What makes social skills different?

[weckar]

So, here's a thing that's been bugging me for a while:

If I were to try and open a lock in most games, I roll some dice to see if I succeed.
I do not actually need to provide a strategy to open the lock - this is a skill my character has, after all, not me.

Yet, if I were to convince someone of something and roll my persuasion, my GM always asks me to come up with an actual argument.
If it is not good enough, I fail regardless of the roll, even though with his skills my character likely could have come up with a good one.

What makes social skills so different that they can be treated this way?

[aberratio ictus]

The fact that we're sitting around a table playing a social game with dialogue. Swinging swords and picking locks is rather hard to do at the table.

Then again, at the tables I've played at, the most you'd get would be a slight bonus or malus to the roll. Most if the time, not even that.

Playing in a game with dice rolling instead of dialogue seems rather sterile and not very engaging to me personally, but to each their own, I suppose.

[Cluedrew]

The fact that it is easy to replicate around the table is part of it. I think the more important issue that what your character does in a "social encounter" is very closely tied to characterization, as in portraying who you character is and something people don't want to mechanically abstract away.

As someone who is making a homebrew system, that tug between allowing people to say exactly who their character is and at the same time the character being as good as the sheet says they are comes up a lot. Actually it comes up everywhere, but social skills especially.

[weckar]

It just feels that for a character to be good socially, it requires the player to be good socially. No other area of gaming really requires that.

[Floret]

This is an old debate - and, at its base, matters of taste make it different.

For some people, sitting around a table, enabling to talk is close enough to the situation in game; whereas for most everything else, it is not. And it is further removed, certainly.
Now I personally think, for numerous reasons that it isn't close enough - the factors pressing on the character being quite different from the factors working on the player (Sitting around the table talking to a friend simply isn't the same as talking in front of a crowd, or in a dark, damp cavern, or at a city gate to a guard; whose relationship to you will be different to the one you and the GM share. Yes, imagination and immersion are a thing, but they only go so far, in my experience; my comparison here being between otherwise largely comparable situations in Tabletop RPGs and LARP.)

There is also, certainly, the question of differentiation of player and character skill. The gulf might just be larger in other situations - everyone can talk to some extent (Well, communicate; there are mute people). Some people, including you, it seems, (rightfully) point out that for the game, if there are stats, those should matter, instead of player skill. After all, if you put points in something... That should have some relevance, no?
(Then there are some things people might not wanna play out in great detail, particularly seduction or hardcore intimidation/torture; without denying it as part of the game. Boiling them down to a roll might just be easier and more comfortable for everyone involved than playing it out.)

One of the bigger points I have also seen to argue against rolling is the inherent unrealism of the mechanics ("This is not how persuasion works IRL"), which... certainly it might not be; or it might be abstracted to a point that people don't intuitively understand it to be abstracted to, leading to the feeling of "diplomancers". Different, more fleshed out rules than most games have might help here, if that is the only point of contention.

There are also some people that argue the mind of a PC is a sacred space that should not be touched by rules - and if rolls work to seduce NPCs, then they might be used on PCs as well, so any mechanisation holds at least some danger there. (One might contend that they'd much rather screw with a PCs brain directly than go through the player's; which one would have to to reach the PC if there are no rules. Your mileage may vary.) In a similar vein, it carries a feel of the dice determining what comes out of a character's mouth for some - which some of those people heartily object to; either because of realism, or because of getting control over their character taken away in an area they deem too holy to touch.

Ultimately, it comes down to taste. I myself would hate to have discussions boil down to simply a roll, like to play out discussions (Because it's just... fun, more immersive, more engaging... a number of things); and call for a roll only if the question "Does the character let slip something/go with the demands/etc." requires me to actually make a decision based on character skills.
Taking it to a roll has benefits, and so has playing it out. I mostly fall down on the play-assisted roll side of things; you sound like you might as well, but your GM does not, or to fall further on the "roleplaying" side of things. Nothing wrong with that per se, but it seems a dissonance. Maybe talk it over with him, voice your arguments for why you feel it should be treated (copy any of my arguments, if you want?), try to understand where he is coming from, and then look if this is something you can deal with in a game you are playing.

I will note that voicing an argument makes it easier to act out not only that the NPC acquiesces, but WHY, and most importantly HOW, so I can understand him asking - then going around and saying "despite your character's skill, nope" is a playstyle I much less agree with, but again, a matter of taste.

(And, I do hope I have not missed any of the arguments for why to treat them differently (And have them presented fairly); I had the feeling I have over the years gathered some insight into the arguments, seeing as they were made in contention to my opinions )

[PhoenixPhyre]

One reason that matters for me (as a DM) is that unlike lock-picking or chasm jumping, the approach really really matters for social skills. I don't require an exact speech, but you have to let me know (at least) the following:

1) what you're trying to get the NPC to do
2) how you're approaching them (beyond simple skill names)
2a) are you trying to bribe? how much are you offering?
2b) threaten (and if so, with what)?
2c) appeal to their sense of justice?
2d) lie? What shape does that lie take?
3) what risks you're asking them to take for you.

All of these go into setting the target number. Some NPCs are easily bribed, but will laugh off threats and have no sense of justice. Others hate being lied to, but respect honesty. Yet others respect chutzpah.

Yes, in reality locks differ, but it's a lot harder to characterize the lock well enough to make a significant difference.

[2D8HP]

....What makes social skills so different that they can be treated this way?

.
The difference is that the GM may judge it more fun to hear some dialog than to watch some dice rolling.

D&D:didn't used to have any "skills" until the Thief class had stuff like "pick pockets".

The closest to social skills was the CHA stat, and there were rules for rolling reactions with a mod for CHA but a PC convincing n NPC of something, then as now, was handled ad-hoc.

There may have been others before, but of games I'm aware of,Traveller (in '77) had skills, as did the more D&D like RuneQuest (in '78), and other games followed, and social skill mechanics became usual, but the tradition of including table talk was already in place, but really it's just about deciding what seems more fun, player patter or dice rolling.

[Darth Ultron]

The fact that we're sitting around a table playing a social game with dialogue.

This is a big part of it right here. There is no point in playing a social game if your going to be anti-social.

Role-Playing is also a big part of the game...for most gamers. It is fun to play a role and inter act with others. It is not fun to watch someone roll play.


And while it is true that most DM's won't ask you ''how'' your character picks a lock....that is not true all the time. A lot of actions in game play depend the player describing a bit more then the boring ''my character does X" and then rolling dice. D&D, for example, does have a circumstance bonus, for example. So if a player can take a couple seconds to descried in detail what they are doing in a role play sense, they can get a bonus to that mechanical roll. The same is true for social skills.

And combat in most games needs a level of detail to it, yet few players complain about it. They will happily count squares, crunch some math and make all sorts of tactical decisions and even write a whole story about their ''gleaming sword strike as it flashes through the air to do a combat maneuver, but then act like it is the end of the world if they are asked ''so what story do you tell the guard?"

When you take a away all role playing from social skills...you might as well just play a video game. The gameplay is then like ''my character talks to people and rolls and 11'' and ''the people talk back to your character and tell your character stuff'' and ''ok, my character uses the stuff he now knows to solve the problem." Wow..exciting gameplay.

And it's not like most DM's what a massive novel every time a character does anything......just look at what the average DM will accept:

Player:"I walk up to the guard and tell him that I'm from HighCity and I'm here at the castle to inspect the tapestries.
DM: Nods, ok, roll your persuasion check.

See, it is not so much to ask.

[Tinkerer]

It just feels that for a character to be good socially, it requires the player to be good socially. No other area of gaming really requires that.

It is an awfully fuzzy line to draw. Myself I draw the line pretty darn hard to fail (maybe a minor bonus or penalty for circumstance and eloquence). If a player is really having a problem with the concept I will often sit down with them after the session and discuss a few different hints I have for them including: having a cheat sheet of pre-written lines for a variety of situations, blatantly ripping off media (then again that's also how wands of lightning bolts get referred to as boomsticks), talking in character more often when they aren't trying to use their skills, etc...

[LordCdrMilitant]

Because it's more important than other checks. NPC's react to them, so I have to have some foundation based on at least a general idea of what you say.

If you tell the guards you're going to inspect the tapestries, they'll expect you to be in the great hall doing that, not in the kitchen adding things to the soup.

[Joe the Rat]

If you're a Scottish Lord, then I am Mickey Mouse!

One reason that matters for me (as a DM) is that unlike lock-picking or chasm jumping, the approach really really matters for social skills. I don't require an exact speech, but you have to let me know (at least) the following:

1) what you're trying to get the NPC to do
2) how you're approaching them (beyond simple skill names)
2a) are you trying to bribe? how much are you offering?
2b) threaten (and if so, with what)?
2c) appeal to their sense of justice?
2d) lie? What shape does that lie take?
3) what risks you're asking them to take for you.

All of these go into setting the target number. Some NPCs are easily bribed, but will laugh off threats and have no sense of justice. Others hate being lied to, but respect honesty. Yet others respect chutzpah.

Yes, in reality locks differ, but it's a lot harder to characterize the lock well enough to make a significant difference.

Perhaps not for locks, but you could easily take this approach to traps:

1) Are you trying to disarm or disable?
2) The trigger, or the mechanism?
3) Be able to reset or repurpose?
4) Hide your tracks?

Then there's the physical obstacles: How do you get past the shifting corridor / huge chasm / frictionless room that drops you into a pit of deadly blades if you can't stop. What you propose certainly decides what rolls you need to make.

---

If you have someone that's really into the oratory side of social encounters, then flip the system - have them say how they're approaching the exchange, roll, then talk out something appropriate to the result. Anyone can say a stupid, but it takes a true master of discourse to lay out an epic failure on the dice.

[Tanarii]

So, here's a thing that's been bugging me for a while:

If I were to try and open a lock in most games, I roll some dice to see if I succeed.
I do not actually need to provide a strategy to open the lock - this is a skill my character has, after all, not me.

Yes you do. At the least, you need to provide Intent (open lock) and Approach (break with strength, pick with expensive thieves tools, pick with makeshift thieves tools, poor acid on it, etc). Each of these gives potentially different chances of success (on a spectrum from automatic to check needed and modified by approach to no chance of success), success and failure outcomes (open lock suddenly, open lock slowly, fail to open lock, jam lock open or closed), and consequences (cannot lock behind you, make lots of noise, quiet but takes time).

What makes social skills so different that they can be treated this way?

They aren't. Either you've got DMs that aren't resolving each correctly, or you're failing to understand where the points are for your decision making vs resolution.

It looks like the latter to me. When you decide to get past a lock, you're making the same decision points as getting the king to hire more troops. If you have to come up with an actual argument to persuade, that's because the DM needs to know the details of the Approach you are taking. Persuade instead of Intimidate is like Pick instead of kick door down. Thieves Tools and using them to pick the lock, as opposed to using makeshift tools or using acid to burn it out, is like argument A that hiring troops will save the Kings daughter's life, vs Argument B that the king will save money in the long run by defending his peasants. One Approach is better than the other.

That said, there is often a difference in complexity between physical and mental/social checks. Often physical checks have rather simple decision points and simple consequences. Just a few for each. Whereas mental/social checks can have quite a lot of complexity or nuance for both Approach and Consequences.

[Airk]

Yes you do. At the least, you need to provide Intent (open lock) and Approach (break with strength, pick with expensive thieves tools, pick with makeshift thieves tools, poor acid on it, etc). Each of these gives potentially different chances of success (on a spectrum from automatic to check needed and modified by approach to no chance of success), success and failure outcomes (open lock suddenly, open lock slowly, fail to open lock, jam lock open or closed), and consequences (cannot lock behind you, make lots of noise, quiet but takes time).

They aren't. Either you've got DMs that aren't resolving each correctly, or you're failing to understand where the points are for your decision making vs resolution.

It looks like the latter to me. When you decide to get past a lock, you're making the same decision points as getting the king to hire more troops. If you have to come up with an actual argument to persuade, that's because the DM needs to know the details of the Approach you are taking. Persuade instead of Intimidate is like Pick instead of kick door down. Thieves Tools and using them to pick the lock, as opposed to using makeshift tools or using acid to burn it out, is like argument A that hiring troops will save the Kings daughter's life, vs Argument B that the king will save money in the long run by defending his peasants. One Approach is better than the other.

That said, there is often a difference in complexity between physical and mental/social checks. Often physical checks have rather simple decision points and simple consequences. Just a few for each. Whereas mental/social checks can have quite a lot of complexity or nuance for both Approach and Consequences.

I don't agree with you very often, but this is a very good explanation. You need to provide the approach/method for a social skill check, but if the GM is asking you to give him a blow-by-blow argument, then from my perspective, the GM is doing it wrong. The blow by blow can be played out a little bit afterwards when you know how things are going to go.

The correct way to resolve any action is pretty much always (i.e. unless your game explicitly says otherwise):

State intent ("I want to pick the lock!" "I want to persuade the king to send troops to the Vale.")
State Approach/Method ("I'll use my high quality lock picks." "I will grandstand about the value of the Vale as the breadbasket of the kingdom and the potential havoc caused by it being overrun.")
Optional (Varies by game): Decide stakes or consequences. ("If you fail, you're not going to be able to open this lock this way." "If you fail, the king will order his guards to arrest you.")
Engage resolution mechanism (Roll dice, draw cards, spend points, whatever.)
Roleplay the result. ("The lock clicks open as the well oiled tumblers roll over." "The king rubs his chin, nodding, and says "Very well, I will dispatch the 2nd infantry to defend the Vale, but if you have mislead me...")

[Aliquid]

Yet, if I were to convince someone of something and roll my persuasion, my GM always asks me to come up with an actual argument.
If it is not good enough, I fail regardless of the roll, even though with his skills my character likely could have come up with a good one.

Say to your DM, "I roll on my persuasion skill to see if my character can come up with an effective argument against this particular person". If the roll is high, you say to the DM "So... that's a good roll, my character observes the person and the social situation and is able to decide what tactic to use. You as the DM now need to tell me what my character figured out about the NPC. Should I lie to him, guilt trip him, offer a compromise, pretend I am an authority and pressure him, flatter him...?"

Lets say that the DM says "flatter", your next response is "what sort of thing would flatter this guy? I'm assuming that flattering him about his strength would be a good bet, does my character (who has a high charisma) agree with that assumption? He would know better than I"

[Anonymouswizard]

There are essentially three schools of thought with regards to social skills.

The Gonard School: 'back in my day social interaction was ad-hoc'. While not every member of this school of thought has actually played old-school games, the idea is that you can talk while sitting round a table, so why do we need to simulate talking.

The Fairness School: While we might enforce roleplaying, we're not going to penalise players. Therefore, a good roll will mean your character worded it better than you did.

The Halfway School: We'll apply modifiers based on your roleplay, but you still might succeed with a bad argument if you get a lucky roll.

Note that I have never seen a GM declare an argument fail because they think it's bad. I've had to suck penalties, sometimes up to my skill bonus or more, but never outright failed without a roll (at least officially, I'm sure I've been penalised into 'not a chance' once or twice).

[Tanarii]

I don't agree with you very often, but this is a very good explanation.

I stole the terms / clarity from Angry DM. He wrote a very good article on it. He's mostly a D&D guy, but the concepts are mostly applicable to other systems:
http://theangrygm.com/adjudicate-actions-like-a-boss/

Say to your DM, "I roll on my persuasion skill to see if my character can come up with an effective argument against this particular person". If the roll is high, you say to the DM "So... that's a good roll, my character observes the person and the social situation and is able to decide what tactic to use. You as the DM now need to tell me what my character figured out about the NPC. Should I lie to him, guilt trip him, offer a compromise, pretend I am an authority and pressure him, flatter him...?"

the problem with this is the player skips a very important step: determining what approach they want to use. That will affect the possible outcomes, as well as the potential consequences.

That's basically half the agency/role playing involved. Using the highest level definition of roleplaying: making decisions for your character in the fantasy environment.

It's certainly possible to have resolution back-fill the approach, but I wouldn't want to do that as a DM, nor have my DM do it as a player. I want the player to make that choice as a DM, or to make it myself as a player.

That's not to say there's something wrong with ways to help narrow down the possible approaches that might bring the outcomes and consequences they want, if there's a way for the PC to determine that in game. Possibly involving some kind of roll. E.g. Wisdom / Intuition in D&D for people, or Intelligence / Investigation for situations (traps, locks, etc).

Note that I have never seen a GM declare an argument fail because they think it's bad. I've had to suck penalties, sometimes up to my skill bonus or more, but never outright failed without a roll (at least officially, I'm sure I've been penalised into 'not a chance' once or twice).

There is a difference between having an attempt to persuade automatically fail without a check because :
- the 'argument' cannot possibly succeed. Ie Approach is invalid with the person you're trying to persuade.
- the players delivery of the 'argument' is poor or poorly phrased.

Similarly, there's a difference between having different arguments (approaches) have different chances of success (DCs in D&D, or modifiers to the roll in various other games), and modifying it based on the players delivery.

It can be hard to separate approach vs delivery/description, especially for social situations that need a roll to affect the outcome where the player is saying exactly what their character is saying. But it's an important distinction for properly resolving the action, IMO.

[Esprit15]

My sort of go-to with talking is to have the person at least lead into how they begin talking to someone, enough that I get which social skill they are rolling, whether they are offering something or otherwise playing to the target’s interests, etc. If they are a charismatic person in real life, I’ll just let them finish out their spiel, but if they are clearly stuttering and uncomfortable, I’ll just ask them if they want to roll for it, and then will give a brief overview of what their character said.

Ex:

Player: “So, we’re a pretty decent... erm, group of adventurers, and would like to– I mean, if you’ll have us, uh, work for you, not for a lot I mean, but... um...”
DM: “Okay, so, OOC, you’re offering the party’s services for a fairly low fee?”
Player: “Yeah, basically.”
DM: “Okay, roll diplomacy, give yourself a +2 for offering to do it at a discount.”
*rolls well*
DM: “So yeah, you all see Leho the warlock work his magic, talking up the strength of Derek and the magical ability of Soren, while offering your services to the king as more of a favor than a true job. The king and Leho eventually settle on 500 gold to each of you.”

It coaxes people just a little out of their comfort zone, allows characterization, but still leaves the dice in control. Meanwhile, I’ll often skip over the diplomacy check for minor encounters if it is well RPed well, though more important checks ultimately come down to luck.

[Bogwoppit]

What makes social skills different is that we're all used to using them, as opposed to fighting skills, breaking and entering skills, arcane skills, and acrobatic skills - so we don't usually try to simulate them, because we hardly think of them as skill tasks.

The suggestions above - declare Intent, apply a Method, adjust DC, roll the dice - have been suggested by people before, even people on this board.
They don't work for everyone, though. Some sort of compromise, based on a rules-set + RPing, is what tends to happen in my group.

[Aliquid]

the problem with this is the player skips a very important step: determining what approach they want to use. That will affect the possible outcomes, as well as the potential consequences.

That's basically half the agency/role playing involved. Using the highest level definition of roleplaying: making decisions for your character in the fantasy environment.

Then what's the point of having a high charisma or a high wisdom if you can't ever use it?

The comparison people are using of "tell me how you approach a lock pick" isn't even a fair comparison. The real comparison is to put a real lock in front of a player and tell him/her to pick it.

That's not to say there's something wrong with ways to help narrow down the possible approaches that might bring the outcomes and consequences they want, if there's a way for the PC to determine that in game. Possibly involving some kind of roll. E.g. Wisdom / Intuition in D&D for people, or Intelligence / Investigation for situations (traps, locks, etc).

I don't use D&D as a system anymore, so I don't know current versions, but with 3.5, I think it would be perfectly fair for a player to use a "gather information", "knowledge", or "sense motive" roll to find out what approach would be best under the given circumstances.

In reality, it shouldn't have to come down to that. This should be a collaborative process, and the GM shouldn't be an ass, and penalize the player for lacking social and diplomatic skills.

A character with a good charisma would be able to carefully test the waters and see if flattery works. They could make a casual flattering comment, and watch the NPC's body language to see if it had an impact or not, and then abort if it isn't working. The player quite literally can't do that, even if he/she had the skills to read subtle body language... the NPC isn't in front of him/her to observe in the first place.

There is a difference between having an attempt to persuade automatically fail without a check because :
- the 'argument' cannot possibly succeed. Ie Approach is invalid with the person you're trying to persuade.
- the players delivery of the 'argument' is poor or poorly phrased.

Agreed. I would be much more annoyed at a GM that punished a player for personally failing at the second, rather than just the first.

But still, I think if the character had a good wisdom, the player could say "I'm going to carefully suss this situation out, and see if this guy would be the sort of person to accept a bribe".
- If the roll succeeds, the DM will say "Yeah, he would take a bribe", or "no, he would take offense to that"
- If the roll fails, the DM could say "your careful probing wasn't subtle enough, he caught onto what you were implying and took offense"

Actually I like adding this step to the process, as it makes the player consider the situation a lot more carefully. A good player might even say

• I chat up the various guards to find out which one is the most gullible and make note of when his shift is and what entrance he guards
• I gather information to find out how legitimate people get past a guard (must be on a list, must show a pass, etc)
• I gather information on what sort of people would be going into that entrance at that point of time (caterer, diplomat, merchant? etc)
• I use disguise self to make myself look like that type of person
• I use my knowledge gained on how people normally get past the guard to help think of a good bluff (or forge a pass... or take on the name of someone on the list... etc)
• I go up to the guard and bluff my way past

This way the player can use their skills to help them decide on the best tactic. Maybe their initial roll showed that none of the guards are gullible... ok then, abandon this plan and think of something new... etc.

[Tanarii]

Then what's the point of having a high charisma or a high wisdom if you can't ever use it?

What I'm suggesting verry much makes a high charisma and high intelligence and high wisdom important. Any time there is a chance an action might fail that involves them, it's relevant. The player can't just talk their way around a problem unless they figure out something that will result in an automatic success. They also can't do something at all if it has no chance of success. They also need to think about consequences based on their approach. It makes a difference, both in your ability score/skill used, the chance of success, and the long term consequences. Examples would be attempting to bully a guard (intimidation) vs bribe them (persuasion) vs remind them they owe you a favor (possibly no check needed due to automatic success, but still an attempt at persuasion). How good you are it if the outcome is in doubt matters, as does the resulting consequences.

What you're suggesting disassociates actions from consequences. Instead of the player needing to consider how they're going to do something, and have to face resulting consequences, you're skipping that and only caring about intended result and if it succeeds or fails. That's fine if you want 'narrative' causality, where you know what's happening now you just need to determine how. It's less fine if you want causality to be action --> consequences.

The comparison people are using of "tell me how you approach a lock pick" isn't even a fair comparison. The real comparison is to put a real lock in front of a player and tell him/her to pick it.

Youre making the same mistake of not understanding where the decision point on intent and approach are, vs description.

Here's the full comparison.

Intent: get through the portal or open the container
Approach 1: pick the lock
Additional decision points: use thieves tools vs use makeshift tools
Approach 2: break open
Additional decision points: use muscles, use axe
Resolutions: pick lock skill (possible penalty) vs success chance, Use strength-type skill vs break open chance, use attack vs object defense/durability/hps
Outcomes: it's open or not open
Consequences for 1: it's quiet and the portal/object can be relocked, but it takes time.
Consequences for 2: it loud, it destroys the portal object partially or completely, it's fast.

Intent: get past guard
Approach 1: bully him
Additional decision points: use physical threats, remind him of information you have over him, pull rank
Approach 2: persuade him
Additional decision points: bribe him, remind him he owes you one, try to reason with him
Approach 3: deceive him
Additional decision points: present falsified documents, pull fake rank
Resolutions: make require checks to fake a document, requires money, guard may or may not be impressed by threats or rank. Checks to be believed may be required. Checks to persuade with an argument/reason of 'please, it's important' might be incredibly hard.
Outcomes: you get past or don't get past
Consequences 1: he's pissed off
Consequences 2: he's not pissed off
Consequences 3: you might get caught later.

The advantage of this is "Player Skill" is relevant in this, and so is "PC skill". The player needs to make the decisions, the PC executes with their skill. It works just like an in-game physical action in that regard. Players don't just get to describe/narrate their way to success, but neither do they get to just roll a die and have all important decisions get made for them.

Edit: also, since this isn't a video game, it doesn't play out like this decision/resolution/consequences tree. The player says what he wants to do and how, possibly in a descriptive or narrative way or by just saying what their charcater says, either approximately or exactly. Then the DM asks for any clarifications on Intent or Approach that are needed, decides if a roll is needed, what it is if so, and then the players resolves. Then DM gives the outcome and either announces or notes the consequences, and if an NPC is involved has them react. Nor does it have to be a constant check for each little thing. This entire cycle might be a simple thing ("I break down the door with my axe") or a complex thing (on-going 5 minute conversation between PCs and NPC, followed by a single resolution check).

[LibraryOgre]

I'd say another aspect is that social skills interact with characters, whereas most other skills interact with the environment.

Open Locks interacts directly with the environment... do a thing and a thing happens to a thing.
Knowledge skills represent knowledge of the environment... do I know a thing? Tell me about ti.
Move Silently interacts with others, but more as features of the environment... if I succeed at a Move Silently check against someone, I'm not altering their character (until I stab them in the back), just reacting to them as a piece of the environment.

But inherent in social skills is the idea that I will change you. I will make you believe something you do not believe. I will shape your way of thinking about things. In this way, social skills are very much like combat... both make you different, but one makes you think different, while the other makes you differently alive.

And, while combat does NOT require you to demonstrate your ability, it does require you to lay your tactical cards on the table. Sure, you can walk up to the orc and start hitting her with a sword until she falls down, but that's a tactic in and of itself... much like convincing the king to do what you want by walking up and telling him the truth. Both are straightforward and honest... but you can also kill the orc by sneaking up behind her and stabbing her in the kidneys, or by poisoning her food. Likewise, you can convince the King to do what you want by lying to him, or by offering him a deal, or what have you.

In a lot of ways, this why using detailed social systems are important... if you fall entirely on role-play and player fiat, you have Bob, who spent no points on Resist Persuasion, ignoring torture because "That's what my character would do", while Sarah, who spent those points, interacts with the system and possibly fails.

[Aliquid]

What you are suggesting disassociates...

no it doesn’t. You are have inconsistent expectations between “unlocking a door” and social skills.

The person does select an approach. They select “diplomacy” or “bluffing” or “intimidation” or “bribing”.

The problem is when the DM asks the player to state a good sub-approach within the chosen approach. Such as “now you have chosen to intimidate, tell me your approach as to how you intimidate” and then the DM punishes the player for not coming up with a clever strategy.

[Quertus]

Ah, my favorite topic is back. Others have already hit on most my points, but I'll make them again anyway.

It's darn hard if not impossible to convince a vegetarian to eat a steak (not threaten, not coerce, just "this is Timmy, try it"). Just like it's really hard if not impossible for a woman to seduce a gay man.

The approach one takes when using social skills greatly impacts one's chances of success.

But isn't that true of all skills? Well, somewhat. I mean, I imagine it would be much more difficult if I tried to program nuclear guidance systems using bubble sort in basic. Or, I would, if I didn't suspect it may well have been written that way in the first place.

Presumably, most of us have a basic sense of goals, motivations, drive, etc. If not, a) why are we playing a role-playing game, and b) how is that even possible?

Right or wrong, most humans have some sense of how things work. They use this to determine what DC defeating a given lock is (or, to turn that around, to describe what a lock with a listed DC might look like). But, as a rule, intuition says that defeating any given lock is possible, sinking any given ship is possible, hacking any given system is possible. But most of us believe that no amount of telling us that Timmy is yummy (darn autocorrect) will convince us to kill and eat our best friend. Right or wrong, most of us believe in social no-sells. And you'll excuse me as I slowly inch away from anyone who says, "nah, I'd totally kill and eat someone if I was told they were yummy convincingly enough"

I mean, heck, when asked to do something against their nature, characters who've already lost to a spell get an automatic saving throw - just how much less effective should someone just using words and telling us our buddy looks tasty be?

Here, unlike questions of whether wood floats or how barns burn, my intuition tells me that "just wing it" will produce better and more consistent results than any system I've seen, under myself and most GMs I've seen (and, for those who don't know, I've seen lots of terrible GMs).

Now, any portion of the game could be micromanaged the same way social skills often are, with the GM asking for a detailed description of how things are done. In fact, some of the earliest examples of Player skills trumping character skills is when dealing with traps. What is the point in putting resources in making one character good at dealing with traps, when another had a player who understands traps, mechanics, and basic physics well enough to handle the GMs traps better than the skilled character can by using their skill?

This is a fundamental RPG issue.

Personally, while I have the player provide the path that the character takes, I have the stats / skills / dice determine how far along that path they can get.

Even back when all there was was a Charisma stat, I had that determine how NPCs responded to the PCs given strategy. For example, someone who isn't interested in bribes may cause trouble for a low Chr character, but may just comment that they're glad the high Chr character was joking, because, otherwise, they'd have to report them.

And, I could say a lot more, but let's not lose the forest for the trees.

[Cluedrew]

Lots of good points, I especially liked Floret's and Mark Hall's posts. But I'm going to address a different issue, which is the approach model brought in by Tanarii. I think the idea is correct, but I think the part it misses, or perhaps just doesn't address explicitly enough. The difference between social skills and others by this model in my mind is that people ask not only for the approach but the details as well.

I think "I attempt to convince the guard to let me pass by acting like I belong." outlines an intent approach well enough as "I'm going to try to force the bandit to surrender by disarming him." However asking for the exact wording of "acting like I belong" is like asking for "I hilt strike the bandit's axe close to the handle then hook around it and spin it against his grip."* To me those are the details of the attempt. The might be fun to fill in for flavour but I do not believe they should be required to make the attempt.

*I have no idea if that is a good way to disarm someone of an axe.

[Aliquid]

Cluedrew gets it.

[Cluedrew]

Well thank-you, I think it is something that a lot of people know, but don't quite have the words for. I had to borrow a lot from up-thread, but I think I got my idea across.

[Segev]

To answer the first thing the OP asks - why can he roll a die and be done with it for lock-picking, but not for "social stuff" - I'll ask a similar question:

"Why can I just roll a die to see if I can leap across this chasm, but I have to roll tons of dice and make all sorts of tactical decisions to see if I can defeat these kobolds in combat?"

The difference in both situations is the same: one is a singular action that is not easily broken down into sub-actions where individual successes and failures pile up to lead to a final outcome; the other is a scene unto itself, with many things that can be tried.

Or, examined another way, rolling to pick the lock or leap the chasm is making an attack roll. The combat result is analogous to whether or not you get past the whole set of obstacles of which the locked door and chasm are parts.

The reason social skills are different is that they are, as presented in most games, essentially the equivalent of having a "fight skill," where the actual dice resolution is a single "fight roll," and the player either "roll plays" it and just says "I roll fight; do I win?" or "role plays" it and describes his entire combat scene and approach, then rolls the dice anyway.

If social mechanics are to live up to what people seem to want from them when they demand that you "role play" it, then they need to have the depth of physical combat. Rolls aren't to "persuade," but to learn about the other socialites, to see if you can sway or tempt them, to stoke anger or greed or lust or joy. To get them into a position where their natural inclination is to go along with what you want. It would be like rolling lots of rounds in combat, but more puzzle-like because you're not just wearing down "will points" to browbeat them into being your mind-slave. You're interacting with their mechanical social statistics to manipulate them into doing what you want.

[Thrudd]

So, here's a thing that's been bugging me for a while:

If I were to try and open a lock in most games, I roll some dice to see if I succeed.
I do not actually need to provide a strategy to open the lock - this is a skill my character has, after all, not me.

Yet, if I were to convince someone of something and roll my persuasion, my GM always asks me to come up with an actual argument.
If it is not good enough, I fail regardless of the roll, even though with his skills my character likely could have come up with a good one.

What makes social skills so different that they can be treated this way?

They shouldn't be treated that way. It is either dictated by the rules or it isn't. If it isn't, then there shouldn't be any character resource expenditure. If it is, then technically you shouldn't need to say anything at all, just like with lock picking, etc. Some people might do a hybrid, where bonuses or penalties might be applied to your roll based on what you say, but your character's skills are still applied and the dice make the final decision.

[weckar]

I do think I like the middle ground approach of describing a rough strategy a lot. I'm here to roleplay, after all, not to act!

[Pelle]

I think "I attempt to convince the guard to let me pass by acting like I belong." outlines an intent approach well enough as "I'm going to try to force the bandit to surrender by disarming him."

If I understand Tanarii right, both would be acceptable. Sorry if wrong. "I roll Bluff!" does not.

The person does select an approach. They select “diplomacy” or “bluffing” or “intimidation” or “bribing”.

For me this would be sufficient to determine what roll to make (ability + skill), but not the DC and concequences. That depends on the sub approach.

The problem is when the DM asks the player to state a good sub-approach within the chosen approach. Such as “now you have chosen to intimidate, tell me your approach as to how you intimidate” and then the DM punishes the player for not coming up with a clever strategy.

I see that this can be difficult for a player when the PC is supposed to be much better. But the fishing for the best approach that you describe can be a fine part of the role playing. That is also not necessarily contrary to Tanariis' method...

My problem with just wanting to roll Diplomacy or whatever, "GM tell me what my character would think and say (to convince the king)", is that it is players giving away their own agency. I want them to make descisions for their characters. This includes player skill in figuring out approach, which I think is a good thing, while PC skill determines if the approach is successfull.