On Skill Deconstruction: The Travesty of Knowledge & Linguistics

Knowledge is a skill reflecting what your character knows. Linguistics is a skill reflecting what languages your character knows. It uses the Intelligence stat as a modifier.

Let's skip the usual format today and talk about what exactly you are doing at your table.

I am confused by the purpose of this skill. Pathfinder has a nice little write up that allows you to use this skill to determine the abilities of monsters you encounter. Roll well enough and you can ask "What's this monsters AC?" or "What type of Damage Reduction does this monster have?"

I am even accepting of the traditional uses listed in the table for this skill (determine aura type, depth underground, common plants or animals) as useless as they are.

I just fail to understand how this is supposed to be used in play. Here are the situations I can imagine.

There is a piece of information. It is either trivial and of no importance, or it is interesting providing some depth and background to the game and not vital, or it is a crucial piece of information.

In any possible conceivable case is the game improved by withholding any of the above information?

The erudite LS of Comma, Blank_ proposed an example in which a knowledge check was made for background information about the crown, when pressed about the benefit of using random chance to determine the item, replied with this:
The benefit is the opportunity for the player to choose their own level of involvement. Setting aside the die roll for a moment, the system you outlined doesn't seem to allow for much flexibility beyond class.
 As discussed recently nearly every edition has some sort of system for allowing customization outside of class. It is totally possible for any class to be a scholar of religion, either using the customization options within your game or simply allowing the player to describe their character. While discussing it, he mentions the same root problem with many of the 3.x skills:
In honesty, though, I will admit that the randomization element isn't very useful for knowledge skills. It can easily be explained, but it's not particularly exciting and the roll doesn't really drive the game forward the way an acrobatics check to dodge a pendulum blade might. I tend to think that there's only a dice roll attached to it because otherwise there would need to be a secondary skills system which didn't require dice. Even I would admit that would be needlessly complicated.
It is convenient to use in 3.x because you're getting all these skill points anyway, but the system sucks.

Should players be allowed to determine their specialties and what their characters know? Yes.

Are there any arguments for the roll improving the game in any way?

Not to be forgotten, languages have always sucked in Dungeons and Dragons. Did anyone ever choose to learn Xorn? How about Elven Cat? There is one stellar improvement in the 3.x system regarding languages from Advanced Dungeons and Dragons, being that they are almost identical systems - it is that in 3.x you have a system for automatically acquiring new languages.

Every time you level up and get skill points.

Even if it only takes you a week.

Regardless it's still an improvement over having no system for increasing the number of languages you know; though I am fond of having a maximum number of languages knowable instead of being able to speak 20+ languages.

Am I being sarcastic here? I don't even know.

(Here's a hint - you can learn a new language by taking the time in game to learn a new language).

Now I know there are some players out there who are just under some compulsion to think a game should model reality - they will try to tax your money in monopoly, politically influence you in chess, and require players to chant and swing their arms to cast spells while playing magic.

So yes, people have varying fluency in language (or crafts. or professions, or etc.). You can represent that using some sort of partial skill system - but in what way does that improve the experience of play?

In spite of my tone, I'm really looking for some sort of spirited defense, some insight about how to make skills tons more absorbing at the table. These d20 skills are certainly my very most least favorite things during play, to the point where 90% of the time it's irritating to even have to roll. I want to be shown what I'm doing wrong and how I'm just not looking at it the right way.


  1. Often there will be multiple sources of most pieces of information. It could be something one of the PCs already knows. They could seek out a sage. They could seek out someone connected to the information. It might be found in a book or letter. Hunting down the information can be interesting play.

    If we were only talking about one piece of information, then a PC knowing it is the least interesting source because it short-circuits the hunt. The campaign, however, is filled with lots of bits of information that the PCs might be interested in. Allowing PCs to know some of them adds some characterization—some connection to the setting—and not at a great cost because there are still plenty of other bits to hunt for.

    Though, I’ve never been a fan of rolling with knowledge skills. The skill score should simply be used by the DM to decide whether a piece of information is within the PC’s scope of knowledge.

    And even still, I can’t really mount a spirited defense here. I prefer to write up some background information. Some I’ll make available to all the players. Some may only be available to some players based on their character. Then it’s up to them to read and recall things. If they want. There’s usually always the other options for them to get anything that might be important.

  2. Yeah, Knowledge skills are also the biggest offender in my own personal pet peeve of skill rolls using Margin of Success while attack rolls and saves are simple Pass/Fail. Which encourages module designers to make at least some interesting and useful information potentially forever out of the reach of the party if the Wizard or Bard doesn't happen to roll really well on the chance to know stuff roll. The Pathfinder modules are sometimes really bad about putting clues that are necessary to fully understand the plot as a DC 30 result on a Knowledge check at low level.

    They also don't really have much of a system for research, which is sort of the major fantasy trope that they ought to be good for (the sage spends hours in the library searching for the solution before time runs out...).

    I frequently have players that wind up taking lots of ranks in Knowledge skills, because they want their characters to be educated in that area, and then they hardly ever need to roll it. Honestly, the two Knowledge skills that wind up getting a lot of use in my game (which is city-based) are Local and Nobility, in which case they wind up being more social/navigation skills than Knowledge skills.

  3. "In any possible conceivable case is the game improved by withholding any of the above information?"

    I believe so.

    Obviously, for plot hooks, key reveals, and such, you're not going to benefit from withholding any information. But in moments where the PCs stand to gain something immaterial--but will have just as much fun without it--then why NOT withhold it?

    Star Wars, for example. A knowledge (dungeoneering) check could be used to reveal that "Hey, don't Dionagas live in trash compacters? If we go down there, we're putting ourselves at risk." If the characters succeed, they have to choose whether to take the risk or fight the stormtroopers. If they fail, hey, fun dionaga fight.

    Not to mention codifying the age-old debate of "Player capabilities vs. Character capabilities" in regards to, as I've often encountered, it's ugliest form: Puzzle solving. Checks provide a way to say, "I'm stumped, but my character would know this like the back of his hand," and have a way to prove it.

    Indiana Jones and the Last Crusade is the perfect example of that. The player has cracked the "name of god" puzzle. "J-E-H-O-V-A," they exclaim! They put their foot on the "J," and it falls through. Rather than have them throw their hands in the air, a knowledge (religion check) reveals "Ah, but in Latin, it's spelled I-E-O-V-A!"

  4. I no longer feel players are "cheating" by using "player knowledge" as opposed to character knowledge. Once a player discovers that green slime will dissolve your weapon if you stab it but will be destroyed if you burn it, it's pretty discouraging to expect the player not to act on that information just because he has a new 1st level character. Justify it however you like --- maybe someone told the noob about green slime. Besides, if you are the DM you can just confuse the issue by changing the color of the slime or adding other qualities (mauve slime that can only be killed by cold?).

    All such skill check problems I would prefer to resolve through 'talking' with the players and then rolling dice --- the better the strategy the players come up with, the better the odds. It even seems fair to 'negotiate' out in the open... "I'll give that a 50% chance of working;" so the player says, "But my dad was a blacksmith, remember, so I helped him out as a lad, and I know a bit about metalwork," so I say, "Okay, a 65% chance then..." and so on. To me, that's part of the fun.

    I know enough Spanish to know that beyond asking for a room or a beer or the toilet, I don't really speak Spanish. In order to make myself into an RPG character, I would have to give myself some (very low) skill rating in Spanish and higher ratings in languages I know --- but language knowledge is often spotty --- I don't know how to speak French, but I recognize the meaning of certain written French phrases since I've done a lot of reading on art history --- how do I model that? A different level of skill points for reading versus speaking? What a pain in the ass.
    Given all that, the 'binary' language system of original D&D(either you know the language or you don't) seems fine to me simply because language, in D&D, is only a means to an end in the game. If the players find a note written in blink dog, they are going to have to find someone who speaks blink dog to decode it, otherwise it will all be "woof woof blink blink" to them. And, although absurd, that's fine to me.

  5. I think like any skill check, the GM should consider the consequences of the pass/fail. In the context of combat the consequences of pass/fail are immediate and obvious. In the "subtler" skills, this isn't the case.

    I love JeffStormer's Indiana Jones example above. A classic fail on the knowledge roll might reveal the "Jehova" clue, prompting the players to trigger the trap. They ask, "What gives man, we figured it out right?" The GM replies, "Perhaps your knowledge roll was not as successful as you thought."

    Information usually comes into play on a longer timeline than other "rolled for" results, but a skill check should always have consequences.

  6. What are your thoughts on using Knowledge skills in a vein similar to 3.X Languages - you either have the skill or you don't?
    Rather than determining the success or failure of a particular course of action, I would think this to have a greater impact on the course of action itself. Having a specific skill in a situation would help the DM answer player questions he/she had not thought of earlier. ("That's a good question: what is the church's stance on the political tension between the two kingdoms? Does anyone have a relevant Knowledge skill?")

  7. @Robert Fisher
    Agreed. PC's knowing little interesting bits, perhaps handed to them on papers or in background notes and reading is good. Saying "You don't know, but here's how you might find out" is also good, but that second part is important.

  8. @Jeffstormer
    I'm afraid I'm having trouble parsing your statments.

    You don't want to withold a plot twist, but do want to withold inconsequential information?

    Your comment seems to indicate to me that players are entitled to bypass puzzles they can't figure out by using skills.

    Remember, there is no railroad or single way things can happen in the OSR, no puzzle will block progress because progress can occur hundreds of different ways.

  9. @Limpey

    I agree 110% with you.


    I am (very) fond of that, or of a simplier system in Skills: The Middle Road of beginner, proficent, expert, master.

  10. I think that you are leaving out the critical type of knowledge that the skill was built around: clues that are helpful but not crucial.

    "Wait a minute! That statue is wrong. Old King Billy was missing his right eye! Try moving the eyepatch."

    "Guys, this culture eats with its right hand and considers the left hand unclean. Don't touch anyone with your left hand, it's an insult."

    "I recognize this. It's devil's snare. You need to go totally limp, and it will think you are dead and drop you."

    All of these represent conflicts that are made easier by a successful Knowledge check, but the check is not necessary to navigate the encounter.

    If you want to just give the brainiac PC all the clues, go with it. Personally, I like teasing it out a bit more.

    Using the Knowledge skill well is definitely a test of good DMing ability. Knowing what information to just give and what information to require a check for is not intuitive, and must be learned. Knowing how to regulate the flow of information so that players stay on the hook and feel rewarded for being smart is also a knack that must be developed. But, the fact that the skill comes with a number of ways that it can be used badly does not mean that the skill is itself bad.

    If you can point to any system anywhere that does languages well, I'd be intrigued. It's very difficult to model "realistically". It is often an area where realism and fun are in direct conflict. I think the d20 variants generally get it right enough.

  11. > In any possible conceivable case is the game improved by withholding any of the above information?

    I'm not for skills, and not addressing your larger question "how make skills more absorbing" Just wanted to point out information is no different than picking a lock, defeating a monster, finding hidden treasure. Why not just auto pick a lock, auto defeat all the monsters? (some games/dms/players/ do exactly that. But then all your doing is story telling.) Each element of the game, skill, class, magic item, level, etc gives the players an opportunity for advantage. **Opportunity**, not guarantee. Thus knowledge skills.

    Also, ignorance / misunderstanding is the source of all sorts of fun. "The Medusa" is just a nickname, you mean the real villion is the stable boy! etc.

  12. Ultimately, the randomization of known knowledge handicapped based on a character's defined skill makes sense if you assume that the GM is not interested in railroading his players.

    If, instead, the GM is interested in creating a situation and then seeing what happens when the PCs interact with it, the question of, "Do they recognize that the heraldry hanging on the wall belonged to the Blood Lord who terrorized the local area 100 years ago?" is an interesting question to ask. (If they do, then they may have some warning about the threats they'll face. If they don't, then they may make a mistake in interacting with the locals.)

    Concrete examples from a recent dungeon I designed include:

    - Can they translate the runes on these marble jars? (If they can, it will them that they're canopic jars. Which may warn them about the mummy coming up in the next room.)

    - Do they recognize the legendary Dagger of Denial? If they do, are they also familiar with the lesser known story of how it was horribly cursed by the Duke of Terror?

    - Are they skilled enough in arcana to figure out how to "jump start" the necromantic shield so that it will be more effective in warding against undead?

    If a character knows some of this stuff, it will result in a certain type of interesting gameplay. If they don't, it will result in another.

    What's interesting for me is discovering in play what will happen. (And, of course, not all of these things are limited to a simple skill check: If they don't recognize the runes, maybe they'll decide to expend a comprehend languages spell to figure it out. Or take the dagger to a sage who can identify it for them. Et cetera.)

    1. I think my question is, why do you need a skill for that? Can't all of those questions be answered without having a knowledge skill?

      In fact, you answer my question in your last paragraph. Your final example is the archetypical example of a die roll eliminating actual play.


Related Posts Plugin for WordPress, Blogger...