Special roles and powers: what are they good for?

What function do special powers or roles (SP&R) serve in game designs? They’re fairly commonly used, examples including Citadels characters, To Court the King cards, Puerto Rico buildings, genes in Ursuppe etc, but what common abstract design problem(s) do special powers and roles address as a game mechanism, and what problems do they also encourage?

I came to this question when I complained to a friend that one of my designs simply lacked arc; there was no change in tempo, narrative, dramatic structure or even the primary problem space during the course of the game and I thought I might want to change that. His immediate answer: “Oh, you need special powers and roles!” This caught me by surprise. Why do special powers and roles necessarily generate arc? Certainly there are plenty of games with developmental, dramatic, narrative or whatever type of arc which don’t have special powers and roles. But every special powers and role game that I’ve thought of certainly doesn’t lack for arc.

SP&R seem to do a few things:

1) SP&R add a developmental complexity curve, err, arc. The game starts either without SP&R or with the value definitions for the SP&R largely undefined yet. During the course of the game SP&R are acquired and their relational values manipulated by the players, creating an increasingly complex problem space. The game starts with a simple value space which becomes increasingly complex during the course of the game: an arc.

2) SP&R encourage player investment. Having invested in SP&R, players are to that extent committed to the matching strategic path. Presumably the paths are (somewhat) different, thus supporting developmental complexity. The investment can be both analytic and emotional, and the development of those investments arcross players and their associated values can provide arc.

3) SP&R can generate tension in games which require SP&R combinatorials and where specific SP&R are of limited supply. Having committed to one selection, the challenge is to also acquire the desired combination of other SP&R, possibly within a timing constraint. As each step is made, or failed and a contingency taken instead there are tension arcs, one arc for the entire strategy, and smaller arcs for each SP&R step.

4) Designs which support combinatorial SP&R and have a large viable permutation space will have a high(er) variance in game patterns from session to session. While not really in-game arc, the session variance may be viewed as a subjective familiarity arc.

5) Combinatorial or time-phased SP&R encourage emergent discovery arcs. The discovery process of learning which SP&R combine well, in what order, and how to best exploit the prime intersections. The ar in this case does not occur within the game, but within the player’s understanding of the game wboth with a single session and across sessions.

…more to be added as I think of them.

The problems caused by abuse of SP&R seem to all fall in the same camps of chaos, complexity, and unbalanced intersections.