A couple of weeks ago, I talked about how we stripped all rolling out of social interactions and made it purely a case of GM fiat. As you might imagine, that experiment didn’t work out very well at all. Not only did that remove the classic cues for players to identify how well they were doing in a conversation, but it also made the system fundamentally uninteresting to interact with or design for.
I’m not a big believer in designing a system which basically just tells the GM to make it up him- or her-self. Design is not “permission to be creative,” design needs to have actual substance.
But the excellent thing about taking that step and making that experiment is that it got us to the heart of how we wanted social interactions to work in Forthright. By having nothing, we were starting from scratch – not from a baseline developed by another game (D&D or Burning Wheel, for instance). And what we found is that we still wanted rolling to be involved, but not as a hard mechanic where we were generating (or subtracting) points for an overall total.
Involving points (or any kind of currency) in social interaction felt like a too-hard mechanic, something that made conversation awkward and not as free-flowing as we wanted it to be. Instead, we began looking at the social roll as a soft roleplaying cue. Fans of the OSR might recognize these as reaction rolls, but performed periodically throughout the conversation in order to gauge how the conversation is flowing.
In essence, the Guide is still sovereign and roleplaying the NPCs just as the players are sovereign and roleplaying their PCs. If too much is requested, the answer is still no – you can’t roleplay the king into giving you his kingdom, for instance (this is our go-to “is it broken?” test). However, whenever a listener is taken aback, surprised, befuddled or would otherwise be inclined to say “no” to a portion of the conversation, the speaker makes a Presence Check to see how his or her words are taken.
Rolling low, of course, brings rejection and perhaps even a cooling in the relationship – the listener reacts badly and is off-put. Rolling high, on the other hand, suggests that the listener is more accepting, and maybe even warms up to the relationship. However, in neither case does the Presence Check remotely act like mind control (as Influence, in Beta 2, kind of felt like it did) because there is no social obligation implied.
So, for instance, the king might not be willing to give up his kingdom, but he might respect the audacity it took to ask (if the roll was spectacular). Alternately, he might be so furious that you’re cast from his presence immediately by armed guards. Rolling somewhere in the middle might leave him thinking you’re not operating on all cylinders.
This allows free-flow and variety, and allows NPCs to impact PCs, as well, without risking player sovereignty. This has actually been a fairly important step for us in the development of Beta 3, and I think this better reflects at-the-table behavior of GMs and Players. Gaining this perspective was well worth the failure of the idea.
Thanks for reading! As always, we welcome comments 🙂