GMMichael
Guide of Modos
Or, (and I know this wasn't one of your options) you let the players in on it, and create a sort of mini-game: if players know that most NPCs have only one prepared line of dialogue, then they know that they've found a story-relevant NPC when they talk to one that obviously has more than one line prepared.So, what, either you pre-plan the entire backstory of every single minor NPC the players may ever decide to strike up a conversation with, or you have them all clam up the moment they've said their one arrow-to-the-knee line and refuse to speak to them again? The former seems like a great deal of work that will go to waste, and the latter a greater breach of veresimilitude than a minor historical inconsistency.
Having something prepared is a boon for GMs who struggle when they run off-script. Because, well, meeting Steve the Farmer or Conan the Warrior is an equal breach of verisimilitude.
Sure, sure. It's an opportunity. And it's even something that a lot of players won't notice. But if you have one or more players who do their homework (another GM, perhaps), then inconsistencies are highly disappointing. (Insert link to ENWorld's "what's fun" article).When the five players sitting around the table cannot all agree on what they had for supper the previous session... the kind of thing you describe is not anything desrving of JJA. its actually a fun little thing (or maybe a *useful* little thing.)
Not exactly the planning that was causing problems. Part of it is the quantity of the planning. Say a Dragonlance reader decides to DM a Dragonlance campaign. Inconsistencies will come up continuously if one of the players is a more dedicated fan than the DM. The other part is the lack of planning inherent in improv. A GM who does his homework doesn't have to worry about the dialogue of major NPCs because part of that homework is knowing their stories and how those stories tie together. Unscripted NPCs never figured into the story, so it's easy to grind gears when you drop them in.One, it is the planning that is causing the inconsistency - ie you already have all this pre-planned "lore of the game" that the GM is not on top of; and you also have this player-authored backstory that the GM is not on top of.
So what your example shows is that if the GM commits to the (ingame) truth of a whole lot of stuff that s/he is not across, then s/he might carelessly contradict it. Which seems obvious but irrelevant to the merits of improvisation.
Obviously, if there is no story, you don't have to worry about inconsistency. But that brings up other problems.
It seems that as your game runs on, "none of it was prepared in advance" becomes "some of it has been prepared in advance." I suspect that as the game history stretches into two or three pages, your unscripted NPCs will become de facto scripted NPCs - as you create the script on the fly by referencing your notes.In my Traveller game, before the fourth session I made a list of all the backstory that had been established over the prior three sessions
. . . There actually was a bit more backstory than that - individual NPC backgrounds were noted with the NPC statblocks - but all-in-all that's not too much to keep track of. Contradiction is not that likely.