The exchange between me and Ron about stakes setting has been on my mind, and perhaps because of that I have been extra attentive to the presence or absence of setting explicit intentions and stakes in my current play. I've been playing a game of the Pool with two players (George, from my Legendary Lives game, and his brother Ted, who has never played an RPG before). The setting is essentially Napoleonic-era military action and intrigue, except with magic, inspired more by quasi-historical sword and sorcery literature than actual history. We wrapped up last night and I noted the following:
At no point did the process of conflict resolution pause for us to explicitly set intentions or to bring up possible stakes of success or failure. Rather, it was obvious to us* both (a) what would trigger a roll of the dice based on the ongoing activity of the characters and (b) what failure or success would look like, again, based on what had just been happening. I wish we had recorded the session because it’s easier to show than explain, but what I mean is that when two characters were having a discussion, it was clear just from what they were saying to each other when we had reached a point where a roll needed to happen, and once the dice were rolled it was clear what a given success or failure could mean. There was no need at any point immediately before, during, or after the die-rolling to clarify specific “stakes” one way or the other. When the traitor sorcerer’s dragon bore down on George's PC and she responded by drawing her sword and charging it, it was clear that the failed roll meant she was dead — even though the possibility of death hadn’t explicitly been brought up before that. It was a powerful, unwelcome and unwanted (to use Vincent’s terms) moment and it would have been less powerful had we paused to spell out the potential outcomes of success or failure before rolling.
An emergent property, that we quickly came to appreciate, was that if I as the GM got to narrate a success, I would limit it to the barest minimum of what that success would mean without undermining the fact that it was indeed a success. This in turn gave monologues of vistory (MOV) real value and led to lots of tense decisions about whether to choose an MOV or an additional die for the pool. Had we set stakes beforehand, this property of the game would have been short circuited. (I think Burning Wheel actuall short circuits in a similar way: some of the more interesting features of its resolution system fall away if you use the "setting explicit intent and stakes" procedure that was screwed onto it during the series of revisions from Classic to Revised).
Now, there were times when the players would take an action, and we would have to pause to orient things or clarify things. For example, at one point George had his character punch someone, and this action was confusing to me so I did stop to ask what the character's intent was — “Oh, she's trying to make a distraction”; that clarification allowed the scene to continue, and though eventually we did roll dice, the clarification of intent was not directly related to or triggered by and nor did it trigger the conflict roll.
And Ted (this was his first RPG) would often state his intent for some longer term goal, and we’d have to dial things back and figure out what the first step would look like and start playing that. So, in those cases, we did have to stop and verbalize the idea of the overall intent, but what we were playing was the immediate action the character was taking. These clarifications were necessary so that we had a clear picture of what was going on, and that clear picture was necessary for us to use the conflict resolution system at all. But we were not using the conflict resolution system to pull us out of the murk or to decide between two predetermined paths.
*To expand on this, sometimes what should trigger a roll was obvious just to one of us at first, and they would point it out and the other two of us would say, “oh yeah, of course”.