Second Person: Role-Playing and Story in Games and Playable Media is a remarkably rich book, covering amateur and commercial projects in a number of media and forms. It contains four essays explicitly about text-based interactive fiction: Jeremy Douglass’ reading of Andrew Plotkin’s “Shade”; Steve Meretzky’s account of the creation of the robot Floyd in “Planetfall”; Nick Montfort’s essay on the player character in interactive fiction; and a short piece by me about the ways the puzzles in Savoir-Faire are designed to bring the player’s skills in line with those of the player character.
Those four essays are by no means the limit of what the book has to offer to IF authors and players. Several other pieces are by people whose names are known in the IF community: G. Kevin Wilson writes about creating board-games using intellectual property from other media, but he is best known to the IF community as the founder of SPAG and the yearly IF competition — arguably the two most important institutions of the freeware IF community — and for his game “Once and Future”. Chris Crawford, Andrew Stern, and Michael Mateas talk about the processes involved in creating interactive characters; though their approaches involve user interfaces unlike the turn-based text parsers of IF (as I mean IF), their ideas are intriguing and potentially valuable.
And much of what remains illuminates broader issues of interest to IF authors: how storytelling can be done in an interactive context; what are the challenges and possibilities of transferring a story from one medium to another; how literary genres project themselves into games; the social perception of computer games; the potential of interactive media for education, art, and political advocacy as well as entertainment.
What follows is an attempt to address the IF-relevant ideas in Second Person. I haven’t much to say about some articles: George R. R. Martin’s history of the invention of the Wild Cards series; a number of pieces on RPG design in various systems; etc. Inasmuch as I’m in a position to evaluate these essays at all, I generally thought them very good, and often quite enjoyable. But here I will be picking out just a few themes. Those interested in a less targeted overview should perhaps read the introduction online, instead.
[Side note: to be consistent with the rest of this website, where I say “interactive fiction” or “IF” throughout this essay, I mean a work presented by a computer primarily if not exclusively in text, accepting commands through a text-based parser, and determining subsequent outcome by means of a programmed model. By “the IF community”, I mean the group devoted to creating and playing this type of game, especially those who communicate primarily through rec.arts.int-fiction, rec.games.int-fiction, and ifMUD. Other meanings of “interactive fiction” are adopted by the various article authors, and I grant that there is some virtue in other definitions, but for clarity I will refer to things that are not “interactive fiction” in my sense as “interactive narrative”, “interactive storytelling”, and the like.]
Narrative and Interactivity
Second Person begins with a discussion of table-top games: board games, card games, or RPGs, but all games that can be played in real life against physically present opponents, or using some kind of solitaire (but physical) game system. Some of the essays here are essentially historical documentation; some theoretical discussions of story-related games (especially role-playing games). Most have some bearing on the question of how a story can meaningfully be said to be interactive, and what types of interactivity apply to narrative.
Greg Costikyan starts with an article on the various ways in which games and stories can come together. His stance is a bit more open, more qualified, than some of his earlier writing, in which he argued that game-ness and story-ness work against one another. He also argues that conventional live role-playing games do best at this, and that adventure games (text and graphical) tend at best to resemble strings of beads:
Rather than being an explicit branching narrative, however, players often return to areas in the world, and an inventory system and set of puzzles provides more gameplay. But the narrative is still quite linear: adventure games tend to be “beads-on-a-string”: small areas where there is some freedom of action until some event occurs, at which point a transition to the next bead is opened. While there is some freedom within the beads, the overall game is a linear progress through the beads.
He goes on to point out that this is partly a problem with the expense of content development; a concern that is much more significant in graphical games than in text-based ones.
There are some good observations here; I’m not quite doing Costikyan justice. Still, this mostly felt like old news to me, from the perspective of amateur interactive fiction. The problems of linear, nonlinear, and multilinear storytelling have been pretty heavily discussed in the IF community, and tentative solutions have been explored:
- Branching structures: IF with significantly branching story lines (to a degree my “City of Secrets”; “Slouching Towards Bedlam”, “Vespers”, “Legion”, “Narcolepsy”, and a number of other pieces); or in which the player is allowed to choose among several goals and work towards whichever one most appeals (“A New Life”)
- Nonlinear structures: Short IF scenes that can play out in a very wide variety of ways without explicit branchpoints (“Galatea”, “Shadows on the Mirror”, “Redemption”, “Whom the Telling Changed”)
- Variant versions: IF intended to be played to several loss endings before the winning playthrough, in which failing story lines nonetheless add to the narrative significance of the successful one (“Varicella”, “Lock and Key”, “Rematch”, and to some extent my “Glass” and “Damnatio Memoriae”);
- Constructed meanings: IF revealing story through an exploration of elements not in a particular order (“Exhibition”, “Ribbons”); or, alternatively, in which the player plays repeatedly through the same narrative from different perspectives (“Heroes”) or with different available tools (“Moments Out of Time”) in order to assemble a meta-narrative comprehension of events; or Victor Gijsbers’ unique experiment “The Baron”, which gives the player the responsibility of choosing his character’s motivations as well as his actions, allowing the player to interfere not only in the outcomes of particular events but also in their meaning.
- Complicity and identification: IF designed to engage the player with the experience, attitude, or perceived limits of the player character, more than in affecting events per se (“Common Ground”); sometimes even making a point of the player’s inability to change the plot (“Rameses”, “Photopia”, “Constraints”) or of the player’s willingness to go along with a story that involves an immoral protagonist (“1981”, “Bliss”, “The Last Hour”). Or, alternatively, IF in which the player is allowed to partly construct a character by selecting the constraints by which he would like to try to play (“Scavenger” in which the player can purchase tools or weapons which will affect his options for problem-solving but also determine how violent his character must be; “Bolivia by Night” in which choosing to be female affects NPC attitudes towards the player)
So it’s true that we haven’t nearly explored all the possibilities, and that some (not all) of our most oft-cited successes in storytelling are quite linear. But there’s quite a bit more out there than Costikyan’s article would imply, and not all forms of interactive narrative require that the player’s agency pertain to the plot as such.
Nor are all of these ideas unique to IF. In fact, most of the ideas I’ve listed above are addressed somewhere in the book. In particular:
Branching structures: Kim Newman writes on Life’s Lottery, a CYOA-style novel; describing, among other things, how several of the branching CYOA-style options must be read in tandem to reveal secrets of what’s “really” going on.
Nonlinear structures: One of the most technically challenging articles in the book is Andrew Stern and Michael Mateas’ detailed discussion of Façade, a work that does drama management and constructs a story around the player’s actions from a collection of short “beats”. I will talk a little more about the techniques later; but Façade, for all its flaws (and it has many), achieves a genuine fluidity: it doesn’t feel like a linear structure, or even a branching structure.
I’m not sure what it means that most of the nonlinear interactive stories I can think of are essentially conversation-centric: perhaps it has to do with the fluid causality of such pieces. In a conversation, it’s reasonable that talking about alcohol could lead to a discussion of marital difficulties, or that talking about marital difficulties could lead into talk about alcohol; conversational events have a significance and part of the significance does derive from the order in which they occur, but it’s rare for one to be a prerequisite to another. By contrast, it would not make sense to order at random a) a scene in which Joe blackmails Sam and b) a scene in which Sam kills Joe.
Variant versions: Kevin Wilson describes his introduction of narrative into the Arkham Horror board game, in which decks of event cards reveal what can happen in a given location:
…with repeated plays, the players learn and remember the story as I intended it, because even if it doesn’t work out every time, more often than not it occurs in the order I intended. (92)
This does imply that the designer has in mind a set story (at least loosely), whose ideal form is experienced more or less successfully by the players during the course of play. But this doesn’t mean that the “failed” versions of the story, those which contain fewer of the plot points or which do not end happily, are irrelevant, or that the “successful” version could be read on its own to the same effect. IF reviewers have noted that in “Anchorhead” (a work of IF that, like Arkham Horror, belongs to the Cthulhu mythos), the terrible deaths experienced in losses of the game reinforce the sense of peril, and make the player character’s risks seem real even when the player successfully guides her to survive until the end of the game.
Constructed meanings: James Wallis and Eric Zimmerman both describe games in which players assemble their own narratives out of story elements that might be randomly selected (often through the mechanism of a deck of cards), and Jonathan Tweet talks about the character creation method in the Everway RPG, which also uses random cards to spark characterization ideas.
IF generally steers away from this level of randomness except in its very most experimental forms (though see “The Tarot Reading”), but the idea of presenting some evocative elements and allowing the player to work out a narrative or philosophical significance appears in several common types of IF: highly surreal works (such as “So Far”, “Dreamhold”, “Shade”, “End Means Escape”); works oriented towards discovering a mystery (“Elysium Enigma”, “Redemption”) or interpreting an out-of-order or ambiguous set of events (“All Roads”, “Primrose Path”).
Some of the interest of these pieces lies in trying to work out their significance as a kind of meta-puzzle; a game played collaboratively, through long discussions on ifMUD or rec.games.int-fiction. It is hard, as an IF author in the modern hobby community, not to be aware that your game will be played by a small, close collective of people who may well discuss its meaning. Sometimes I have written in elements specifically to enhance the communal experience rather than the individual one. There is more incentive to add special plot branches or quirky easter eggs if you feel reasonably confident that someone in the community will encounter the variant and report it to the rest of the players. (In this respect, IF authorship sometimes resembles the experiments Jane McGonigal describes, in alternative reality games and online puzzle games designed to be solved by social communities of players.)
Complicity and identification: Nick Montfort argues that the player character in IF is not used by the player for performance or dramatic play, but:
My concept is that the player character in interactive fiction is not played at all, but is a constraint and possibility defined by the author, within which the interactor is bound to a particular perspective and a particular set of capabilities… (145)
Montfort works out this suggestion at considerably more length than I can summarize here, with illustrations from his own IF and other people’s. It’s a thesis that bears some digging into, but I think I essentially agree with it.
Also shedding some light on complicity and identification is John Tynes’ article on “engagist” games. He defines this category a little more narrowly than I might, saying
An engagist work is one that uses the modern world or the recent historical past as its setting and that provides tools and opportunities for participants to explore and experiment in that setting in ways that real life prohibits or discourages. (221)
His point is that such works can teach us about other environments and the struggles faced by other people; thus far I agree. (I was an avid player of Oregon Trail as a kid, a game he cites with approval, and I certainly retained more of it than of any given chapter of my high school American history textbook.) I’m not sure I buy into the restriction that engagist games must involve the present or the recent past, though; it strikes me that one could just as validly write a game set in the middle ages that allowed the player to grapple with issues of faith seen differently at that point in history; or a game set on a near-future Mars colony that simulated with some accuracy the challenges humans might face in such an environment. The rebuttal might be that we have enough evidence to write an engagist game about the present or recent past, but that the distant past or (worse) the future are too unknowable. I’d counter, though, that — however much it may draw directly on the newspapers of the moment — a game like Peacemaker is also speculative, since it posits a resolution that has not yet come to pass.
IF engaging deeply with difficult issues or unusual personal experiences are not very common, but there have been some essays in that direction (“Jane” on domestic violence, “Who Created That Monster?” and assorted others on war and foreign policy). These are interesting if not universally successful; and I think this is something of which we will see more, as the medium matures. One of the common failings of many existing pieces is that they tend to take simplistic or one-sided views of their issues and not, in fact, to simulate the real pressures very accurately. Perhaps the worst offenders are the games of religious intent (“Jarod’s Journey”, e.g.). My recent “Floatpoint”, though not engagist in Tynes’ sense, was intended to be a somewhat less simplistic approach to some issues of immediate concern: the ethics of genetic engineering, the problems of reconciling deeply dissimilar cultures, our responsibilities to the environment.
So do these essays reveal forms of interactive narrative that don’t occur in existing IF? Yes, many. Among them:
Dramatic performance and improvisation. As just noted, IF doesn’t offer much opportunity for the player to perform a role, in the sense of creating a characterization from scratch. There are two challenges about doing this that would be hard to surmount: first, that such performative behavior tends to involve very diverse, freeform kinds of input which would be hard for the parser to interpret meaningfully; and second, that most of the pleasure of performing comes from having an audience to entertain. Even if one could embellish one’s PC’s behavior with an assortment of quirks, gestures, facial expressions, and witty asides, the computer is in no position to appreciate these actions. (Actually, I can think of an exception: when I am playing a game as a beta-tester, I do sometimes give nonsense instructions like >LAUGH WICKEDLY, even though the parser won’t respond meaningfully. In these cases, though, I am performing for the benefit of the game’s author, to whom I will eventually send the transcript. But as a rule, if no one is reading over my shoulder and no one is going to see the transcript, I do not indulge in much of that kind of play.)
I might cite a few IF works where the game recognizes and rewards the player’s in-genre attempts to make the PC behave in character (“Tale of the Kissing Bandit”, “The Act of Misdirection”), but these are still cases where the author has designed the character entirely and rewards only one style of characterization.
The social aspect also tends to strengthen moments of dramatic choice in a story. My favorite moment from an RPG comes from a Seventh Sea campaign. I played a courtesan who secretly also possessed magical powers — powers that, if revealed, might result in her death, since persons of her class were not supposed to have such abilities. One of my comrades in the game was an unsavory male character; he tended to leer and wink, and I tended to ignore him, since he couldn’t afford my character’s services commercially. In the final showdown of the game, the RPG rules permitted me to use my magic to bestow special combat skill. Things were going so badly for us that I gave the unsavory fighter this magic kiss.
The *mechanics* of this moment would be possible to translate into IF, given appropriate rules. But the other story-telling nuances could only be appreciated by humans: that my character didn’t want to kiss the fighter at all, that she’d put herself in his power by revealing a life-threatening secret, that the whole thing was a desperate gamble. In general, it’s possible to let the player make meaningful choices and discover creative solutions in IF, but only within the predefined interaction domain of the game; it’s not possible to let the player invent whole new thematic content or add character nuance. Which brings us to
Participation in inventing the story at a high level. IF doesn’t allow much scope for the player to be a collaborative storyteller. The player doesn’t get to propose new events, as players do in some kinds of RPG. He doesn’t get to invent bits of setting. He doesn’t get to engage creatively in the environment in the same way that the author does. There are occasionally places where a player may be allowed to customize some aspect of the game world, but these tend to be narrow (selecting a name and gender for a character) or unimportant (writing messages or otherwise injecting text into the game world in a way that has no effect on the narrative). The player doesn’t mostly get to mess with pacing or decide what kind of scene will happen next.
The closest approximation I’ve seen was Mousechief’s independent game “The Witch’s Yarn” — not IF in my sense, though its designer calls it interactive fiction and it shares many of the constraints of an adventure game — in which the player is allowed to “cue” new characters to come on stage and take action. At its best, this allows the player to direct the story towards conflicts and events he finds interesting; but that happens relatively seldom. Much of the time the player is working to solve a puzzle rather than shape the narrative. Another possible contender is “Delightful Wallpaper”, in which the player plays a kind of supernatural arranger of events and can solve the puzzles in any of several ways to bring about an internally consistent narrative about the non-player characters; but all the same, the system is fairly tightly constrained, without the freedom of a live RPG.
“Shade” and IF coding
Jeremy Douglass’ article on “Shade” grapples with the meaning of second person narrative in IF, the choices about what to communicate to the player, the manipulation of player experience, and IF’s peculiar obsession with light sources.
The content of the article will not mostly surprise anyone who has hung around the IF community a long time and studied “Shade” already. I did find it interesting to read, though, just after I had finished writing a technical document comparing the entry point routines in Inform 6 and Inform 7. “Shade” is written in Inform 6, and what Douglass says is correct: the library is very much concerned with light sources and light transmission. The code of “Shade” reflects this.
Inform 7, as it turns out, makes different assumptions: it makes rooms lit by default, and it does not contain as many hooks for the author to manipulate the behavior of light and darkness. What’s more, this is one of the few areas where I7 is less capable than Inform 6 and yet hasn’t generated many author complaints. (It would be a trivial matter to create an extension reintroducing the missing abilities. But they are no longer part of the standard library set.)
This reflects a real change in the attitude of IF authors since the early 90s when Inform was first developed. Light source puzzles are much less common, but also, in general, IF authors have moved somewhat away from the idea that a single standard world model is appropriate for most or all works. The latest generation of IF languages approach this idea in different ways — I7 by abolishing many of the more esoteric features of Inform 6’s world model, TADS 3 by offering a library with greater abstraction and more hooks for customization than the TADS 2 library.
Genre and Medium; Translating Stories
Periodically IF authors write works retelling existing stories.
Because the IF community is an amateur community producing freeware, this happens much less often than it probably would in a commercial environment: on the one hand, no one has the budget to buy a license to popular intellectual property, and on the other hand, there is no real marketing incentive to do so. And while some authors might want to produce derivative IF just because they like a given story or game universe, certain institutions of the IF world discourage the “if we’re not making money, we probably won’t be sued” approach found in fanfic communities. The IF Competition, for instance, accepts no copyright-infringing works, so anything entered there must be of the author’s own creation, based on a public domain source, or created with the permission of the copyright holder.
The few translated-to-IF stories have met with mixed reactions. In my reviews of IF Comp 2006 games, I commented on “Tower of the Elephant” (adapted from a Conan story) that I thought it was one of the better translation efforts I’d seen from static to interactive fiction; but it still didn’t offer a huge amount over the static version of the story, I thought. It did offer a strongly-styled prose that one might not have found in IF otherwise — but that just means that it’s better in one dimension (and worse in other dimensions) than other IF works, not that it’s better than the written story. That particular case, and two Dunsany adaptations released shortly afterward by Peter Nepstad, have provoked some discussion in the IF community about whether good adaptation of static fiction is possible, and if so, how. (And for that matter whether it’s desirable, and why.)
Second Person sheds some light on this problem, since it includes essays on efforts to translate existing stories or story environments to RPGs, collectible card games, and board games. Eric Lang’s essay notes:
A helpful mantra is: Respect the Medium. Collectible card games are at their root combinatorial exercises; players fall in love with possibilities as much as they do with strategies, and no other type of game offers more options. As long as players could conceive of the possibility that an event might take place in the world of Westeros, I had to make sure it could be done in the game. And in many cases, things that could never happen in the books could also be done. (88)
Some IF adaptations of stories do this to a limited degree: “Tower of the Elephant”, for instance, offers several alternate, plausible solutions to a situation, even though only one of them occurs in the original story from which it derives. Nonetheless, the urge to get the player to recapitulate the elements of a specific story may not always play to IF’s strengths. Lang also remarks that collectible card game design tends to focus on trying to evoke the qualities of a setting and genre, rather than specific events. Admittedly, IF allows for much more coherent narrative than the average CCG; but I suspect that successful IF adaptations of stories will generally tend to build heavily on the setting of the original and on the protagonist’s abilities, rather than on the specific chain of events in the story.
Possibly related is James Wallis’ point about genre:
Intentionally or not, [Edward Gorey’s shuffled story The Helpless Doorknob] hits one of the four cornerstones that underlie any successful game or system that allows players to actively manipulate a story: a clear understanding, encapsulation, and communication of its genre. (73)
This does not necessarily mean that interactive storytelling needs to conform rigidly and unimaginatively to the cliches of some existing literary or cinematic genre; and indeed Jordan Mechner elsewhere argues effectively that game design is strengthened if the designer combines previously uncombined genres (115). But genre, in this context, means an understanding, shared by the author and the player, of the kinds of events, actions, and atmospheres that are appropriate to the kind of story that is being told. (On this point, see also Rebecca Borgstrom’s article on the understood genre and player contract within “Exalted: The Fair Folk”.) Genre defines what we expect the player character to do (something quite different in detective noir than in a romance novel). It ties into Nick Montfort’s concept of the player character as a product of the constraints and perspectives at work on him.
Infocom’s “Hitchhiker’s Guide to the Galaxy” adaptation works in part because it does follow this guideline: without precisely recapitulating the events of the book, it adopts the book’s essential sensibilities about what can and cannot happen. Within the universe of the book, it is appropriate for the character to spend a lot of time on frustrating tasks (such as getting the babelfish from its dispenser).
The pessimistic theorist might say at this point that we’ve just proven that IF cannot yield stories as artistically satisfying as static literature does, because it cannot reproduce the same intricacies of plot. Some already have argued this, in fact: adding that the repetition and computer-generation of IF text undercut the aesthetic merit of interactive fiction, which can never approach the taut, crafted pacing of static fiction; that the concern of IF with player agency makes it impossible to introduce nuanced characterization; that IF is by nature doomed to puerile content and mediocre presentation. Several remarks in Second Person might be read that way as well:
The goal of a storytelling game isnt to produce a good story; it’s to participate in good storytelling… The point of fiction as a medium is to yield a product — a story — worthy of being read. That is not a kind of storytelling that most games are good at, as a medium. Games are good in the moment. Games are anecdotal. Stories that develop over the course of gameplay are personally exciting and meaningful in a way that movies and novels aren’t, but they achieve this level of personal meaning at the expense of secondhand meaning. No one but the players are included in the excitement. (52, Will Hindmarch)
or, in Chris Crawford’s essay about a story told in his interactivity-supporting language Deikto,
This example demonstrates just how different interactive storytelling is from conventional storytelling. In terms of conventional storytelling, this story is wretched: there is no color, no subtlety, no nuance. However, this kind of storytelling boasts one advantage — and only one — over conventional storytelling: it is interactive. (175)
As it happens, I disagree with Crawford that interactive storytelling need be quite as bare and nuance-free as his example (an argument for elsewhere); and I don’t think Hindmarch’s point is necessarily as gloomy as it might at first sound. If IF is an engine for generating story experiences that are personally meaningful to the players, what’s wrong with that? Is it necessary that a reading of the output be as good as the playing experience, for the work to have aesthetic value? As Paul Czege points out in this volume (though in another context),
Story happens only in retrospect, when you pick from all the details of the game session and organize them for yourself…
and there is, I think, something in this: the story (as opposed to the text) is constructed in the mind of the reader by the work.
The Problem of Interactive Character
I had originally meant to attempt a detailed summary of three chapters on interactive characters: Steve Meretzky on Floyd in “Planetfall”, Chris Crawford on Deikto, Stern and Mateas on “Façade”. I’ve changed my mind. Meretzky’s piece is amusing but brief and primarily of historical interest. (Second Person intersperses long essays with shorter pieces; these are mostly by authors about some small aspect of their own work, and the word limit encouraged concision. I think the overall effect is good and makes the book more diverse and more readable, but there were a few of these short pieces where I would have been glad to read more from the author.) Conversely, there is too much content in the other two chapters — especially the one on Façade — for even a long summary to do them any justice.
That last is a long, technical, and detailed article on how the narrative of Façade was constructed from small elements and an over-arching drama manager. Stern and Mateas are clear-eyed about the weaknesses of the final product — the somewhat unsympathetic characters, the lack of global agency (it is hard for the player deliberately to direct the whole course of the story), the times when the natural language input fails to convey anything to the computer at all, the feeling that the story/game/drama is not always much fun. They’re also clear-eyed about the strengths to be taken from it. They have sensible things to say about interpreting the player’s physical or spoken acts as socially significant behavior, and responding to that behavior contextually. Anyone interested in the technical details really should acquire the article and read it. Not everything they say would apply to IF (which is not real-time and is also not usually trying to interpret natural language conversation); all the same, I could see some of their drama management ideas being of use for an IF conversation engine.
I expected to be — and was — somewhat less enthused about Deikto. I have reservations about the grand experiment of interactive storytelling Crawford is currently working on, which I’ve expressed already elsewhere; and when it comes to Deikto, his simplified language for talking to characters, I tend to share a lot of Stephen Bond’s reaction. I’ve already quoted Crawford’s line about how nuance is lost from his Deikto story; my notes add, unsympathetically, “One might also say there’s no structure, no theme, and not enough intentionality for the player.”
I know I said a little earlier that IF’s strengths don’t lie in quite the same places as the strengths of static fiction, and that IF could be excused for producing streams of text that would be crap in a novel. But that is not quite the same as saying that the language doesn’t matter. I’ve enjoyed a number of IF games with inferior interaction on the strength of their good writing alone; while good writing for an IF game is not always the same as good writing for the printed page, that doesn’t mean that no such thing exists. There are plenty of plays that are dull to read but riveting when staged, too. (It’s a pity that so much literature-class exposure to drama takes the form of reading. I was early convinced that I didn’t like drama that much — at least, not nearly as much as I liked novels — until I actually saw some live.)
I’m not crazy about Crawford’s model of people, either. His base class of actors, he reports, have the following characteristics:
location (which stage the actor is occupying)
spying on (whom the actor is spying on, if anybody)
strength (overall physical strength)
wealth (how rich the actor is)
loquacity (how talkative the actor is)
This strikes me as a deeply peculiar collection, especially “spying on” — though I suspect it may just mean something along the lines of “the person or people this actor can currently see”, and thus represent something to do with scoping, rather than implying that all storylines incorporate some espionage. Nonetheless, surely some stories would call for a different set? If we (like Stern and Mateas) were drafting a bit of pseudo-Albee, maybe instead we’d want
degree of drunkenness (how much the actor has had so far)
hitting-on (which character the actor is currently trying to get into bed)
wit (overall ability to score points in debate and repartee)
flippancy (how much the actor values jokes over sincerity or kindness)
boredom (how likely the actor is to walk out of the room)
In other words, I think Crawford is committing just the mistake that IF languages are currently moving away from: the mistake of assuming that one specific model is going to provide an adequate basis for all the kinds of procedural storytelling he wants to do. For any given kind of story, it might be possible to enumerate a set of five or ten procedurally significant character traits which should enter into calculations about a character’s behavior. But which traits those are, and how they affect character reactions, must surely vary depending on the type of story it is.
Even with all that, though, I was a little, and grudgingly, intrigued by Crawford’s example in which his player is able to bargain with an NPC that she should kiss a third character in exchange for the gift of a candlestick. It’s a silly suggestion, but clearly the system allows much more detailed social bartering than IF is currently able to express. I’m wistfully drawn toward the concept of a game incorporating such complex personal interactions. I’m just also discouraged again by Crawford’s examples of the kinds of storytelling that would result, and the ways in which these stories would be presented to the player.
If there’s ever an example written in Crawford’s Storytron system for me to play with, I reserve the right to change my mind and be impressed. But I’m also not expecting this to happen. In the meantime, is there anything cool we can do with social negotiations in interactive fiction? I earnestly offer you candlestick in exchange for you write that game.
The Cultural Place of Games and New Media
Second Person ends with a section on “The Real World”. The essays here are even more diverse than those that come before. Sean Thorne writes on using a role-playing game — John Tynes’ Puppetland — to give creative writing students something to write about. Ian Bogost and Gonzalo Frasca describe the creation of their game “Howard Dean for Iowa”, meant to support Dean’s campaign for president by teaching players about methods of grassroots activism; Kevin Whelan talks about role-playing scenarios used to prepare door-to-door activists. Jane McGonigal writes about “I Love Bees”, flash mobs, and what she calls power plays — games that involve getting groups of people to act out unusual scenarios in concert. There are also essays on soundscapes players explore by moving through physical spaces with a GPS device; chatbots; LARPs with real-world consequence; improv comedy games and interactive theatrical performances; social communities arising from online play.
The presence of this section at all is as interesting as its content: the essays here assume that role-playing, gaming, and interactive story-telling are not trivial, time-wasting, or culturally valueless, but that they have the power to persuade, educate, and enlighten.
IF is under less pressure than commercial video games in this respect; the video game industry is constantly being attacked for corrupting the youth of America. (Like Socrates, they must be doing something right.) But the amateur community certainly has its fits of self-doubt and uncertainty about whether IF can be artistic, or educational, or meaningful as something other than an escapist diversion. I think we’ve already proven that in some respects it can. Still, there’s a great deal of inspiration in this sampling of real-world applications of other games and new media.
There is much else that could be said: I’ve only been able to point out a few of the highlights, and have had to omit much that I found relevant, entertaining, informative, or even moving. This is fine: the book is worth buying and reading in full. To be honest, Second Person blew me away. I was expecting good things when I saw the contributor list, but I wasn’t expecting anything quite this good. The book explores abstract theoretical and political questions that come with the kind of work we do; it also provides thorough, detailed discussions of design problems and solutions, illustrating many articles with images, diagrams, code snippets, and even full-length sets of RPG rules. It spans the production spectrum: professional games intended for large commercial audiences, start-up games by indie designers, hobbyist and fan freeware, academic experiments, pedagogical tools, one-time art installations.
The result is both grounded and inspiring. Of all the books on game design or new media I’ve read, Second Person is the one I would most strongly recommend to IF authors seeking a deeper understanding of their own work and of the work being done in related fields.