.

> kill nymph

You slay the nymph and receive 20 gold pieces.

What makes this an instance of monologue is that the decisions open to the player have been clearly divided into one right and one or more wrong decisions, where the wrong decisions are immediately punished. What is punishment? How does it work, in interactive fiction, where the author can hardly do any real harm to the player?

Let us look at he example again. Kissing the nymph leads to death, and dying is presented as a failure to complete the game, that is, as a failure to play the game as it ought to be played. What happens when the game tells us “You should have known better! You die a horrible death.” is that the game tells us that what we did was wrong, which is the same as telling us that by choosing the action we did we showed that we do not (yet) understand the game. Our action of choosing to kiss the nymph has been denied validity as an interesting narrative move within the work of art.

Terminologically, I will say that the decision is not taken seriously by the game. This is going to be the key term I wish to start exploring in the rest of this post. Let me suggest at this point that the game engages in dialogue with the player whenever it gives the player multiple choices that it takes seriously; and it doesn’t engage in dialogue whenever if gives the player only one (or even fewer) choices which it takes seriously.

Let us modify the example a bit, to start exploring the notion of being taken seriously:

> kiss nymph

Risking all on the mysteries of love, you embrace the nymph. Her kiss is everything that you expected, and more, much more than a human body can withstand; but while your material form dissolves in seconds, your soul experiences an eternity of bliss.

What is the difference? Most obviously, this is a positive instead of a negative ending. This is important, but its importance can be overestimated, and I want to spend a few seconds to talk about that.

Making the outcome of a choice positive is not a sufficient condition for taking a choice serious. For instance, if, in the above example, the player had been given every reason to believe that kissing the nymph would simply lead to a horrid death, making the outcome of the choice positive would actually mean that the choice has not been taken seriously. By taking away her capacity to predict (more or less) what will happen, we have robbed her of the capacity to make informed choices.

Conversely, making the outcome something that the protagonist could conceivably choose, is a necessary condition for taking the choice seriously–given that the player is expected to identify at least partly with the protagonist. Nobody is going to choose a horrid death, so if we are expected to identify with the protagonist, kissing the nymph when we know that this is going to lead to a horrid death will never be a serious choice. We could still choose to do it, but only in jest, to see what the author of the work has programmed.

This leads me to a first and, I suppose, very provisional definition:

For a game to take the choice of a player seriously is to offer the player a choice about which she can form reasonable expectations as to the outcome, which she can reasonably choose because of her identification with the protagonist or for other (puzzle-solving, artistic) reasons, and to which the game then responds in a way that honours those expectations (and thus the reasons for choosing it).

As I said above, the game engages the player in dialogue whenever the game offers the player more than one choice that it takes seriously.

This is only a beginning, but I hope I will be useful. Harry, the floor is yours.

 VG

Understanding Techniques for Co-Authorship

The idea for the IFdialogue project came from an e-mail exchange between myself and Victor on the subject of his essay on Co-Authorship and Community. I’d recently written some notes for an essay on the concept of the dialogical in interactive fiction (IF) that were in part inspired by Victor’s work, and was looking to discuss some of the ideas which had come up in the two pieces. This open blog discussion seemed like a good way of brining the issues out and sharing them with the IF community. So I’d like to begin by analysing the key aspects of Victor’s original essay – what it is that I found inspiring in it, but also what it was about it which bothered me.

Victor identified two paradigms which currently dominate IF authorship – first, giving puzzle-solving a central place in the interactivity of IF; second, a player-author paradigm in which the author sets up a world for the player to explore. This second paradigm is, he argues, a necessary product of the first – if IF is based around an author setting puzzles for a player to solve, then even relatively puzzle-less IF is going to be dominated by a strong author-player dichotomy. Similarly, the artistic effects of the first paradigm – trapping the player in a means-ends psychology of “What am I supposed to achieve in this situation?” – underlie the motivation for challenging the second.

Victor’s solution is to seek to unleash the potential of co-authorship. I followed and enjoyed his arguments to this point, and agree that co-authorship – giving the player some kind of authorial control over the game – is a strong desideratum that will lead into exciting new areas of interactive fiction. What bothered me about Victor’s essay, though, was that it presented a restricted view of what that co-authorship could entail.

The essay proposed to give the player exactly the same powers as the author. “If the player is to effectively comment on the vision of the author,” he writes, “she needs to have the same powers as the author”. The player must be allowed to “change the objects in the world, add and remove locations, change the rules that govern how the world works, and change the texts generated by the work”. In Victor’s vision, IF can become a community art project, with the player becoming co-author in a radical way – with multiple versions of the same work competing for popularity. (Recognising the potential “emptiness” of complete authorial freedom, Victor claims that community curatorship and rating will become essential to such a project’s success.)

Such a project sounds fun, exciting, and has enormous potential. But I am unsure if it really addresses the problem of the limits of the IF paradigms Victor carefully outlined. The problem is that this vision of co-authorship is essentially world-external: the authorial debates it raises are not those of the problems within the world of the text, but rather those of the quality of the text itself. Victor writes:

Different versions [of the game] are offered, and you can explore them and think about the different visions they incorporate. Which one do you prefer? And then, a slight change here, a good idea there: they are made so easily, why not make them? Of course Margareth should have alluded to having a boyfriend in that first conversation; only that gives Charles’ moral choice real poignancy.

What Victor’s suggested project produces, then, is not a new way of being an IF player, but rather a new way of being an IF author. Playing and authorship remain different activities: it is simply that we have a new authorship system that some players may appreciate. The project would open up IF authorship, and would produce, I think, high-quality works – but playing those works need not feel any different.

It is for this reason that I felt the need to explore the concept of the dialogical in my own notes. This concept is drawn from Mikhail Bakhtin’s dialogic, which refers to the intertextuality of texts (the shaping of any given text’s meaning by other texts). Taking a “text” to be an imaginative mind as much as a book (or computer game), we can also take the dialogic to refer to the shaping of the meaning of individual experience – and from here we get to the dialogical, which, refering to the dialogic, specifically concerns active dialogue between entities. In my notes I was analysing works of IF as dialogues between author and player.

What I found was that while in some works – I chose the classic Photopia as an example – the author assumed total control of the game-world, guiding the player’s actions and attempting to induce specific meanings and interpretations – there was a recent trend for games to treat meaning-production as a collaborative effort between author and player. I chose Galatea as an extreme example of this (mainly because the game’s structure was explicitly a dialogue), but Victor’s own The Baron features such dialogue: in these games, the author sets up manifold potential narratives, and the player authors these narratives for emself by playing the game. More significantly, the authors of these works allowed the player interpretive freedom as well as authorial freedom – The Baron achieved this by specifically asking the player to choose what meaning to attribute to eir choices, while Galatea achieved this (in the manner of enigmatic and ambiguous works of authors like Andrew Plotkin) by leaving unanswered questions and mysteries for the player to solve.

I believe that this kind of co-authorship is significant – and perhaps far more significant for developing new artistic realms in interactive fiction than Victor’s own suggestion. The co-authorship in Victor’s essay is world-external, and so is really just a different way of producing the same kind of works. What I’m interested in exploring is how to write different kinds of works. The key questions for me are: How can we increase the interactivity of IF to allow players to produce manifold narratives? How does ambiguity aid dialogical interpretation? What new writing techniques can we explore to produce narratives dialogicially? Overarchingly, what is the nature of the relationship between author and player in interactive fiction which allows for these questions to be asked?

I’d like to explore some of the answers to these questions as this blog progresses. First, however, I’d like to see what Victor has to say about my analysis, criticisms and proposals. It is entirely possible this blog will go off in a direction I simply wasn’t expecting. It is, after all, a dialogue.

HG

Welcome to IFdialogue

Welcome to IFdialogue, a new project in interactive fiction theory and discussion.

The idea of IFdialogue is simple: to set up conversations on IF between authors, players, and theorists. We hope to have wide-ranging discussions on all aspects of IF theory — but at the moment we’re just a speculative project.

The opening dialogue is between Victor Gijsbers (The Baron, Fate) and Harry Giles (The Chinese Room) on the subject of co-authorship in interactive fiction. We aim to update the dialogue at least weekly. We hope you enjoy it, and we look forward to your comments on it!

If you find the IFdialogue project exciting and you’d like to get involved in starting up a new dialogue thread, do get in touch.