GNS, Intent and Motivations

[Valamir]

I'm not sure where to begin to explain why I think you're wrong Mike...I'm not even sure if there's a point to it.

But I shall try.  

You are saying that your baby kobolds example is a behavior.  It isn't.  It can't be.

"Player killed baby kobolds":   THAT'S the behavior.  That's what's observable.

"Because they aren't worth any XPs alive"  This is NOT part of the behavior.  It can't be because it isn't observable.  We only know this is the reason he killed the kobolds because you told us, because he actually said it in the game.  So if its not part of the behavior what is it?  Its part of the REASON for the behavior.  Its the part of the reason thats relevant for our purposes.  This is the piece I called motive, and later called Game Motive to distinguish it from other issues that were muddying the discussion.

"Because they aren't worth any XPs alive" is the REASON...the "because" gives it away.  Behaviors don't involve "because".  Motives involve "because".


And this "because" is the key piece of the whole thing.  "Player killed baby kobolds" tells us nothing about GNS mode.  As has been pointed out above and won't be repeated again this sentence could be ANY mode.

GNS can only be distinguished from the "because".  Without the "because" there is absolutely no way whatsoever to categorize any behavior as being G, N, or S.  Playing in Character is a behavior.  It does not equal Simulationism.  Using the rules to help the character perform better is a behavior.  It does not equal Gamism.  Only if we know the WHY of these things can we determine one or the other.  Without the "because" GNS is devoid of all meaning whatsoever.


But now we get to the good stuff.  How do we uncover the "because" when players aren't so accomodating as to announce it?  This is why, while the theory is about decision making, we can't look at decisions on an atomic level.  Because by themselves...we can't.  This is why Instance of Play becomes important.  Because over time and enough observations we can (in a mental process much like trying to guess the right pattern in Master Mind based on vague clues) deduce or infer the "Because" even when its not explicitly stated.

What started me off on this whole road to begin with is Ron's claim that motive (i.e. the "because" part) isn't part of the model at all and if we put it there is our own customization.  To which I say "Rubbish and Bollocks".  The motive part is there; has been there since the beginning, and to say otherwise is to me complete and utter nonsense.

Including the idea of motive doesn't make the model harder to understand.  Its the only thing that makes it possible TO understand.


Unless you, Ron, et.al. are prepared to make a laundry list of every possible roleplaying behavior and ascribe to that behavior a corresponding GNS mode, behavior by itself its insufficient for determining GNS.

Unless you are prepared to document a pattern of behaviors such as "in character dialog", followed by "ignoring a possible positive modifier", followed by "whatever"  is and always is a Simulationist event without exception; behavior by itself is insufficient for determining GNS.

Behavior is what is observed.  Motive is what is inferred from observing a sufficient amount of behavior.  Motive is what translates to a GNS mode.

GNS is WHY a player chose to do A instead of B.

This is what I've said since the beginning.  This is what I remain convinced of now.

[Valamir]

I think the crux of the debate for me is: "How can you identify, say, Narrativist behavior without a dead-giveaway (the player goes "mmm ... this will be good for the story")?"

If you can't then using GNS to analyze anyone else's play is an exercise in seeing what you want to see.

If you can, I'd like to know how.

If you're pretty sure you can most of the time, what are the objective-tells?

-Marco

Good question Marco...and one you've been hinting around at for a while.

For me the answer is simple.

The same way a jury decides whether a suspect is innocent or guilty  (only in the case of GNS there is no "guilt" just different positions).

We look at the evidence as its presented, piece together the puzzle and reach a conclusion.  As in all discussions that occur in a forum, everyone throws in their two cents and some consensus is reached.  There will usually be some dissenters, who, like in the Supreme Court, will write up their minority opinion on the matter.  Since this is hardly a life and death situation, there is no need for a unanimous verdict.

Like the justice system, no doubt sometimes the analysis will be wrong.  As Ron said this is where being "rigorous" comes in, in a scientific sense.

In a management sense its the difference between making a wrong decision vs. making a bad decision.  A decision can still be a good one, even if it winds up being wrong, if it were made intelligently based on the information available at the time (i.e. rigorously).

I sense a touch of "but how can you be sure when you perform your GNS analysis that you're not wrong?" in your question.  My response can only be...we can't be.  Any more than a jury can be sure they didn't just let a criminal off or send an innocent man to jail.  Only since we don't have peoples lives riding on the outcome, right or wrong is still a learning experience.

[damion]

Here's my definition for GNS modes.

(you can substitute any mode in the example, I just got tired of people always using N and G)

Observer sees behavors from player Y that cause the observer to think Y is making simulationist decisions.

Observer then  introduces Y and the rest of the group to a game that
promotes simulationist decisions.  

During ths game the Observers see's more behaviors from Y and the rest of the group that  the observer believes are indicative of simulationist decisions. (doesn't have to be, and probably won't be the same behaviors as originally observed).

Thus the observer can conclude that these players are making simulationist decisions, because
1)The observers induction of the Game Motives from the behaviors are consistent with the definition of simulationism.(This requires the observer to understand the definitions of course).

2)Promoting these decisions has increased the behavior.


I don't think there ARE objective tell-alls, other than a player explicitly vocalizing their modivation, and even then you could be wrong.  I think just having a good understanding of the people involved and the situation is the way to go.

[Mike Holmes]

You are saying that your baby kobolds example is a behavior.  It isn't.  It can't be.

"Player killed baby kobolds":   THAT'S the behavior.  That's what's observable.

"Because they aren't worth any XPs alive"  This is NOT part of the behavior.  It can't be because it isn't observable.  We only know this is the reason he killed the kobolds because you told us, because he actually said it in the game.  

How is something that someone said in a game not an observable behavior? He only said that he killed the kobolds. And the GM only said they died. And the player only said that he was happy to record them on his character sheet. How is that not all observable behavior?

"Because they aren't worth any XPs alive" is the REASON...the "because" gives it away.  Behaviors don't involve "because".  Motives involve "because".

The behavior of Gamism is described as making decisions based on in-gme metrics yadda yadda. Yes, that's because of the existence of that criteria. If a rat hits the red bar "because" he know's he'll get a shock from the green bar, that's a description of a particular behavior. His motive is probably avoiding pain. But we really can't ask the rat, can we?

Again, you are saying that something that some people term behavior is called motive. I disagree. Purely semantic. In the end, what is it that we disagree on other than the definition of these terms?

Playing in Character is a behavior.  It does not equal Simulationism.  Using the rules to help the character perform better is a behavior.  It does not equal Gamism.  Only if we know the WHY of these things can we determine one or the other.  

Simulationism has nothing to do withy playing in character. That's not in any way a definition of it. Simulationism could be stated as making certain decisions "because" those decisions prioritze nothing beyond basic exploration. Without these Beacuses as part of the definiton of the described behavior all you have is Players make decisions. So again, the only thing we disagree on is whether or not the "because" can be used as part of a description of a behavior (my contention) or must indicate a motivation (your contention).  

Either way it does not undermine our actual (IMO) agreement on the purpose or use of the theory.

But now we get to the good stuff.  How do we uncover the "because" when players aren't so accomodating as to announce it?  This is why, while the theory is about decision making, we can't look at decisions on an atomic level.  Because by themselves...we can't.  This is why Instance of Play becomes important.  Because over time and enough observations we can (in a mental process much like trying to guess the right pattern in Master Mind based on vague clues) deduce or infer the "Because" even when its not explicitly stated.

Again, we're in total agreement. We can't look at the decisions by themselves. The definition of the behavior of Gamism includes the conditions under which the decision was made. How could we create a system that catered to suh a player if we could not control these conditions. But they are conditions, IMO, not motives.

What started me off on this whole road to begin with is Ron's claim that motive (i.e. the "because" part) isn't part of the model at all and if we put it there is our own customization.  To which I say "Rubbish and Bollocks".  The motive part is there; has been there since the beginning, and to say otherwise is to me complete and utter nonsense.

Well, that's what you may see, but I don't. He says that players are high -fiving about "Winning". Because they have won. How else could he make the determination. What we don't know is why they want to win. Which is what I'd call motivation. The condition of there being a winning metric (the death of the baddies) is what they are responding to with their Gamist behavior.

Including the idea of motive doesn't make the model harder to understand.  Its the only thing that makes it possible TO understand.

Doesn't hamper me. If you need to state it this way, then I'm guessing that's why Ron said, "Hey, go ahead." I think that there may be people who are only going to see it this way. But consider that I have as much trouble seeing it from your perspective as you do seeing it from mine.

Unless you, Ron, et.al. are prepared to make a laundry list of every possible roleplaying behavior and ascribe to that behavior a corresponding GNS mode, behavior by itself its insufficient for determining GNS.

No, we only need to make a list of three behaviors. They are called Gamism, Simulationism, and Narrativism.

Unless you are prepared to document a pattern of behaviors such as "in character dialog", followed by "ignoring a possible positive modifier", followed by "whatever"  is and always is a Simulationist event without exception; behavior by itself is insufficient for determining GNS.

This is silly. A player makes a decision to kill baby kobolds because there are experience points available. He tells me this. This is falls precisely under the definition of the behavior of Gamism. How could I miss it?

I have now come to where the previous poster was before. We must agree to disagree. Ron has much more wisely seen the end of this discussion before it began, I think, and essentially declined to comment. He says that if you want to call it motive, that's fine, he just will not. Which is just a way of saying before the conflict erupts that there is no solution to this problem likely, so why should I waste my time debating.

I only wish that I had seen the wisdom of this before as well.

Mike

[Valamir]

Behavior:  The actions of a person in response to stimuli

Stimuli:  Something that incites or rouses to action

Motive:  Of or constituting an incitement to action


Abridged from Dictionary.com

Seems to me that Motive and Stimuli share pretty similiar definitions as being somthing that incites to action.

Ergo

Behavior:  The actions of a person in response to Motive.



I'm done now.  

As long as we agree on the ultimate application, argueing over definitions is indeed pointless.

[jdagna]

Damn, I had this great long response written when I figured out where the disagreement stems from.

Ron, your essay does not speak in terms of behaviors.  It speaks in terms of "decisions."  There is a vast difference between behavior and decision - since, for example, you could decide to do something, yet be prevented from expressing it as a behavior.

Perhaps I am missing it, but I always see GNS modes talked about in terms of priorities and decisions, never behavior alone.

Decisions include motive.  "Killing baby kobolds beacuse they aren't worth EXP dead" is a decision.  The behavior is merely "Killing baby kobolds."  And the motive is merely "because they aren't worth EXP dead."

Thus, my point seems to be in keeping with the definition of GNS modes, except that I've been using the wrong terminology.  Behavior + Motive = Desicision, and decision is the language of the essay.

Am I missing something here?

[M. J. Young]

Here's my definition for GNS modes.

(you can substitute any mode in the example, I just got tired of people always using N and G)

Observer sees behavors from player Y that cause the observer to think Y is making simulationist decisions.

Observer then  introduces Y and the rest of the group to a game that
promotes simulationist decisions.  

During ths game the Observers see's more behaviors from Y and the rest of the group that  the observer believes are indicative of simulationist decisions. (doesn't have to be, and probably won't be the same behaviors as originally observed).

Thus the observer can conclude that these players are making simulationist decisions, because
1)The observers induction of the Game Motives from the behaviors are consistent with the definition of simulationism.(This requires the observer to understand the definitions of course).

2)Promoting these decisions has increased the behavior.

I'm quoting Damion because again we have the hidden trap.

Observer sees behavors from player Y that cause the observer to think Y is making simulationist decisions.

But what are simulationist decisions? Are they not decisions made to achieve simulationist results? Is not the desire to achieve simulationist results a motive or intent?

People are getting confused because of two problems:
1) The fact that GNS is about motives doesn't mean that it is about deep subconscious causal factors or something of that nature. It can be about motives and not be about all possible motives.
2) The fact that someone tells us why they do something may be behavior, but as behavior it is only relevant as evidence of motive.

This latter. deserves a bit of clarification.

Let us suppose that the player said, "I'm going to kill these kobolds because my character perceives that the only good kobold is a dead kobold, and so it would be true to the nature of this world and my character for him to do so." Now, that statement puts the killing of the kobolds in an entirely different context. Arguably now the character is killing them for simulationist purposes. We would now be justified in reading both the statement and the action as evidence that the player was performing simulationist actions.

But it is only evidence; it is not proof. The player may well have said this because his real motive is that he wants the experience points, but he knows that if he says "let's kill these kobold babies for the experience points" others at the table will object, or his girlfriend will think he's being sadistic. It could be that his real motive for killing the kobolds is that he's building a story about his character in which this will become an important factor, but doesn't think it fits the story to say "my character is going to kill the kobolds babies now because further in the story this act is going to come back to haunt him." The statement made by the player that he is doing this to retain verity in the world is no less behavior than the action of killing the kobolds; and it is equally subject to interpretation, as to whether it actually represents his reasons for the action or not.

But those "reasons" for the action are precisely what GNS is trying to infer from the actions. We might all say, "Oh, yes, Bob is always being true to the world in which play occurs, and that is exactly what his character would do." We might as likely say, "You liar, you're just doing it for the experience points and you know it," only to have him blush sheepishly and say, "well, it was worth a try."

In the end we don't care whether or not Bob's character killed the baby kobolds, and we don't care what Bob said was his reason for doing so. We are only concerned with Bob's actual motives--"Reasons", if you prefer, as long as you don't confuse these with excuses--as a player for having his character take certain actions.

Intent cannot be known from observation; yet it is usually inferred from observation. In law, we distinguish between general and specific intent. Let us consider the difference between Burglary and Breaking & Entering, and throw in something that is no crime at all.

If there is a car hurtling out of control at you on the sidewalk and you dive through the glass window of a storefront to get out of the way, you never formed any intent to enter that store illegally; it was an accident, and you are not guilty of a crime.

If you're cold and have no shelter, so you break that same window in order to get inside and avail yourself of some protection in a storm, you clearly intended to break the window and enter the building. You are guilty of breaking and entering. But you are not guilty of burglary.

Burglary requires a double intent: you must have intended to break into the building and intended to steal something. If you accidentally got locked in the building at night and decided as long as you were there you would steal something, that is still theft, but it's not burglary. If you broke into the building for warmth and did not commit any other crime while in there, that is not burglary either. You must have both intentions.

Now, whatever motivations players have when they make character decisions, some of these motives are gamist, and some are simulationist, and some are narrativist. Some aren't any of these (such as venting your frustrations, or showing off for your girlfriend, or making Bob look bad). But just because we're not interested in the wealth of motivations and intentions that make up the psychological complexity of human decision does not mean that "why did he do that" is either not relevant or not about motivation.

You can't talk about in-mode actions; you can only talk about actions which appear to imply mode motivations. Someone suggested a sort of graphing approach by which you place each action in one section of the graph and see where the majority of them fall, but even this requires that you first infer mode motivations from each of the actions in order to place them on your graph.

I'm with Ralph. My problem isn't that I feel anyone is saying I can't use the ideas of motivations and intentions in my understanding of GNS; my problem is that I can't see any other way for the model to have any meaning at all, unless you are inferring why someone did what they did from what they did. The entire taxonomy seems to be defined by "to win a game--to create a story--to explore a reality", which are themselves motivations which may be implied by actions. But "gamist decisions" are nothing more or less than "decisions which stem from a gamist motivation in play", and I don't see how this can be avoided and still have any meaning left to the concept.

--M. J. Young

[contracycle]

There is a vast difference between behavior and decision - since, for example, you could decide to do something, yet be prevented from expressing it as a behavior.

In which case I would expect to see "I am frustrated" behaviour.

[contracycle]

Is not the desire to achieve simulationist results a motive or intent?

Is that question relevant or useful?

1) The fact that GNS is about motives doesn't mean that it is about deep subconscious causal factors or something of that nature. It can be about motives and not be about all possible motives.

I do not regard the GNS as being about motive.  I regard it as being about behaviour.  I do not care a fig for what motivation produces a given behaviour, in the general case.

2) The fact that someone tells us why they do something may be behavior, but as behavior it is only relevant as evidence of motive.

Correct.  This is substantially safer than baseless speculations as to unknowable motive.

We would now be justified in reading both the statement and the action as evidence that the player was performing simulationist actions.

Tentatively, yes.

The statement made by the player that he is doing this to retain verity in the world is no less behavior than the action of killing the kobolds; and it is equally subject to interpretation, as to whether it actually represents his reasons for the action or not.

Exactly so.  This is why all claims as to motive can only be provisional.

But those "reasons" for the action are precisely what GNS is trying to infer from the actions.

Why is GNS doing this?  I do not see it doing this: it describes the three forms of behaviour in which all such motives are expressed (a slight extension admittedly).  WHY a player is Gamist is irrleveant to me, only THAT they are Gamist.

In the end we don't care whether or not Bob's character killed the baby kobolds, and we don't care what Bob said was his reason for doing so. We are only concerned with Bob's actual motives--"Reasons", if you prefer, as long as you don't confuse these with excuses--as a player for having his character take certain actions.

This is extremely confusing.  You have just demonstrated that Bobs motive is unknowable, or not reliably knowable.  You are now trying to construct a theory on data which you concede is both hard to get and unreliable!  Does not sound like a very useful theory to me.

Furthermore, I feel the legal analogy is misplaced.  If enter a place illegaly, and are discovered by the police, you will likely be arrested REGARDLESS OF YOUR MOTIVE.  The forces of "law and order" will leave determining your "true" motivation at a later date and act, now, on your behaviour alone.

Secondly, not all schools of legal thought have even considered that motive was relevant at all.  The famous "ethical" dilemma used to identify emerging morality in chidlren posits the cotradiction between theft and necessity; if you could steal a drug that would save a friends life, where does your social obligation lie?  The conventional answer is that you should not steal: your motivation in no way alters the criminality of your behaviour.

[damion]

Again, you are saying that something that some people term behavior is called motive. I disagree. Purely semantic. In the end, what is it that we disagree on other than the definition of these terms?

If you include WHY a person acts a certian way in behavior, rather than just HOW they act, then I'll agree the GNS is about behavior.  Works for me.

from contracycle:
This is extremely confusing. You have just demonstrated that Bobs motive is unknowable, or not reliably knowable. You are now trying to construct a theory on data which you concede is both hard to get and unreliable! Does not sound like a very useful theory to me.

There is a large difference between not being able to know something directly and knowing nothing about it. I'm sure there is someone here who can come up with a great quantum phsyics example. In generaly people can infer peoples motives from their decisions.
I thought the legal analogy was great. If the police find you standing over a dead body holding a gun, they are gonna arrest you,  but the crime your charged with depends on motive.(in increasing order of severity)
1)Self-defence--No crime
2)Crime of passion, argument out of control--Manslaughter
3)You followed them and ambushed them, you planned it ---Murder

In all these cases you can look at the persons behavior and get pretty good idea of why they did what they did. The fact that you can do so is important in our legal system and is the basis of alot of interpersonal interaction. 'put yourself in their shoes'.

[jdagna]

There is a vast difference between behavior and decision - since, for example, you could decide to do something, yet be prevented from expressing it as a behavior.

In which case I would expect to see "I am frustrated" behaviour.

Yes, you probably would see such behavior.

But that wasn't my point.

Where in "GNS and Other Matters of Role-Playing Theory" does it say that GNS modes are represented only by behavior?  My point is that the essay speaks in terms of decisions, which are a much deeper concept than behaviors, because they deal with reasoning - what I call a motive, what others call intent, and what the GNS essay itself calls a goal.

[Ron Edwards]

Justin,

"Behavior" and "motive" mean such different things to you and me that we're not going to be able to discuss this stuff in a medium like this one.

GNS is about decisions. To me, that's not necessarily motive and it is behavior. To you, it's necessarily both. Obviously, we are using "motive" to denote different phenomena, rather than disagreeing about a fundamental issue.

If you read all of my posts in this thread carefully, you'll see that no shred of discord needs to arise in our discussions of GNS due to this difference in terms.

Best,
Ron

[Ron Edwards]

Hi Ralph,

Gareth is correct in challenging your use of the courtroom analogy. That analogy properly belongs to my presentation of GNS.

It's surprising to many people, but "motive" is not a central issue in criminal law. The question in law is not examining beyond-reasonable-doubt of motive, but examining beyond-reasonable-doubt of commission. "Motive" gets brought in as a piece of that examination when physical evidence is lacking and it comes down to jury opinion of circumstantial stuff, but it doesn't have to be brought in if physical evidence is more conclusive.

Court methods "prove" nothing, despite the courtroom or Hollywood rhetoric. They provide, at best, rigorous defense of one interpretation of a given set of past events.

My judgment of what GNS modes are operative in a group exactly parallels the courtroom method, and for the same goals - not "true knowledge," but beyond-reasonable-doubt inference.

Best,
Ron

[Valamir]

Hi Ralph,

Gareth is correct in challenging your use of the courtroom analogy.

In the interest of precision, I believe it was MJ's courtroom analogy he was challenging.

That analogy properly belongs to my presentation of GNS.

I am unable to process the context of the this sentence...

Court methods "prove" nothing, despite the courtroom or Hollywood rhetoric. They provide, at best, rigorous defense of one interpretation of a given set of past events.

My judgment of what GNS modes are operative in a group exactly parallels the courtroom method, and for the same goals - not "true knowledge," but beyond-reasonable-doubt inference.

Best,
Ron

That's pretty much how I view it.  I think I even explained it that way in a response to Marco somewhere around here.

[contracycle]

There is a large difference between not being able to know something directly and knowing nothing about it. I'm sure there is someone here who can come up with a great quantum phsyics example. In generaly people can infer peoples motives from their decisions.

Oh indeed.  If motive was completely opaque to us we would never be able to have consensual sex, and would not be here.  We can infer motive motive from what people do - based on the collection of behavioural evidence.  In the legal analogy you provide, of the differeing degrees of sanction, that evidence is partly interrogation - does the perps version of events stack up?  If they say it was a crime of passion, can we find the third in the menages a trios to confirm the scenario?

Similarly with the kobolds scenario, we pretty much have a signed confession.  When asked what their motive was they said they did it for the XP - there is no real reason to doubt their word.  If they are later seen hunting for random encounters because, they say, they are only 10xp from levelling, this would be supporting evidence that it would be perverse to deny.

Speech and self description are also behavoirs.  There is nothing preventing us from drawing conclusions about their motivations FROM THEIR BEHAVIOUR, including self expressive behaviour - which arguably all of RPG is.  In fact we should do so, because it will enlighten our understanding of the behaviour.  None of that prevents us chopping the behaviour up into categories if there seems to be reason to do so for descriptive and discursive purposes.  If as claimed, the observed categories are actually distinct, and really exist, in that they are discrete, then we don't need to know what the actual motivations of any given player are to still make games.  We can still think about how to do gamism well regardless of the individuals actual personal motivation.

What would be a Bad Thing, IMO, would be to fancy we can just figure out how everyone thinks.  Sure, we could sit down and pull a bunch of "motivations" out of heads; but all these will be is projections of our own motivations.  We will try to imagine how other people think... how reliable is that?  And then we can procede to categorise people based on how WE think THEY think.  How can we do that, given the differences in everyones personal life experience?  I say thats impossible.

But I think we get a lot of milage out of looking at behaviour.  For one thing, we dont have to speak the same language - RP in French is still G or N or S, (or so goes the claim).  And RPG by 15 year olds and 45 year olds is still G, or N, or S despite the fact that there will probably be very real divergence between their actual motivations.  We could still make a reasonable game recommendation.