Look at the different ways we have chosen to handle it. We try to fix every event, not merely what is in each room or behind each door, but pre-determining how the monsters or the devices within will react once triggered. Alternately, we try to fabricate a story that will establish an arc of purpose, allowing some latitude in player behaviour but ultimately intended to propel the party from event to event. This failing, we offer the players agency - but in removing our own direction, the players find themselves without a direction of their own, so the campaign withers and dies.
The trick is ultimately to hybridize all these approaches into one:
- There are events, but if at all possible we struggle to increase the profligacy, the free rein of these events.
- There is a 'story,' but we struggle to ensure that is it a story that has been created through events, without predetermination.
- We give the players agency, but we aid their struggle through creating events and story.
I'm operating under an epiphany here, inspired by a comment I've just received, so I want to go very slowly. Let's handle them one at a time.
We should think of these events as predicaments rather than blueprints. We take a simple circumstance, such as a party moving along an empty road in the mountains. We introduce the predicament - the party finds a sphinx, cut to pieces, stabbed by a giant-sized spear that remains in the sphinx's body. The sphinx is near death, unconscious, itself immense in size and terrifying to the party. All the better if the party is not high enough level to take on a sphinx.
Okay, remove the backstory for a moment. Empathically place yourself in the predicament; presume there will never be an explanation for this. Eliminate from yourself a prediction of the party's response. Establish in your mind that no matter what the party does, it will be the right thing. Because there is no story, because the life or the death of the sphinx serves no specific purpose for the character's experience, whatever the characters do changes nothing about the DM's world.
The is the foundation of the predicament. It is a difficult situation, perhaps unpleasantly difficult. The lack of information perplexes. The beast itself, whatever has nearly killed the beast, the size of the spear, the presence of the sphinx in the open, helpless, where it can be found, all these things scream danger to the party. It is so weird, so strange, it defies immediate explanation. Anything that is so clearly not understood produces a wide range of emotion.
There is no need for an adventure here. There is no guarantee the party will ever receive an explanation. IF they want an explanation, they will have to risk exposing themselves. That is the beauty of the predicament. There is no knowledge without investment.
Being the DM, it is up to you to conceive of predicaments such as the one I've described. Literature and film are full of them. Steal if you must. But first, you will have to strip away everything else about the story or the character in order to understand the fundamental nature of the predicament. The plans for the destruction of the empire are missing. We may never find them. Nazis have knowledge about the whereabouts of the Ark of the Covenant. No plan of action exists. A magic ring is in your possession. We have no idea what it is or what it does. We're on our own.
Let's look at the next piece. Think of the story as a set of specifications. Wash out all the motivations, all the narrative, all the right, wrong, intention, meaning and interpretation for the explanation of the sphinx's appearance on the road and concentrate ONLY upon its actually being there. What series of events in the past brought its body to this place, at this time, in this condition? Eliminate the motive; what this creature thought or what were the thoughts of the other creatures involved is a matter of total immateriality. Right now, in this moment, all that is wanted is a series of plausible instances that produced the scene as the party discovers it.
This chain of events we keep to ourselves - until the party takes steps to deconstruct the specifications. We must understand this chain of events a perfectly as possible - so that as we relate clues about the scene, the party will perceive continuity in our description. Any specification that is related inaccurately or in conflict with some other specification will ring bells, immediately. So whenever possible, keep your specifications simple.
As time passes, as the party investigates, as more predicaments arise in relation to the party's actions, more of the story will come to light and the party will have the opportunity to reconcile what they've seen in the past with the campaign in the party's present.
None of this requires a predetermined future! The circumstances that nearly killed the sphinx will suggest actions for all those associated with the event that can occur simultaneously with the party's choice of action. If the party chooses not to investigate or involve themselves in the predicament, the specifications behind the predicament may well never come to light. Alternately, the party may encounter another, connected predicament, another mile down the road. As DMs, we are free to create these predicaments as we see fit - but as mysteries, and ultimately as mysteries for which no future is absolutely established. The past, yes, this we know. The present, yes, we can produce a present for other beings as the party's present progresses. But the future - ah, this we should simply ignore, letting things play out as they will, should the party decide to be involved.
If the party refuses, well, we can simply toss the whole matter aside, as all we have actually created to this point is the predicament, yes? As DMs, we've made no long term plans. As DMs, we've made no investments. We are here to wait for the players to invest. We don't until they do.
WHICH MEANS, create the predicament alone. Rely upon yourself to create the specifications for the predicament when and if the party invests; until that time, don't bother! Save your creative juices for the next predicament. Cross the bridge of creating a specification when you need one.
I regularly create predicaments without specifications. Now and then, I do find myself painted into a corner - but in such cases, the answer is really very simple. If it is impossible to create a set of specifications for a given predicament (and this happens very, very rarely, as I have a fertile and quick imagination), then there is NO explanation. Too bad. The world is very strange sometimes. Not every mystery has an answer.
Let us look at agency then. Agency is the player's choice to pursue the predicament. That is all.
We therefore do not wait for the players to 'invent' an adventure. They find themselves on one once they've chosen to pursue a given predicament. The specifications they discover produce a happenstance where they encounter others, learn things, discover motives (which we can delicately add when the party begins to inquire into the reasons behind what they've seen) and then see how it all plays out according to what the players do, how well they succeed, who they choose to speak to, etcetera, etcetera. As they dig deeper, I create more things for them to find, from session to session adding to the backstory that has occurred in order to give them things to discover in the future about what is happening. When I wrote that I had revealed things to the party recently, this is what I was describing - specifications that remained unknown to the party for seven years.
I play the deep, long game.
My players pick up on hints, make wrong guesses, go storming off . . . then I create a predicament that pulls them back in, drawing them into further investigations and revelations. That is how my campaign unfolds - one predicament after another.