Knowledge (Full Description on PFSRD)(-C’s Post): I’ve always liked knowledge skills, though I’m not certain why. Perhaps it’s because most games, particularly video games, consistently put the player in the position of the most ignorant person in the room. The rationale for doing so is obvious. Representing the player with an avatar who is ignorant of what is going on around them is a useful tool for justifying exposition without breaking the fourth wall. It’s lazy storytelling, but it works well enough in a pinch. Problem is: I don’t like being ignorant. I think that, perhaps, when I first encountered knowledge skills I was excited by the prospect of having a measure of control over what my character knows about the world.
I’ve always maintained that a knowledge mechanic is beneficial to the game. However I must confess, -C’s arguments forced me to seriously reflect on whether or not I was right. He presents a legitimate concern: does the inclusion of a knowledge check lead GMs to restrict information in a way which is not conducive to creating a fun game? Is rolling dice to determine a character’s knowledge of a subject beneficial even slightly, or is it nothing more than a side-effect of the knowledge mechanic being shoe-horned into the skills system? At the time I even said as much myself: “the randomization element isn’t very useful for knowledge skills […] it’s not particularly exciting and the roll doesn’t really drive the game forward.”
My thinking on this matter began to change after a recent experience at the table. My players were investigating a dungeon which had been specifically built to keep gnomes out. In one room, there were a number of statues of gnomes, which were part of a larger statuesque depiction of a battle. Nearly all of the gnomish statues, however, were actually real flesh-and-blood gnomes, who had been turned to stone hundreds of years in the past. My intent was that the players would not realize this until the end of the dungeon, where they would encounter a large number of gnomish ghosts. Once the ogre who had murdered those gnomes was killed, the ghosts would be able to tell the players about their petrified kin on the floor above. My GM notes for that gaming session were actually posted on this blog, if you’re interested.
When my players entered the room with the petrified gnomes in it, one of them actually guessed immediately that the statues might simply be petrified, and asked if there was any way they could know for certain. At first, I was a little flustered. I did not want to deny my player the right to investigate his hunch–particularly because he was right! But I also did not want to hand him information which he had no reasonable way of obtaining. The builders of this room had gone to some lengths to disguise the statues as genuine, even going so far as to include a number of real statues. And if I set a precedent of answering player’s guesses with accurate information, I rob the players of the ability to take actions based on an incorrect guess.
I then recalled that the player had put a number of points into Knowledge(Dungeoneering). I asked him to roll that to determine whether his character knew of any method to differentiate between petrified beings, and finely crafted statues. I set the DC at 15, and the player succeeded. I then confirmed for him that, yes, these statues were in fact petrified gnomes. A fact which the party managed to learn long before I had intended them to, and which proved to be quite relevant when they continued through that dungeon in the next session.
In this case, the Knowledge skill did exactly what I want a skill to do. It took a situation where the outcome was uncertain, and it resolved it with a minimal amount of wasted time. It allowed the players to succeed in a way which I had not intended them to be able to succeed. Even if the roll had failed, demonstrating that none of the players were able to determine petrified creatures from statues, it would have simply created a different interesting situation. One in which the players needed to seek out the answer to their question
Judgement: Like crafting, knowledge works well in its current state, but isn’t good enough. Also like crafting, I plan to devote some of my upcoming time to a complete redesign of the knowledge skill, in the hopes of making its ideal function more apparent, and shaving some of the fat from it.
Linguistics (Full Description on PFSRD)(-C’s Post): To my knowledge, linguistics is unique amongst skills, in that its primary use (learning languages) is accomplished by putting points into the skill, rather than rolling skill checks. Which isn’t to say skill checks are never made with linguistics, simply that none of those checks would (on their own) justify this skill’s existence in my eyes. Aside from learning languages, you can create or detect forgeries, and you can also decipher writing in an unfamiliar language or dialect. Officially, that’s all you can do with a skill check, and neither of those two is very likely to come up frequently enough to justify the skill’s existence.
Regarding the creation and detection of forgeries, how often will this actually show up in a game, honestly? In all my years of playing and running both Dungeons and Dragons, and Pathfinder, I don’t believe I’ve ever had cause to forge a single document. I’ve never been to a city which required me to carry identification papers, nor have I tried to deliver false orders to a group of soldiers. I’m not saying these occasions will never arise–they might! But will they arise often enough for this use of the skill to provide a significant frequency of use? No, probably not. That’s why Paizo combined the old D&D 3.5 forgery skill into Pathfinder’s linguistics skill, and that’s good. But if there is no other justification for having the skill, then creating/detecting forgeries wouldn’t provide that justification.
The other use of the skill which calls for rolling dice is deciphering writing. Like forgery, this was also a stand-alone skill in D&D 3.5, called decipher script. The idea is to provide players with the chance to interpret written languages they don’t know, or writing which is otherwise incomprehensible to them. It’s like reading Chaucer for you or I; it may be English, but understanding it is a difficult skill to master. Indecipherable writing is a challenge players may face somewhat more often than forgeries, but still not something they’re likely to encounter often. And even in those cases where it does come up, there is more often than not going to be a simpler and more entertaining way to solve the problem: finding someone who is fluent in the language for example.
I propose a new feat; Master of Writing. Players with this feat can decipher any written language with perfect accuracy after 1 hour of study. They are also able to create and detect forgeries with perfect accuracy, so long as they have seen a copy of the “real” document at least once before.
Which only leaves one purpose for this skill: learning new languages. Since that doesn’t require any rolling, it seems obvious to me that this skill has no justification for remaining in the game. But how, then, should a character learn new languages? In D&D 3.5 this was handled by spending a skill point, but that hardly makes sense. Normally, spending a single skill point allows you to become roughly 5% more likely to succeed at a given task. So why should skill points suddenly become so potent when it comes to learning new languages? And as I learned with The Owlbear, there aren’t even enough official languages in the game to support this system to its logical conclusion.
Additionally, why should characters be allowed to learn languages so quickly? I’m not suggesting that we force players to waste in-game years learning languages in pursuit of realism, but why not attempt a middle ground? A character who wishes to learn a new language must either be traveling with someone who speaks the language fluently, or must purchase a book for 10gp which teaches the language. The character must spend 8 hours of light activity studying the language each day (during which time they cannot craft magical items or perform other tasks). After 1 month, they gain enough of an understanding of the language to read, and hold conversations.
Judgement: House rule this out of the game. Languages can be learned by spending time resources, rather than skill resources.
Knowledge and Linguistics both ran a little long, and the next skill on the block is a doozy, so only two skills for this post!
See I would argue that your example of the knowledge skill working perfectly is an example of how it also becomes, as C- puts it, a skill tax. Rather than have the players come up with some explanation as to how or why they would determine a petrified gnome from a statue (Stone to Flesh spell, dispel, perhaps the petrified gnomes are far more detailed than the gnome statues, or the petrified gnomes look perfectly crafted while the rest of the statues look somewhat rough or unfinished), someone simply has to pay the skill tax to have the right skill to reduce discovery to a mere die roll. Of course the players had to come up with the idea to check in the first place, but if that’s the critical action taken, then why bother with the roll at all?
That isn’t to say that this isn’t a perfectly valid way of doing things, and certainly if there is going to have a knowledge skill and check, this is where it would be used, but I think I would have asked the players to come up with some sort of test or examination they could do to determine if any of the statues were petrified gnomes and allowed success or failure on that. If a player might have reasonable experience with stone work, I might suggest hints, or allow a plan that wouldn’t normally succeed to work. I suppose though this all boils down to rules for adjudication vs DM adjudication.
I think this is a really good example of a knowledge skill roll enriching the game, and actually makes me question my anti skill dogma somewhat. If this were my game, if they asked directly, I probably would have said something along the lines of the statues look so incredibly lifelike that they certainly could be petrified gnomes, but I probably would not confirm or deny anything absolutely without something like a detect magic (which would actually be an arcana check in my current 4E hack game, but if I was running something like B/X it would require a prepared spell). And even the, I would probably just say that they are lifelike and radiate magic and let the players draw their own conclusions.
One thing I would ask: could this not be something resolved by an intelligence check, assuming the PC has a somewhat appropriate background? For example, is a gnome or is a magic-user.
Languages is a good example of something that probably deserves its own little subsystem rather than being forced into the skill system. Older D&D used to just give you a number of languages based on your intelligence score, which seems to work pretty well in practice. The LotFP method is even more elegant. Your intelligence determines the max number of languages, and you roll during play when you want to determine if your character knows a particular language! If several languages make sense for a character’s background, they would start with those, but all others would be determined in the game. From the free Rules & Magic book, page 37:
When a character comes into contact with another language, their chances of knowing the
language is 1 in 6, with the character’s Intelligence modifier applying. If a character has a Languages skill at a greater level than 1 in 6, use that as the base chance instead.
…
A character gets one attempt to know any particular language. If that one attempt fails, the character simply does not know the language.
Personally,I disagree with the language per skill point thing, it’s ridiculous. Personally I say keep the Linguistics skill, minus the language per skill point, and give each individual language it’s own ‘Speak Language’ skill, keeping it at one skill point per language. In this way, you don’t NEED a lot of skill points to learn new language, but you also can’t just speak a whole ton of languages just because you are really good in ‘English’ class (Linguistics skill.) You gain one free skill point in your ‘base’ languages (that is, one skill point for each racial (like common and elven for elves) and one free skill point for your free language per intelligence modifier number.)
I think Linguistics can have some fun aspects – I played a rogue in one game, and our party ran into some guards who spoke another language that none of us knew. I rolled Linguistics, and learned that the guards planned on arresting us for some crime committed.
Reasoning?
“Rogues know a few words in every language: crime, escape, and plausible deniability.”
I actually really like the how you get something immediately with Linguistics. I’m seriously considering in designing my classless RPG to have a skill system where every rank in a skill grants you a feat related to that skill.
I love Knowledge skills and totally agree. I also find them a great way to give exposition without boring players. I can casually mention a painting on a wall and a curious player might ask if they know anything about it. They feel clever for investigating and they get to roleplay their character being educated.
I have a problem with -C’s argument that withholding crucial information generates bad gameplay. If a GM is using Knowledge to bar information crucial to the success of the players, that’s not bad skill design — that’s bad GMing. Like how The Alexandrian mentioned in several of his articles, a GM should almost never design his adventure where the entire game hinges on the binary success of a single skill check. It doesn’t matter that skill is.
-C’s argument (and I’m pulling this from memory, so forgive me if I butcher it) is essentially that information is either crucial (in which case, withholding it is bad GMing) or it’s fun and interesting (in which case withholding it prevents the players from fully experiencing the game, and having a mechanic JUST for fun tidbits isn’t really an effective way to design a game).
That said, I still think there’s room for Knowledge skills. I wrote more about it earlier this year, though looking back on that post I’m not entirely happy with it. It’s an issue I still ponder now and again.
Very late on commenting, but I thought I’d say it anyway.
I think the problem is more that the player wanted to ask a question their character is unlikely to immediately ask without having previously encountered petrified creatures or creatures that can petrify before in some way.
I’m not trying to be overly critical of your style, but I think you should have said “your character has no reason to have that idea immediately, but you can examine the statues with a perception check to find out if you see anything noteworthy” for example an surprising level of detail or a difference in general detail between statues.
Should they succeed at the check, allow them to make an engineering check (if they wish) to determine their opinion on the impressiveness of the craftsmanship and/or their perceptions on the viability of believing that the statues were crafted by normal means. Likely the most they could discern is that the statues are of better quality than any they have ever seen before. If they are trained in engineering, their check may also indicate whether they believe the statues were made through mundane means or not.
If they are a character that would reasonably have been exposed to myths or stories of petrifaction they may make a dungeoneering or arcana check to try to recall if they’re even aware that petrifaction is possible, otherwise (or if they fail the check) they should simply think “that’s an extremely detailed staue, never seen one so well made”.
If they successfully recall a means to petrify creatures then the player may decide what their character thinks happened, but they certainly can’t be sure.
Unless, their character has previously been exposed to petrifaction either in person or through study and could reasonably be expected to try to identify the difference between a petrified creature, an ordinary statue, and a magically crafted statue, they would have no means of knowing the difference. Their check may still result in an inaccurate conclusion between the three options.
It would also depend which check they choose as their means of deciding (they may choose the wrong check to determine the true means).
Fundamentally, it seems – and I think – the issue was that you allowed a player to use knowledge they had to influence how their character reacted to a situation too much. If as the GM/DM you don’t think a player’s character should be able to think up an idea (especially one that influences the story in a way you don’t want to allow) don’t even let them try, because they’re meta gaming.
Hope you take this as constructive criticism, and I’d be happy to hear back from you or anyone, whether you agree, disagree, or somewhere in between.