I've written something. It's been inspired by Victor Gijsbers's work
and commentary on interactivity in IF, and by the recent coversations
between Stephen Bond, Emily Short and Nick Monfort on player freedom
-- essentially, it's an exploratory philosophical study of the
concepts of authorship and meaning creation as they relate to
The essay comprises a brief review of historical conceptions of
authorship, four case studies of interesting approaches to IF
authorship (Photopia, Shade, De Baron, and Galatea, all quite heavily
spoliered), and some ending comments. It is neither scholarly not
conclusive, but rather intended as material for discussion. If anyone
would like to discuss it. It may be that, er, a beta-version of the
essay, which is more scholarly and systematic, might be written, but
I'd value any comments people have.
Anyway, it's called "Greatly Exaggerated", which is a moderately
obscure literary pune, or play on words, and can be found at the
following address, if this newfangled Google Documents thing works
I hope someone finds it interesting. Some extracts follow.
>From the introduction:
"A postmodern view of aesthetics (the study of art) calls into
question the concept of the author as sole creator and determiner of
meaning of eir artwork. The view both considers the influence of the
artist's sociocultural background in creating eir artwork, and the
importance of the audience's role in determining an artwork's meaning.
If we ask "What does this artwork mean?", can only the author give a
full answer? must the audience attempt to fathom the author's
intention? or is interpretation the privilege of the audience only? If
we ask "Who created this artwork?", does the notion of the artist-as-
creator hold? can the artist be considered a conduit for the
expression of sociohistorical assumptions? or can it be said that it
is the audience's interpretation of the artwork is what truly creates
the work as an artwork?
"In any case, asking such questions indisputably highlights that the
processes surrounding artistic creation must be considered as an
interaction between the artist/s, the audience/s, and their world/s.
If interaction plays a role in all art, then artwork which is
explicitly interactive, such as IF, is going to have real significance
for these questions. If this essay has an argument, then it is that IF
as a medium amplified the effect of these questions - that the
artistic effects of postmodern aesthetics, which explore the
involvement of the audience in the artwork, are exaggerated by
explicitly interactive art."
>From the study of Photopia, including a MAJOR SPOILER:
"Cadre, then, relinquishes very little authorship to the audience. In
fact, that Photopia is a work of interactive fiction in fact decreases
audience control. In static fiction, lacking IF's realtime element,
the audience is free to treat the work in a wider range of ways:
skimming, repeating passages, skipping to the end, and so forth. More
significantly, that the audience is always playing as a particular
character, and that that character is always given motives and
reactions by Cadre, gives the audience far fewer degrees of freedom in
interpretation: the audience is always given a strong motive to do a
particular action and react a particular way. Through this, and
through thematic lucidity and beautiful prose -- in short, skilful
writing --, Cadre is largely in control not only of the text but also
of the audience's reaction to the text. In terms of meaning creation,
then, it is Cadre's intentions which become important to a degree
usually not seen in static fiction. As expected, interactive fiction
exaggerates what static fiction traditionally performs.
"There is an identification between Cadre and Alley. Just as Alley is
a character we desperately want to know, Cadre is an author we want to
accurately interpret -- Cadre has written his work in such a way that
we are in the thrall of his story. (Moreover, to borrow from Barthes,
just as we cannot truly know Alley, nor can we truly know Cadre: his
conscious intentions remain technically unknowable to us.) Alley and
Cadre are Authors supreme. And while it is Alley who dies in the
story, she remains in control of our emotions for a long time
afterwards. In Photopia, it is not the author but the audience who
I've been thinking about this since you posted it, and am still
puzzling over this bit:
> dialogical IF need not restrict itself to NPC programming:
> the landscape, objects and worlds of IF works can also be
> given a sort of "personality", or at the least a complexity of
> narrative depth allowing the audience to dialogue with the author.
I can imagine a landscape with "personality" (though I think I would
call that "atmosphere", which I'd say is already found in plenty in
works such as "Anchorhead" and "Sunset Over Savannah"). But that seems
like something different from "complexity of narrative depth", and it
doesn't really allow for "dialogue with the author" in any major
sense. Do you envision something in which (say) the order and
thoroughness with which the player examines items changes the way
they're described? Or affects the plot itself? Or am I on the wrong
Well, I've very glad you thought it worthwhile to do so. It'd been
sitting in my head for a few days, and on that day it decided to come
out. I just reread it and was struck by just how very lightly sketched
(and badly typed) it is -- very much an alpha-test. With this reply
you've encouraged me a little to take it to beta, so thank you.
(I've never applied the testing metaphor to essays before, but it's
really quite apt, isn't it?)
> and am still
> puzzling over this bit:
> > dialogical IF need not restrict itself to NPC programming:
> > the landscape, objects and worlds of IF works can also be
> > given a sort of "personality", or at the least a complexity of
> > narrative depth allowing the audience to dialogue with the author.
> I can imagine a landscape with "personality" (though I think I would
> call that "atmosphere", which I'd say is already found in plenty in
> works such as "Anchorhead" and "Sunset Over Savannah"). But that seems
> like something different from "complexity of narrative depth", and it
> doesn't really allow for "dialogue with the author" in any major
> sense. Do you envision something in which (say) the order and
> thoroughness with which the player examines items changes the way
> they're described? Or affects the plot itself? Or am I on the wrong
> track entirely?
I rather threw that in there, didn't I? Let's flesh it out a bit.
Anchorhead the town certainly has a personality -- if a rather
psychotic one. In fact, Anchorhead as a location could be considered
the meta-puzzle of the game: one meets the town, gets to know it, has
to deal with its behaviour, learn about its past, and finally vanquish
it. (Aside: Anchorhead's NPCs, apart from perhaps Michael, are quite
thin, really -- as if they were part of the whole town's personality.)
In order to do all this, one has to work out what is going on in
Anchorhead's mind -- which is a construction of the author's mind. So
we poke and prod at Anchorhead, trying to get it to tell us its
secrets. At the same time, Anchorhead reacts to what we do it,
although in this case in a linear plot kind of way. The town can hurt
us, imprison us, kill us. There is a lot here that resembles dialogue,
and it's of the kind we find in all good IF.
But that's cheating: I'm taking a classic, thoroughly fleshed-out game
and using dialogical terms to describe its merits. So let's think
about something more unusual. "For a Change", I think, provides some
more meat to what I'm talking about. Because the game uses a
surrealistic non-standard English, one is forced to actively engage
with the game in order to get it to divulge its secrets. From your IF
background, you have a bank of things you can say to the game, and the
game responds to you. In order to progress, one has to accommodate
oneself to the game until one understands the way it works -- has
restructured one's mind in order to deal with the game. At the same
time, the game is accommodating itself to you, restructuring itself
for you -- because part of dealing with the game is developing a
picture/idea of what is going on. This idea is going to be unique to
every individual because of the surrealistic language: everyone will
interpret it differently, more so than perhaps any other IF. So the
classic elements of a dialogue are here: the entities involved are
interacting with and reinterpreting each other in no trivial way.
(Aside: much of this paragraph could also have been said about "Bad
Let's think about what *wouldn't* be diaological as well. I think the
Earth and Sky trilogy, especially the first and last games (the second
game similarly has a surreal world one accommodates oneself to and has
to interpret in an individual way), provide some material. One doesn't
explore these games. Action takes place linearly at a comic-book pace:
one situation follows another, building up in rapid succession.
Moreover, the puzzles are generally quite didactic: one is told what
the suits can do; one is always given very clear and direct motives.
The game tells us a great deal. Sure, we type in input, but we are
generally told from either convention or the game's text what to type
(for this particular factor, Photopia again is an extreme case).
There's no real engagement with the game -- and the game certainly
doesn't need to accommodate itself to us in any way. (Note: none of
this means I think they are "bad" games -- simply that they are non-
dialogical in the sense I've been trying to explore.)
A couple more brief examples: "Symbolic Engine" consists of a series
of lectures, and it's quite difficult for the player, despite
technically being able to set which lecture is given, to really have
much understanding of how to elicit particular responses. And imagine
what the game would have been if one could have argued with it, or
asked it specific questions! In the other direction, puzzlefests like
the Muldoon Legacy make the player do all the legwork: they give the
player a series of tough brainsqueezers to solve, requiring the player
to twist eir mind to fit the game and work out what to do. The game
doesn't give much to the player -- doesn't adapt itself to the player.
This may partly be my playing style: I know some people approach
straight up puzzles in a very dialogical way: asking the puzzle
questions by trying various things; seeing what the puzzle has to tell
them. With me it's more either me getting the idea straight away, or
after a little thought, or turning straight to the hints. So another
example of non-dialogical puzzles are guess-the-verb style puzzles,
puzzles where, unless you've got the right answer, the game will tell
you nothing to help you get it.
Finally, if an idea of "dialogical" IF is beginning to form, and while
I hope it is I may still be being hopelessly vague, let's think about
what its potential might be. I can see a lot of IF Art Show entires
striving for the kind of thing I've been talking about -- indeed, the
brief for the competition positively encourages it. "The Fire Tower"
exhibits a lot of the elements I've been talking about: the player
explores the game, and the game adapts itself to the player -- in this
case by providing multiple different narratives of the walk: one can
charge through the scenery for a bracing hike, dawdle in just a few
places, examining everything, one can treat it as a wildlife-spotting
expedition, or as a botany lecture. Time-limiting helps in providing
all of this, as does restricting backtracking. How could the effect
have been heightened? Perhaps by the game realising how you were
treating it -- changing the way things were described if you rushed
through, for example. Perhaps by giving dawdlers internal reflections
as well. Just as "The Fire Tower" involves a dialogue with a
landscape, so to does another IF Art piece, "Swanglass", strive for
something dialogical with an object (and a landscape. And a player-
character.). The author has stories to tell, but the audience has a
large interprative role in understanding those stories. What could
improve things here? Well, perhaps *less* enigma -- so that it's less
as if the author is holding something back from you, and more as if
you can work with the game to reveal things. (This is what happened in
Right. Let's try and round things up. What I'm talking about, in
essence, are games which feel like a conversation. There is a
traditional understanding of puzzle design which can inform this, but
that's not all I'm talking about (if all there are is puzzles to poke
and prod at, the game is still not adapting itself at the meta-level
to the player.) Multiple plotting possibilities helps here, because
then the game adapts itself to the way the player decides to play it.
Having a narrative also involves limiting player freedom in crucial
ways: generally, a game which allows the player to do anything,
explore every possibility, isn't dialogical either, because then the
player doesn't have to accommodate emself to the text at all. Another
thing which helps is giving the player material to interpret --
allowing the player mental freedom to understand things in eir own
I'm going to mention what I think is key again: for there to be
dialogue, the entities involved must change themselves for each other.
For there to be dialogue, the player must change to understand/solve
the game/entity, and the game/entity must change for the player,
through allowing interpretation, reacting in its text, providing
narratives depending on the player's behaviour.
This idea of dialogical writing encompasses a lot of what other IF
theory has covered. It's essentially an approach to IF thoery rather
than an element within it. More things to explore here include looking
at *why* this kind of gaming provides particular kinds of
satisfaction, and probably more on what the potentials of it are.
That's probably a lot more answer than you were looking for. Er. Sorry
about that. I find it difficult to stop sometimes, and without a
concrete idea of what I'm talking about, a lot of waffle is involved
as the idea shapes itself in front of me. I hope you (or others!)
still have things to say -- of so, I really look forward to your
I admit I am a bit bothered by your use of "dialogical" when you are
citing examples of what is already called "adaptive".
For example, if City of Secrets detects a beginner is playing, it will
automatically switch to an easier difficulty level.
There are games which are literally a dialogue (try "Astronomy Without
a Telescope" at http://www.ifarchive.org/indexes/if-archiveXbooks.html)
but do not match well with the ideas you are presenting above. It's
essentially a labyrinth rather than a conversation.
Finally, the word itself is not that great an invention. I've heard
use "dialectical" but not "dialogical". Something like IF-as-dialogue
cause less eyebrows to curl.
-- Jason Dyer
Just to situate myself a bit: I'm still fairly new to really working
with IF, and my background is in academic philosophy, the performing
arts (especially CIRCA clowning, from which I get most of my art
theory and practice), and, a little tangenitally, in the radical
politics of the newest social movements. This means that my language
and approach has a particular sort of background -- which I try to
tone down a bit, but still leads me to waffle and use particular
On 31 Jul, 21:29, Jason Dyer <ditt...@gmail.com> wrote:
> On Jul 31, 11:46 am, Lodestone <harry.lodest...@gmail.com> wrote:
> > This idea of dialogical writing encompasses a lot of what other IF
> > theory has covered. It's essentially an approach to IF thoery rather
> > than an element within it. More things to explore here include looking
> > at *why* this kind of gaming provides particular kinds of
> > satisfaction, and probably more on what the potentials of it are.
> I admit I am a bit bothered by your use of "dialogical" when you are
> citing examples of what is already called "adaptive".
> For example, if City of Secrets detects a beginner is playing, it will
> automatically switch to an easier difficulty level.
Well, I'm indeed citing adaptive examples, but I'd hope it's clear
that adaptive systems in games are only a part of the general approach
to IF I'm trying to sketch.
Certainly I ought to be using some of the more understood terms when
talking about understood concepts, but as I said, I'm still fairly
new, so am not always on the ball in this regard -- so thank you for
pointing it out.
> Finally, the word itself is not that great an invention. I've heard
> use "dialectical" but not "dialogical". Something like IF-as-dialogue
> cause less eyebrows to curl.
"Dialectical" is something else entirely -- it refers to processes of
the development of ideas. Marxism involves "dialectical materialism"
in that marxist theory incorporated a particular theory of how ideas
and socities evolve, for example.
"Dialogical" is certainly not my own coinage. In aesthetic theory its
use can be traced to Mikhail Bakhtin and a work usually translated as
"The Dialogic Imagination", contrasting what he called the
"monologic" (single, authoritative, closed) with the "dialogic" (open,
relational, dynamic). "Dialogic" is the term used, but "dialogical" is
the more common current usage, largely due to its currency in Freirean
politics and teaching methods, which are much-loved by postmoderns of
many stripes. There's a subtle difference between the two (think about
ironic/ironical), but there it is.
In any case, the term still has a great deal of currency in aesthetic
theory and practice (and in politics), which is how I came to be
thinking about IF in those terms. I think the concept has a great deal
to offer IF, though certainly what I've sketched out is minimal at
Thanks for commenting,
> "For a Change", I think, provides some
> more meat to what I'm talking about. Because the game uses a
> surrealistic non-standard English, one is forced to actively engage
> with the game in order to get it to divulge its secrets. From your IF
> background, you have a bank of things you can say to the game, and the
> game responds to you. In order to progress, one has to accommodate
> oneself to the game until one understands the way it works -- has
> restructured one's mind in order to deal with the game. At the same
> time, the game is accommodating itself to you, restructuring itself
> for you -- because part of dealing with the game is developing a
> picture/idea of what is going on. This idea is going to be unique to
> every individual because of the surrealistic language: everyone will
> interpret it differently, more so than perhaps any other IF.
Have you tried "The Gostak"? It's even more open to this process than
"For a Change", but (I have to admit) I didn't get very far in it --
perhaps because I felt as though I was having to do so much
interpretation that it was becoming an exercise in invention, instead.
I appreciate the cleverness of the piece, but felt that I'd gotten the
basic idea out of it within a few minutes of play, and from there on
it was just going to be hard work. Other people really enjoyed it,
played through to the end, and had conversations about what their
different interpretations were, so clearly my reaction wasn't
universal. Still, my own feeling is that it's possible to go too far
in the direction of making the player provide meaning for the piece.
(See also "On the Other Side".)
> Let's think about what *wouldn't* be diaological as well. I think the
> Earth and Sky trilogy, especially the first and last games (the second
> game similarly has a surreal world one accommodates oneself to and has
> to interpret in an individual way), provide some material. One doesn't
> explore these games. Action takes place linearly at a comic-book pace:
> one situation follows another, building up in rapid succession.
> Moreover, the puzzles are generally quite didactic: one is told what
> the suits can do; one is always given very clear and direct motives.
> The game tells us a great deal. Sure, we type in input, but we are
> generally told from either convention or the game's text what to type
> (for this particular factor, Photopia again is an extreme case).
I'd say this kind of highly linear, highly guided design is relatively
uncommon in IF: by your definition (as I understand it, anyway) most
IF *would* be at least moderately dialogical.
> I know some people approach
> straight up puzzles in a very dialogical way: asking the puzzle
> questions by trying various things; seeing what the puzzle has to tell
Yes; I think that's fairly common among people who like puzzle-heavy
games at all. You try various things, you poke at the puzzle, you make
guesses about what's really going on; if the puzzle is well-designed,
it has good feedback for incorrect attempts. This back-and-forth is
part of what keeps the puzzle fun even when I'm stuck and can't solve
it. The puzzles in Zork III come to mind here: most of them are very
abstract set-pieces which have little or no narrative framing and do
not participate much in an overall plot, but I still enjoyed working
through them because they were in general very fairly designed and
responsive to wrong guesses.
(I wonder what you would make of Spider and Web here -- the puzzles in
it have more than the usual claim to being dialogue-like, I'd say.)
Mm, the Gostak. There's an awful lot of material there.
Your "too far" comment comes back to something important I've been
trying to emphasise: that in dialogical art, it's important the all
the entities involved to be engaging with each other. (In classic
terms, there must be both give and take). It's possible for the author
to dictate to the player -- either if the purpose of the interactivity
is purely to aid identification, as in Photopia, and so the game keeps
things highly linear, or if the text makes all the puzzles so obvious
and easy that the player is never required to give anything for
emselves, and so on. Similarly, it's possible for the author to really
remove themselves from the text (as I argued Victor Gijsberg attempts
in De Baron), or for the author to make the game so wilfully obscure
that too much is demanded from the player (part of what I felt the
problem was with Swanglass). In Shade, I argued one could either see
it as the author asking us to puzzle him out or the author offering us
something to allow us to puzzle ourselves out (I prefer the latter).
> > Let's think about what *wouldn't* be diaological as well. I think the
> > Earth and Sky trilogy, especially the first and last games (the second
> > game similarly has a surreal world one accommodates oneself to and has
> > to interpret in an individual way), provide some material. One doesn't
> > explore these games. Action takes place linearly at a comic-book pace:
> > one situation follows another, building up in rapid succession.
> > Moreover, the puzzles are generally quite didactic: one is told what
> > the suits can do; one is always given very clear and direct motives.
> > The game tells us a great deal. Sure, we type in input, but we are
> > generally told from either convention or the game's text what to type
> > (for this particular factor, Photopia again is an extreme case).
> I'd say this kind of highly linear, highly guided design is relatively
> uncommon in IF: by your definition (as I understand it, anyway) most
> IF *would* be at least moderately dialogical.
Certainly we're talking about things on a continuum, here. As well as
that, if it is useful to consider dialogical art theory as an approach
to IF overall, rather than something within IF theory, then we're
going to be able to apply it in different ways to different things.
You outline the conventions of a good dialogical puzzle below -- and I
think Galatea is a sublime exercise in making NPCs dialogical. I've
tried to describe the way objects and landscapes can have some of
these qualities overall, and then there's the business of the games
overall and some of the more adventurous techniques. And so games may
display all, some, one or none of these aspects, to a greater or
lesser degree. Goodness, how informative.
I like that "as I understand it", by the way. I've kind of talked
around the concept rather than right up to it, so I'd be interested to
hear what you think I'm talking about. I'm sure I don't know!
> > I know some people approach
> > straight up puzzles in a very dialogical way: asking the puzzle
> > questions by trying various things; seeing what the puzzle has to tell
> > them.
> Yes; I think that's fairly common among people who like puzzle-heavy
> games at all. You try various things, you poke at the puzzle, you make
> guesses about what's really going on; if the puzzle is well-designed,
> it has good feedback for incorrect attempts. This back-and-forth is
> part of what keeps the puzzle fun even when I'm stuck and can't solve
> it. The puzzles in Zork III come to mind here: most of them are very
> abstract set-pieces which have little or no narrative framing and do
> not participate much in an overall plot, but I still enjoyed working
> through them because they were in general very fairly designed and
> responsive to wrong guesses.
> (I wonder what you would make of Spider and Web here -- the puzzles in
> it have more than the usual claim to being dialogue-like, I'd say.)
Heh. I'll think about that. I think I'm not in the best position to
approach it because I found it so very hard (I'm not a great puzzler),
and so that clouds insight.
As far as the puzzles themselves go, they certainly fit some of the
patterns you've described. They give feedback in the game's unique and
special way, as well as in the usual poking-and-prodding way. They
don't go so far as you do in Metamorphoses, though, to adapt to
different players and allow multiple solutions. The plotting, also, is
beautifully responsive to the player, allowing some aspect of the
"moral choice" genre as well as in the various sub-optimal endings --
but Andrew Plotkin is throwing some of his own ideas out there for you
to work with as well. The game's meta-puzzle and its psychological
switch involve this important give-and-take aspect too. But on the
other hand, there's still a lot of linearity there -- especially
present in the conversation with the interrogator, which is the only
part I wish had been more creatively implemented. While I think the
yes/no parser is justified, I would have liked to have seen, for
example, the way the player responds to some of the interrogator's
more philosophical questions to have had more impact on the arc of the
So I don't know. What do you think?
> I like that "as I understand it", by the way. I've kind of talked
> around the concept rather than right up to it, so I'd be interested to
> hear what you think I'm talking about. I'm sure I don't know!
Well: so far it seems to me you've attempted to describe "dialogical"
experientially (as that quality that makes a game feel more like a
conversation); by feature (as the use of any of a list of techniques
such as plot branching, openness to interpretation, and response to
"play-style"); by example (as that quality exhibited most strongly by
the games you call most dialogical, and least strongly by those you
identify as least dialogical); by effect (as that quality that allows
the game to respond to the will of the player and engage the player in
creating meaning); and by reference to the meaning of dialogical in
other theoretical discourse (as that quality that allows any work of
art to engage other works of art and create a meaning that arises
meaningfully from the relation between them, or insert your own
summary of that Wikipedia article here).
Some of these strike me as more useful forays than others. The
experiential is, of course, subjective and likely to vary from person
to person; there are probably some who would say that *all* IF is
"like a conversation", and indeed a Spanish term for interactive
fiction is "aventuras conversacionales" (see http://es.wikipedia.org/wiki/Aventura_conversacional
). I also find the reference to Bakhtin, et al., minimally
enlightening in this context, since the kind of relational meaning
you're talking about with IF is that which develops from dialogue with
the reader/player, rather than through intertextual reference and
dialogue with the meaning of other works of art. There is certainly a
great deal of reference and intertextual exploration in IF games --
too much, I sometimes think -- but that is not what you have been
Definition by example communicates what games you want to talk about,
but since you've named such a broad spectrum of work, it's hard to
condense this into a specific set of criteria. Definition by effect is
clearer. I suspect not everyone agrees that it's desirable for IF to
let the player participate in shaping the meaning of the work, and I
don't know that you've compellingly argued for this point (as opposed
to stating a personal preference); but there are enough people who do
agree with you to justify a study of how it might be done.
That leaves the approach by technique -- and here I become most
interested, because we are able to talk about the craft of real games.
At the same time, there has already been a fair amount of discussion
over the years of how a work of IF might be responsive to the player's
contributions; there have been various experiments in offering the
player the freedom to choose narrative paths, or play a role, or
select an ending from many, or impose his own interpretation on
events, or select a harder or an easier level of puzzle challenge, or
make a moral decision. So any new theoretical approach to this
material, I think, needs to offer more than just a list of such
One of the things I'm trying to elicit here is how a dialogical theory
is useful to us. Many of your individual observations make sense to
me, but when I try to extract a thesis statement from the whole of
what you've written, I get something like "there are many ways for an
interactive work to let the player affect (or interpret) the meaning
of that work, and I really like it when that happens." I'd like to
think there's more there, though: perhaps you are looking to analyze
the various techniques of adaptive and responsive IF design in order
to determine in what ways, and how well, they balance meaning-creation
between the player and the author.
This paragraph made me laugh :-)
> Some of these strike me as more useful forays than others. The
> experiential is, of course, subjective and likely to vary from person
> to person; there are probably some who would say that *all* IF is
> "like a conversation", and indeed a Spanish term for interactive
> fiction is "aventuras conversacionales" (seehttp://es.wikipedia.org/wiki/Aventura_conversacional
> ). I also find the reference to Bakhtin, et al., minimally
> enlightening in this context, since the kind of relational meaning
> you're talking about with IF is that which develops from dialogue with
> the reader/player, rather than through intertextual reference and
> dialogue with the meaning of other works of art. There is certainly a
> great deal of reference and intertextual exploration in IF games --
> too much, I sometimes think -- but that is not what you have been
> talking about.
Mm, well the links were the original just to give some of the
genealogy of the term. But I think Bakhtin's theoretical construct of
the dialogic is more useful here than you say -- because
intertextuality is not a matter simply of reference and dialogue with
"other works of art" in the shallow sense of making explicit
references to other works or of including tropes from other works
(like the "twisty little synapses" in H2G2), but the dialogical
relationship between all texts, which include not only artworks but
thoughts, speeches, shopping lists . . . Expanded to this level,
Bakhtin's dialogic certainly applies to the relational meaning
developed between author and reader, artist and artwork, implementor
But. Clearly in the context of what is useful to us it's probably not
worth vanishing up into jargon-heavy theory too much. Still, if it's
interesting to you, there's what I think is a rather beautiful
politico-theoretical essay from Grant Kester in Variant here:
And in looking for that, I found a lighter sketch on the topic here:
> One of the things I'm trying to elicit here is how a dialogical theory
> is useful to us. Many of your individual observations make sense to
> me, but when I try to extract a thesis statement from the whole of
> what you've written, I get something like "there are many ways for an
> interactive work to let the player affect (or interpret) the meaning
> of that work, and I really like it when that happens."
Mm. The work in this thread seems to have been largely taxonomic, as
you say. This is partly, I think, because it grew out of an aside in a
paper intended to be almost entirely taxonomic! But I'm glad there's a
stimulation here to push things further.
> I'd like to
> think there's more there, though: perhaps you are looking to analyze
> the various techniques of adaptive and responsive IF design in order
> to determine in what ways, and how well, they balance meaning-creation
> between the player and the author.
That's certainly one of the things I'd now like to do. Another is to
explore what you pointed to in this post: developing a case for making
IF dialogical, or at the least explaining why it can have particular
artistic merits. And a third is to see if I can come up with new
techniques for it. (Sadly, that last will almost certainly have to
stay on the back-burner until my programming skills get better. While
the lovely I7 is empowering me to cobble together a game at the
moment, it's likely to exhibit precisely none of the theory talked
about in this thread.)
On that last point, though: did anything ever become of OpenFate? I've
been reading quite a bit of Victor GIjsbers stuff and got a little
excited about that possibility, but couldn't find anything more of it,
or indeed anything from Victor in the last couple of months.
> Mm, well the links were the original just to give some of the
> genealogy of the term. But I think Bakhtin's theoretical construct of
> the dialogic is more useful here than you say -- because
> intertextuality is not a matter simply of reference and dialogue with
> "other works of art" in the shallow sense of making explicit
> references to other works or of including tropes from other works
> (like the "twisty little synapses" in H2G2), but the dialogical
> relationship between all texts, which include not only artworks but
> thoughts, speeches, shopping lists . . . Expanded to this level,
> Bakhtin's dialogic certainly applies to the relational meaning
> developed between author and reader, artist and artwork, implementor
> and player.
Okay, yes, point taken; I'm used to intertextuality being talked about
in classics primarily to refer to the interplay of meaning between
texts (or, sometimes, other cultural artifacts and artworks), but I
appreciate that Bakhtin has something broader in mind.
For that matter, it would be easy to point out elements of IF works
that draw their meaning in relation to newsgroup posts -- not just the
games that parody RAIF personalities (though there are some of these)
but also games that were designed specifically to respond to or
challenge theoretical arguments advanced here. And I think it is
useful (if we're thinking about meaning construction) to bear in mind
the relationship between the author and the community of players,
since especially the most interpretively open works (e.g., Shade, So
Far, Schrodinger's Cat, The Gostak) tend to elicit a lot of
discussion, with players trying to decide on a consensus reading.
There is even sometimes communal play through ifMUD, where multiple
players share a game session and take turns issuing commands and
discussing what they should do next.
> On that last point, though: did anything ever become of OpenFate? I've
> been reading quite a bit of Victor GIjsbers stuff and got a little
> excited about that possibility, but couldn't find anything more of it,
> or indeed anything from Victor in the last couple of months.
He posted some comments on my blog a couple of days ago, so he's not
gone. But I don't know what happened with OpenFate. I'm a little
concerned that he might have decided not to go ahead with it due to a
lack of clear response to his proposal on the newsgroup, but I had the
impression that quite a few people were interested in seeing where
such a project might lead.
I'm still mulling over all this, but I'd like to add that one of the
connections textually between different works of IF is in the default
parser messages. ("Real adventurers do not use such language.")
-- Jason Dyer
Indeed, this IF intertextuality was something I referred to towards
the end of the notes I originally posted.
I'm going to make a try at application, taking adaptive
difficulty as an example.
I'm not sure if the "dialogical" concept is really
necessary and might even cloud the issue, but I
wanted to at least attempt this.
First, two non-IF examples. In the arcade game Parodius
the game gradually gets harder as you survive longer.
This has the effect of being highly irritating to the
player -- essentially success is penalized to the point
of being overwhelming.
A more recent game with the effect is Max Payne. This
game has the extra wrinkle of save games, so a player
who meticulously saves after each section (and reloads
upon death) will appear very skilled to the game, which
will correspondingly adapt with more enemies. This can
also end up irritating, but also invisible; but players
who figure it out will sometimes die intentionally just
to make a certain section easier.
So while both cases 'adapt to the player', in practice
increasing the difficulty with success can cause
irratation and gaming of the system.
Now, IF with adaptive difficulty could potentially have the
Suppose a game gets harder (or extra) puzzles if a player
solves a puzzle in few enough turns. This might cause
completist players (who want to see every puzzle) to reload
after solving a section, then solve it in as few moves
Suppose, inversely, a game gets easier when a player seems
to be stuck. This could also cause the player unhappiness;
a particularly stubborn player may be wanting the highest
difficulty, and want to bang their head on things a while.
I can't think of any IF that replicates the two scenarios
above exactly. In the original Adventure, if stuck on
an early puzzle the game will prompt "do you want help";
the player can respond YES or NO, so he or she is under
no obligation to take the difficulty adaptation. City
of Secrets will detect if the player has made multiple
"novice typos" at the start of the game and switch to
NOVICE difficulty if that is the case; however, the
difficulty can be switched back by the player.
Automatic unprompted difficulty adaptation often seems
particularly unnecessary in that multiple paths can
be accounted for in the game structure; there might
be "easy paths" and the "hard paths", with perhaps
the "hard paths" offering more points and a more
satisfying ending. (One could counterargue the novices
also deserve a chance to see the "best ending".)
Now, in terms of having a "dialogue" with the IF, the
likely pitfall is presuming that just because the player
is struggling, he or she is "asking" for help, just like
the vice versa situation would be "asking" for more
So, one might worry about "explicit dialogue" and
"implicit dialogue" and try to account only for those
parts of implicit dialogue that do not harm the game
experience, because the game might interpret things
badly. (As a harmless example, to go back to Adventure:
if you type WEST a bunch of times, it will prompt along
the lines of IF YOU PREFER YOU CAN TYPE W INSTEAD OF WEST.)
-- Jason Dyer
My understanding of "dialogism", as used by Bakthin, is that words or
utterances are rooted in particular contexts (for ex., a formal,
academic style vs. the slang of a subculture). There are different
genres, professional jargons, etc., and the growth of literary style,
especially in the novel (as opposed to poetry, which for Bakthin often
tends towards stylistic homogeneity) involes the mixing of different
contexts within the same utterance or literary work, through parody,
imitation, collage, etc. An author can suddenly imitate the language
of a legal document, for instance, and then shift to a parody of a
medical report, and then imitate a particular slang or dialect, etc.
Joyce's Finnegans Wake is an extreme example, but Bakthin also
mentions other, less experimental authors like Dickens. I would add
the novels of Luther Blisset and the Wu Ming group.
The use of multiple registers is essential.
Here is a quote I found:
"Style organically contains within itself indices that reach outside
itself, a correspondence of its own elements and the elements of an
alien context... The word lives, as it were, on the boundary between
its own context and another, alien, context... In artistic prose, and
especially in the novel, this [internal] dialogization penetrates from
within the very way in which the word conceives its object and its
means for expressing itself, reformulating the semantics and
syntactical structure of discourse. "
Bakthin speaks of dialogized heteroglossia, to indicate the multiple
voices and registers, as opposed to any efforts at cultural
centralization and uniformity.
Dialogism also refers to the idea of an utterance that invites, calls
for, anticipates, an answer. It is thus inherently directed to some
Other. This aspect perhaps suggets a work of interactive fiction: but
I wonder whether we can speak of IF as a genuine conversation, in
Bakthin's sense. For Bakthin wants to highlight that language and
meaning are not homogeneous but multiple and open to change through
living interaction between people.
I think Victor's ideas on Community in the Innovation Comp are very
close to what you are saying. In a sense, I read your notes as calling
for the same sort of revolution in the field of IF that he is also
calling for. At the moment, most IF works are highly closed: they
allow for no stylistical or ideological input from the user: the
style, narrative structure, tone, etc., are all determined in advance,
at least to a large extent.
I am not sure about this, because I have not read the Dialogical
Imagination in at least three years.... I remember that one of my
professors, Robert Stam, had written very clearly on Bakhtin. I cannot
remember the name of his book, but you might want to look it up.
A theoretical discussion of the difference between dialogical and
adaptive could be very illuminating. I find them, at least
intuitively, to be very different.
Sorry if this post is unclear, but I am writing in a rush because I
must (ooops) get back to work...
That's a great analysis of this partiucular area. Naturally, I'd
disagree and say that the approach of thinking about what is
"dialogical" can and does help us here -- or, at least, it helps me.
What I think you've demonstrated is the limitations of limited
artificial intelligences making assumptions about their interactors --
in this case, showing that adapting difficult levels based on player
behaviour can have highly adverse effects. So what I would say is that
what's going on is the programme taking control away from the player
-- not engaging with the player at all, but making decisions for the
player. A far more dialogical approach is simply, as you suggest, to
think about the problem in terms of "explicit dialogue"--the best way
of adapting difficult level to the player is to allow the player to
explicitly choose a difficulty level. That's fairly basic stuff.
It certainly doesn't apply do all implicit adaptation, though. (And
even within this kind of adaptation, Emily's Novice mode provide a
useful solution) What happened in Metamorphoses was far more subtle,
certainly *improved* gameplay for many, and had a definite literary
effect. Implicit plot-branching and "moral choice" IF similarly offers
improving possibilities. Victor's games, interestingly, have involved
*explicit* plot-branching and moral choices, which has produced
varying reactions in players.
Moreover, I'd like to emphasise something I'm worried might get lost
in the tech-specific analyses, because it's more literary than
technical, and that's writing in a fashion that gives the reader some
form of interprative autonomy, or *requires* that meaning be produced
dialogocially. This can be done through unusual language (For a
Change, the Gostak, Bad Machine), through ambiguity (Shade -- though
it can be taken too far, as I've argued in other posts in this
thread), or through removing authorial moral stances as much as
possible / engaging the reader in a moral argument.
Hector made a point in his most about the difference between
"dialogical" and "adaptive", which I think is important, as this
thread threatens drifting towards the latter (which is fine, actually;
that it went off into "dialogical" was a coincidence of original
questioning from Emily), and which applies to the above. So I'll go
and reply there now.
Thanks for this illumination and Bakhtin for the thread; I think it's
rather better than any I could have provided.
As for "whether we can speak of IF as a genuine conversation, in
Bakthin's sense", I think the answer really depends on the IF in
question, my OP notes suggested.
I do think there's something inherently weird in a dialogical sort of
way in the medium of IF, though. While the process of entering things
at the prompt only shallowly represents a conversation, we are still
always interacting, we are still always engaging with the game, and
there is something special about this. What I was suggesting in the OP
notes was that that dialogical weirdness has the potential both to
greatly increase authorial control *and* to greatly empower the
reader, depending on how it is used. (Loaded political analogy: when
the IMF negotiates with an African country regarding funding, it has
the shallow characteristics of a negotiation, but that it is a
negotation greatly increases the ability of the IMF to exercise
control over the country.) This thread has explored some of the
possibilities here, and gone off on a couple of tangents, inclusing
into analysing the adaptive.
> I think Victor's ideas on Community in the Innovation Comp are very
> close to what you are saying. In a sense, I read your notes as calling
> for the same sort of revolution in the field of IF that he is also
> calling for.
Mm, I was very much inspired by that essay. I wouldn't have a
revolution, though, as I'm a bit more relativistic about artistic
merits -- I just think there's more to explore. I'm also -- this may
be the place to mention this -- uncertain about a couple of the
aspects of the way Victor has approached the issue: firstly, I think
he has focussed to much on making the co-authorship explicit (his
games ask questions bluntly), which is a blunt instrument approach to
dialogue, ignoring the potential of subtle, implict ways of increasing
audience freedom (see my reply to Jason above); secondly, I think he
has focussed too much on giving the player *total* control over the
The latter problem is interesting to analyse. While OpenFate (and the
speculative gaming system suggested towards the end of Victor's essay)
would essentially be a Live Art approach to IF, and while that has
great potential, I think other methods offer more -- not least
artistically. One of the major objections voiced to these open
projects was a fear that the "artistic merit" of the "resulting" works
would be limited. I disagree, on the grounds that this narrows the
definition of "artistic merit" to an individualistic construct about
authorial skill and the quality of rational authorial ideas, and also
on the grounds that the projects' aims need not be solely "artistic"
in these terms. However, I do think there is an argument for the
games in which the author maintains some form of control, in whic the
author does not die but rather *engages with the player*. This is the
essence of what I've been talking about in this thread. I want author
and player *both* to be able to work together to contribute to
constructing meaning dialogically. There are some established ways of
doing this (see Galatea, the Gostak, and comments in this thread
thereon &c.), but I am musing now on new systems which could allow for
it. Hopefully more to come on that.
> A theoretical discussion of the difference between dialogical and
> adaptive could be very illuminating. I find them, at least
> intuitively, to be very different.
Ha. This is what I intended to reply to you about, before I realised
there was much more in your post.
Yes, they are very different. But they also overlap. Adaptive systems,
as found in IF so far -- through changing difficulty, altering
puzzles, providing moral choices, providing plot branches, observing
player behaviour -- can, I think, be employed to improve author-player
or game-player dialogue. (Though they don't, of course, necessarily do
What is it about the adaptive that allows this? Put simply, that a
game responds to the player allows the player to exhibit control over
the style and themes of the game. This is basic. But why is it that
the adaptive doesn't always allow for this?
The first reason is a shallow one: that some adaptive systems aren't
intending to allow for this. Some adaptations are of absolutely
minimal effect, not giving control to the player at all. (See the
Novice Mode in City of Secrets.) These are clear cases of using
adaptive functions for reasons other than to promote dialogue.
The second reason, I think, is something in the nature of the IF
community, and how IF players tend to play games. That is, we often
treat a game with multiple possibilities like something to "complete"
-- we want to see all of the different ways the game could have played
out. This is not dialogical -- in this case, the author's text simply
becomes something for the player to discover, as in traditional
reading. The only way I can see for adaptive systems to avoid this is
to bring the dialogical desire to the forefront -- as in De Baron, to
explicitly state that decisions are there to allow for player
exploration of particular themes, and not for the player to complete
the game. (Branching games for the player to complete have their own
merits, of course.)
The third reason may be the most interesting. As I suggested in my
reply to Jason, I think the quality of the artificial intelligence
involved is important. A game which tries to adapt to the player but
instead makes decisions for the player is not dialogical. In order for
it to be dialogical, the player must be consciously trying to cause
particular things in the game, and the game must have at least the
semblance of conscious response.
Galatea is the most advanced AI I know of in IF, and of course she is
not particularly advanced. But conversation with Galatea has more of
the character of dialogue than any other event in any other IF I know
of. Her complex moods, large database of responses, and the method of
new conversation topics being revealed by past responses, make it feel
as though the meaning of the piece is being co-operatively constructed
between the PC and Galatea, between the player and the author. So the
AI is essential to that.
But that kind of AI is still incredibly limited, and also was so much
for an IF that it was the entire game. I mentioned in the original
notes that that kind of dialogue could be created with other game
entities (landscapes, objects), but that, too, would take up the
majority of the game. While IF remains small-scale and amateur, this
is likely to remain the case -- even with that, there is enough to
fill out IF Art Shows with attempts at dialogical games, of course.
But the difficult of putting AI in IF brings us back, I think, to some
of Victor's ideas: if there is to be intelligence in IF, it must be
provided by humans, and so here dialogical IF requires the presence of
authorial intelligence in a Live Art kind of way. (Victor's background
in roleplaying is coming out through me here, I think.) What are the
potentials to explore here? I'm only beginning to think about
them . . .