probability and consistency
Mark Grundy
[Warren writes in response to my comments about `sufficiently large,
sufficiently diverse' populations:]
| No offense, Mark, but it's just this kind of reasoning that I'm
| objecting to.
| The probability of the string of coincidences I'm referring to is
| one part in 10^20: one part in 100 quintillion (American usage of
| 'quintillion'). In contrast, the probability of plucking a
| particular person from the population of the earth is more than one
| part in 10^10: one part in 10 billion (American usage of 'billion'),
| and I think most roleplaying backgrounds have a smaller population
| than that.
None taken. I don't know what kinds of game events you're
thinking about specifically. So far, I've assumed they're events like
the unlikely `deus ex machina' events you sometimes see -- eg, passing
hero dropping out of a parachute, saves the bacon of beleaguered
protagonists. Is this the sort of thing? If not, please give an
example. But anyway, for this post I'm gonna go ahead and defend the
probability of this improbable event using an indomitable mixture of
logic, maths and humour. :) I do so as an exercise to show why it
messes with your mind to try to hold sensible conversations about
probability in fiction.
Firstly, I'm going to rely on the premise that in fiction, you
never really know what state the world's in. Even assuming that the
author has kindly given you rules of physics that you understand, a
geography you know, characters that make sense to you, and a detailed
political and military history, there's no way that you can know what
everyone in the world is doing, where they are, or what motivates them
at the time the story starts. The story only gives you a narrow lens
into the world for this kind of detail.
Now, at any point in the story, we can imagine a population of possible
`what comes next's. This population can be thought of as a set E of individual
events, describable in a single sentence (possible quite a long sentence),
each with a non-zero probability of occurrence. Now we can ask, what
kinds of elements are likely to be in E, and what kinds of probabilities
do these elements have?
>In other words, the probability that a specific person on earth will be
> randomly selected is much closer to a certainty than it is to the posited
> string of coincidences. Put another way, when asked the question, which of
> the following items doesn't belong in the following group:
>A. probability of a string of ten 1% coincidences
>B. probability of me being the next person to die on earth
>C. probability of sun rising tomorrow
>A person with no feel for probabilities will say "C", because it's a certainty
> and both "A" and "B" seem unlikely, while a person with an intuitive feel for
> probabilities will say, "A", because compared to "A", "B" and "C" are both
> virtual certainties. Confusing "A" with "B" is worse than confusing "B" with
> "C".
> Our intuitive feel for probabilities is built
> in part from prejudices arising from our
> experiences and our own personal codes, and this
> plays a affects what we expect, and what we'll
> accept in a story.
>I see I haven't explained what I mean properly. I'm not talking about our feel
> for assigning probabilities to events. I'm talking about the calculation of
> the probabilities themselves. What I mean when I say an 'intuitive feel for
> probabilities' has to do with having spent a lot of time calculating compound
> probabilities from simple probabilities. It's math, not personal codes and
> preferences.
> ... part of the charm of fiction is that you can
> reflect reality in unrealistic ways.
>Here we agree. As long as we're agreed that events of probability 1 in 10^20
> are 'unrealistic'.
>Warren Dew
--
Dr Mark Grundy, Dept. Comp. Science, Ph: +61-2-6249 3785
Researcher, Education Co-ordinator, Fax: +61-2-6249 0010
CRC for Advanced Computational Systems, Web: http://cs.anu.edu.au/~Mark.Grundy
ANU 0200 Australia Email: Mark....@anu.edu.au
Mark Grundy
[Apologies for a scabby-looking version of this news article,
which might be floating about. My news server chose to interpret my
dying modem as a mandate to post whatever I'd been editing at the
time. Anyway, please ignore it -- MG]
In response to my comment:
: If you take a sufficiently large, sufficiently diverse population,
: and pluck just one unique member from it, the chances of any member
: being plucked is infinitesimal.
the worthy Warren Dew writes:
| No offense, Mark, but it's just this kind of reasoning that I'm
| objecting to.
| The probability of the string of coincidences I'm referring to is
| one part in 10^20: one part in 100 quintillion (American usage of
| 'quintillion'). In contrast, the probability of plucking a
| particular person from the population of the earth is more than one
| part in 10^10
No offense taken, Warren. The kinds of populations I was thinking
about when I wrote weren't human populations, but populations in the
statistical sense. In fact, I was thinking about populations of
stories and story events -- but more on that later.
I guess that the value of this discussion really depends on what
kinds of improbable events you're thinking of, and how they're used.
The kinds of thing I've had in my mind as I write are the usual `deus
ex machina' ideas of `passing hero crash-lands into lost valley, just
in time from saving scantily-clad heroine from being devoured by a
dinosaur'. These kinds of events are about as unrealistic as I've
ever seen, and if you're thinking of some *other* kind of events, then
please let us know what they are. :)
I want to go on and say that what fails with unbelievable events
isn't probability so much as communication. There's always some
possible explanation that justifies the improbable and makes it
probable, but it may need additional assumptions. A good author needs
a knack for simulation to get the assumptions right, but he needs a
knack for communicating them properly too.
Fiction is different from reality in that the audience *never*
knows the state of the world anywhere but where the narrative is
currently focussed. Even when the world has no-nonsense physics, a
complete geography and political history, the lens of the narrative is
a microscopic focus on just some few characters in some tiny part of
the world. This is why fiction is so good at messing with the
audience's sense of the reasonable.
`What comes next' is a popular game for the author to play with
the audience, but it's a game that's rigged worse than the games in
the most crooked casinos. In real life, you'd expect that if I drop
an uncooked chicken egg from a height of six feet over a hard kitchen
floor, it will fall and splatter. If I do this a thousand times, I'll
have five hundred omelette's worth of broken cackleberries. In
fiction it doesn't happen that way, and it's for the simple reason
that:
* the audience doesn't get to choose the qualities of the environment
in which the egg gets dropped.
In real life, if I do this experiment, I'll do it under conditions
I control. I'll make sure that my brother isn't nearby to dive and
catch the egg. I'll make sure I'm on earth, and haven't teleported to
a space-station. I'll make sure that the egg isn't full of alien
protoplasm, about to hatch and grow wings. I can do this because I
can keep my environment more-or-less independent of the experiment.
As an audience of someone else's story, I have a much narrower
window into their environment. When I try and guess `what happens
next', I must trust that the description has been complete in all
relevant details, and that nothing important has been omitted or
misrepresented. This is entirely an act of trust on my part. The
author is under no obligation to tell me all the relevant facts before
inviting me to guess what comes next. He may emphasise facts that
aren't relevant, or omit facts that I might need in my guesswork.
You can easily construct a story where a man drops an egg three
times and it doesn't break. If you make the right assumptions, you
can even make it the *most probable* event in the story. And what's
worse, you don't have to even tell the audience *why* the egg doesn't
break. You can leave them to just wonder. Since they can't read your
mind, they'll never know what assumptions held in your world, so they
can never know whether the event was likely or not. This issue comes
up in Tom Stoppard's play `Rosencrantz and Gildenstern are Dead',
where one character flips a coin hundreds of times, and it comes up
heads each time. This fact terrifies him, and makes him doubt his own
free will. Stoppard never explains to the audience why the coins fall
as they do, but the audience is led to guess the reason:-- that Rose
and Gild are figments of someone else's imagination (they're actually
the names of minor characters from Hamlet).
The upshot is that it's only meaningful to talk about
probabilities in fiction if the author is utterly scrupulous in what
he shows the audience before inviting them to guess. If the author
builds up that kind of trust, then probability can be used in the
story as accurately as it can in real life. But for this, it's not
sufficient that the author is a good simulator. He must also be a
good communicator.
My playgroup doesn't complain much about probability when the
story gets implausible. They complain about communication instead.
If the hero crash-lands in the valley to save the girl, the issue for
us isn't `what are the odds of that?' -- the issue is `why couldn't we
(the audience) predict such a surprising event?' Or put another way,
`how come the author didn't telegraph this earlier?' We've long since
learned to accept that you can come up with sufficient justification
for *anything*, just by adding assumptions.
Back to populations of stories: If you imagine the set of all
possible stories anyone could ever tell, what are the odds of randomly
picking a realistic story? Surely, just as composite numbers are
easier to find than primes, absurd stories should be easier to find
than sensible ones. Why then do we encounter more realism than
absurdity in fiction? Surely, when confronted with a long string of
realistic stories, we should scoff and say, `Three realistic stories
in a row? That's awfully improbable, you know. Are you *sure* you're
designing these stories impartially?' :)
Cheers,
Mark
Mary K. Kuhner
> My playgroup doesn't complain much about probability when the
> story gets implausible. They complain about communication instead.
> If the hero crash-lands in the valley to save the girl, the issue for
> us isn't `what are the odds of that?' -- the issue is `why couldn't we
> (the audience) predict such a surprising event?' Or put another way,
> `how come the author didn't telegraph this earlier?' We've long since
> learned to accept that you can come up with sufficient justification
> for *anything*, just by adding assumptions.
This is something that varies from player to player and group to group.
It is not true for me.
My branch of statistics is based on comparing the likelihoods of
competing hypotheses, not on estimating plain probabilities. This is
also the way I naturally think, and the way I tend to approach gaming.
In many games, there are two hypotheses competing in my mind for each
in-game event:
(1) The GM did that to further his story.
(2) The world did that by its nature.
If I decide that the preponderance of support is for explanation 1 most
of the time, I tend to lose interest in the game: it doesn't feel real
to me, I don't believe in it, and at most it's a diversion. I may
accept this if I was only looking for a diversion, but I'll be disappointed
if I was hoping for more.
It doesn't matter how probable explanation 2 is, if explanation 1 is
*more* probable. If you take a situation like the hero falling out of
the sky just in time to rescue the lovely maiden, there is a *huge*
weight of likelihood for explanation 1, and it is almost impossible for
explanation 2 to compete no matter how good the in-world justifications
are made to be.
However, if it was a PC decision that led to the hero falling out of
the sky, or a random process such as dice, there's much less support for
explanation 1, which helps. I can handle quite a lot of plain
improbability (I accept that improbable things do happen) but I can't
handle improbability that seems to be harnassed to the service of plot.
One of the reasons we value Immersion so highly is that we recognize
its presence as an absolute defense against claims of story-based
decisions. I know from experience that when Immersion is working well
my PCs don't give a hoot about what's good for the plot, so I can trust
their decisions to be cleanly world-based. Similarly with NPCs--though
Immersion is a bit harder to obtain, when it does happen it stands
as clear evidence for explanation 2.
Experience with and trust in the GM also helps: I know that Jon does
not think much about what kind of story he wants, so there's less
weight on explanation 1 when he's running. However, I also know that
he has some reflexive prejudices (for example, given the chance he'll
always escalate the size of a conflict) so even knowing that he tries
to run in a simulationist fashion isn't enough to kill explanation 1
dead in some cases.
If I were playing under Warren I could accept even very odd coincidences
because I know how Warren makes his decisions, and that plot does not
enter into them. If I were playing under David Berkman I would tend to
look at even rather ordinary events as contrived.
I know, Mark, that you'll be asking yourself at this point why I *care*
whether explanation 1 or explanation 2 is correct--why can't I just
accept that the GM is telling a story? All I can say is, my pleasure
in the game is much more acute if I don't have to be aware that the
GM is telling a story--it's the difference between stage magic and real
magic. Not all players feel like this--I think it's a key difference
between players who enjoy dramatist games and ones who find them
unsatisfactory.
Gee, I never thought I'd be explaining likelihood theory on rgfa, but
it does describe my experiences much more clearly than consideration of
flat probabilities.
Mary Kuhner mkku...@genetics.washington.edu
Psychohist
Mark Grundy posts, in part:
No offense taken, Warren. The kinds of populations
I was thinking about when I wrote weren't human
populations, but populations in the statistical
sense. In fact, I was thinking about populations
of stories and story events -- but more on that later.
Sorry for the misinterpretation. Okay, your position is different than the
(common) one that I objected to. I still think there's more to it, though.
... what fails with unbelievable events isn't
probability so much as communication. There's
always some possible explanation that justifies
the improbable and makes it probable, but it
may need additional assumptions. A good author
needs a knack for simulation to get the assumptions
right, but he needs a knack for communicating them
properly too.
Talking about stories for a moment before getting back to role playing, I'd
point out that there are two possible situations when such a failure occurs.
One is, as you point out, that the author has failed to properly communicate
explanations. (Whether or not the explanations are realistic is another
issue.) The other is that the author has no real comprehension of the
improbabilities involved, and thus doesn't even understand how much
justification is needed. The latter is far more common.
Fiction is different from reality in that the
audience *never* knows the state of the world
anywhere but where the narrative is currently
focussed.
A tangent, but this is true for reality, too.
In real life, you'd expect that if I drop
an uncooked chicken egg from a height of six
feet over a hard kitchen floor, it will
fall and splatter.... In fiction it doesn't
happen that way, and it's for the simple
reason that:
* the audience doesn't get to choose
the qualities of the environment
in which the egg gets dropped.
In real life, if I do this experiment, I'll
do it under conditions I control....
As an audience of someone else's story, I
have a much narrower window into their
environment.
Agreed. However, this difference between fiction and the player world is also
a difference between fiction and role playing games. A character in a role
playing game has just as much free will to act on his environment as does a
player in the player world.
... in Tom Stoppard's play `Rosencrantz
and Gildenstern are Dead', ... one
character flips a coin hundreds of times,
and it comes up heads each time. This fact
terrifies him, and makes him doubt his own
free will.
In which he is correct, insofar as he is a figment of Stoppard's imagination.
But the whole point of playing a role playing game rather than reading a story
is that the characters do indeed have free will, unlike characters in stories.
In the player world, a player could check the coin to see if it has a head on
both sides; if not, he could reasonably conclude that someone else had
switched coins on him. In a realistic, world oriented campaign, the player
characters could do likewise. Games in which the characters are not permitted
to make such tests and draw such conclusions with reasonable assurance cannot
make a claim to being world oriented or 'simulationist'.
They may still be good games, just not world oriented ones. Games in which the
laws of arithmetic don't always work can be good games, too.
I personally prefer games in which my characters can solve their problems by
thinking about them, rather relying on fate or a strong right arm.
... Surely, just as composite numbers are
easier to find than primes, absurd stories
should be easier to find than sensible ones.
Why then do we encounter more realism than
absurdity in fiction?
Again a tangent, but - I don't.
Warren J. Dew
Brett Evill
>Mark Grundy posts, in part:
> Why then do we encounter more realism than
> absurdity in fiction?
"Truth is stranger than fiction: because fiction has to make sense."
--
Brett Evill
To reply, remove 'spamblocker.' from <b.e...@spamblocker.tyndale.apana.org.au>
Brett Evill
In article <632m5j$otv$1...@nntp5.u.washington.edu>,
mkku...@phylo.genetics.washington.edu (Mary K. Kuhner) wrote:
>My branch of statistics is based on comparing the likelihoods of
>competing hypotheses, not on estimating plain probabilities. This is
>also the way I naturally think, and the way I tend to approach gaming.
>In many games, there are two hypotheses competing in my mind for each
>in-game event:
<snip>
>Gee, I never thought I'd be explaining likelihood theory on rgfa, but
>it does describe my experiences much more clearly than consideration of
>flat probabilities.
How many of us are statisticians? (My field is econometrics, which is to
say multivariate regression.)
Mary K. Kuhner
In article <b.evill-2810...@tynslip1.apana.org.au> b.e...@spamblocker.tyndale.apana.org.au (Brett Evill) writes:
>How many of us are statisticians? (My field is econometrics, which is to
>say multivariate regression.)
I'm a geneticist, but in a very statistical corner of the field.
As I recall, Alain does something similar.
Most of the gamers I know are computer professionals of one sort or
another, but then again that's probably the circles I move in.
(My husband's GURPS group has so many Microsoft people in it, they
had trouble keeping the game going right around the time Win95
came out.)
ObRPG: In fantasy games I haven't found my biology background to make
too much difference, except that I like realistic treatments of animals
and plants, not hokey ones. In SF, however, it's both very helpful
(lots of raw material for making interesting aliens) and a big SOD
problem (can I really suppose that people will be able to eat things
that grow on other worlds? Raise crops without having to create soil
from scratch? Even breath the air?)
I have a fun book, _What if the Moon Didn't Exist?_ which plays what-if
games with astronomy, meteorology, climatology and a bit of biology.
Unfortunately the author is not a biologist and he's way too cautious--
he asserts that animals can't do without a day/night cycle, for example,
despite the apparent success of Earthly creatures such as polar bears
and penguins. Does anyone know of similar speculation games done
by a biologist?
One lovely image from the book: on a world where the seasons are very
extreme and most land animals migrate, cities are built in the shape
of ship's hulls so that the great herds are channelled past, rather
than breaking against the city like a tsunami.
Mary Kuhner mkku...@genetics.washington.edu
CHRISTOPHER S JACKSON
Brett Evill (b.e...@spamblocker.tyndale.apana.org.au) wrote:
:
: "Truth is stranger than fiction: because fiction has to make sense."
No, but you're close. Fiction has to seem /plausible/, while
fact does not. The fact that the odds of the Sun blowing up
spontaniously and erradicating mankind while you read this are greater
than, say, someone developing super-powers is undoubtable. Which of the
two make for a better story? As authors (or GM's. Authors get to control
the main characters, and there-by the plot), it is our duty and
priveledge to create a world that is both plausible and interesting.
Face it people, reality is usually dull, and when it isn't, we wish it
was. Trying to grind down imagination to conform to some pathetic model
of numbers is like saying that there should have been no difference
between Botachelli and Bosche, and both should have been painting by
numbers.
I strive for plausable realism in my games, but only so far. I
have made a life of the study of our past, and know, with a certain
order of likelihood just how badly life stank. Does this
mean that I simply knuckle under to history and force the female
characters into completely submissive roles? No, and the ladies of my
group are greatful. Instead, I search for ways in which something
extraordinary (a woman sheriff in 1860's Ft. Worth, Texas, etc) could
find its way into the logic of my game. Sometimes, these over-ride other
precepts of the game. So what? If I want historical accuracy, I'll read
and write essays. I wand fun, damn it! And I'm not having fun if I spend
all my time worring about the fact that you're X more times to get hit
by lightning that win the lotto. Realistic situations only do a little
for a story. Realistic *CHARACTERS* with real *PERSONALITIES* do.
OK, I've vented my spleen. You can finish ignoring me now.
"...I don't know how to tell you this. You've got harpies."
-Atrolicus, "Herculies, the Ledgendary Journies"
-=CJ=-
csj...@omega.uta.edu
scott....@3do.com
In article <632m5j$otv$1...@nntp5.u.washington.edu>,
mkku...@phylo.genetics.washington.edu (Mary K. Kuhner) wrote:
>
> whether explanation 1 or explanation 2 is correct--why can't I just
> accept that the GM is telling a story? All I can say is, my pleasure
> in the game is much more acute if I don't have to be aware that the
> GM is telling a story--it's the difference between stage magic and real
> magic. Not all players feel like this--I think it's a key difference
> between players who enjoy dramatist games and ones who find them
> unsatisfactory.
I agree with you strongly, Mary. I value immersion, and I value an
"undirected" sequence of events, because I specifically do not want
things to follow a traditional narrative arc. I prefer as a GM to base
as much on #2 as I can. #1 sorts of decisions, no matter how well hought
out seem somehow contrived, too "neat'. I do not think "gaming", and
"Fiction" are as close a match as people believe. Well perhaps for some
people, they are, That they derive entertainment from genre emulation,
wheras i do not.
>
> Gee, I never thought I'd be explaining likelihood theory on rgfa, but
> it does describe my experiences much more clearly than consideration of
> flat probabilities.
>
Well, Mary, if the explanation is as clear as this one, maybe you should
do it more? :-) Once again. Nice post.
Scott
-------------------==== Posted via Deja News ====-----------------------
http://www.dejanews.com/ Search, Read, Post to Usenet
Mary K. Kuhner
In article <635n17$3...@flash.anu.edu.au> ma...@cs.anu.edu.au (Mark Grundy) writes:
> Back to the `crashlanding hero saves lost babe from dinosaur' plot
> device, in a pulp game the explanation usually comes out through play:
> ``My engine was sabotaged by the evil Dr Cretaceous. Presumably he
> wanted to test my mettle against his evil megareptiles. But then when
> I saw you fleeing the dinosaur in that glade, I thought `waste not,
> want not' and steered my doomed craft right into the creature's
> gullet''.
Outside of a pulp game, though (and I see pulp as having an implicit
contract "Don't ask for things to be probable") this isn't going to be
enough for a lot of players: the cliche is so strong, and the sense
that the GM needed this to happen for story purposes is so strong, that
no amount of world justification will overcome the presumption that it's
GM fiat at work. (And, after all, why was his engine sabotaged just
now, and not one minute too early or too late?)
In such cases, the only tactics that have worked for me involve reducing
the evidence for GM fiat, not increasing the evidence for world
justification. Randomizers and rules can help here, especially if the
players have access to them. So can timetabling in advance--a
timetabled game can accomodate wonderous coincidences, because the
players have evidence that they were real coincidences and not
contrived.
Also helpful to me is seeing the GM react to the event so as to show
that it wasn't his preferred choice. "Doggone it," says the GM in
post-mortem (or even in play), "I wish he hadn't crashed there, but
that's just how it happened." This doesn't happen often in our
games, but having it happen now and then creates a strong presumption
that the NPCs' actions are not driven by fiat.
(I'm using "fiat" here as shorthand for "this is what improves the
story" because I can't think of a better word.)
Immersion also helps, though in the given situation it's hard to see it
helping enough (the hero's state of mind has too little to do with
when he arrives).
> In an investigative game, weird effects often can't be justified
> in this way. If the villain has the heroes' telephone bugged, but the
> heroes haven't considered that, then they may assume that his `magic'
> knowledge is a result of sloppy GMing. The only defense the GM has is
> to give away the bug information, which then may change the course of
> the plot. So it can be tricky to maintain confidence in the play
> contract while preserving the storyline.
I don't think that's the only defence (though it probably is in a
convention game). The players can say forthrightly: "Hey, how did he
know that? Are you cheating or what?" The GM can say "No, I'm not
cheating." If there is enough trust--which means in this case
enough previous experience that says the GM will admit it when he
screws up--the players may well accept this. (It helps if the event is
not too stunningly inexplicable to them, though--most players have
limits of trust in this direction, and there's also the problem that
if the world suddenly becomes incomprehensible most of us don't cope
well.)
One problem here is that the players may well, if sloppy GMing is ruled
out, next turn to the hypothesis of assumption clash. Thinking that
an assumption clash is happening can paralyze players--they *have* to
ask what's going on, because if they keep plowing forward with wrong
assumptions a disaster is almost inevitable. The only remedy here
is enough shared world understanding to minimize the chance of
assumption clash.
I have a real-game example, from _Paradisio_. We had a discussion of
world-laws early on, and one thing we noted was that teleportation by
spell wasn't possible. (You could set up standing gates from place to
place, that's all; and that was very difficult.) Now, one day the
PCs cornered some enemies on the waterfront. The enemies dropped a
smoke bomb and...disappeared. The PCs had enough technological goodies
that they were pretty sure they hadn't missed an invisible enemy.
They were also badly demoralized, and so was I, so I jumped to the
conclusion that it had been teleportation--a contract violation. I
growled at the GM, who growled back. He didn't want to say what had
happened, since he thought it reasonable that the *characters* would
suspect teleportation. As a player, though, I didn't want to deal with
that possibility--I felt the PCs would become so demoralized the game
would collapse. The possibility of assumption clash also got in here--
I began to suspect the GM envisioned the NPCs running away under cover
of the smoke, ignoring the fact (as I saw it) that the PCs had them
triangulated. We had a long argument.
In retrospect three things were going on. One is that the player was
intensely demoralized, and GMs get little slack in such situations--
to my tastes, they're worth avoiding if at all possible. The second
is that the GM, though pulling a lot of tricks out of his sleeve earlier
in the campaign, had somewhat degraded the player's trust that she
understood what was possible and what wasn't. (A few impossible things
had turned out to be possible earlier on, after all--why not
teleportation?) And the third one was that there wasn't clear enough
communication to rule out assumption clash.
I don't have a clear conclusion to draw here, except to note that
player expectations vary with circumstances--what will work in one
situation may not work in another--and I don't think it's just an
investigation/action dichotomy.
Mary Kuhner mkku...@genetics.washington.edu
Mark Grundy
I wrote:
: My playgroup doesn't complain much about probability when the story
: gets implausible. They complain about communication instead. If
: the hero crash-lands in the valley to save the girl, the issue for
: us isn't `what are the odds of that?' -- the issue is `why couldn't
: we (the audience) predict such a surprising event?' Or put another
: way, `how come the author didn't telegraph this earlier?' We've long
: since learned to accept that you can come up with sufficient
: justification for *anything*, just by adding assumptions.
Mary replied:
| My branch of statistics is based on comparing the likelihoods of
| competing hypotheses, not on estimating plain probabilities. This
| is also the way I naturally think, and the way I tend to approach
| gaming. In many games, there are two hypotheses competing in my
| mind for each in-game event:
| (1) The GM did that to further his story.
| (2) The world did that by its nature.
| If I decide that the preponderance of support is for explanation 1
| most of the time, I tend to lose interest in the game: it doesn't
| feel real to me, I don't believe in it, and at most it's a
| diversion. I may accept this if I was only looking for a diversion,
| but I'll be disappointed if I was hoping for more.
Stepping outside the reaches of stats for a minute, isn't this
really a group dynamics issue? If we join a game on the expectation
that we're going to contribute to story, we'd like some confirmation
that we're having an impact. If we feel that the GM is ignoring our
ideas and their consequences, no matter how cleverly it's being done,
we'll feel like we're not getting what we were promised. Even kids
with no training in stats have their own ways of working this out,
yes? So math stats isn't the issue here. It's how we believe the
world should work, how we perceive it reacting to our responses, and
the problems inherent in relying on a GM.
In an earlier followup to an article of Warren's, I said that
players depend on the GM to interpret for them, and that the bandwidth
of narrative is lower than the bandwidth of our real-life experiences
-- the information comes through slower, and more heavily processed.
Because of this, we need to know that the GM is a reliable filter into
the imaginary world, and isn't tricking us through misrepresentation,
or shadings of emphasis. We also need to know that what we give the
GM is enacted faithfullly in the world, and that the world is reacting
to it in some way. Otherwise, we'll feel manipulated.
There are games where I'd feel affronted if the hero crashlands to
save our group, and games where I'd be delighted. The difference is
in my expectations of how the world works, and what guides the
creation of consequences.
In a game where I'd come to expect that the world was hostile and
inimical (eg, a Cthulhu game), a crashlanding hero would seem
insulting. But in a 30's pulp fiction game, it could be great. The
difference is that I'd be half-expecting it in the pulp game. I'd be
expecting it not at all in the Cthulhu game, and more importantly it
would cheapen the experience I felt I was promised.
| I know, Mark, that you'll be asking yourself at this point why I
| *care* whether explanation 1 or explanation 2 is correct--why can't
| I just accept that the GM is telling a story? All I can say is, my
| pleasure in the game is much more acute if I don't have to be aware
| that the GM is telling a story--it's the difference between stage
| magic and real magic.
Er.. I wasn't asking myself such a question -- the reason for
caring seems obvious, and I explain my take on it above. What
interests me most about this account is that it shows why a strictly
immersive view of world sim can sometimes get you stuck.
Confronted with a surprising event, it makes sense to ask `could I
have predicted that?' If you couldn't have predicted it then then you
have a choice -- to hypothesise extenuating circumstances that you
don't yet know about, to presume that the GM's sim is faulty, or to
presume that the communication is faulty.
But you can't know for sure which it is without checking with the
GM somehow. You can come up with your own extenuating circumstances,
but they might not be shared by the GM. You can accuse the GM of
sloppy design, but as in the case of the secretly bugged room, there
are times when the GM can't defend against this accusation without
compromising the story. All you *can* do safely at the time is to let
the GM know: I didn't expect *that*; I don't understand why it should
happen. I think that the issue is best resolved through a
communication check. Depending on how the group communicates, this
might be done in character or out of character.
Once the GM knows the player is having credibility problems, he
has to decide what to do -- if there's a sim problem, it can be
revoked or fixed with additional premises. If it's a communication
problem, then additional scenes or clarifications can help. But if
it's part of some forthcoming, well-justified revelation then the GM
may elect to do nothing. If a GM does nothing, you can't tell at the
time whether he's being clever, or lazy. Ultimately the GM's track
record and the story itself will decide for you.
Playtesting is an extremely useful tool for honing a GM's sim
skills, and in developing better communication in the group. To
playtest a game effectively, it really needs to have a beginning and a
middle and an end -- so it's easier to playtest a one-off game than a
campaign session. By discussing how you felt and what you believed at
different stages of the game, and by finding out directly from the GM
which bits of the game were planned, and how, and which bits were
improvised, you get a much clearer picture for how the game play
develops.
Brett Evill
In article <634uft$6...@news.uta.edu>, csj...@omega.uta.edu (CHRISTOPHER S
JACKSON) wrote:
>Brett Evill (b.e...@spamblocker.tyndale.apana.org.au) wrote:
>
>:
>: "Truth is stranger than fiction: because fiction has to make sense."
>
> No, but you're close.
I was quoting. That's what the quotation marks are for.
Mark Grundy
I wrote:
: Back to the `crashlanding hero saves lost babe from dinosaur' plot
: device, in a pulp game the explanation usually comes out through
: play: ``My engine was sabotaged by the evil Dr Cretaceous.
: Presumably he wanted to test my mettle against his evil
: megareptiles. But then when I saw you fleeing the dinosaur in that
: glade, I thought `waste not, want not' and steered my doomed craft
: right into the creature's gullet''.
Mary offered the opinion:
| Outside of a pulp game, though (and I see pulp as having an implicit
| contract "Don't ask for things to be probable") this isn't going to
| be enough for a lot of players: the cliche is so strong, and the
| sense that the GM needed this to happen for story purposes is so
| strong, that no amount of world justification will overcome the
| presumption that it's GM fiat at work. (And, after all, why was his
| engine sabotaged just now, and not one minute too early or too
| late?)
I thought I'd add some more examples for breadth. Any game with a
notion of karma, with outcomes created by moral or emotional as well
as material considerations, or with subtle divine manipulations, or
that are driven by allegory or symbolism or even comedy are also
exempt from material probability analyses.
For instance, in my Pendragon campaign, the success of your
character's lands depends partly on your relationship with your
family, your liege and your oaths of knighthood. If you encounter
difficulties in these things, you may suffer a crop failure, or a
plague in your village or similar. There is always a material reason
for these calamities too -- insects in your crops, or a plague-ridden
beggar visits, or so on. The players know of this relationship, and
can predict it qualitatively (they know something bad will happen),
but not statistically (they can't say exactly what, or quantify its
impact). The characters normally interpret these calamities as divine
reaction to their own moral disquiet.
This superstitious relationship between heart and lands is part of
the mind-set of the game. If you want to know what kind of a man is
the liege of some lands, you can check to see the health of his crops
and the attitudes of his peasants. The correlation is qualitative but
not quantitiative: the crops may be plentiful, but with small yields
(he might be a wealthy miser). Or the crops may be lush, but untended
(perhaps he is pixie-led, or has an unruly disposition). Reading
these symbols is an art, not a science; it's a symbolic language you
can develop with your players. With my group, rather than being an
encumberance to the play contract, it's a valued feature of this
particular game.
In another example, sometimes a game features a prophecy or
prediction. It is often more fun to stick to the prophecy, even at
the cost of probability, than to break it by blind application of
statistical mechanics. The tragedy of _Oedipus Tyranus_ is that he
was ridiculously, implausibly unlucky in the man he fought on the road
and the woman he eventually married. It's a damn good play, even if
the probabilities are hard to swallow.
None of this is to suggest that these kinds of games should ignore
physical cause or material probabilities. Far from it -- it's much
more exciting to deliver a game with symbolic cohesion *and* material
causality. But if you take on this kind of game as a group, you'll
quickly find that players would rather flex a bit on probability than
lose the rich symbolism. The unlikely is part of the wonder of the
game.
Nor am I suggesting that every game should have these components.
For instance, the _Delta Green_ game I've been running adheres
strictly to material causality, because we think it works better that
way.
| In such cases, the only tactics that have worked for me involve
| reducing the evidence for GM fiat, not increasing the evidence for
| world justification. Randomizers and rules can help here,
| especially if the players have access to them. So can timetabling
| in advance--a timetabled game can accomodate wonderous coincidences,
| because the players have evidence that they were real coincidences
| and not contrived.
I'm keen on using randomizers and rules to help offset the GM's
greater influence in the kinds of games I mentioned above. As well as
dice and the usual mechanics, we find it useful to add tarot for this
purpose. Also, please recall my ancient comments about sharing NPCs
and story with players. This also helps maintain player trust and a
sense of player contribution.
My experience with scheduled (timetabled) plots is more
ambivalent. If the players have a good idea of the schedule in
advance, then it can work well to add suspense and enhance the story
focus. If the players don't know the schedule then the results are
less impressive. In the worst case, the GM can lose story flexibility
at the times when the players are taking the game in interesting,
unplanned directions, and the players gain no especial story benefits
from the experience. I saw a big spate of scheduled _Call of Cthulhu_
games in the late eighties. For my money, they contributed to too
many random PC deaths, dramatic non-events and wasted opportunities,
largely because the schedules were too independent of character
actions.
Nowadays I'm more inclined to use schedules flexibly as a check on
power balance, or sometimes I give them to players to enhance the
suspense. The flexible use works more like a decision tree: ``At
twelve o'clock, if the PCs haven't yet visited the graveyard, then
Count Loogie will move his coffin. If they've already been there,
he'll try to spy on them in their homes. If they've set watches, and
he spots that, he'll go off and steal a new coffin.'' This balances
better than having Count Loogie always in the players' faces, and it
does help generate some interesting and unusual coincidences, of the
sort that Mary mentions.
| Also helpful to me is seeing the GM react to the event so as to show
| that it wasn't his preferred choice. "Doggone it," says the GM in
| post-mortem (or even in play), "I wish he hadn't crashed there, but
| that's just how it happened." This doesn't happen often in our
| games, but having it happen now and then creates a strong
| presumption that the NPCs' actions are not driven by fiat.
I think that it is useful to share thought processes with players
after the event. Aside from bringing the group closer together, it
does help avoid the kinds of misunderstandings Mary mentions. Very
often, players interpret heavy GM interventions as a sign of their own
failure. By being clear when something isn't an intervention, it
helps keep players confident about their own contributions.
| (I'm using "fiat" here as shorthand for "this is what improves the
| story" because I can't think of a better word.)
I dunno what is a good word either. GM fiat may be plausible or
implausible, so the word has more scope than we really need. What's
more confusing is that even an intervention that initially looks
implausible can be rendered plausible by various story means, so what
begins as fiat at one moment may end up being a crucial logical
consequence minutes later. This seems to be the old problem of
`plot-in-progress' versus `plot-after-play', come back to haunt us.
: In an investigative game, weird effects often can't be justified in
: this way. If the villain has the heroes' telephone bugged, but the
: heroes haven't considered that, then they may assume that his
: `magic' knowledge is a result of sloppy GMing. The only defense the
: GM has is to give away the bug information, which then may change
: the course of the plot. So it can be tricky to maintain confidence
: in the play contract while preserving the storyline.
| I don't think that's the only defence (though it probably is in a
| convention game). The players can say forthrightly: "Hey, how did
| he know that? Are you cheating or what?" The GM can say "No, I'm
| not cheating."
Mmm. It's a tricky issue, but I didn't mean to suggest that it
was impossible. Here's how I deal with it in my groups.
I've never had an ``Are you cheating?'' question, but if I did get
one I'd have to answer, ``There's no cheating if we're all on the same
side.'' [If the play contract is any use at all, it should be to make
sure the group *is* all on the same side!] I'd also want to ask,
``What is it that you want that you're not getting so far?'' [You
don't get cheating accusations unless the player thinks she's in
contention with you for something or other. My blanket policy for
contention in the group is to resolve it out-of-play.]
This brings us to a question of the sort I *do* sometimes get:
``I'd like to find out how he knew our secret plans.'' My response to
this would be, ``It's odd, isn't it? The reason hasn't come up in the
game yet, but it's something that we really should find out by the
end.''
Note that this approach *isn't* a defence against accusations of
cheating. If a player is naturally suspicious and hostile to your
GMing then there's no way to convince the player you're honest until
the facts come out. What I try to do instead is to make a commitment
to deal with the issue later in the game. What I need from the player
is a renewal of her suspension of disbelief -- an agreement to defer
the question until later.
| One problem here is that the players may well, if sloppy GMing is
| ruled out, next turn to the hypothesis of assumption clash.
| Thinking that an assumption clash is happening can paralyze
| players--they *have* to ask what's going on, because if they keep
| plowing forward with wrong assumptions a disaster is almost
| inevitable. The only remedy here is enough shared world
| understanding to minimize the chance of assumption clash.
Yep. If this kind of question comes up, I think it's worthwhile
for the GM to review the current state of play before getting back
into it, and ask if there are any comments or questions. The GM can't
be sure that there's *not* a misunderstanding, and the player can't be
sure either way. So it pays to check, whenever you get `Are you sure
that's right?' questions.
| I have a real-game example, from _Paradisio_. We had a discussion
| of world-laws early on, and one thing we noted was that
| teleportation by spell wasn't possible.
[Evidence of a possible contract violation, or possible fiat
breakage of sim understandings as an NPC got away]
| I growled at the GM, who growled back. He didn't want to say what
| had happened, since he thought it reasonable that the *characters*
| would suspect teleportation. As a player, though, I didn't want to
| deal with that possibility--I felt the PCs would become so
| demoralized the game would collapse.
Mary, this is an interesting example -- how did it resolve? Was
the GM acting scrupulously and in good faith after all?
| I don't have a clear conclusion to draw here, except to note that
| player expectations vary with circumstances--what will work in one
| situation may not work in another--and I don't think it's just an
| investigation/action dichotomy.
The play contract is the most important organ of the game. If
it's broken, intentionally or inadvertantly, that has to be
acknowledged, and the contract renewed before play can proceed. Else
the contract is worthless.
The most common example of contract breach in my GMing is that I
sometimes forget a player's stated interests or commitments, and
accidentally marginalise them. This happens between players too. In
my `care and feeding of play contracts' article, I mentioned that
stating interests and stakes was an important part of the play
contract, and this is why.
If a mistake like that is pointed out to me, I make a point of
apologising, explaining why I made the mistake, affirming the
correctness of the player's complaint, and suggesting a course to take
from here on. It doesn't matter if the event seemed in character or
consistent with the world, or right for the story over-all. The play
contract is more important than these issues, and as long as we all
have the same goals, we can make the right changes to find a way
through.
Another related thing -- if any member of the group sees another
group member breaking the play contract, they should point it out --
even if it doesn't affect their own interests. In one case, a PC
slipped a love potion into another PC's porridge, creating a romance
plot that would change the second character forever. As GM I felt
obliged to point out that the second player didn't have to accept
this. The player was happy to go ahead, and the plot ended up being
good fun. But I think we needed the option of refusal at that point,
or else the door would have been opened for recriminations later.
How this relates to the current probability and consistency
discussion is that I believe that play contracts are more important
than sim issues, and shouldn't be confused with 'em. I think that
Mary's example is a good moral lesson as to why.
Another thought about probability, play contracts and cheating --
I've yet to make a play contract with players that includes the clause
`we'll stick with whatever the dice show'. As a GM, I use the dice to
generate ideas for what to do next and how to do it. I figure that if
that's how I'm using them, then players should use them in the same
way. So dice rolls aren't binding in our games -- theyre just
indicative, and while we try to adhere to dice outcomes whenever we
roll (because that's half the fun), we're also all free to reject
outcomes that we don't like, or to waive rolls in the first place.
This is normally done by proposal and consensus. Eg: `What!? You
rolled a fumble? Given the circumstances, I think that's an
inappropriate result.' Or: `Let's not roll. You're bound to succeed
anyway. It's not even a question of degree.'
William Clifford
b.e...@spamblocker.tyndale.apana.org.au (Brett Evill) wrote:
>>Mark Grundy posts, in part:
>> Why then do we encounter more realism than
>> absurdity in fiction?
It's a matter of technique. You can "not make sense" and be
"realistic" and maintain SOD in a fiction. I've seen it done. It was
one of those Japanese animated things, "Wings of Honeymayonaise" or
something like that. The main character was right next to a big
explosion and survives with a few scratches. His teacher, who dived
behind a pile of sandbags was killed by a stray piece of shrapnel.
When I figured out what happened (I try to be alert for this
phenomenon) I rewound so I could see it again and see how they did it.
I think I know how to do it in an RPG context too. It's not a matter
of dice but a matter of when the PCs find out whether or they are
alive (or whatever the case may be).
-William Clifford
from fields foiled! (you figure it out)
Psychohist
Mark Grundy posts, in part:
It's not too hard for designers to learn to
challenge their own storylines for causal
links and logical structure. A few good
after-game discussions can hone these skills
quite well.
True. However, it's much harder for them to learn to challenge their games for
accurate statistical treatment of alternative events, which is a very
different thing.
Warren Dew
Psychohist
Mark Grundy posts, in part:
For instance, in my Pendragon campaign, the
success of your character's lands depends
partly on your relationship with your family,
your liege and your oaths of knighthood. If
you encounter difficulties in these things, you
may suffer a crop failure, or a plague in your
village or similar.... The players know of this
relationship, and can predict it qualitatively
(they know something bad will happen), but not
statistically (they can't say exactly what, or
quantify its impact).
It strikes me that this is an excellent example of where the players can indeed
predict something statistically - "they know something bad will happen", but
"they can't say exactly what". This is a classical statistical conclusion.
What the players can't do is to predict the event deterministically.
Nor am I suggesting that every game should have these
components. For instance, the _Delta Green_ game
I've been running adheres strictly to material
causality, because we think it works better that way.
Mark, we're not in disagreement that 'strict material causality' is unnecessary
to a 'world oriented game'. Our disagreement seems to be that unlike you, I
don't see probability and statistics as implying strict material causality.
For example, there are certain highly magical areas in Laratoa which do not
meet even your requirement that there be a specific cause and effect
relationship, even if secret. Yet, the players can still apply probabilistic
reasoning - they can accurately observe that 'weird things are likely to
happen when you get too close to the tower on Misty Point'.
Warren Dew
Mary K. Kuhner
In article <636fsv$5...@flash.anu.edu.au> ma...@cs.anu.edu.au (Mark Grundy) writes:
>Mary wrote:
> | I have a real-game example, from _Paradisio_. We had a discussion
> | of world-laws early on, and one thing we noted was that
> | teleportation by spell wasn't possible.
> [Evidence of a possible contract violation, or possible fiat
> breakage of sim understandings as an NPC got away]
> Mary, this is an interesting example -- how did it resolve? Was
> the GM acting scrupulously and in good faith after all?
I believe in retrospect that he was acting in good faith, and that
there was no teleportation (though at other points in that game he
did slip and put in things which should not have been allowed, so it
was not a totally unreasonable concern). There was almost surely a
serious assumption clash, either about how the Stealth mechanic
worked, or about the PC's lines of sight, or both. There was also a
morale problem which eventually had to be tackled in the metagame.
The PCs never found out how the NPC got away. The game had no contract
provision that the player would ever find these things out; at the end
of the campaign a number of key questions were unresolved, and in most
cases I still don't know the answer. Sometimes I like this quality,
sometimes I don't, but that's the way Jon runs his gaes.
About three years after the game I suddenly said to the GM "Jayhawk
was dead from the middle of the campaign on, wasn't she?" He
grinned and said "I wondered if that would occur to you."
> How this relates to the current probability and consistency
> discussion is that I believe that play contracts are more important
> than sim issues, and shouldn't be confused with 'em. I think that
> Mary's example is a good moral lesson as to why.
SIm issues can be an important *part* of play contracts: my major beef
with Jon was that I felt he was violating sim and thereby violating
contract. I think this is a false opposition for many groups.
> This is normally done by proposal and consensus. Eg: `What!? You
> rolled a fumble? Given the circumstances, I think that's an
> inappropriate result.' Or: `Let's not roll. You're bound to succeed
> anyway. It's not even a question of degree.'
We did just what you describe for _Radiant_. It has, in our hands, a
good side and a bad side.
Markus put himself in a position where he thought he was about to die,
and the GM asked for a Constitution check. I rolled awfully low. The
GM went ahead and described the outcome, which involved Markus
sustaining some weird changes, but not dying. I've never been able
to look back on that scene with any more than dead neutral feelings.
"Okay, it happened, it's part of continuity." But there's no sense of
excitement or rejoicing that he lived (though I would surely have
been dismayed if he'd died). I know the roll was awful, and that
Markus' survival was fiat. I can't get excited about it as a result.
(I can, though, get excited about still having a favorite character
to play. Like I said, a good side and a bad side.)
I do strongly agree with your "Let's not roll" example. Far better
not to roll at all if the result is obvious in advance, than to roll
and discard all unwanted results.
Mary Kuhner mkku...@genetics.washington.edu
Mary K. Kuhner
> I thought I'd add some more examples for breadth. Any game with a
> notion of karma, with outcomes created by moral or emotional as well
> as material considerations, or with subtle divine manipulations, or
> that are driven by allegory or symbolism or even comedy are also
> exempt from material probability analyses.
They are exempt from real-world probability considerations, but in my
experience such settings have an internal logic of their own, and this
must be kept in mind. There are still probabilities: if the players
see someone cheat karma time after time, the question "Why?" will become
more and more pressing until it gets an answer or breaks SOD.
The middle part of _Jayhawk_ all takes place somewhere intensely
allegorical, where mundane laws don't apply and states of mind translate
directly into "physical" reality. None the less, the GM had to abide
by his understanding of how that "world" works, and had to convince the
player that he was doing so: if I had felt that he was doing things
so as to make the story better, I would not have been as engaged with
Jayhawk as I was. (I don't expect this to be a universal, by the
way. Many, probably most, players don't care about this point at
all.)
This meant that some of what happened was frustrating and anticlimatic.
In particular, Jay got "stuck" at one point, unable to figure out the
symbolism or make any progress, and the GM just let that happen, saying
No to reasonable try after reasonable try, for (if I recall correctly) a
couple of sessions straight. This was the price we paid, and were
willing to pay, to keep the sense of internal reality as strong as
possible.
> For instance, in my Pendragon campaign, the success of your
> character's lands depends partly on your relationship with your
> family, your liege and your oaths of knighthood. If you encounter
> difficulties in these things, you may suffer a crop failure, or a
> plague in your village or similar. There is always a material reason
> for these calamities too -- insects in your crops, or a plague-ridden
> beggar visits, or so on. The players know of this relationship, and
> can predict it qualitatively (they know something bad will happen),
> but not statistically (they can't say exactly what, or quantify its
> impact). The characters normally interpret these calamities as divine
> reaction to their own moral disquiet.
I agree with Warren: you are using "statistical" to mean "precise" and
that is not what he or I meant by it. The characters have a good
working understanding of what happens and what doesn't, and it is
correct. The GM would get in mondo trouble (at least with my group)
if he started violating the world-logic--say, having one character's
lands suffer because of his sins, but another's lands stay fine, for
reasons of pacing or spotlight or some such. World-logic is no less
important when it is magical logic rather than material--in fact it
can be more important, since such games are trickier to start with.
> In another example, sometimes a game features a prophecy or
> prediction. It is often more fun to stick to the prophecy, even at
> the cost of probability, than to break it by blind application of
> statistical mechanics.
For some people, the emotional reaction to the prophecy's fulfillment
is much greater if we *know* that it was not made to come out by
GM contrivance. I am willing to put up with some unfulfilled prophecies
in order to be able to have the experience I had with Jayhawk, where
a prophecy came true with no volition at all on GM's or player's part.
It amazes me: I can still feel, six years later, a strong echo of
just how amazed I was. It was like being touched by the hand of
Fate. Sticking to the prophecy via manipulation is more reliably
fun, but it is not the same *kind* of fun.
Again, not a tradeoff that's even worth considering for a one-shot,
where you'll just end up with a lessened session and never reap the
long-term SOD payoffs.
> My experience with scheduled (timetabled) plots is more
> ambivalent. If the players have a good idea of the schedule in
> advance, then it can work well to add suspense and enhance the story
> focus. If the players don't know the schedule then the results are
> less impressive. In the worst case, the GM can lose story flexibility
> at the times when the players are taking the game in interesting,
> unplanned directions, and the players gain no especial story benefits
> from the experience.
In other words, there is a sim/drama tradeoff here, and you are (quite
reasonably) unwilling to make it. You don't want to make the story
worse to make the sim better. I might, though.
Again, it matters a lot whether this is a short game or a long one, and
what its expectation are. I agree that if you want each arc to work
out smoothly, timetables are risky. But if you are willing to botch
some arcs in return for more overall sense of simulation, they can help
a lot.
Mary Kuhner mkku...@genetics.washington.edu
Mark Grundy
Mary writes:
| By no means is this a bad outcome from a story point of view (it
| sure beats Black killing all the PCs, which I think she had the
| wherewithal to do). However, my emotional reaction to it is going
| to greatly depend on whether I think it's a decision made by Black,
| or by the GM for plot purposes.
| It happens that we got to the end of that session, and the GM said
| feelingly, "Gosh, that was scary. I thought for a long time that
| she was going to go through with it." In other words, he was playing
| Black Immersively, and only the girl's own character kept the PCs
| from dying an ignonimous death.
| The event packs a lot more emotional voltage for me under these
| circumstances than it would if I believed that the GM chose to have
| Black confess out of a desire for a better story, or to preserve
| script immunity.
The proof of the pudding for me is whether the GM is willing to go
the whole way on these issues. Ie: does the GM kill off characters at
times? A track record is more convincing to me than protestations of
motive. I also dislike Hollywood-style happy-ever-after-isms, and so
script immunity is not guaranteed in our play. What is assured is
that each player's *stake* will be looked after -- that stake might
or might not be caught up in the continuity of one specific character.
In my own games, I've killed off PCs at times with no forethought
-- they were in the wrong place at the wrong time, and I knew darn
well that anything else would break sim credibility, and the group's
trust in me. After a major event like that though, I need to check
that everyone's stakes are intact. If they're not, I'll have a think
about how to recover or further them, and if necessary I'll consult
the group about it.
| It is not a flavor that I believe worth trying for in a short game,
| especially a convention game. Nor is it compatible with tightly
| crafted stories or guarantees of good dramatic structure.
I'm unclear why you think this, Mary. Short stories often suit
character death quite well, because there's less concern about
long-term viability. Dramatic stories needn't suffer from character
death either -- they can actually benefit from them, as long as you
protect the drama from loss of character continuity. And tightly
crafted stories can be robust or fragile, depending on how you spent
your effort.
| I think your refusal to accept the classification of players and
| games has some good points, but it sometimes leads to an apparent
| unwillingness to grasp that other players are really looking for
| *different* things in their games, not just different proportions of
| the same things you are.
I've rephrased this paragraph into a question which I think is
more constructive and less hurtful:
``How can you give players what they want in terms of drama and
sim without pigeonholing them, or marginalising the interests of other
players?''
The way I see it, each player comes to the game with certain
interests and enthusiasms, which end up becoming stakes in the game.
They also come with different skills and commitments, which become
part of the resources of the game. It's up to the group to work out
how to make the play compatible and rewarding for everyone. Part of
doing so means identifying stakes and commitments. I advocate doing
this on an individual basis, because it's more responsive than using
pre-made labels.
This approach works well if the players all know what they want,
and have the experience to articulate it. If they're unsure, then
it's important to draw them out with questions and examples, and so
examples that highlight areas that may be exciting or problematic are
important.
So let's assume that you have stakes and commitments from each
player. Now what?
Since play is often improvised and can go anywhere, it happens
from time to time that the flurry of play kicks over a player's stake.
If that happens, you have to redress the problem immediately, because
a broken contract means a broken group. To avoid this, a trick is to
formulate the stakes in as robust a way as possible, so that you're
not forced into positions of retcon too often, and don't have to wimp
out on the sim or the drama to protect the crockery. :)
For this reason, I try and avoid contracts that say `don't kill my
character'. Instead, I aim at contracts that express goals in the
positive, rather than the negative. This gives you more flexibility
in events like character deaths, because char death isn't ruled out,
as long as the stake is somehow preserved.
I believe that if it's prepared properly, the play contract gives
you everything you need to guide your game decisions. You don't need
to poke players into special boxes, or to ask them to accept play
labels. You just need to know what they want, and what they're
prepared to commit. If you get past the contract stage, then you
should be confident that your group has the skills and the
understanding to play together, even if they have different play
preferences and game stakes.
If a character is dead, then I think it's a creative challenge to
come up with a way to preserve a player's interests. Some solutions
we've used in the past:
* Use the opportunity to review the stakes, and see if they can be
improved on with a new character.
* Create a new character who embodies the same stakes, maybe in a more
exciting way.
* See if death is really the end of the character. (Sometimes it
isn't).
* Find out where you went wrong, work out whose responsibility it is
to avoid that problem in future, retcon and learn from the
experience.
| In my experience, only a more or less simulationist player feels
| that the game is diminished if a decision was made for drama:
| dramatist players not only don't feel that, they have no clue why
| *anyone* would possible feel that.
I guess you must think of me as a simulationist then. I'm very
dissatisfied when a GM or player breaks character just to protect me
from the hurt of losing my character. Good drama and good sim are two
essentials for game enjoyment. While ever they're in conflict the
game loses, no matter which one wins. Rather than playing favourites,
I say: fix the conflict.
Mary K. Kuhner
In article <345848...@iol.ie> mano...@iol.ie writes:
>Somewhat on a tangent, but one thing I'm curious about: Have any of you
>ever had your *characters* realize that the world they live in is
>failing to obey the normal laws of probability?
My PBeM character Catalina had this experience.
I don't think it worked out the way the GM intended. I suppose he
meant for Cat to decide that she was being fooled and manipulated,
and work on finding out who her tormentor(s) were: either that, or
he simply didn't think about how the sum total of events would appear
to her. (I agree with you, Russell: it's not enough to explain each
event, if they add up to an inexplicable pattern.)
However, Cat had mental illness in her backstory, and rather than
remaining resolutely rational and trying to find the cause of her
delusions, she slowly and reluctantly concluded that she was crazy.
As a result, by the end of the campaign she *was* crazy--not just
in the sense of behavior discordant with world expectations, but
flat-out insane by her own standards. (The last scenes of the game
involved her systematically losing her belongings, her money, and
finally her clothing, and throwing herself into the sea. It made
intuitive sense, but no rational sense at all, even to her.)
The descent into madness was fascinating, but painful. In the end it
didn't leave us with much of a game. In a multi-player game I'd have
abandoned the character as she was totally dysfunctional with little hope
of recovery.
If one's going to run this kind of scenario with Immersive players, I
think it's a good idea to explicitly ask for characters with a lot
of mental stability and self-confidence--the kind of person who keeps
believing in herself no matter *what* happens. Or else to be prepared
for a substantial proportion of characters to self-destruct.
Mary Kuhner mkku...@genetics.washington.edu
Mark Grundy
| They are exempt from real-world probability considerations, but in
| my experience such settings have an internal logic of their own, and
| this must be kept in mind. There are still probabilities: if the
| players see someone cheat karma time after time, the question "Why?"
| will become more and more pressing until it gets an answer or breaks
| SOD.
I call these `expectations', rather than `probabilities'. I agree
that expectations need to be honoured, or the `cognitive dissonance'
[wanky arts term meaning: irritating disappointment :)] needs to be
explained.
Something it's worthwhile to point out here -- everyone has
expectations, and it doesn't matter whether they're right expectations
or wrong ones, or whether the player is an authority in the field or
just an average joe. The issue is whether the expectations are met,
and what happens if they're not. That's a communication and
interpersonal issue first and foremost. The sim details are
secondary.
[An account of playing several sessions of blocks until the group
tumbled onto the GM's original idea.]
| This was the price we paid, and were willing to pay, to keep the
| sense of internal reality as strong as possible.
It must've been frustrating. It's the right decision though if
it's what the play contract requires, and nobody has any better ideas
for how to progress it. I guess that you wouldn't have wanted to
continue doing this indefinitely though -- eventually, someone would
have wanted to change the play contract, or question the story premise
or something. Have you ever had this happen?
| I agree with Warren: you are using "statistical" to mean "precise"
| and that is not what he or I meant by it. The characters have a
| good working understanding of what happens and what doesn't, and it
| is correct. The GM would get in mondo trouble (at least with my
| group) if he started violating the world-logic
For generality again, I'd say `expectation' here, and not
world-logic. There are games when the GM reveals every rule for
deciding what happens next, and the players share this, and so there
really is a world-logic, and all that can vary the outcomes are the
premises. In my Pendragon game, the world-logic is meant to be
mystical, elusive. Just as you think you've got the truth, it slips
out of your hands like a fish. There are still player expectations
though, and the obligations Mary discussed above still apply. It's
just that the players are never in a position to authoritatively say,
`That's impossible!', because in the course of the world's history,
the ground shifts.
On the other hand, this elusive quality doesn't mean that anything
goes. If something would ruin the flavour of the game, we don't do
it. For instance, a character might have a dream about a future
twenty years from his own time, but he wouldn't have a dream about a
future in the twentieth century. Why? Not for any special
world-logic reason, but because the imagery would clash too much. Put
another way, it doesn't break any admissible world simulation rule,
but it would break the scope of the play contract.
I could go ahead and invent a world-logic reason to bar dreams
about far future events, but I wouldn't do that. Instead, the players
would know the play contract reason (I'd tell 'em), and the characters
would be left to see the failure as a mystery. They'd come up with
their own in-character explanations for any failure, which is fine.
From this you can see that I separate play contract from world
logic. I know of groups who lump them together. The problem with
doing this is that sim disagreements can then spill over into play
obligations, which confuses the objective with the interpersonal.
Also, you run the risk of getting the argument `if it's alright in the
world, it's alright between players', which is generally false. I
think that world logic should support the play contract, but shouldn't
have to embody every part of the play contract. The world logic can
then change without changing the play contract, and (often) vice
versa.
| For some people, the emotional reaction to the prophecy's
| fulfillment is much greater if we *know* that it was not made to
| come out by GM contrivance.
Handling prophecies well is an article in itself. :) Given that a
lot of prophecies are self-fulfilling, talking about player and GM
volition can be tricky. I wanna let this one slide for another day.
| Again, not a tradeoff that's even worth considering for a one-shot,
| where you'll just end up with a lessened session and never reap the
| long-term SOD payoffs.
Most of my prophecies are done in my Pendragon campaign, rather
than in one-offs. I hate forced prophecies too, and a campaign gives
you the scope to let them progress at their natural pace. I don't
like being put in the situation of having to choose between forcing a
prophecy or abandoning it -- neither is a desireable choice. As I
mentioned earlier, I'm happy to tweak a few minor probabilities to
help a prophecy along, but I don't like shoehorning. Somewhere in the
dim, dark past I have an article posted to aus.games.roleplay about
tarot and prophecies and making them work. I'll resurrect it if this
is a topic of great interest.
: My experience with scheduled (timetabled) plots is more ambivalent.
: If the players have a good idea of the schedule in advance, then it
: can work well to add suspense and enhance the story focus. If the
: players don't know the schedule then the results are less
: impressive.
| In other words, there is a sim/drama tradeoff here, and you are
| (quite reasonably) unwilling to make it.
Yep. The second case is a sim/drama tradeoff, while the first
case gives you better sim for no cost in dramatic flexibility. I'm
all for better sim, but as I mentioned in another article, I don't
like putting sim and drama into conflict. So in cases where secret
scheduling is useful, I prefer to go with a more flexible `contingency
schedule', which I described in the article quoted above. It gives
you more sim benefits than a flat schedule (it's more realistic about
changing NPC motives and knowledge), and doesn't hurt the drama.
| Again, it matters a lot whether this is a short game or a long one,
| and what its expectation are.
I'm not sure why game length counts here. A schedule either works
by the last event, or it doesn't. So it should integrate equally well
(or badly) into a long game or a short one. I've used it in both
kinds of games, and have found no noticable difference.
Given my predisposition for impro, I prefer flexible schedules to
rigid ones. Pendragon comes with its own campaign-length schedule,
which is listed on a year-by-year basis. The first thing I did with
this schedule was rub the years off, and convert it into an
event-by-event schedule. This gave me the flexibility to expand and
compress some bits. Then later, I began swapping the order of events
around, moving bits of some events into other events, inserting my own
events, deleting some of the boring, repetitive events and so on.
There's still a schedule there, but now there's enough flexibility to
satisfy my impro impulses, enough surprises to intrigue my players,
but still enough detail and ideas to structure the story and help
players plan for things six months down the play track. This is my
only GMing experience with a long-term campaign schedule, and it seems
to be quite favourable.
Travis Hall
Psychohist (psych...@aol.com) wrote:
:
: And sometimes, you can't even do this without, say, compromising
: immersion and thus seriously reducing your enjoyment of the game.
: Which is why, as Mary points out, prior experience with the
: gamesmaster's adjudicatory style is so important.
Now, I know I'm not as heavily into immersion as some people here, but
something strikes me as odd here. If you are questioning the GM's actions,
hasn't your immersion already been compromised? In which case, wouldn't it
be better to check with the GM and allay your fears, thus making it
possible to restore the integrity of your immersion?
--
Why is it that when I do finally get around to creating a .sig file, I
can't think of a single witty thing to say in it?
The Wraith
Sea Wasp
Mary K. Kuhner wrote:
>
> In article <345848...@iol.ie> mano...@iol.ie writes:
>
> >Somewhat on a tangent, but one thing I'm curious about: Have any of you
> >ever had your *characters* realize that the world they live in is
> >failing to obey the normal laws of probability?
>
> My PBeM character Catalina had this experience.
Virtually all PCs in my world believe in "destiny". That is, there are
heroes and there are those who sometimes try to BE heroes and fail.
Those who are the true heroes are lucky as well as determined, smart,
and skilled. That is, anyone can be heroic, but if you're going to
survive to enjoy the results of your heroing days, you'll need more than
a bit of luck as well. This doesn't make them feel that there's anything
wrong with the WORLD, just that destiny is on their side... or against
them, at times when everything's going bad.
--
Sea Wasp
/^\
;;;
Psychohist
In response to Mary Kuhner:
In my experience, only a more or less
simulationist player feels that the
game is diminished if a decision
was made for drama: dramatist players
not only don't feel that, they have no
clue why *anyone* would possible feel that.
Mark Grundy posts:
I guess you must think of me as a
simulationist then. I'm very
dissatisfied when a GM or player breaks
character just to protect me
from the hurt of losing my character.
Good drama and good sim are two
essentials for game enjoyment.
While ever they're in conflict the
game loses, no matter which one wins.
Rather than playing favourites,
I say: fix the conflict.
Mark, I don't think you understand what Mary means here.
What she said was that 'simulationist players feel that the game is diminished
when a decision is made for drama'. This has nothing to do with 'the hurt of
losing a character' - it's at a much more fundamental level than that.
The idea is that for certain 'simulationist' players, any, and every, decision
made for dramatic reasons inherently diminishes the game.
You can't fix this conflict - it's inherent. The fact that you haven't
acknowledged the possibility that this position even exists is probably why
Mary seems to consider you more 'dramatist' than 'simulationist'.
Warren Dew
Mary K. Kuhner
In article <b.evill-3110...@tynslip2.apana.org.au> b.e...@spamblocker.tyndale.apana.org.au (Brett Evill) writes:
>How that GM ever thought we had the confidence to take on EPIC beats me.
>Players get information about their characters' competence from two
>sources: the character representations in game mechanics, and the
>characters' performance in adventures. Guess which is the more compelling.
This story is *so* familiar. It seems to be a common mistake.
One common cause is probably the "make everything challenging" GM bias.
It's easy for the GM to get into the mindset that the game will be
the most fun if every single challenge is really proportioned to the
PCs--so if they attack a pair of orcs it will be a tense, close fight,
and if they attack a red dragon it will be a tense, close fight.
This may work for gamist players who can just accept it as a given,
but if the players are trying to build up a view of the world and
their place in it via in-game events it causes all sorts of trouble.
We just about lost a campaign because the GM, early on, suppressed
dozens of trivial encounters and only presented the four challenging
ones: the player concluded that everything out there was as tough
or tougher than the PCs, and thus the PCs acted like small potatoes when
they were supposed to be big ones. This could have been avoided quite
simply (we both realized afterwards) by saying "During the following
month you overhaul five merchant ships who give you a wide berth,
warn off two tax extorters, see three pirates veer off when they
recognize the Imperial flag, and slaughter four manticores who try
to scoop sailors off the deck. Then, one night...." before introducing
the hard encounter.
If the PCs are highly competant, some of their conflicts should
logically be trivial. You can play these out (sometimes they're more
fun than the GM thinks they will be) or abstract them, but what you
can't do for many players is delete them: it screws up the
probabilities and leads to false conclusions about the gameworld.
(It can also rob the player of any sense of his PC's competance.
What fun is it being the best swordsman in Alarre' if you never get
to fight anyone you can trounce?)
Mary Kuhner mkku...@genetics.washington.edu
Sea Wasp
Mark Grundy wrote:
>
> I've rephrased this paragraph into a question which I think is
> more constructive and less hurtful:
>
> ``How can you give players what they want in terms of drama and
> sim without pigeonholing them, or marginalising the interests of other
> players?''
>
> The way I see it, each player comes to the game with certain
> interests and enthusiasms, which end up becoming stakes in the game.
> They also come with different skills and commitments, which become
> part of the resources of the game. It's up to the group to work out
> how to make the play compatible and rewarding for everyone. Part of
> doing so means identifying stakes and commitments. I advocate doing
> this on an individual basis, because it's more responsive than using
> pre-made labels.
>
> This approach works well if the players all know what they want,
> and have the experience to articulate it. If they're unsure, then
> it's important to draw them out with questions and examples, and so
> examples that highlight areas that may be exciting or problematic are
> important.
>
> So let's assume that you have stakes and commitments from each
> player. Now what?
But what if my "stake" *IS* my character -- that is, I'm playing to
enjoy the interaction of my character with others and the group? This is
sometimes a motive for me, and it's almost ENTIRELY the motive my wife
plays with.
> Since play is often improvised and can go anywhere, it happens
> from time to time that the flurry of play kicks over a player's stake.
> If that happens, you have to redress the problem immediately, because
> a broken contract means a broken group. To avoid this, a trick is to
> formulate the stakes in as robust a way as possible, so that you're
> not forced into positions of retcon too often, and don't have to wimp
> out on the sim or the drama to protect the crockery. :)
>
> For this reason, I try and avoid contracts that say `don't kill my
> character'. Instead, I aim at contracts that express goals in the
> positive, rather than the negative. This gives you more flexibility
> in events like character deaths, because char death isn't ruled out,
> as long as the stake is somehow preserved.
Again, this assumes that the character isn't INTEGRAL to the stake.
Case in point: my character Kyrie Ross, the Phoenix Saint of Myrionar
(aka, a Paladin/Holy Warrior of Myrionar). Kyrie's entire purpose was to
avenge the deaths of her parents and her brother at the hands of the
False Saints (a group who had corrupted the entire order of Myrionar and
who were pretending to be Myrionar's servants) and to restore the true
Order of Saints. This was HER mission. HER oath. No one else COULD
fulfill that "stake", because it was focused on what she had to do. You
couldn't repair this one after whacking her.
Since you say that you don't like Happy Endings, of course, it's
obvious that you wouldn't have players of my stripe IN your games, but i
find it hard to believe that there aren't some grimmer players who still
design "stakes" which require the individual character.
> | In my experience, only a more or less simulationist player feels
> | that the game is diminished if a decision was made for drama:
> | dramatist players not only don't feel that, they have no clue why
> | *anyone* would possible feel that.
>
> I guess you must think of me as a simulationist then. I'm very
> dissatisfied when a GM or player breaks character just to protect me
> from the hurt of losing my character.
Yep. That's a sim all right. In my own games, I give people of that
type two options:
1) You don't have any karma or Fortune points; if you die, you die. In
other words, I eliminate the special protections I have on all the other
characters. This gives you the sim for YOUR character, but on the other
hand it means that you're about ten times more likely to pop off than
any other character, unless you're considerably better at playing and
tactical thought.
2) You give up your sim point of view during my game and think more
along movie lines. Some sim players can do this. Others can't. Those who
can't abide either choice don't play in my games, because I'm CERTAINLY
not about to take all my other dramatist players and start whacking
their characters dead right and left for the sake of "realism".
CONSISTENCY is a different matter. As long as I can JUSTIFY the
character's survival, there's no loss in consistency.
Good drama and good sim are two
> essentials for game enjoyment. While ever they're in conflict the
> game loses, no matter which one wins. Rather than playing favourites,
> I say: fix the conflict.
Sometimes there IS no fix for it. If you consider sim to require that
only the probable happens, then this WILL run head-on against good
storytelling, from the HEROIC point of view, very often. I have a strong
requirement for good simulation... in the sense that whatever happens is
CONSISTENT with the world's powers and laws. It DOESN'T mean that I
can't choose to use very improbable events to acheive a desired dramatic
goal. If a player spends a fortune point to save his butt in a really
bad situation, it's MY job, as a GM, to come up with some way, within my
universes' rules, to save him, NO MATTER HOW IMPROBABLE THAT EVENT MAY
BE. I try to choose the least improbable effective means of acheiving my
goal, but if the only way to save his butt is a conveniently timed
meteor infall, I can do that. But in cases that extreme, I have to
consider the results on the campaign. If I think (as in the example
case) that the event might turn high heroics into campy comedy, I'll
take the player aside and let 'em know that if I save the character it
could damage the campaign. It's THEIR decision -- do they feel that the
campaign will recover from a rather comedic incident of this nature?
(well, if meteoric infalls were common, it wouldn't be comedic, I
suppose) And is their character worth the risk TO THEM?
I'll note, however, that I've never actually HAD to have such a
discussion. There was always a perfectly reasonable, if improbable, way
to save the characters' butts without breaking the consistency or the
flavor of the campaign.
--
Sea Wasp
/^\
;;;
Mary K. Kuhner
In article <6397no$d...@ra.anu.edu.au> ma...@cs.anu.edu.au (Mark Grundy) writes:
> Mary writes:
> | It is not a flavor that I believe worth trying for in a short game,
> | especially a convention game. Nor is it compatible with tightly
> | crafted stories or guarantees of good dramatic structure.
> I'm unclear why you think this, Mary. Short stories often suit
> character death quite well, because there's less concern about
> long-term viability. Dramatic stories needn't suffer from character
> death either -- they can actually benefit from them, as long as you
> protect the drama from loss of character continuity. And tightly
> crafted stories can be robust or fragile, depending on how you spent
> your effort.
Pronoun trouble. My "it" was not meant to refer to "killing a PC." It
was meant to refer to "doing something that is bad for the story, just
to preserve simulation." If this story is meant to stand on its own
feet, with no campaign to look forward to in front of it or behind it--
especially if the play group is also transient--there is no point in
ruining the story in order to preserve simulation. You'll never reap
the rewards. On the other hand, with a stable play group and,
especially, a stable campaign, one may sometimes want to sacrifice
short-term story goodness for long-term SOD.
Character death works fine in short stories, and fine in (some) dramatic
games, as long as it is coupled properly to the dramatic goals of the
game. I wasn't trying to imply otherwise. I do not see character
death as a drama/sim issue, though some forms of script immunity are.
> For this reason, I try and avoid contracts that say `don't kill my
> character'. Instead, I aim at contracts that express goals in the
> positive, rather than the negative. This gives you more flexibility
> in events like character deaths, because char death isn't ruled out,
> as long as the stake is somehow preserved.
What do you do to accomplish this? I would much rather not have script
immunity for _Radiant_, if I thought there was any way the game could
continue enjoyably without it.
Mary Kuhner mkku...@genetics.washington.edu
Mary K. Kuhner
In article <6391eg$8...@ra.anu.edu.au> ma...@cs.anu.edu.au (Mark Grundy) writes:
> In the case of my players predicting events statistically, out of
> character, any clever model they produce for `what happens next' is
> guaranteed to (eventually) become weak. Why? Because part of my job
> is to surprise them! The more I find them anticipating the outcome --
> especially dramatic outcomes -- the more I shift the models I use to
> generate story. :) Of course, this sort of thing is a nightmare for
> statistical analysis. It's analogous to having a demon breathing over
> your shoulder while you flip coins, taking note of your current
> calculations, and then adjusting the results to mess you up. :)
> This is not to say that I always want the players to lose, or feel
> stupid, or to have no say in the outcomes. It's to say that I want to
> avoid a situation where a player's plot-model is so strong that
> there's no surprise left. So, there's a limit to how accurate I'll
> let players' world- and plot-modelling get.
What do you do to communicate this detail of your play contract (an
important one, I think) to your players? What do you do to reassure
them that you aren't breaking their models in order to make them lose?
We've been in situations where the player's understanding is so strong
that there is no surprise: for example, I've seen PC plans work out
perfectly down to the last detail. My experience is that the GM is
almost always displeased with this result, but player reactions vary
widely: some players are frustrated when things work out perfectly,
others are delighted. (One of my personal high spots from _Paradisio_
was a plan that worked perfectly: the NPC did *exactly* what I
pegged him to do, and died like a dog as a result.)
Are there games for which "surprise the players" is not defined as part
of your job, or is it a universal for your play style?
Mary Kuhner mkku...@genetics.washington.edu
Mark Grundy
Sea Wasp <sea...@wizvax.net> writes:
| But what if my "stake" *IS* my character -- that is, I'm playing to
| enjoy the interaction of my character with others and the group?
| This is sometimes a motive for me, and it's almost ENTIRELY the
| motive my wife plays with.
Having a lot of stake in the character is valid too. There are
games or scenes that will support play of a single character
indefinitely, and games or scenes that won't. The stakes help select
the plot and the interplay. If character continuity is important,
then I believe that's a responsibility that the GM and players should
share.
: For this reason, I try and avoid contracts that say `don't kill my
: character'. Instead, I aim at contracts that express goals in the
: positive, rather than the negative. This gives you more flexibility
: in events like character deaths, because char death isn't ruled out,
: as long as the stake is somehow preserved.
| Again, this assumes that the character isn't INTEGRAL to the stake.
Yes it does. But in cases where a character is integral to the
stake, then it can't be avoided. At those times, you're better off
keeping a tight control on risk.
| Since you say that you don't like Happy Endings, of course, it's
| obvious that you wouldn't have players of my stripe IN your games,
| but i find it hard to believe that there aren't some grimmer players
| who still design "stakes" which require the individual character.
I don't like Hollywood-style happy endings for games that aren't
derived from Hollywood-style movies. The Hollywood-style happy ending
is the one where character and plot integrity are abandoned just to
deliver a sacharine end. There are plenty of happy endings that don't
work this way of course. These work well in a wider variety of games
than the Hollywood-style can support.
[With regard to some comments of Mary's]
: I guess you must think of me as a simulationist then. I'm very
: dissatisfied when a GM or player breaks character just to protect me
: from the hurt of losing my character.
| Yep. That's a sim all right. In my own games, I give people of
| that type two options:
| 1) You don't have any karma or Fortune points; if you die, you die.
| 2) You give up your sim point of view during my game and think more
| along movie lines. Some sim players can do this. Others can't.
I also find it distasteful to lose drama from the loss of a
character. I see it as a group failure when this occurs. I'm happy
for my character to die if it makes good story. I don't like
characters to die needlessly because the mechanics or the
storyline are running riot.
An attraction of a movie-style genre game is that it's often part
of the genre to have implausible rescues and last-minute miracles.
There are tasteful versions of this, and tasteless ones. The tasteful
versions may stretch probability, but not the story fabric itself.
The characters remain in character; some plot-device, mentioned at the
very start and forgotten, gets activated, and while the audience knows
it's pulling a rabbit out of a hat, they also know that the author has
had the foresight to keep his basic story elements intact with it.
Those are the ones to aim for, I think -- except in comedy games,
where all bets are off.
There are other kinds of games when it breaks expectation to do
pull even this kind of rabbit out of a hat. It doesn't work in grim,
hard stories for instance. So the methods you use will depend on the
play contract, and are affected by the player stakes, and the content
agreed on for the game itself.
| CONSISTENCY is a different matter. As long as I can JUSTIFY the
| character's survival, there's no loss in consistency.
This thought has been kicked around a bit before, I think. Mary's
already said that consistency isn't enough for her. She wants
integrity in the play relationship between GM and player too.
If you look at the list of elements we've got that can possibly
conflict, it looks daunting:
* character continuity as a possible stake
* world sim credibility as a part of the play contract
* good drama as a goal
* integrity and authenticity in the group's relationships and
contributions
If GMs get into a situation where two or more of these conflict,
they're jammed, and may have to sacrifice something and disappoint
someone. It's useful if the group prioritises what's important among
these things in the play contract, so that if you're in such a
regrettable situation, at least the group will be agreed on what the
right sacrifice should be. These prioritisations can be done as
blanket principles (eg `Sim first!'), or on a case-by-case basis (eg
`Sim first, except when it kills characters').
Better still is to find ways to avoid these conflicts. One way of
doing that is to anticipate possible conflicts as you make the play
contract, and work out ways of reducing risk. Another way is to not
heap the responsibility for all these issues on the GM. By having
players be more aware of these issues earlier, some difficult
situations can be avoided.
Some posters have said before that they don't want to share
responsibility for these game issues; it messes with their fun, and
they'd much rather take on additional risk than lose their particular
interaction with the game. This is why it pays to know what the
player is willing to commit when you organise the play contact. That
way, you know who's willing to contribute what, and it gives you a
picture for how to get the most out of the interests, needs,
enthusiasms and skills in your group. Sometimes it's of benefit if
only a few players are thinking about the play contract and the story
development some of the time.
: Good drama and good sim are two essentials for game enjoyment.
: While ever they're in conflict the game loses, no matter which one
: wins. Rather than playing favourites, I say: fix the conflict.
| Sometimes there IS no fix for it. If you consider sim to require
| that only the probable happens, then this WILL run head-on against
| good storytelling, from the HEROIC point of view, very often.
There is always risk. The benefit of a good play contract is that
you have the opportunity to anticipate it, to balance the risk level
against the potential rewards as a group. Because sometimes there
aren't any perfect solutions, I believe that the acceptable risks, the
contingency plans and the prioritisations for what to save, should be
decided on a game-by-game basis, according to the feelings, skills and
preferences of the people in the group. To this extent, I don't
advocate blanket simulationist or dramatist resolutions of these
conflicts.
Notwithstanding that, I have my own personal preferences for what
to aim for as a player and as a GM, but they're not rigid preferences.
They flex according to what I think is achievable with the game and
the group.
| I have a strong requirement for good simulation... in the sense
| that whatever happens is CONSISTENT with the world's powers and
| laws. It DOESN'T mean that I can't choose to use very improbable
| events to acheive a desired dramatic goal.
I agree.
| If a player spends a fortune point to save his butt in a really bad
| situation, it's MY job, as a GM, to come up with some way, within my
| universes' rules, to save him, NO MATTER HOW IMPROBABLE THAT EVENT
| MAY BE.
If I don't have any good ideas, I often admit as much to the
group, and ask them to contribute their own. Sometimes we've had some
doozies that have made for better story, and left the contributing
player feeling like a hero. :) Sometimes, none of us have had any
good ideas, and we've just gone with the the best thing we can think
of.
If a player does contribute a good idea, I'll try to work it a bit
-- add value to it, or find an interesting take on it -- so that the
players get the reward of spontaneity as well as authorial rewards.
An important insight for me as a GM is that a GM doesn't have to
be alone. The players have stakes vested in the story, and that means
that if you're in a bad jam, you can expect them to want to help out.
You don't have to carry the problem just by yourself. The trick to
using the group as a group is to communicate the problem effectively,
get comments quickly, minimise the game down-time spent in discussion,
and not to abdicate responsibilities as narrator and creative sounding
board for their play.
| If I think (as in the example case) that the event might turn high
| heroics into campy comedy, I'll take the player aside and let 'em
| know that if I save the character it could damage the campaign.
| It's THEIR decision -- do they feel that the campaign will recover
| from a rather comedic incident of this nature? (well, if meteoric
| infalls were common, it wouldn't be comedic, I suppose) And is their
| character worth the risk TO THEM?
That makes sense. If you have other players too, then they also
have stakes in the campaign. I'd advocate talking to them too if the
decision was likely to jeapordise their stake in the game.
| I'll note, however, that I've never actually HAD to have such a
| discussion. There was always a perfectly reasonable, if improbable,
| way to save the characters' butts without breaking the consistency
| or the flavor of the campaign.
The best resolutions are when you don't have to discuss stuff --
when the group organically chooses something that's acceptable to
everyone, and there are no recriminations afterward. That's the right
thing to aim for, I think.
If I know I've made a weak or questionable decision, I like to
discuss it with the group afterwards, and explain what the issues
were, and why I made the decision. Sometimes a good postmortem
teaches you a lot as a GM, and in any case it helps keep the
communication flowing in the group. At other times, I may think that
a decision that seemed questionable at the time is in retrospect quite
a strong one. At those times I might keep quiet about it, to preserve
the spontaneity and story mystique.
Mark Grundy
I wrote:
: In the case of my players predicting events statistically, out of
: character, any clever model they produce for `what happens next' is
: guaranteed to (eventually) become weak.
Mary asked:
| What do you do to communicate this detail of your play contract (an
| important one, I think) to your players? What do you do to reassure
| them that you aren't breaking their models in order to make them
| lose?
| Are there games for which "surprise the players" is not defined as
| part of your job, or is it a universal for your play style?
It's not universal, and I over-generalised in the post you quoted
-- sorry for misleading folks.
There are times when leveraging surprises off the players'
expectations makes sense. Most commonly it's for improvised games
where the expectations aren't settled to begin with. In an impro
horror game for instance, you can spend a lot of time gauging player
reactions and working to that. That's one case when it can be useful.
My Pendragon game is another case. It's a game of passion; the
world rules shift naturally over time anyway, and the magic is
supposed to be elusive and unpredictable. If players aren't leaning
heavily on world rules then I generally leave their expectations
alone. But if they're playing the intellectual side too hard at the
expense of the passion in the game, then it's appropriate for me to
have the world give them some shocks -- challenging their sense of the
probable while reinforcing the notion that the game is driven by
passion, spirit and symbol, not by probability and material cause.
An example was with a relatively new player. The group were on a
Quest, and came to a bridge made from the edge of an enormous sword.
To cross the bridge would risk cutting your feet, so the new guy got
the idea to fashion shoes of wood with clefts in them, so that the
tips of the wood would touch the canted sides of the blade, rather
than the edge. Normal world probability tells you that this should
work -- you can test it on scale models and everything. :) The
problem was, the bridge was created by Merlin as a test of courage,
not ingenuity. The new guy's clever attempt to cross it attracted the
attention of a knight who'd stand on the bridge and beat him back and
humiliate him, while the other characters crossed a razor edge by the
strength of bravery alone. The same knight could walk on water, but
almost drowned in earth.
The characters couldn't predict that you could cross a sword
bridge without being cut, they couldn't predict that an armoured man
could swim in earth or walk on water. There was no statistical
evidence for such stuff, and they'll never be able to repeat the
results. So any formal notion of probability is useless in these
cases. Naturally, the characters wrote it off as `weird faerie magic
stuff', and the players realised that it was just one of those
counterintuitive things that makes up magical Pendragon stories.
I don't run all games like this. In _Delta Green_, I leave player
expectations alone. :)
There are also times, even in the more malleable games like
Pendragon, where I don't mess with player expectations at all. If the
stakes are high and there's risk of players losing out, then I try and
act as consistently as possible, so that they have the best chance of
looking after their own interests.
Hope this helps clarify things.
Mark Grundy
Mary wrote:
| Pronoun trouble. My "it" was not meant to refer to "killing a PC."
| It was meant to refer to "doing something that is bad for the story,
| just to preserve simulation." If this story is meant to stand on its
| own feet, with no campaign to look forward to in front of it or
| behind it-- especially if the play group is also transient--there is
| no point in ruining the story in order to preserve simulation.
| You'll never reap the rewards. On the other hand, with a stable
| play group and, especially, a stable campaign, one may sometimes
| want to sacrifice short-term story goodness for long-term SOD.
There are some good reasons to preserve sim in short games --
including, maintaining the trust of the players. :) But I agree that
you can take liberties with consequences if you know that the game
will end before you play 'em out. :)
: For this reason, I try and avoid contracts that say `don't kill my
: character'. Instead, I aim at contracts that express goals in the
: positive, rather than the negative. This gives you more flexibility
: in events like character deaths, because char death isn't ruled out,
: as long as the stake is somehow preserved.
| What do you do to accomplish this? I would much rather not have
| script immunity for _Radiant_, if I thought there was any way the
| game could continue enjoyably without it.
When a player wants script immunity for the character, it's often
not an end, but a means to an end. The end itself might be stuff
like:
* Not wanting to see the history go to waste
* Wanting to pursue specific story threads
* Not wanting to have to repeat all the effort to amass resources and
opportunities with a new character
* Being attached to exploring the personality and attitudes
* Seeing loss of character control as personal failure, and being
afraid to fail in this way
* Not wanting to deal with the grief and loss of a favourite character
If you can break down the player's wishes in terms of component
goals, it's sometimes possible to preserve what they want without
having to guarantee full and permanent script immunity. Instead, I
try and find ways of negotiating *temporary* and *limited* script
immunities. These come up for review as the circumstances and stakes
change.
There aren't any formulas for making it work -- you really have to
do it on a case by case basis. Here are some options which can work
for some players in some games:
* Don't want to waste the history: Consider playing a child of the
former character, or promoting a close NPC to player status. It
helps to keep NPCs around to facilitate this.
* Wanting to pursue specific story threads: Ensure that those story
threads will get closed in reasonable time, or find some way to
transfer them to a new character.
* Not wanting to have to lose all those resources: Transfer the
resources, or find a way of generating a comparable amount of
interesting new resources for the new character.
* Being attached to exploring the personality and attitudes: Consider
reusing the interesting parts of the character in other games.
* Being afraid to fail: Work with the group on this, and settle on
what success and failure really is for this game, and what are the
consequences.
* Being afraid of grief and loss: Consider generating multiple
characters, and playing them interchangably. (I dunno if this works
for other players, but I found it useful to dilute character
attachment when I was playing on MUSHes)
Paradoxically, if you make these kinds of offers, players may be
concerned that you're gunning for their characters, or won't take due
care with their interests during the course of the game. This is why
it's useful to explain under what conditions you *will* assure script
immunity, and to establish a play contract that clearly identifies
player stakes, and to stick to it, and to be seen to stick to it. It
also helps if the player has a positive reason to *not* want full and
permanent script immunity -- eg, the player realises that script
immunity competes with risks and story thrills.
In practice, even with temporary and limited script immunities, we
don't have a high character death rate in our campaigns. More
campaign characters are written out by players than killed by the GM.
I think that the main benefit of temporary and limited script
immunities for us is exactly that immediacy and thrill which Mary was
talking about earlier -- the realisation that right now, the GM
doesn't *have* to look after your character, and so you can afford to
be there, playing in the moment.
Psychohist
In response to my comment:
The idea is that for certain 'simulationist' players,
any, and every, decision made for dramatic reasons
inherently diminishes the game.
Brett Evill posts, in part:
The same is true for some dramatist players!
... So the art of the dramatist GM is to set
things up so as to produce dramatic results ...
How do you "set things up so as to produce dramatic results" without having any
"dramatic reasons" for it?
Warren
Warren
Mary K. Kuhner
> The characters couldn't predict that you could cross a sword
> bridge without being cut, they couldn't predict that an armoured man
> could swim in earth or walk on water. There was no statistical
> evidence for such stuff, and they'll never be able to repeat the
> results. So any formal notion of probability is useless in these
> cases. Naturally, the characters wrote it off as `weird faerie magic
> stuff', and the players realised that it was just one of those
> counterintuitive things that makes up magical Pendragon stories.
Hm. It sounds to me like both the characters and the players *could*
predict, perhaps not the exact details, but the general run of things:
otherwise they would never have even tried walking across a sword
blade. The events you describe make very good folklore sense, and
I would expect the characters to have plenty of resources for reasoning
about them.
It would not, I think, have been out of milieu for another PC to say to
the one who proposed the wooden shoes: "You know, you're adopting the
guise of the Devil to pass an obstacle, and meeting what is likely a
test of virtue with low cunning: I don't think that's going to work."
This is an assessment of probability based on past experience and
world understanding.
I don't see what you did as inconsistent (rather the contrary, given
the setting--you practically had to do something of the sort) nor as
noticably dramatist, at least as presented.
Both Warren and I use folklore logic when it's appropriate, and don't
see that as at all contrary to simulation, given a world which works
that way. The question is whether you use it even when it's
inconvenient and undramatic. (One of the times we had to invoke Script
Immunity for Markus involved folklore logic, and it would have been
really nice to just pretend it wasn't there, but it was.)
Mary Kuhner mkku...@genetics.washington.edu
Brett Evill
In article <19971030154...@ladder01.news.aol.com>,
psych...@aol.com (Psychohist) wrote:
>The idea is that for certain 'simulationist' players, any, and every, decision
> made for dramatic reasons inherently diminishes the game.
The same is true for some dramatist players!
Good drama requires that the world be consistent, as much as it requires
that characters be consistent. So the art of the dramatist GM is to set
things up so as to produce dramatic results with simulationist
adjudication. This is hard, and we somethings fail. When we fail to do
this, we sometimes get ourselves into situations where either sim or drama
must break, and it is a matter of choosing the lesser of two evils.
More and more, of late, I have been letting drama break, especially as I
felt that I had got into the habit of letting players exploit script
immunity. But this does not, I think, mean that I am turning into a
simulationist. I remain of the opinion that RP is the collective,
extemporary, collaborative recounting of a narrative, and that the point
is the enjoyment we get out of the literary/dramatic experience thus
produced.
Brett Evill
In article <19971031223...@ladder02.news.aol.com>,
psych...@aol.com (Psychohist) wrote:
>In response to my comment:
>
> The idea is that for certain 'simulationist' players,
> any, and every, decision made for dramatic reasons
> inherently diminishes the game.
>
>Brett Evill posts, in part:
>
> The same is true for some dramatist players!
>
> ... So the art of the dramatist GM is to set
> things up so as to produce dramatic results ...
>
>How do you "set things up so as to produce dramatic results" without having any
> "dramatic reasons" for it?
You do it with the decisions that are obviously world-creating decisions,
not with adjudications during play.
Obviously, at some stage the GM has to decide what sort of person the
Constable of Thekla is, taking into account the sorts of people available
and the selection process for the job. Out of the range of people who are
plausible in the part, he chooses a ferocious widowed avatar of Luciphage
with a willful and very attractive teenaged daughter who is his only human
connection. There has to be some decision made as to who is Constable, and
this one is made because the GM feels that such a character will offer the
most interesting interactions and dramatic possibilities. A simulationist
or a gamist GM has to make similar choices in setting up the world.
But when it comes to adjudication, fooling with world simulation damages
SoD, and therefore is bad from a dramatic point of view. One does it only
when previous failures have driven one to choosing the lesser of two
evils.
I guess I presumed from context that by 'decision' you mean an
adjudication, or at least a decision made during play. I may well have
misunderstood.
Neelakantan Krishnaswami
In article <19971031223...@ladder02.news.aol.com>,
Psychohist <psych...@aol.com> wrote:
>In response to my comment:
>
> The idea is that for certain 'simulationist' players,
> any, and every, decision made for dramatic reasons
> inherently diminishes the game.
>
>Brett Evill posts, in part:
>
> The same is true for some dramatist players!
>
> ... So the art of the dramatist GM is to set
> things up so as to produce dramatic results ...
>
>How do you "set things up so as to produce dramatic results" without having
>any "dramatic reasons" for it?
I think the idea is that if you do your campaign setup cleverly enough, then
there will be a strong story that emerges from the natural logic of the
setting and the characters' motivations.
One metaphor I use to visualize this method of plotting comes from physics.
Rather than applying a force on a particle to get the desired trajectory, the
world-based dramatist sets the initial conditions so that the particle moves in
the desired path without any metagame interventions.
Another way of looking at it, I picked up from David Berkman. The idea is to
stop worrying about what the story will be exactly, because as long as you
make sure the ingredients of a strong story are present -- a vivid, memorable
setting that ties strongly to the motivations of the characters -- then there
will be a good story. Knowing what will happen is irrelevant at best, so
forcing the plot isn't needed.
There are downsides to this approach -- players come to expect a strong plot
and a simulationistic resolution ethic, so it becomes very hard to introduce
another driving conflict to the campaign after the first has been exhausted
without it smacking of contrivance. But it can work very well for a finite
duration game.
Neel
Psychohist
In response to my comment:
How do you "set things up so as to produce dramatic
results" without having any "dramatic reasons" for it?
Brett Evill posts, in part:
You do it with the decisions that are obviously
world-creating decisions, not with adjudications
during play....
[example] There has to be some decision made as
to who is Constable, and this one is made because
the GM feels that such a character will offer the
most interesting interactions and dramatic
possibilities....
I guess I presumed from context that by 'decision'
you mean an adjudication, or at least a decision made
during play. I may well have misunderstood.
I certainly don't mean to limit it to adjudications involving player
characters. As for decisions made during play ...
I can see your description for the Constable of Thekla, I probably wouldn't
have a problem with it if you decided it before first running the game (Though
I have known players who objected to even that amount of dramatism - more on
the philosophy here only if you're interested). On the other hand, if you're
having to fill out details of Thekla only because it now seems, after starting
the game, that the player characters might go there, I would consider the
decision process you describe to be distinctly not world oriented.
Warren Dew
Travis Hall
Mary K. Kuhner (mkku...@phylo.genetics.washington.edu) wrote:
: In article <345848...@iol.ie> mano...@iol.ie writes:
:
: >Somewhat on a tangent, but one thing I'm curious about: Have any of you
: >ever had your *characters* realize that the world they live in is
: >failing to obey the normal laws of probability?
:
: My PBeM character Catalina had this experience.
: I don't think it worked out the way the GM intended. I suppose he
: meant for Cat to decide that she was being fooled and manipulated,
: If one's going to run this kind of scenario with Immersive players, I
: think it's a good idea to explicitly ask for characters with a lot
: of mental stability and self-confidence--the kind of person who keeps
: believing in herself no matter *what* happens. Or else to be prepared
: for a substantial proportion of characters to self-destruct.
Interesting. I've just realised, I did something somewhat similar (I
think) to the entirely group in my regular AD&D game a while back. I ran
an adventure in which the very world-rules were breaking down around the
characters. That break-down was supposed to be something of a clue towards
figuring out what was going on. It was a bit strange, I thought, that the
players refused to allow their observations of the break-downs to impact
upon character decisions for quite a while - though it worked fine,
because the characters came out acting just as if they were in denial
about it, until the sheer improbability of things could not be overlooked.
Upon reflection, it isn't really that strange - I mean, if what you
observed led you to question your own sanity (and, in this case, perhaps
even existence), would you tell people about it?
However, I didn't have to worry too much about self-destruction. The
entire adventure was a sort of dream-sequence, with each character being
put into the dreamscape in a distorted form. Thus, the characters
themselves were not completely part of it, and thus were somewhat
insulated from madness. In fact, for most of them, the only way they could
be harmed (other than the psychological impact of a really weird
nightmare) was to (effectively) commit suicide in the dream, and there
were also strong clues indicating that they should not do this.
Brett Evill
In article <19971101135...@ladder01.news.aol.com>,
psych...@aol.com (Psychohist) wrote:
>I can see your description for the Constable of Thekla, I probably wouldn't
> have a problem with it if you decided it before first running the game (Though
> I have known players who objected to even that amount of dramatism - more on
> the philosophy here only if you're interested). On the other hand, if you're
> having to fill out details of Thekla only because it now seems, after starting
> the game, that the player characters might go there, I would consider the
> decision process you describe to be distinctly not world oriented.
I guess, then, that that is what makes me a dramatist. Where is it clear
to everyone that *something* has to be made up by the GM, because nothing
definite can be extrapolated from what is already established, then I
don't mind being [seen to be] dramatic. If players object to the world
being rather over-full of 'interesting' things, then I just guess that
they won't like my GMing much.
If, like Mentor the Arisian, you can deduce the whole world, down to the
length of the shaving cut Vergil Samms will get on a particular morning
ten years from now, from the handful of facts generally established in RP
background and during play, you will find your inventiveness more
constrained than I find mind. I only aspire to be Sherlock Holmes' perfect
reasoner, who can deduce the *possibility* of a Niagara or an Atlantic
from a single drop of water.
On the whole, I find the contrast between myself and simulationist GMs or
in-character players is that they apprehend single definite certainties,
while I tend to imagine ranges of possibilities, and choose between them
for dramatic or depictive reasons. You might say that they are better at
deduction, or possess a faculty of direct apprehension that I lack, or
that I am more imaginative, or that our preferences simply differ. I
prefer not do commit myself on the point.
Brett Evill
In article <63e3g1$3...@senator-bedfellow.MIT.EDU>, ne...@athena.mit.edu
(Neelakantan Krishnaswami) wrote:
>Another way of looking at it, I picked up from David Berkman. The idea is to
>stop worrying about what the story will be exactly, because as long as you
>make sure the ingredients of a strong story are present -- a vivid, memorable
>setting that ties strongly to the motivations of the characters -- then there
>will be a good story. Knowing what will happen is irrelevant at best, so
>forcing the plot isn't needed.
Well put!
Mary K. Kuhner
In article <63fmrl$q07$2...@nargun.cc.uq.edu.au> zzt...@mailbox.uq.edu.au (Travis Hall) writes:
>It occurs to me that what you are doing with this "folklore logic" (and
>what I do with both folklore logic and similar concepts in many of my
>fantasy games) is using simulation to depict a dramatic world. In other
>words, the world acts as dramatist theory would have it work, but this is
>a part of the world, and we then wrap up the world+drama package in
>simulation. Thus, our desire for drama is provided quite happily by our
>simulation techniques, as we simulate drama.
>While often simulationism and dramatism are presented as opposing view, in
>this group, I don't think they need be so. I think there are ways that
>the two can meet, and that this is one of them.
Hmm.
I am currently running a fairly dramatist Feng Shui game and prepping to
run a fairly simulationist Ars Magica game. Both of them use magical
metaphysics--neither is true to real-world physics or causality--but
there is still a striking difference in my mindset as GM. In the Feng
Shui game I don't just rely on the world-laws to produce the kind of
game I want: I actively manipulate things, for example by looking for
ways to drain NPC allies out of a scene so that they don't swamp the
PCs or make the fight anticlimatic. While I could try to get that
effect by world-laws alone, it would require drastic ones--the usual
approach is to say "No one but the PCs is both heroic enough to tackle
this situation and willing to get involved." I'm not willing to use
that law, so I use contrivance instead. It's an understood part of the
game contract that I will contrive as necessary: all I'm required to
do is find reasonably plausible justifications.
The Ars Magica game won't have magical worldlaws mainly for drama. Most
of our design-for-story-value stuff involves choice of initial
characters and scenario. The magical worldlaws are there to make the
place interesting and different, and to make the magic feel like real
magic rather than renamed physics. I expect they will fight against
story values now and then, and I plan to let them do so.
I'm not using the kinds of meta-laws that say "Right prevails" or
"Virtue is rewarded" (nor their opposites): I'm using ones like
"Something transformed retains a touch of what it was" and "If you
use a given power, you open yourself to its masters." These may
make for good stories, but there is no guarantee that they will.
By the contract of that game, I *won't* be free to contrive or fudge
very much, certainly not to get the kind of tight pacing and every-
week-a-new-challenge plot that the Feng Shui game has. I'll be required
to play the magical laws resolutely by their internal logic, whereas
in Feng Shui I can bend them as needed for plot or pacing. ("Normally
Priss needs a physical link to find someone, but she knows Clyde real
well, and besides it would slow the pace to have her go back for a
link right now, so I'll just let it work.")
We are hoping for a bit more emotional kick in the long term, but
I know the AM game will also have more flat sessions, anticlimaxes,
and distressing events. I do see it as a tradeoff. Designing your
world to make it a dramatic place can help, but I've never seen a world
design that came anywhere close to guaranteeing smooth storylines
every session.
Mary Kuhner mkku...@genetics.washington.edu
Mary K. Kuhner
In article <345eb...@nunki.anu.edu.au> ma...@cs.anu.edu.au writes:
> F'rinstance, my Feng Shui games use a few basic folklore
> principles. (They're really just genre principles):
I'd call them genre, definitely: by "folklore" I meant things like
the Law of Contagion--dream-logic, magical logic--not drama principles.
> * Only Bad Guys can escalate the violence
> * The more inspired the stunt, the more likely it will succeed
> * Serious Heroes never look incapable, except in relationships
> * Boss Bad Guys can always escape, except at the end of the game
Interesting list. My player won't stand for #4 and is ambivalent
about #3, but we do use the other two (in fact, it was an article of
yours that led me to formally adopt #1) along with:
* Fights always happen in interesting places.
* The Heroes can always find a way to engage their opposition.
* Problems are solved by individual heroics, not mass action.
* There are no problems you can't solve by some combination of
combat and conversation.
I don't feel that these laws describe a coherent world, magical or
otherwise; but they do make for a fast-moving game with the flavor
my player likes. We find that neither of us cares very intensely
about these characters and their world, but that's a welcome change
now and again from intensity-fests like _Radiant_.
It will be fun, though, for me as a GM to get back to a more
simulationist game; at their best they run themselves in a way
that dramatist games just don't do for me.
Mary Kuhner mkku...@genetics.washington.edu
Mark Grundy
Travis Hall wrote:
| It occurs to me that what you are doing with this "folklore logic"
| (and what I do with both folklore logic and similar concepts in many
| of my fantasy games) is using simulation to depict a dramatic world.
| In other words, the world acts as dramatist theory would have it
| work, but this is a part of the world, and we then wrap up the
| world+drama package in simulation. Thus, our desire for drama is
| provided quite happily by our simulation techniques, as we simulate
| drama.
| While often simulationism and dramatism are presented as opposing
| view, in this group, I don't think they need be so. I think there
| are ways that the two can meet, and that this is one of them.
Folklore logic is indeed very nifty because it does help give a
sense of a coherent, sensible world (a sim goal), while at the same
time offering great drama.
The problem in using it for simulation is, folklore logic is often
full of contradictions, and it doesn't always apply to every
situation. So if players want a highly predictable world, you'll need
something to add to folklore logic to resolve those circumstances
where it's ambiguous or incomplete.
F'rinstance, my Feng Shui games use a few basic folklore
principles. (They're really just genre principles):
* Only Bad Guys can escalate the violence
* The more inspired the stunt, the more likely it will succeed
* Serious Heroes never look incapable, except in relationships
* Boss Bad Guys can always escape, except at the end of the game
There are lots of situations where these can conflict (eg an
inspired stunt against a boss bad guy early in the game), and plenty
of situations these principles don't cover. (Eg: two NPCs fighting).
These are resolved through other means -- I use a combination of world
sim and drama, but you could use just one or the other, and you might
get different results.
---
Brian Gleichman
Mary K. Kuhner <mkku...@phylo.genetics.washington.edu> wrote in article
<63b2i8$f46$1...@nntp5.u.washington.edu>...
> We've been in situations where the player's understanding is so strong
> that there is no surprise: for example, I've seen PC plans work out
> perfectly down to the last detail.
Just out of general interest, how often does this occur?
Travis Hall
John H Kim (jh...@konichiwa.cc.columbia.edu) wrote:
: Travis Hall <zzt...@mailbox.uq.edu.au> wrote:
: >It occurs to me that what you are doing with this "folklore logic"
: >(and what I do with both folklore logic and similar concepts in many
: >of my fantasy games) is using simulation to depict a dramatic world.
: >In other words, the world acts as dramatist theory would have it work,
: >but this is a part of the world, and we then wrap up the world+drama
: >package in simulation. Thus, our desire for drama is provided quite
: >happily by our simulation techniques, as we simulate drama.
: Hmmm. I find that folklore can make many standard dramatic
: techniques more believable -- but you still have to treat it as a
: story for the _players_. That is, if you are really trying to
: "simulate" drama, you have to take your audience into account.
: Drama is written to an audience.
:
: In contrast, the Secrecy game was often run to be objective.
: Consequently, in practice it was *nothing* like the standard fantasy
: that the Lore codified -- rather it was more like an overly serious
: parody. As characters, we saw and would comment on the dramatic
: storyline approaching -- i.e. like in-character trying to lighten
: the tone of the story so that the ending won't be so tragic for us.
: This often made us sound more like literary critics than literature.
I think what you were doing here is perhaps a little different from what I
was thinking, though it is still an interesting example in its own right.
In the Secrecy game, the audience for which the drama was written was the
characters themselves. Right? Now, that's a very interesting way of
looking at things, and I might have to try that sometime, but it wasn't
what I was getting at.
Instead of performing a simulation of drama watched by the characters, I
was thinking of a simulation of a world which would appear dramatic to an
outside observer. Note that this outside observer need not be the players,
though they would often still observe the simulated drama themselves, and
generally would perceive it as real drama. The characters, while living
within a dramatic world, would not be observers of that drama themselves.
Let's look at an earlier example - say, that Pendragon game with the
sword-bridge. To the players, this sounds like quite a dramatic game, what
with the unlikely challenge of the bridge and the actions required for the
players to overcome that challenge. Well, even if the players didn't find
it to be dramatic, it certainly seems that way to me, so if nothing else,
I am the audience that the drama requires. Remember, the players need not
be the audience. (In fact, the audience could be entirely hypothetical -
postulate the existance of a simulated audience for the simulated drama,
if you will.) However, we know that the game was based on quite definite
world laws: our "folklore logic". The characters, meanwhile, are
completely unaware of the drama taking place around them, despite the fact
that it is observable to an outside (and perhaps only imagined) audience.
Thus, a simulated world with simulated drama which poses as real drama.
Now, this might not be how the game originally came out, but I could see a
game having that sort of a feel.
Hmm... Now I'm starting to move towards simulations of games, and the game
within that simulation is itself simulating drama for an audience which is
only imagined by the imaginary players in our simulation. This recursion
has the potential to get really ridiculous.
Mary K. Kuhner
Hard to say, since it depends on what you count as a "plan". Some
campaigns play out routine engagements, others abstract them; obviously
the first kind has a lot more opportunity for things to go smoothly.
In games where we play out routine encounters, I'd say half of them go
smoothly.
For challenging encounters it's pretty rare: maybe a couple of times
per campaign. In our first Honolulu Feng Shui game I recall one where the
villains had a magic back door, but by a feat of inspired detective
work the heroes already knew that and had secured the other end. The
fight was trivial as a result; the player seemed quite pleased.
In _Paradisio_ the PCs were badly outclassed, and as their player I
treasured the one fight which went as planned (at least, the first
half did). It was by usual standards an anticlimax, since a major
villain simply died on impulse 1 with no chance to react, but having
months of hard thought pay off was extremely gratifying.
_Radiant_ has had a couple, though Markus is overly fond of
improvisation. The vampire they lured into attacking Markus did
*exactly* what she was supposed to, which was very nice (and
Markus did turn out to be able to survive the attack, which was
the weak spot in that particular plan....) Chernoi's scheme to give
a speech at the Convocation worked out perfectly, though she hadn't
quite forseen all the consequences.
Our second Feng Shui game has a lot of martial arts tournaments and
planned duels, and the PCs' record at having those go as planned has
been extremely good--partly because they were newcomers and people
underestimated them. I guess many of these fall under "routine
encounters".
One famous plan (took five hours to construct) in _Sunrise War_ went
absolutely like clockwork until one of the PCs got greedy and tried
to turn a little profit on the side (oops).
We did a mini-game in Shadowrun, basically one carefully planned hit,
where the entire game was the clockwork accomplishment of the PCs'
plan--I was a bit surprised, but the player seemed quite happy.
For me as a player, it's not something that has to happen very often, but
I really resent it if the GM takes the tack "No PC plan will ever work
out perfectly." I enjoy planning, and I really want to be able to
see my hard work pay off now and then. I recall a PBeM game where I
started suppressing my character's planning thoughts from the game
transcript because I learned that if the GM thought I'd planned
something, it wouldn't work, whereas if he thought it was improvised
it had a chance. I found this infuriating.
As a GM myself, I don't worry about it unless a large proportion of the
PCs' plans are trivially straightforward. I try not to get committed to
any particular vision of how an encounter will turn out, and if it
happens to be smooth and sweet, that's fine.
Mary Kuhner mkku...@genetics.washington.edu
scott....@3do.com
In article <b.evill-0211...@tynslip1.apana.org.au>,
b.e...@spamblocker.tyndale.apana.org.au (Brett Evill) wrote:
>
> In article <19971101135...@ladder01.news.aol.com>,
> psych...@aol.com (Psychohist) wrote:
>
> >I can see your description for the Constable of Thekla, I probably wouldn't
> > have a problem with it if you decided it before first running the game
(Though
> > I have known players who objected to even that amount of dramatism - more on
> > the philosophy here only if you're interested). On the other hand, if
you're
> > having to fill out details of Thekla only because it now seems, after
starting
> > the game, that the player characters might go there, I would consider the
> > decision process you describe to be distinctly not world oriented.
>
> I guess, then, that that is what makes me a dramatist. Where is it clear
> to everyone that *something* has to be made up by the GM, because nothing
> definite can be extrapolated from what is already established, then I
> don't mind being [seen to be] dramatic. If players object to the world
> being rather over-full of 'interesting' things, then I just guess that
> they won't like my GMing much.
It's not that we in the In character, and simulationist (World Oriented)
camp do not likeintereting things. It is that we do not like "interesting
things" happening at illogical or noticeably frequent occurances. Or
occuring because it aould be "pat" to do so. Beliefe n the world is
more important to us than good drama.
>
> If, like Mentor the Arisian, you can deduce the whole world, down to the
> length of the shaving cut Vergil Samms will get on a particular morning
> ten years from now, from the handful of facts generally established in RP
> background and during play, you will find your inventiveness more
> constrained than I find mind.
Yes, and the Ideal of that is what some of us strive for. Consistent
chemical reactions were discovered through observations over time that
gave us the benifits from this observation as we applied what we
observed, there by "learning, and applying "The Rules" of chemistry to
make what we have today.
>I only aspire to be Sherlock Holmes' perfect
> reasoner, who can deduce the *possibility* of a Niagara or an Atlantic
> from a single drop of water.
>
> On the whole, I find the contrast between myself and simulationist GMs or
> in-character players is that they apprehend single definite certainties,
> while I tend to imagine ranges of possibilities, and choose between them
> for dramatic or depictive reasons. You might say that they are better at
> deduction, or possess a faculty of direct apprehension that I lack, or
> that I am more imaginative, or that our preferences simply differ. I
> prefer not do commit myself on the point.
>
It's not certainies. If it were certainties we would be railroading. No,
possibly reduced number of probabilities for a particular action, traded
off by a wider selection of total actions or options our characters can
take. But to each their own...
Scott
-------------------==== Posted via Deja News ====-----------------------
http://www.dejanews.com/ Search, Read, Post to Usenet
Mark Grundy
I wrote:
: F'rinstance, my Feng Shui games use a few basic folklore principles.
: (They're really just genre principles):
Mary opined:
| I'd call them genre, definitely: by "folklore" I meant things like
| the Law of Contagion--dream-logic, magical logic--not drama
| principles.
They seem more genre'y to me too, but I figure that's because I'm
a child of the era in which they were written. Today's drama
principles are tomorrow's myth. Consider Casablanca. :)
: * Only Bad Guys can escalate the violence
: * The more inspired the stunt, the more likely it will succeed
: * Serious Heroes never look incapable, except in relationships
: * Boss Bad Guys can always escape, except at the end of the game
| Interesting list. My player won't stand for #4 and is ambivalent
| about #3, but we do use the other two
My Feng Shui games are all one-session stories, so a boss escaping
is only a very short-term measure for story progression, and not too
onerous. It also rarely comes up. Because the characters are
normally one-use, #3 actually helps story quality. If we played
campaign Feng Shui, I'd probably want to drop it as too taxing.
Mary's additional rules:
| * Fights always happen in interesting places.
| * The Heroes can always find a way to engage their opposition.
| * Problems are solved by individual heroics, not mass action.
| * There are no problems you can't solve by some combination of
| combat and conversation.
My group digs all these, but would amend #4 to include `occasional
sabotage' too. Additionally:
* Hit or miss, every major shot breaks *something*
| I don't feel that these laws describe a coherent world, magical or
| otherwise; but they do make for a fast-moving game with the flavor
| my player likes.
It adds cohesion to the extent that it helps players know what
consequences to expect, which lets them play to desireable outcomes,
avoid undesireable ones, and avoid false expectations.
Are these rules for consequences any less mythical than those of
the ancient Greeks? I feel that they're not. Two hundred years from
now, I reckon people will view (some of) these tales with the same
indulgent delight that we view Grimm stories: as the articulated
morality and concerns of a bygone era.
woodelf
In article <633ug9$bor$1...@nntp5.u.washington.edu>,
mkku...@phylo.genetics.washington.edu (Mary K. Kuhner) wrote:
> I have a fun book, _What if the Moon Didn't Exist?_ which plays what-if
> games with astronomy, meteorology, climatology and a bit of biology.
> Unfortunately the author is not a biologist and he's way too cautious--
> he asserts that animals can't do without a day/night cycle, for example,
> despite the apparent success of Earthly creatures such as polar bears
> and penguins. Does anyone know of similar speculation games done
> by a biologist?
well, Wayne Douglas Barlowe write and illustrated a lovely book:
Expedition: Being an Account in Words and Artwork of the 2358 AD Voyage to
Darwin IV. it's certainly not too conservative, but i don't know if his
boilogy is really up to snuff. it can fool me, but that's not much of an
accomplishment.
After Man: a Zoology of the Future by Dougal Dixon. also, The New
Dinosaurs by him. these books each run a what-if evolutionary sequence;
the first looks 50 million years into the future, minus mankind, and the
second picks up where we might be today if the dinosaurs had never been
wiped out. Dixon is a zoologist (i believe) for his dayjob, so the
biology is quite good, if not impeccable. the illustrations are also very
lifelike. of the two, i think that After Man is the more interesting.
The Science in Science Fiction, by Peter Nichols is not so much a what-if
book as a what-if debunker. while a bit out of date on some of the more
esoteric fields of science (copyright '82), it looks at most of the major
themes in SF and addresses how much basis there is for them, and where SF
gets it wrong. things like FTL, fusion, black holes, time travel, psychic
powers, shrinking rays, and silicon lifeforms. a good overview, though
you'll want more detailed books in those areas you want to really dig
into. very readable for the non-scientist, but not insulting to the
scientifically-minded (though it's unlikely to deal with anything that is
news to you in your field).
woodelf <*>
nbar...@students.wisc.edu
woo...@cs.wisc.edu
http://www.upl.cs.wisc.edu/~woodelf
Rules of combat older than contact with other races. Did not mention
aliens. Rules change caught up in committee. --ex-Green Drazi Leader
scott....@3do.com
In article <01bce920$0d1e8320$627c5ecf@emet>,
"Brian Gleichman" <glei...@ix.netcom.com> wrote:
>
> Mary K. Kuhner <mkku...@phylo.genetics.washington.edu> wrote in article
> <63b2i8$f46$1...@nntp5.u.washington.edu>...
>
> > We've been in situations where the player's understanding is so strong
> > that there is no surprise: for example, I've seen PC plans work out
> > perfectly down to the last detail.
>
> Just out of general interest, how often does this occur?
For me. in the superhero game, it occured about 33% (A third of the time.
The plans worked as planned. Sure their would be unforseen complications
but delays were planned for, but about 10% of those operations were
flawless walk in, never fire a shot cakewalks. about an additional 40% of
the time they didn't precisely work as expected, but we planned for
contingencies, or brought enough equipment to adapt to the new situation.
about 10% were complete disasters where we had been fed bogus information
and were out of position or luredinto a trap. These were comfortable
ratios, butthey occured because we took the time in game to perform
recon, talk to people and gather intelligence, and practice practice
practice. For me it was the execution of teamwork and seeing a plan fall
together in operation that was the thrill for me, not the individual hero
overcoming nigh unto insumountable odds. But I recognize this is not a
common taste. You see, the GM wanted to reward thinking and intelligence.
Mary K. Kuhner
> Mary opined:
> | I'd call them genre, definitely: by "folklore" I meant things like
> | the Law of Contagion--dream-logic, magical logic--not drama
> | principles.
> They seem more genre'y to me too, but I figure that's because I'm
> a child of the era in which they were written. Today's drama
> principles are tomorrow's myth. Consider Casablanca. :)
I think you're missing my point. The things I was thinking about as
folkloric laws--like a knight in your Pendragon game possibly saying
"Don't try to overcome a moral test by base cunning"--exist inside the
gameworld. The characters can talk about them and it doesn't make
the game feel self-referential or hypermodern.
I recall Channa, who was a Shadowrun mage, lecturing Jay, who was a
Shadowrun decker, about her blase' disregard for magical issues like
intention and symbolism. "It's not safe to pretend to be your enemy,
not in a magical context. Image becomes actuality."
By contrast, if Feng Shui characters are actually made aware of the
kinds of laws you cite, like "The villain will always escape till
the end of the story", it makes a huge difference in the feel of
the game, tending to move it towards satire or comedy. There is a
section in the Feng Shui rulebook about exactly this: one should
probably not let one's characters talk about movie conventions if one
wants the game to seem like a standard action movie and not _The Last
Action Hero_.
These are two different kinds of "magical" laws. One is inside the
gameworld and can be known by the characters; one is purely metagame
and cannot, in general, without turning the whole game into a metagame
exercise.
I've played in games with a wide assortment of internal magical laws.
I've never seen one which would consistently, using only simulationist
GMing tactics, produce games with the kind of stuff Feng Shui is
aiming for. I think you need the metagame there.
Mary Kuhner mkku...@genetics.washington.edu
Mark Grundy
Mary wrote:
| I'd call them genre, definitely: by "folklore" I meant things like
| the Law of Contagion--dream-logic, magical logic--not drama
| principles.
> They seem more genre'y to me too, but I figure that's because I'm a
> child of the era in which they were written. Today's drama
> principles are tomorrow's myth. Consider Casablanca. :)
| I think you're missing my point. The things I was thinking about as
| folkloric laws--like a knight in your Pendragon game possibly saying
| "Don't try to overcome a moral test by base cunning"--exist inside
| the gameworld. The characters can talk about them and it doesn't
| make the game feel self-referential or hypermodern.
Ah. Okay. In my Pendragon game, this is mixed. Sometimes
players trot out things like `The Land and the King are One', or `If
you have faith, God will provide', and this is really an in-game
equivalent of a play contract arrangement. At other times, the
players know that I've got a Principle at work, but can't quite tie it
to any folklore. At those times, they may find personality reasons or
invent superstition to adapt to the Principle if they can. Here's a
synthetic example:
By misfortune, a knight comes to be involved in the suicide of a
lady renowned for her beautiful long hair. Later on, he comes to a
ford where a challenge knight awaits, and sees there a long, weeping
willow, its branches swaying like tresses. The dramatic principle
involved is that the knight's neglect of the lady will eventually come
back to haunt him, and I've chosen to telegraph this scene with a
motif echo. The character could duck the combat from a nameless sense
of foreboding, or could tell a story he heard once about how willows
house the ghosts of forlorn maids... but this is a player choice.
Either way would result in a check on the character's Cowardly trait.
| By contrast, if Feng Shui characters are actually made aware of the
| kinds of laws you cite, like "The villain will always escape till
| the end of the story", it makes a huge difference in the feel of the
| game, tending to move it towards satire or comedy. make the game
| feel self-referential or hypermodern.
The same can occur in our Feng Shui games, where as you've
mentioned, there isn't really any folklore guidance at all. If the
players encounter the big villain early in the game, they may come up
with reasons not to press him too hard, or they press him too hard and
he'll elude them anyway. They're still free to (if they wish), make
stuff up like `I don't like this. He's acting like he has an ace up
his sleeve. Let's investigate further.'
Travis Hall
: mirroring-emotions-in-environment thing, both deliberately and
: intuitively (i.e. the GM just thought there was something right about
: having a willow there). But we don't tend to think of the reaction as
: a player choice:
A thought: Is there any good reason why sometimes such
sympathetic-environment factors cannot be observed by the character, and
thus become a legitimate factor in character decision-making? For example,
could not the knight of the example see the tree and think, "'Tis passing
strange, but that tree doth remind me strongly of that poor lady who,
tragically, took her own life."?
Obviously, this won't always be so. Sometimes, it is only due to player
knowledge that such factors will be recognised, but in that particular
example, it seems to me that the character could well make the
observation, unless that would not be in keeping with the genre and spirit
of the game.
: in fact, when I'm playing I tend to reject things
: outright if I am clear that they are player and not character choices.
I recently had a player in my Torg campaign try to reject a decision which
was not a character choice. He was attempting something that the character
could legitimately choose to attempt, but there were two similar but
distinct methods to make the attempt, and the character had no means to
distinguish between the two. So, I asked the player which method was going
to be used. He complained that his character was making a decision he had
no means to make. I replied that it didn't matter - I wasn't asking the
character to make the choice. I just wanted to know which mechanics to
apply, and if he didn't want to make the decision, we could flip a coin.
The reason I mention this is because it seems to me that sometimes (again,
definitely not all the time), a non-character decision can be made by the
player without affecting the legitimacy of the portrayal of the character.
Does this sort of concept directly contradict your sense of gaming
aesthetics, Mary?
Mary K. Kuhner
>A thought: Is there any good reason why sometimes such
>sympathetic-environment factors cannot be observed by the character, and
>thus become a legitimate factor in character decision-making? For example,
>could not the knight of the example see the tree and think, "'Tis passing
>strange, but that tree doth remind me strongly of that poor lady who,
>tragically, took her own life."?
I'd be happy with this, if that was the way it actually hit me in play.
If I'm in character and the GM's description makes me feel foreboding,
I'm happy to assume that the observed scene makes my character feel
foreboding: he may or may not be able to say why.
[player must choose among character-invisible mechanics]
>The reason I mention this is because it seems to me that sometimes (again,
>definitely not all the time), a non-character decision can be made by the
>player without affecting the legitimacy of the portrayal of the character.
>Does this sort of concept directly contradict your sense of gaming
>aesthetics, Mary?
The given example would annoy me, certainly: I'd feel that the GM
dragged me out of character to bother me about coin-flipping and
mechanics-picking. In general it's an esthetic principle for me that
the mechanics should not offer multiple ways to do something unless
the character perceives multiple ways to do it, and should not ask
for choices that have no meaning in character.
What purpose are the dual mechanics serving in your game? Wouldn't
it be simpler to stick to just one, if there is no character decision
involved? Or couldn't the GM do all the picking, if you must have two?
On the other hand, which mechanic to use is pretty neutral, so it would
just be an annoyance, not a serious problem. It's much more serious if
the GM wants me to use information the character lacks *in a way the
character would notice*. My great fear is of looking at the past,
in character, and asking "Why did I do that?" and having a metagame
answer rather than an in-character answer pop up. I tend to lose my
ability to play characters for whom this happens too often, so I'm
very picky about it.
The most painful example I remember is looking back at a disasterous
combat involving one of my early AD&D characters, and asking "Why did
I do that?" The real answer was "Because the pushiest of the other
players said we'd upset the GM if we didn't go along with her plot, and
I was too tired to argue with him." My character couldn't possibly
take that as an explanation--but it was so obviously true that all the
in-character explanations I tried to come up with felt like wet tissue
paper, no substance to them. There probably was a reason that would
have satisfied an outside observer, but there were none that satisfied
me. Luckily the game folded shortly afterwards, because that character
was a dead loss.
Mary Kuhner mkku...@genetics.washington.edu
Mary K. Kuhner
In article <64e2kg$1b7$1...@nargun.cc.uq.edu.au> zzt...@mailbox.uq.edu.au (Travis Hall) writes:
>Hmm... I think I perceive a slight stylistic difference between us here,
>Mary. I've never been terribly bothered about things happening which might
>"drag me out of character", I think largely because I view the game on
>multiple levels almost all the time, anyway.
I recall from a previous discussion (several years ago) that some players
are happy to have multiple levels all running at once, and others tend to
be more obsessively interested in only one level at a time. I tend to
deal mainly with two, the character POV and the mechanics-manager, but I'm
happiest when the mechanics-manager is running totally on autopilot.
That means no decision making on that level, just implementation of
rules. This is why I have a prejudice against rules that require OOC
decisionmaking--they bring the mechanics-manager up into foreground
where it gets in the way.
>Thinking about it, I've just remembered that the player who had a problem
>with it wasn't even the one playing the character concerned. The player of
>the character involved didn't seem to have any problems with it, while the
>one who complained had had his character hospitalized earlier in the game.
Maybe it was the general principle he was objecting to: after all, it will
be his character in that situation sooner or later.
>In actuality, the two mechanics did have meaning for the character, but
>the character didn't have enough experience to really know what he was
>doing. The choice really came down to "fast or slow?", and the IC response
>should have been (and was) "what you talking 'bout, Willis? I don't know,
>I don't care, I'm just doing it".
Ah. I think I'd try to solve this one IC, in the following way: envision
the character doing what he's doing. Does it look more like "fast" or
more like "slow"? Then just spit out an answer. If there's really no way
to tell at all, I'd still feel it was a bad mechanic.
>However, OOC choices don't bother me, as long as what the
>character can notice isn't affected at all.
I'm not too picky on emotionally-neutral OOC choices: I really don't
want to make emotionally charged ones, as for me strong player emotions
that don't come from the character seriously disrupt the player level.
I'm curious: how do you feel about cut-scenes and other means of feeding
players OOC information? David Berkman wrote a scenario where the players
learn at the beginning that if their characters straightfowardly complete
the mission, their employer will kill them; but the characters don't know.
Some players like this fine--I'd actually walk out of the game at the
start, as soon as I saw what was happening, because it would so greatly
detract from my enjoyment.
Mary Kuhner mkku...@genetics.washington.edu
R. Serena Wakefield
My spies inform me that, on 13 Nov 1997 19:08:43 GMT,
mkku...@phylo.genetics.washington.edu (Mary K. Kuhner), hiding under
the bedcovers, scribbled in the following in their diary:
>In article <64e2kg$1b7$1...@nargun.cc.uq.edu.au> zzt...@mailbox.uq.edu.au (Travis Hall) writes:
>>In actuality, the two mechanics did have meaning for the character, but
>>the character didn't have enough experience to really know what he was
>>doing. The choice really came down to "fast or slow?", and the IC response
>>should have been (and was) "what you talking 'bout, Willis? I don't know,
>>I don't care, I'm just doing it".
>Ah. I think I'd try to solve this one IC, in the following way: envision
>the character doing what he's doing. Does it look more like "fast" or
>more like "slow"? Then just spit out an answer. If there's really no way
>to tell at all, I'd still feel it was a bad mechanic.
That is the major problem I had with Amber's mechanics -- the notion
of Good Stuff, Zero Stuff and Bad Stuff governed a great many
decisions in the game, but was based entirely on OOC decisions
involving points and auctions rather than anything the character could
understand or interpret.
I suppose for me there's a strong separation between player and
character -- my character is not going to know or care who I am or how
good I am at mathematics or bluffing. If a system forces me to step
out of character to make a decision competently, there's something
very wrong.
>I'm curious: how do you feel about cut-scenes and other means of feeding
>players OOC information? David Berkman wrote a scenario where the players
>learn at the beginning that if their characters straightfowardly complete
>the mission, their employer will kill them; but the characters don't know.
>Some players like this fine--I'd actually walk out of the game at the
>start, as soon as I saw what was happening, because it would so greatly
>detract from my enjoyment.
I would do the same. If cut scenes are used at all, they should be
extremely vague, used as foreshadowing only -- "Miles away, a shadow
falls over Schloss Drachenfeuer. A scream rings out into the still
night, and then all is silent once more. Meanwhile, our heroes ..."
The problem with dramatic irony in RPGs is that any information given
to the audience is also given to the protagonists. The perfect
example is Columbo. In every episode of Columbo, the audience gets to
witness the murder and Detective Columbo does not. The tension comes
not from the mystery itself but from whether or not Detective Columbo
will be able to outwit the murderer. If Columbo knew everything the
audience knew, it would be a mighty short show and not an especially
interesting one.
It's possible to rig up a cut scene IC through scenes and visions, but
I've found that players respond VERY negatively to such heavy-handed
manipulation. I find it best to avoid them entirely, and let them go
looking for prisonersf they want to know what their enemies are up to.
R. Serena Wakefield
rai...@ROSECOLOREDGLASSES.gate.net
(remove rose-colored glasses to e-mail)
Serena's Sanctuary: http://www.gate.net/~raistw
Thomas Cook
R. Serena Wakefield (rai...@ROSECOLOREDGLASSES.gate.net) wrote:
: It's possible to rig up a cut scene IC through scenes and visions, but
: I've found that players respond VERY negatively to such heavy-handed
: manipulation.
I've had success with short stories that are written about major events
that the characters are not present to witness. Since its a
self-contained story, the players find it entertaining on its own.
Since they are read outside of the game, the players have an easier time
seperating what they know from what their characters know (I do avoid
major spoilers though and once held back a few pages until they no longer
were spoilers). And if they don't want to risk any spoilers, they simply
don't read it.
Thomas.
[The Thol-Far RPG]
Psychohist
Travis Hall, discussing the handling of mechanics choices not visible to the
character, posts, in part:
In actuality, the two mechanics did have meaning
for the character, but the character didn't have
enough experience to really know what he was doing.
This changes things significantly. If there were no way to tell the difference
between the choices on the character level, I'd agree with Mary - no point in
having the choices. In this situation, though, I as a player might sometimes
want the choice because I might know something about my character that would
cause him to pick one choice or the other, even if the character itself
couldn't tell the difference.
Ideally, of course, the gamesmaster would decide based on perfect knowledge of
the character. In general, though, the player knows the character better than
does the gamesmaster, and so is better qualified to make the decision.
As gamesmaster in these situations, I tend to throw these choices onto the
player, describing them in nonmechanical terms - for example, 'fast and sloppy
or slow and sure?' If the player seems indecisive, I'll pick a choice and
mention my reasons so that the player gets a chance to contradict me - 'well,
that character usually does things as fast as possible, so....'
I'm curious, Mary, about what method you'd prefer for handling situations like
this.
Warren
Mary K. Kuhner
>As gamesmaster in these situations, I tend to throw these choices onto the
> player, describing them in nonmechanical terms - for example, 'fast and sloppy
> or slow and sure?' If the player seems indecisive, I'll pick a choice and
> mention my reasons so that the player gets a chance to contradict me - 'well,
> that character usually does things as fast as possible, so....'
>I'm curious, Mary, about what method you'd prefer for handling situations like
> this.
It comes up in our homebrew not too infrequently. Skill rolls are
modified by stats, and we use a mix-and-match approach so it's not
always obvious which stat to use. Usually the GM decides based on his
knowledge of the context of the roll, but sometimes he'll pitch it
at me: "You could use either Will or Talent to try that; which one?"
If I can tell from thinking about the character, I'll do so: "He's
going on raw desire and determination: that's Will." If I can't,
I'll describe what I think the character is doing, and
let the GM make a call from that: "Well, he's using the flame to
enhance his concentration, the way he does in sword practice, and
then...."
Pretty much what you're doing, I think. Like you, we find that the
GM sometimes doesn't know enough about the PC to make an accurate
call.
Mary Kuhner mkku...@genetics.washington.edu
Mary K. Kuhner
In article <346b6dcc...@news.gate.net> rai...@ROSECOLOREDGLASSES.gate.net (R. Serena Wakefield) writes:
>In every episode of Columbo, the audience gets to
>witness the murder and Detective Columbo does not. The tension comes
>not from the mystery itself but from whether or not Detective Columbo
>will be able to outwit the murderer. If Columbo knew everything the
>audience knew, it would be a mighty short show and not an especially
>interesting one.
There's also the fact that RPGs are more active than TV. The player
can get real pleasure from *solving* a problem, but there is no way
to recapture that pleasure if you're handed the solution right off.
Our "in-house" play style makes it hard to keep secrets (one player,
multiple PCs) but we certainly don't go out of our way to spoil them.
There are ways to reduce OOC knowledge even with this style of play:
for example, when the PCs are separated one can do the most
significant one last, so that the player won't know the result of
that scene when she's playing out the others.
>It's possible to rig up a cut scene IC through scenes and visions, but
>I've found that players respond VERY negatively to such heavy-handed
>manipulation. I find it best to avoid them entirely, and let them go
>looking for prisonersf they want to know what their enemies are up to.
Here I disagree: we have had pretty good luck with dreams and visions.
I think the difference is that our game contract requires them to be
things that arise out of the game world, not plot devices: we don't
put in a vision just to cue the players onto a plot, we put it in
because it's what would happen. In my more dramatist Feng Shui
game I tried vision-as-plot-device and was not impressed with the
player's reaction.
My absolute favorite along these lines is *having* dreams in-character,
which has happened to me a couple of times: one quite recently, where
I got to see a kind of warped sketch of the main villainess' experiences
from her point of view. Left me feeling a lot more sympathetic with
her, it did.
Mary Kuhner mkku...@genetics.washington.edu
Travis Hall
: extremely vague, used as foreshadowing only -- "Miles away, a shadow
: falls over Schloss Drachenfeuer. A scream rings out into the still
: night, and then all is silent once more. Meanwhile, our heroes ..."
I've had some success with cut scenes. When I was running the original
Dragonlance modules (ala Travis - I never run anything straight :) ) I
wrote (or outright stole from the novels, for early sessions before our
campaign diverged too much from the novels) an introduction for each
module. These introductions showed things that the characters could not
possibly have seen. Of course, much of that information I did not expect
to be used - in fact, there was actually no opportunity whatsoever to
apply that part of the information. It was just tone-setting. However, I
did weave little bits of character-relevant information into the intros.
This was stuff that the characters should know, but often stuff that the
players might not normally notice. (They often still didn't.) It was just
there to subtly direct attention to that character knowledge, although it
was never essential to the modules.
For example, the first intro described a scene in the Inn of the Last
Home, a few hours before the arrival of the PCs. It described the
interaction between Tika (the barmaid and old friend of the PCs) and the
old storyteller. It set the tone for the players by telling them that
dramatic events were afoot ("A party? Yes, girl. It will be a party such
as the world of Krynn has not seen since the Cataclysm. Be ready, Tika
Waylan. Be ready!" Or something like that), but such information had no
relevance to the module. For the characters, though, it gave an idea of
what to expect from Tika, who was an old friend whom they had not seen in
years, and a good idea of the layout of the Inn, including mention of the
back way out, which the characters knew of from their previous visits.
The second module's intro again showed Tika in the bar, now blackened with
soot from the sacking of Solace. Again, this filled in details for when
the PCs arrived. A later intro gave hints of the plight of the refugees
the heroes were trying to find a passage through the mountains for. The
characters would have known damn well that the refugees would be worried
about them, and what would happen if they did not return, but the players
might well overlook that. Another intro described a dragon's difficult
decision to break the Oath of the Good Dragons and get involved. The
description was helpful to the person who was supposed to play that
dragon, as it helped him understand the character a little better (this
game used pre-gen character - don't worry, he still had plenty of room for
interpretation), while it never had the slightest bearing on the game for
anyone else, except from the POV of their appreciation as an audience. A
final intro gave brief descriptions of what other people the PCs had met
along the way (or ex-PCs now separated from the group) were doing, just
before the events of the final module. This let the players realised that
they were a part of something much larger than themselves, a fact plainly
apparent to the characters, who would be hoping with all their hearts that
all their friends are alright and that they succeed in their little parts
of the war effort. Also, I described these scenes as dim, half-remembered
dreams, so the information (which was virtually nothing) was available to
the characters.
: The problem with dramatic irony in RPGs is that any information given
: to the audience is also given to the protagonists. The perfect
: example is Columbo. In every episode of Columbo, the audience gets to
: witness the murder and Detective Columbo does not. The tension comes
: not from the mystery itself but from whether or not Detective Columbo
: will be able to outwit the murderer. If Columbo knew everything the
: audience knew, it would be a mighty short show and not an especially
: interesting one.
Yes, if cut scenes are to be used, they will have to be much more subtle
than that. The information given in them should probably usually be
non-essential, in case the players prefer to ignore it, and it should not
be anything blunt enough to give the game away.
Ken Walton
In article <64g4gv$8...@knot.queensu.ca>, Thomas Cook
<3t...@qlink.queensu.ca> writes>: It's possible to rig up a cut scene IC through scenes and visions, but
>: I've found that players respond VERY negatively to such heavy-handed
>: manipulation.
>
>that the characters are not present to witness. Since its a
>self-contained story, the players find it entertaining on its own.
>Since they are read outside of the game, the players have an easier time
>seperating what they know from what their characters know (I do avoid
>major spoilers though and once held back a few pages until they no longer
>were spoilers). And if they don't want to risk any spoilers, they simply
>don't read it.
In the GURPS Celtic Myth/Harn campaign I ran, I ended up with two
seperate sets of PCs, one lot playing the main characters, the other lot
involved in occasional cut scenes. One of the characters in the main
game had been found abandoned on the steps of a temple with no knowledge
of his past (player's story). One week, when we had less than enough
players for the main game, I started a game set in a kingdom to the
south that was falling apart into civil war. The new PCs saw the king
assasinated and were given the task of finding the long lost rightful
heir to the throne. Every few weeks I'd have another session with the
second group, in which it slowly became apparent that the heir was the
PC from the first group (who had enough troubles to start with!). In the
grand finale of the campaign, I had seven players and 11 PCs involved in
a massive seige, with the action cutting between one group and the other
until they finally met.
It worked very well with that group, since they were very good at
keeping player and character knowledge seperate. In fact, they would
even use player knowledge to do things harmful to their characters if it
made the story better. For instance, Jo's character was a bard who had
the gesa that she could never praise anyone in front of them. Now Jo
*knew* that one of the characters was *really* the great hero Jordin
(though he was going under an assumed name), though her character
didn't. So in the last battle, she deliberately sang a song in praise of
Jordin, *in front of him*. Then, when she found out that she'd broken
her gesa, took her clothes off, painted herself blue and went out of the
fortress to die in battle. The scene where Jordin carried her bloody
body back through the rain to the citadel was one of the most moving in
a very moving campaign. (Celtic Myth makes for good tragedy :-) )
Having said all that, there are very few groups of gamers who I could
rely on to behave like that.
--
Ken Walton
"Do not meddle in the affairs of wizards, for they are subtle and quick to
anger." - Tolkien
"Complex projects are made simple with wizards." - Microsoft
woodelf
In article <346b6dcc...@news.gate.net>,
rai...@ROSECOLOREDGLASSES.gate.net (R. Serena Wakefield) wrote:
> My spies inform me that, on 13 Nov 1997 19:08:43 GMT,
> mkku...@phylo.genetics.washington.edu (Mary K. Kuhner), hiding under
> the bedcovers, scribbled in the following in their diary:
> That is the major problem I had with Amber's mechanics -- the notion
> of Good Stuff, Zero Stuff and Bad Stuff governed a great many
> decisions in the game, but was based entirely on OOC decisions
> involving points and auctions rather than anything the character could
> understand or interpret.
really? while the Stuff arises from game mechanics, i've always just seen
it IC as "unlucky" or "lucky," a self-conception that i can certainly see
someone applying to themselves.
> >I'm curious: how do you feel about cut-scenes and other means of feeding
> >players OOC information? David Berkman wrote a scenario where the players
> >learn at the beginning that if their characters straightfowardly complete
> >the mission, their employer will kill them; but the characters don't know.
> >Some players like this fine--I'd actually walk out of the game at the
> >start, as soon as I saw what was happening, because it would so greatly
> >detract from my enjoyment.
that's an interesting idea; it never occurred to me that anyone would
*deliberately* feed the players plot-related info that their characters
couldn't know. sounds like cruel torture to me ("ok, now i want you to
play your characters as if you don't know what the future will be. oh, and
you're all going to die, even if you succeed.") what's the point? it
seems to me that all it will accomplish is making the roleplaying harder.
if you want the characters to behave differently according to the info
(act with foreboding, frex), then you need to give the info to the
characters. if you don't want them to behave differently, then why inform
the players?
> The problem with dramatic irony in RPGs is that any information given
> to the audience is also given to the protagonists. The perfect
exactly! while one can ideally separate the audience and protagonist
functions, in practice OOC info tends to make the protagonist role less
enjoyable at the expense of the audience role, IMHO.
R. Serena Wakefield
My spies inform me that, on Sun, 16 Nov 1997 14:38:57 -0600,
nbar...@students.wisc.edu (woodelf), hiding under the bedcovers,
scribbled in the following in their diary:
>In article <346b6dcc...@news.gate.net>,
>rai...@ROSECOLOREDGLASSES.gate.net (R. Serena Wakefield) wrote:
>> My spies inform me that, on 13 Nov 1997 19:08:43 GMT,
>> mkku...@phylo.genetics.washington.edu (Mary K. Kuhner), hiding under
>> the bedcovers, scribbled in the following in their diary:
>> That is the major problem I had with Amber's mechanics -- the notion
>> of Good Stuff, Zero Stuff and Bad Stuff governed a great many
>> decisions in the game, but was based entirely on OOC decisions
>> involving points and auctions rather than anything the character could
>> understand or interpret.
>really? while the Stuff arises from game mechanics, i've always just seen
>it IC as "unlucky" or "lucky," a self-conception that i can certainly see
>someone applying to themselves.
The concept of luck is a highly questionable one, specifically in
Amber -- where you can make your own luck by taking a single step
through shadow. There's also the matter of Stuff being used as
alignment (Good Stuff = nice guy, Zero Stuff = mysterious, Bad Stuff =
nasty guy).
Usually, when playing Amber, I just use the superlative combat systems
and background information and ditch the rules ... just like it tells
you to at the end of the book. ;)
>> >I'm curious: how do you feel about cut-scenes and other means of feeding
>> >players OOC information? David Berkman wrote a scenario where the players
>> >learn at the beginning that if their characters straightfowardly complete
>> >the mission, their employer will kill them; but the characters don't know.
>> >Some players like this fine--I'd actually walk out of the game at the
>> >start, as soon as I saw what was happening, because it would so greatly
>> >detract from my enjoyment.
>that's an interesting idea; it never occurred to me that anyone would
>*deliberately* feed the players plot-related info that their characters
>couldn't know. sounds like cruel torture to me ("ok, now i want you to
>play your characters as if you don't know what the future will be. oh, and
>you're all going to die, even if you succeed.") what's the point?
It's somewhat reminiscent of players taking the Dark Fate flaw in
Vampire. If you take that Flaw, you get a pile of extra character
points, but you *know* that your character is going to die horribly at
some point. Real maturity is required to handle that correctly ...
... but, then, that's player-initiated. Nobody makes you take that
Flaw. Forcing the group into that situation using OOC information is
incredibly heavy-handed and horrible.
I'd prefer to tinker with the plan to increase the drama. They return
successfully, and are sent up to bed, but the druid is restless
sleeping indoors as ordered and goes to spend the night under the
stars. While descending the stairs, she overhears the party's
employer instructing his minions to perform some ambiguous but
worrisome-sounding task. Does she awaken the others? Does she
investigate further, risking discovery? Oooh ...
Mark Apolinski
woodelf wrote:
>
> In article <346b6dcc...@news.gate.net>,
> rai...@ROSECOLOREDGLASSES.gate.net (R. Serena Wakefield) wrote:
>
> > My spies inform me that, on 13 Nov 1997 19:08:43 GMT,
> > mkku...@phylo.genetics.washington.edu (Mary K. Kuhner), hiding under
> > the bedcovers, scribbled in the following in their diary:
>
> > That is the major problem I had with Amber's mechanics -- the notion
> > of Good Stuff, Zero Stuff and Bad Stuff governed a great many
> > decisions in the game, but was based entirely on OOC decisions
> > involving points and auctions rather than anything the character could
> > understand or interpret.
>
> really? while the Stuff arises from game mechanics, i've always just seen
> it IC as "unlucky" or "lucky," a self-conception that i can certainly see
> someone applying to themselves.
>
> > >players OOC information? David Berkman wrote a scenario where the players
> > >learn at the beginning that if their characters straightfowardly complete
> > >the mission, their employer will kill them; but the characters don't know.
> > >Some players like this fine--I'd actually walk out of the game at the
> > >start, as soon as I saw what was happening, because it would so greatly
> > >detract from my enjoyment.
>
> that's an interesting idea; it never occurred to me that anyone would
> *deliberately* feed the players plot-related info that their characters
> couldn't know. sounds like cruel torture to me ("ok, now i want you to
> play your characters as if you don't know what the future will be. oh, and
> seems to me that all it will accomplish is making the roleplaying harder.
> if you want the characters to behave differently according to the info
> (act with foreboding, frex), then you need to give the info to the
> characters. if you don't want them to behave differently, then why inform
> the players?
The point is that sometimes it's fun. I've tried it and I've liked it.
If you don't like it, then don't use it. But for the record, it is
definitely *not* "cruel torture".
Mark
Travis Hall
Mark Apolinski (apol...@ix.netcom.com) wrote:
:
: This, however, misses the point that both Detective Columbo AND the PCs
: do *not* know what the audience knows. If your PCs are using the
: information in Cut Scenes in-character, then they are failing to
: properly firewall the information. Cut Scenes are meant to be
: firewalled.
Not always. In my example, I often used cut scenes to feed players
information which their characters had from other sources - like how my
description of the cut scene in the Inn of the Last Home mentioned the
back way out, which the characters had used in previous visits (not
played, as this is campaign pre-history).
In some cases, yes, the information in the cut scene will have to be
completely firewalled, in which case, it may be best not to give anything
away. However, the cut scene could still be used to direct player thinking
without ever giving any actual information - say, a cut scene involving a
chess game in the park might lead the players into thinking along the
lines of a spy-thriller, in which the various people are like the pieces
on the chess board. No information, but good tone-setting.