• D&D: Bluff, Diplomacy, Intimidate; Sense Motive
  • Fate: Deceit, Rapport, Intimidate; Empathy
  • FSuns: Knavery, Charm, Impress; Empathy
  • NWoD: Subterfuge, Persuasion, Intimidation; Empathy

Somewhere along the line, it became accepted fact that RPGs with social skills break them down in very similar ways: you have a skill to lie, a skill to persuade without lying, a skill to be scary, and a skill to tell when you’re being manipulated. The occasional game mixes it up a little, either by combining a couple of the skills or by adding one or two others for specialized uses (e.g., NWoD’s Socialize for being the life of the party). But, by and large, it’s common to arrange skills in a way that you will use a different dice pool for social encounters depending on whether you’re lying, being honest, or being scary.

Social combats are hard to model in general. Even in games like Smallville and Technoir where social combat is functionally identical to physical combat, the cadence of the fight is hard to pick up at the table. We’re used to modeling trying to kill people with whatever pacing mechanism any particular game requires, because it’s never going to get in the way of us standing up and just roleplaying out the fight. But for social conflict, it’s natural for most gamers (particularly those that grew up on systems with at most one die roll being involved in any argument) to just roleplay out the scene in character. Adding a social conflict mechanic with a back and forth similar to combat requires you to break up your roleplay in a way that doesn’t feel natural, only roleplaying out a small portion of your speech before pausing to roll and then giving the other side a chance to rebut when you weren’t even done talking. Or you completely abstract the social combat (as the examples in a lot of indie games seem to suggest) by merely stating your intent in the scene and barely saying a word in character. That latter method is a really hard sell to gamers used to doing everything in character that’s at all practical at the table.

But, getting back to the initial point, I believe that part of the difficulty in making a really workable social conflict system is over-reliance on the standard four-skill model of social skills.

From a purely simulation perspective, these skills are weird. I finished reading The Big Con last week, and one of the things it hit home is that there’s not a bright line between any method of persuasion. There may be poor liars that are good at persuading people with the truth (though politics makes that dubious), but there are probably not many excellent liars that suddenly become unpersuasive if they’re telling the truth. Sure, you can buy multiple social skills in most systems, but the game system doesn’t often reward you for avoiding specialization. So you get situations like my paladin in a previous game: as soon as my attempts to convince NPCs drifted from pure fact to concealing information or being evasive, the GM was suddenly justified in asking me to roll my +5 Bluff skill instead of my +17 Diplomacy skill. Despite the fact that, in my mind, it was all part of the social patter required of being the party’s Face, the system didn’t see it that way. If we’d needed the NPC to buy a lie, I should have deferred to the party rogue (which, itself, would have been suspicious).

The inclusion of a social detection skill (empathy/sense motive) creates further complications to roleplay. The active social character is often at a severe disadvantage against a target with a high rank in this skill, because even the simplest and most believable lies can be detected in a way that borders on the magical. Even experienced police detectives often have to have ideal circumstances (such as an interrogation room and facts to catch the target off guard) to reliably detect lies, but RPG characters specialized in empathy can do it in any scenario (frequently against an opponent in the seat of his power). The converse is also true: a character that doesn’t specialize in this skill doesn’t allow its player to make up his or her own mind about a lie, but can be told that even the most bald-faced lie sounds perfectly correct (not that players regularly actually proceed with that assumption when they know they blew their sense motive check or are up against a character that they expect outclasses them in bluff).

Whether or not you believe that the divide in skill is a realistic simulation, the fact remains that it’s part of the problem in making social conflict similar to physical combat. While games typically silo physical combat ability in a similarly weird way (how many martial arts experts are useless with a weapon?), the typical combat offers the ability to contribute with any relevant skill (except in situations where firearms would be too loud or the PCs were disarmed before entering). The martial artist, the fencer, and the pistoleer are each contributing in their own way. But social conflicts are frequently constructed in a way that makes it way more obvious that certain types of skill can’t contribute. We’re trying to befriend this guy, so the party heavy with intimidate can’t help. We’re not actually lying (or we’d blow everything if we were caught in a lie), so the rogue is cooling his heels. Not only do you have the roleplaying need to just let one guy do the talking for simplicity’s sake, there’s little game benefit in the other players trying to contribute anyway.

So what’s the solution?

The easiest answer is to just have one really broad social skill, but that’s probably not something many game engines want to do (if you thought Diplomancers were bad now…). It is worth noting that the standard methods for adding more skills to what a game is focused on creates a paradox of incentivizing the “wrong” builds. For example, it’s way easier to get good at social conflicts in D&D than in combat and it’s way easier to become great at combat in any flavor of WoD than to become adept at intrigue.

The real answer probably starts with the intrigue tactics available in the Song of Ice and Fire RPG (which itself uses pretty close to the standard four skills, except with intimidate as a subset of persuade, so it isn’t a full solution). Specifically, cutting up social skills not by lie/truth/fear/detect but by whatever the verbs are in your social conflict system. For example, if “convince” is a tactic in your system, perhaps that’s the skill and it doesn’t matter whether you’re using lies, facts, or intimidation to convince the target. The trick would be creating a conflict system where any particular social encounter could map victory conditions to more than one such verb in a natural way. And I haven’t quite gotten to that particular holy grail yet.

So am I missing implementations with the standard four skills where they’re a natural fit to the social conflict mechanic? Can anyone come up with a list of verbs that would be applicable to a broad range of social conflicts?