This is a crucial issue, so let's revisit the original definition of Agency as a refresher.
Player Agency (n.): “the feeling of empowerment that comes from being able to take actions in the [virtual] world whose effects relate to the player’s intention” -Mateas, 2001
So, everything you do as a Dungeon Master is valid for empowered except for one very special case.
Are you taking an action that invalidates the players expectation of the result of their action?
That is what each of the many techniques in the Quantum Ogre series revolve around.
- Explaining the consequences of action
- Letting the players know about the effects of their choices
- Reminding the player of what they might have forgotten, so that their expectations are rational
- Not shutting them down by saying no, but instead letting them know what to do to achieve their intention
- Giving them warning of dangers ahead of time
Know that the answer to that question is not a simple, objective, clear-cut answer. It is one of emotion. Does the player feel invalidated? Do they feel railroaded? This is why so many of the solutions provided involve not hiding information, and reducing that information gap necessary for illusionism, instead of taking advantage of it.
Let go of any preconceived notions about how things should play out.
There is no flaw in a general course set for the campaign, but if the sequence of events isn't in doubt, then players aren't necessary. The organic creation is going to be more innately satisfying to the players, and provide greater opportunity to you.
Look, here's a secret. You're the Dungeon Master. You can make anything happen, anywhere, that isn't in the vicinity of the players, however you want. You can even achieve your outcome without invalidating their choices, depending on how you approach it.
Deciding certain things have to happen a certain way in play and then using blunt, clumsy, techniques to make it happen is just poor technique and style.
Figure out how to make what happens awesome, instead of how to make an awesome thing happen.
This is the correct stance of Flexible Progression.
Vet or clear the basic premise before the game begins.
"You guys are going to be travelling around with an ancient good lich, who acts as your mentor or teacher, does that sound cool?"
You know, I'm running a story focused campaign right now. I said, "can we do Spelljammer?" Everyone said yes. Then I ran the campaign as normal, starting in a village, with a festival. They kept expecting me to shove the kicker down their throat, which caused them to second guess themselves the whole time. When they finally found the ship, they were motivated both in-game and out to take it, of their own free will.
This is addressing the question directly. By doing so, they explicitly make a choice who's effects are respected.
Building up to things in the weeks before they occur, make the players feel as if they are discovering what is happening instead of something being forced on them. Have them hear a rumor about a dead wizard that still lives given up all hope at being reunited with their love three sessions before this one. Then the next session, have people talk about a ghost ship that flies through the sky. Finally, have the adventure before talk about a falling star nearby that no one has been able to find.
By the time they discover the ship in play, they are armed with the information that makes them want to react in the way that allows the campaign to travel forward in the way you have conceived.
This transforms railroading and illusionism into player driven discovery. By foreshadowing, you set expectations beforehand. This by the structure of play aligns player intent with the result of their choice.
Pre-plan and allow for the other option to be as interesting as the one initially presented.
Assume and plan for agency. So what if the players reject your expect action out of hand? The option that they pick should be designed so that it is interesting no matter their choice. Reject the good lich? Perhaps he becomes the nemesis, or vanishes leaving them the ship only to return later. Whatever they should pick, interesting play should flow from that.
The only way that works, is if you spend just a few minutes ahead of time thinking about possible options and what can be interesting depending on which is selected.
This is the effective use of Multiple Outs and Flexible Progression. This is a technique ripe for abuse, and requires thought to insure that the results of the various options don't disregard the intent of the player.
Do not depend on a singular option to move the game forward.
Anything that can stop play should be optional. Full stop.
Puzzles, riddles, clues, mysteries -- you must consider that what seems simple to you may be very complicated to your players. It is very difficult to accurately communicate what is in your head to players of your games.
If it doesn't stop play, but simply changes the outcome, then this option doesn't apply.
The original series on Player Agency "The Quantum Ogre" is there.
Part 1 of looking at Illusionism, "On the Tacit Acceptance of Play" is there.
Part 2 of looking at Illusionism, "On is Ed Greenwood the Devil" is there.
Part 3 of looking at Illusionism, "On Theory Defined: Illusionism" is there.
You're reading part 4 of the series examining illusionism.