There are two ways to look at this.
From the ingame causation point of view: even if the vessel is there, the PC may not notice it in the hubbub of the situation (decapitated mage, assassin trying to flee, two wizards trying to stop said assassin fleeing and one of those two also trying to stop the other from killing the assassin).
But that just establishes the possibility of a vessel not being noticed.
The more important perspective, for me, is the real-world, pacing-the-game-at-the-table perspective. I GM along the lines of "say 'yes' or roll the dice". Here is how Luke Crane describes the idea (BW Gold, p 72):
Unless there is something at stake in the story you have created, don’t bother with the dice. Keep moving, keep describing, keep roleplaying. But as soon as a character wants something that he doesn’t have, needs to know something he doesn’t know, covets something that someone else has, roll the dice.
In the episode of play I described, the character wanted a vessel to catch the blood so he could take it back to his dark naga master. Something was at stake in the story we have created; the character wanted something (a vessel) that he didn't have (his equipment list was clothes, shoes and a stick carved into a staff); and so the dice had to be rolled.
I'm not 100% sure who the "you" refers to - someone GMing for you and your friends? Or any GM in general?
I look at it this way: if the group isn't dissolving, then more adventures are going to be played. If that particular campaign isn't done, then some of those adventures are going to involve these PCs. That means that something is going to happen that frames those PCs into their next adventure.
Such framing can be done more or less deftly, but it has to be done. A meteoroid falling through the sky doesn't seem particularly worse than any other way of doing it.
Where does "the overall thrust of the campaign" come from?
If it comes from the players, then the implicit (or explicit) agreement runs in exactly the opposite direction from what you seem to be suggesting: the GM agrees to frame the PCs into situations that engage the thrust that the players have established. (This is what [MENTION=99817]chaochou[/MENTION] mentioned in the final paragraph of his post upthread.)
So it is not the case that in every campaign the GM either "pushes or pulls the PCs in a particular direction" or "sets the NPC plotting in motion". In my games, the main way that I work out NPC plotting is either (i) in response to the resolution of player action declarations for their PCs, or (ii) in the context of framing the PCs into some sort of conflict. In the latter case, it's not sitting there in the background as something the players might (via their PCs) take a peek at should they be so inclinde. It's a core part of the action.
I still think Paul Czege articulates this best:
I frame the [player] character into the middle of conflicts I think will push and pull in ways that are interesting to me and to the player. I keep NPC personalities somewhat unfixed in my mind, allowing me to retroactively justify their behaviors in support of this.
(Why was there a dark naga in the campaign at all? Because one of the players wrote it into a PC's backstory. Why did this particular PC end up ensnared by the dark naga? Because the player chose to create a snake-handling healer who is obsessed by all things snake-y, who therefore went into the cave partly in hope of meeting the naga. Etc)
