Leading A Horse

I listen to a lot of RPG podcasts (take a look at my Roleplaying & Resources page for my favorites), and there's one topic that every single one of them, without fail, seems to come to eventually: social mechanics.

Normally these topics are brought up by listeners emailing in, asking the podcasters for their advice on how to handle social conflicts in a game. There are legitimate concerns here about player agency, verisimilitude, and the extent to which the game should encroach upon the roleplaying or vice versa. it's not an easy thing to answer.

I've been on both sides of this debate. For years, I was against all social skills in games, all for roleplaying it out. Then I switched, and quite suddenly found myself singing the praises of social mechanics. A few years later, I took a step back towards my previous beliefs, but didn't quite go all the way. That's where I am now: for them, but not as they are typically used.

I've found a solution that works for me, and I hope it works for you too. But before we get too far into that, let's take a look at what people tend to say on this topic.


The most common argument I've heard for social mechanics in games is one about gaming versus roleplaying. It goes something like this: if a player is allowed to roll a die rather than stand up and demonstrate how they swing a sword, why shouldn't a player be allowed to roll a die instead of trying to persuade the king?

Yes, there are a lot of things we allow the dice to take care of in an RPG, but there is a fundamental divide in gaming that this argument completely misses: that between decision and outcome. While we do allow the dice to determine outcomes, we (ideally) should never allow them to determine our decisions. We decide which way to go in the dungeon, what to do in town, whether or not to help this NPC, etc.

Social conflict should take more than one decision. Just saying "I want to persuade him to my side" or "I lie" isn't good enough. People have the back-and-forth of dialogue so deeply ingrained (it is, after all, one of the first things we learn as toddlers) that we don't think about how many decisions go into an ordinary conversation. What do you say? When? How do you say it? These are all important, and can drastically change the outcome. Rather than go through each one of those decisions, many systems/groups just say "roleplay it out" because we can make these decisions and react to them without conscious thought.*

That leads into the next argument: what if a not-so-social player wants to play the party face? Should a player's social awkwardness make it impossible for them to play their character effectively, despite what their character sheet says? Especially when an invalid can still play a fighter, and a child can still play an ancient wizard?

This argument actually holds water, for the most part. While I do think there are some things about a character which will simply never be divorced from their player (it's the player's mind coming up with everything, after all), social mechanics shouldn't be afforded some special status as the one form of conflict resolution in that set. There are people who say it's unfair because, for instance, they are bad at tactics but cannot make some kind of roll to help them with that aspect, so why should others get that help? In those cases, I would say that those people should get help too, and point to things like World of Darkness' "Common Sense" merit or the "Spout Lore" move from Dungeon World as possible ways of making the mechanic work. The answer to that problem isn't to punish unsociable players, but to help players having problems in other areas.


Just like with the pro-mechanics side, I see two main arguments here. The first one is simple, and easy to dispel: the people who say that it's a roleplaying game and therefore should involve roleplaying whenever possible. I agree, it should. But while this argument stresses the part of roleplaying which involves acting as one's character, it dismisses the part of it which involves taking on a role that could be far removed from one's own personality and/or abilities. If we applied this outlook to other aspects of the game, GMs might start expecting their players to all have a working knowledge of the workings of a feudal system and proper medieval etiquette. There are legitimate places for the skill of the player to shine through, but the nitty-gritty of social mechanics isn't one of them.

While that argument can be dismissed out-of-hand, the other argument is one that is incredibly important because it touches on something at the core of what makes roleplaying games possible: player agency. In essence, the logic here is that social mechanics are used to make characters do something (believe this, do that, etc). But what happens when these mechanics are turned on the player characters? Having a roll of the dice to determine what a PC believes or does takes away their agency. if that happens, the dice are making decisions instead of resolving outcomes, which is the exact thing I said we were supposed to avoid.

This is the crux of the problem: we're okay with combat mechanics being able to damage a PC because the player retains their agency, but social mechanics can take that away. And we can't simply say "these skills are only for PCs" because what happens when one PC tries to influence someone else at the table? Then we would have to say "PCs are immune to these skills, and must be convinced the old-fashioned way," but then we get right back into the problem of players being less convincing or charismatic than their characters...

My Solution

I've been on both sides of that debate at one point or another, and for a while, I simply decided to go with whatever the system recommended. That worked pretty well until I decided to make my own system. Creating a game forced me to meet this problem head-on and come to a decision. I'm saying all this because what follows is going to look like an advertisement for my system, but it's not meant to be. The only reason I use it is because I can't think of a better example of what I'm talking about. At the end of this article, I'll discuss ways that the solution in my game can be taken and applied to others.

In Sarenteth, all conflict is handled the same way. There is no mechanical difference between a battle of blades and a battle of wits, except what sorts of things one might find useful. That solves a number of issues right away: someone in a fight is given the same number of decisions, and has the outcome determined in the same way, as someone trying to convince the king to take their side. Any kind of conflict is exactly as robust as any other kind, and is decided in exactly the same way. There should be no concern for one player being "better" at something than their character would be, because we have mechanics for that. It is by virtue of the player's understanding of the system (which is a valid place for player skill to be important) and the character's stats and circumstances that they win or lose. Plus, the way the conflict resolution system keeps the narrative at the fore means that those who want to roleplay are still very much rewarded for doing so.

So that takes care of all problems but one: player agency.

This is the bugbear, and its solution has to do with changing how one thinks about an outcome. Sarenteth allows for very few times when a player isn't in full control of their character. Taking damage to an Approach may change how the character acts, but only because the player is adapting to having different scores. Having an Approach drop to 0 gives an opponent the opportunity to narrate what happens to a character, but not what that character does about it. And if anything too terrible or game-changing happens, the player is given further opportunities to "save" themselves.

The key here is that an outcome in Sarenteth determines what happens to a character, but not what a character does. Let's take as an example a player character and a guardsman they are trying to bribe. They get into a social contest, and perhaps the PC attacks the guard's Empathy; trying to make them think selfishly, only about what they would gain through the bribe. The guard doesn't have a very high Empathy to start with, and so the score is brought to 0. The player is allowed to narrate the outcome... But they can't say "the guard gives in and accepts." Accepting a bribe isn't an outcome, it's a decision based upon an outcome.

So the player needs to think of this a little differently: the outcome of their contest should lead the guard to believe that accepting the bribe is the best course of action. Narrating the guard thinking of some luxury he has been wanting (and which the bribe would pay for) could work, or perhaps the guard feels like the money might make for a good rainy dya fund. The guard will still have the ultimate choice of accepting the bribe or refusing it (and perhaps his selfishness will lead him to conclude that it would be even more lucrative to arrest the PC and collect a reward). We don't choose what we think or what we feel, but we do choose how we act. That means the latter is off-limits, but the former is fair game.

Although the above is wrapped up in one particular system, the basic idea behind it can be adapted to any game. All it takes is for the table to come to a slightly different understanding of what the social mechanics do: rather than making characters do something, they inspire certain thoughts and/or feelings within them, in the hopes of pushing them towards a course of action.

So the next time a player rolls Persuasion, don't ask what they want their target to do. Ask what they want them to feel, or what they want them to think about. The next time they roll a natural 20 on Intimidate, just think about what that NPC does when they're terrified and have them act in accordance with it.

You can lead a horse to water, but you can't make it drink.

* Before someone points out that there is more than one decision behind attacking someone: if your system is so limited that your only mechanical options are attacking or doing something which is not an attack, you need to find a more robust system.




Sheet music on the homepage!