Role-roll, roll-role, just role, just roll, please read the poll.

[Man in the Funny Hat]

Depends entirely on the edition I'm playing but also on the circumstances and in particular on my perception of the roleplaying skills of the player vs. the capabilities of his character.

I would always prefer roleplaying interactions to be governed ENTIRELY by roleplaying and I couldn't care less what your "persuasion" score was. But, I do recognize that not all players are good roleplayers but there should be no eggregious limitations on their choice of creating a character who IS good at what the player himself is NOT. For example, I have a player who is not outdoorsy, has the most crippled sense of direction and generally is NOT all the things that a Ranger IS - so, she habitually plays Rangers as her choice of character. Should she be penalized because she, the player, has no knowledge of wilderness survival or tracking and instead relies upon making die rolls? Should her character be repeatedly outshone by the "bookish, pale, shut-in wizard" because the player of that character DOES know a great deal more about the outdoors and animals? I'd say no, and that's why I'm willing to let one player rely on die rolls from "roleplaying interaction" skills and abilities awarded to the PC to manipulate what I would require other players to do with actual character interaction without rules interference on the outcome.

 

[1Mac]

The case for 2: Like it or not, DnD has codified social encounters with the same essential dice-rolling mechanic as combat. We don't allow an attack to succeed just because it is described well, so we shouldn't allow a social encounter to succeed for the same reason. In either case, we want to reward characters that invest in an ability, and that reward is mitigated if it can also be achieved by clever descriptions. At the same time, this is a roleplaying game. So for both combat and social encounters, let the dice fall first, then let the roleplaying be based off the results.

The case for 1: First, may I say that my interlocutor states the case for 2 quite convincingly! But surely roleplaying should be more a part of the game than illuminating cold die results after the fact? Roleplaying should shape the results as well. The dice should still be paramount, but a clever tactic or evocative description, whether it applies to a swing of the sword in battle or a turn of phrase in argument, should increase the odds of success in the player's favor. That way both die rolls and creative roleplaying have a part in success.

I'm in favor of 1, but I think that same approach ought to be taken with combat, social encounters, and any other sort of encounter. In this way of looking at it, the roleplay-before-the-roll is really just a way of earning a stunt bonus peculiar to social encounters. It's still a stunt bonus all the same.

 

[S'mon]

I'm pleased to see that even with all the other options available, a full 63.75% of voters are still on the side of righteousness.

BTW I really hate the "I, the DM, declare that average-CHA PCs can never be persuasive" - average CHA could indicate someone like me who is above-average in certain situations and below-average in others, as well as somebody who is routinely mediocre. I have something like +5 Intimidate, +0 Diplomacy and -5 Bluff.

 

[Umbran]

The case for 2: Like it or not, DnD has codified social encounters with the same essential dice-rolling mechanic as combat. We don't allow an attack to succeed just because it is described well, so we shouldn't allow a social encounter to succeed for the same reason.

Codified does not imply that there must be random dice rolling.

If you're playing 3e, for example, you can use the mechanic and still not roll dice - assume the players Take 10 on social skill checks! You still reward those who invest in the options, but leave the randomness out of it.

 

[the Jester]

Roleplay and roll dice, in either order.

 

[Rogue Agent]

So my new attitude is "Screw social checks for most situations, because they're lame and binary."

Interesting. I'm increasingly coming to the conclusion that social checks are most useful when (a) they're resolving specific goals and (b) they're binary (or, at most, give a narrow range of flavored binary results -- exceptional success, success, failure, exceptional failure, for example).

Even complex computer programs suck eggs at modeling complex social interactions, so the idea that any tabletop mechanic is going to succeed at producing plausible results is just silly. This rules out any kind of specificity or complexity in the model, IMO.

Using mechanics to check for the success or failure of specific goals -- can I convince the king to give us troops?; does Barda belief my lie? -- is useful. The simple binary (or flavored binary) result can then be interpreted through human creativity to produce plausible results.

The big problem with 3E Diplomacy basically boils down to the fact that the skill does something completely unlike any other skill. If it instead modeled the success or failure of specific actions (just like every other skill), the skill would be just fine.

In general, I think the poll is flawed. At my table, the method is generally "roleplay first, then roll for some specific action which needs to be resolved, then more roleplay". Roleplaying has to happen first because that's what establishes the action being attempted and the conditions of the attempt (which may modify the roll), but accomplishing a social goal rarely means "this conversation is over" -- it usually just defines how the conversation will proceed.

 

[airwalkrr]

Roleplay first then roll the skill check

Generally, I let the players role-play the situation then call for a skill check and use the DM's best friend (apply a +2/-2 modifier depending on the situation) to help determine the outcome. Of course, I feel the need to qualify that statement a bit. It depends on the campaign. I run one campaign which is geared towards powergamers who are more interested in the tactical game than the role-playing. In this campaign, if a player says "I am going to roll a Diplomacy check to convince the guard to let me pass," I let the player roll the dice and let the dice decide the outcome without any role-playing necessary. But my longer ongoing campaign is more laid-back and casual. If the players present a compelling case or find a win-win solution through role-playing, the dice are not necessary unless the NPC is hostile or unfriendly for certain reasons. So my true answer is: it depends.

 

[airwalkrr]

The case for 2: Like it or not, DnD has codified social encounters with the same essential dice-rolling mechanic as combat. We don't allow an attack to succeed just because it is described well, so we shouldn't allow a social encounter to succeed for the same reason. In either case, we want to reward characters that invest in an ability, and that reward is mitigated if it can also be achieved by clever descriptions. At the same time, this is a roleplaying game. So for both combat and social encounters, let the dice fall first, then let the roleplaying be based off the results.

I would note here that players are rewarded in combat for using tactics. The simplest example is flanking an opponent, which provides a +2 to hit. So a player who says the right thing in a social interaction should expect to receive a similar bonus because the player used tactics in that situation. I am not talking about rewarding flowery speech and bluster (unless it is specifically a tactic, such as when addressing a parliamentary session known to respect such talk). What I am talking about is a situation where a player sees an advantage and seizes it, such as recognizing the signet ring of a noble belongs to a house that is destitute and indicating in their speech that if the noble goes along with whatever the player proposes it will increase the wealth of the noble in question.

 

[Plane Sailing]

My top preference is just to handle it via roleplay if at all possible. If that works out well, we can just get to a result which obviously works quickly. Especially good if the player is able to role play his PCs capabilities well.

If we have a situation where, for whatever reason, rolling is necessary, my preference as both a player and a DM is to roll the check first, then roleplay the result. It takes good, willing roleplayers to do this though!

Otherwise it breaks our sense of verisimilitude when someone makes a brilliantly impassioned plea and then rolls a 1. Giving a tiny bonus for some good role playing doesn't really help out there, and suddenly we have to come up with some off the wall reason why it didn't work out at all.

 

[Plane Sailing]

Addendum: as a DM I fact in the PCs capabilities when responding to the player. If a silver tongued player attempts a diplomatic speech from his 6 Cha half orc, and a stumble-tongued player attempts the same from their charismatic and skilled bard, I will role-play a better response to the latter than the former; I expect that players will do their best to RP their character, and I have the NPCs respond to the character, not to the player (if that makes sense)