If you're examining a theme of some kind with a character, then I think there needs to be risk involved. I think this came up a hundred or so pages ago, but I think it's really relevant. The risk of failure needs to be real.....if your theme is one of redemption, then that's something that the character must be able to either succeed or fail at. It shouldn't be as simple as a player deciding at character generation "My character is trying to atone for his horrible past" and then simply plays the character as they would just about any other, engaging in the group's goals, and occasionally monologuing about their sordid past. If there is no struggle for the....if it's all simply up to the player, then it's not much of an examination of redemption.
Now, this doesn't require mechanics; it can be done simply through play, with the player deciding things for the character. But in the absence of mechanics that can help here, then things need to be framed by the GM so that there are consequences that matter at stake. There needs to be some kind of drawback, whichever way the player decides to go. There needs to be a challenge.
I think it's only through that kind of challenge where we get to actually examine a theme instead of simply using it as some kind of character trait to help us roleplay.
Anyone can play a character with a haunted past and use it in their portrayal of the character. But if you want to actually examine trauma and its effects on someone through play, if you want to see if a character can overcome their past and gain some sense of peace or redemption, then you need to go beyond just playing the role. That theme needs to be a focus of play (shared as play may need to be among the group).
I think, fundamentally, if it is a dramatic narrative type of experience, then it must be explorative in some way. We must learn something, about humanity. This means there will be change, some 'impulse' must exist which exerts some pressure on the character, or on the group. Normally this will lead to an attempt to solve some sort of problem, to relieve that pressure, or accommodate to whatever forces are at work. This may have a price. Certainly it will engage the character, and normally be challenging.
In thinking about
theme in RPGing, I tend to start with comics (especially X-Men!), and movies that are comparable. If I run a RPG session that is (in narrative terms) half-way as compelling as an episode of Claremont X-Men I count that a success. If it even has an echo of a film like Hero or Ashes of Time I count that a triumph!
I like
@hawkeyefan's contrast between characterisation (
I have a haunted past) and actually engaging with a theme
(What will I have to do to find redemption?). I think RPGing can struggle with this, especially if the model for the "adventure" starts with White Plume Mountain or the like. And the problem isn't solved by making the motivation for entering White Plume Mountain being to rescue a friend rather than recovering the stolen treasures, if the action is still just moving from room to room and dealing with the orcs or giant scorpions or whatever inhabitants are detailed in the dungeon key.
I don't know if it's true that there has to be a chance to fail. I think on that I tend to agree with
@AbdulAlhazred. In Star Wars, for instance, I don't think we every really worry that Luke will be killed by Darth Vader, and what makes the end of Empire Strikes Back so ripe for another episode is precisely that it leaves Luke's arc unresolved. But the idea of "cost" or "loss" or "risk" seems important (and failing to realise a character's goal as a protagonist might be a special case of that). Luke loses his family; his landspeeder; his homeworld; Obi Wan; his father; his "innocence". He finds things, too, that he didn't expect - his force powers, Yoda, his father, his sister. These aren't all things that he wanted to find, either- learning that Darth Vader is his father is itself a cost at the same time as it is a source of growth and self-realisation.
How do we introduce this sort of material into a RPG? Does the player conjecture and the GM decide (say as part of consequence narration)? Does the GM decide and then the player learns via an action that obliges the GM to reveal that part of his/her notes? What is fair game to put at stake? Of RPGs I know I think Burning Wheel brings the most structure to these questions, because of its various moving parts - Beliefs, Relationships, Circles etc. But it still requires subtle judgement. Eg if I've paid build points for my PC to have a relationship with his mother, how far is the GM allowed to go in turning that against my PC's interests (say by calling for a Duel of Wits as my character's mother tries to persuade my character not to leave her again) or elaborating relevant backstory (say by introducing fiction that strongly implies that the demon-summoner Evard was my character's mother's father)?
Doing nothing of this sort means there is no struggle of the sort hawkeyefan points to - I'm just playing a guy who cares for his mum. But too much the other way and now it's the GM rather than me who's playing my character and writing his story! To move from BW to PbtA-type concepts, I think there's a lot to be said for sticking to "soft" moves in framing, and even initial consequences, and only following through with hard moves if the player keeps pushing with his/her character and fails a check - now the character has invited the foreshadowed "hard" move (it's not just that Darth is trying to trick you by declaring he's your father -
you've searched your feelings and you know it to be true!).
I think what can be tricky, whatever game we might be talking about, is that there's going to be a call to action of some sort, and a general need to work as a group. These things can be challenges to examining theme through an individual character.
<snip>
The fact is that the group element is already a strong constraint on what you can do with a character thematically. And this is one that most games struggle with, to some extent; RPGs are (mostly) a group activity, after all.
This is where I find Claremont X-Men helpful as a model - individual characters have their themes and arcs even though much of the activity involves the group.