Social Interaction Part 2: The Problem with Social Combat

Momentum versus Attrition

In Social Interaction Part 1, I discussed the general issues with how persuasion is approached as being primarily about the skill of the speaker, when it should more accurately be more about the openness of the target to the specific approach being made. I ended by mentioning my intent to look into a number of social combat systems; that spread further than I’d expected, and I’ve now read through a large number of social combat systems and other interaction frameworks (and if anyone has other notable additions, I’m interested in hearing about it). Almost all of these fall cleanly into two camp, what I’m calling Difficulty Based and Defense Based systems.

While there’s no fundamental mechanical reason I can see, the set of games that would generally be considered to have social combat systems seems to have near-complete overlap with those that are Defense Based. Also notable, and against my initial expectations, this grouping did not break along GNS lines, with a firm mix of all three on both sides.

Difficulty Based Systems

In Difficulty Based Systems, interaction is usually resolved with one roll against a target number set based on the difficulty of the action presented. When difficulty is quantified, it is usually reflected in cost and risk.

Key Advantages: variable difficulty by action, quick resolution

Key Disadvantages: lack of rigorous difficulty assignments, often binary outcomes

Key Examples: D&D, Dungeon World, Ironsworn, Blades in the Dark, SWN, Traveller

D&D from B/X and AD&D up through 3.5e achieves a fairly good instance of this framework. Persuasion rolls are made against a table of outcomes, which creates nonbinary possibilities. D&D 3e improves this further by indexing the possible outcomes and difficulties against the target’s current attitude toward you. Sadly, this appears to have been largely lost after 3.5e, and 5e goes back to binary skill checks.

B/X rolls 2d6 instead of 1d20, which sensibly weighs outcomes to the center so that most people respond reasonably and extreme friendliness or hostility is uncommon in the absence of modifiers. ACKS takes this a step further by allowing multiple rounds of interaction and including tables of circumstantial modifiers, though it falls short of integrating individual characters’ values and favored appeals (though I’m not sure any system does by default).

AD&D has uneven spacing on a d100 roll, that strangely seems to weight outcomes towards the edges and make strong reactions more likely than moderate ones.

Dungeon World uses static target numbers that undermine the flexibility to have different attempts have different difficulties. Ironsworn initially appears to do the same, but structuring an interaction as a track instead of a single roll resolves this.

Traveller does a good job mitigating the issue of rigorous difficulty assignments by providing plenty of examples to justify a claim, and in this way does perhaps the best job of capitalizing on variable difficulty by action and circumstance. BitD could be argued to do the same but in my opinion lacks sufficient granularity and the ability to stack advantages (which is really a more general critique of the system, though it plays fast for it).

Defense Based Systems

Defense Based Systems tend to treat social interaction in very similar terms to combat, with some sort of “social hit points,” mental defenses that are rolled against over multiple rounds of interaction, social initiative, and mechanically defined actions.

Key Advantages: rigorous mechanics, complexity of interactions

Key Disadvantages: attritional

Examples: FATE, ASOIAF Roleplaying, FFG Star Wars, Legend of the Five Rings, Burning Wheel, Ascendant

FATE takes this paradigm to its logical extent and treats social interaction exactly the same way it does combat, with different skills but otherwise identical in all regards (save only that there’re optional rules to add armor and weapons that have no correlate here). Like all things FATE, this ends up feeling fairly generic.

ASOIAF Roleplaying and Burning Wheel do a better job of making this interesting, though I think ASOIAF would get repetitive. As for Burning Wheel, well, it would have to be repetitive for the group to actually figure out how it’s supposed to work, but it does seem to create some interesting options, though it could learn from En Garde about forcing telegraphed maneuvers to make play more tactical and less rock-paper-scissors.

Legend of the Five Rings, Burning Wheel, and Ascendant do some interesting things to ameliorate the issues of attrition by reframing the interaction, discussed below, but first it’s worth understanding why attrition is a bad model for social interaction.

The Problem with Attrition

The issues with Check Based Persuasion are fairly obvious. The issue with Social Combat is less so, and worth expanding on. At it’s heart, it comes down to the fact that virtually all RPGs model combat as attrition. You fight back and forth with an opponent, gradually wearing down their physical health and stamina until they hit 0 hp/mark their last Stress/etc., at which point they make a critical error and you incapacitate them.

The issue is that’s not at all how interaction works, and yet it’s how most Social Combat Systems I’ve read represent it. “You argue back and forth with an opponent, gradually wearing down their mental fortitude and stamina until they hit 0 mental hp/mark their last Stress/etc., at which point you make a brilliant argument or offer and they agree with you.”

In real life, when you wear someone down through argument, they get frustrated and your relationship worsens. They’re unlikely to adopt your position or yield unless they have no means of simply leaving (interrogating a prisoner, or nagging someone you interact with often, can both be effective for this reason).

Furthermore, people can have widely different amounts of “mental hp” based on the topic of conversation, and this is only sporadically modelled. But it is generally dramatically more difficult to persuade a lord to offer his daughter in marriage, or grant you land and vassalship, than it is to persuade him to host you for a few nights.

Legend of the Five Rings makes this attritional framework appropriate by its emphasis on honor and shame - losing means breaking your composure and departing in dishonor, or being cornered into a duel to regain it, or the like. This is attritional, but is very setting dependent; for a game where it works though, the premise seems sound.

Burning Wheel likewise makes the attritional framework appropriate by focusing it’s mechanics on public debates, where the object is to persuade the audience and not the opponent. I’ve done a lot of public debates, and while attrition isn’t perfect, the general idea of refuting your opponents’ points and undermining their argument while preserving your own isn’t so far off, and is very gameable.

Ascendant takes a novel approach by splitting out different subsystems of interaction: attracting and influencing a crowd, befriending and influencing NPCs, intimidating NPCs, and interrogating NPCs forming the main pillars of that. This lets it use Determination as mental hp for Intimidation attempts, while acknowledging that dealing psychic damage doesn’t actually help you make friends. There are a lot of interesting and subtle interactions here, but it’s very much a superheroes game and the mechanics enforce that (at human-scaled levels of charisma, a brilliant speaker may still be outweighed by circumstance; when the speaker is a few dozen times more charismatic than the most influential and persuasive humans to have ever lived, circumstance reasonably matters less).

Working Towards Solutions

Social combat systems tend to be attritional, and they tend to assign characters concrete defenses against certain approaches that bias them towards certain modes of interaction. They also tend to be more mechanically rigorous, but like combat, lack a quick resolution mechanic (which seems a more substantial omission, given that trivial conversations happening with somewhat more frequency than trivial fights).

The opposite of an attritional system for conceptualizing social interaction would be a momentum-driven one, where each success builds into further success, and failure makes future failure more certain. In combat, such systems (e.g. Star Wars d20’s Condition Track) are unfavored because they lead to death-spirals. I am not persuaded that’s a problem for social interaction; it certainly seems to match how social interaction tends to actually work. People who already like you are easier to get to know better, more likely to agree with you, and more open to nuanced positions (and vice versa). But it is worth more investigation of whether that’s a good outcome for gameplay.