Monday, April 20, 2020

Dungeoncrawling: Languages

The Problem With Languages

Here's how a lot games use languages.  

Choose a language at random from this list of languages.  You don't know which ones will be useful.  If that language crops up in a game, then you get some bonus information.

That's a pretty shitty implementation.

1st Problem: Choosing languages is usually a blind choice.  You might try to predict what languages will be most useful, but if you're choosing it at character creation, you don't have enough information to make an informed choice.

2nd Problem: Groups will arbitrarily have different experiences in dungeons.  Due to random chance, one group of players will be able to talk to the kobolds in the dungeon, while another group will never have that option.

(Not that I'm against different groups having different experiences, but they should be the result of players choosing to not talk to the kobolds, not being excluded from the possibility.  And the dungeon designer should have a clear answer to this question "is this dungeon better or worse if we can talk to the kobolds?" and then design a dungeon that supports that answer.)

3rd Problem: Information is one of the resources a smart party stockpiles.  It is fuel for their schemes.  It is salvation when they are on their last legs.

Locking information behind an arbitrary language barrier is counter-productive.  The players never have a chance to earn the information--it's either available or it isn't.  Without the chance to earn the information, it turns into a coin flip.  Either the module is better with the information or it isn't.  Pick one.  (Hint: usually, the answer is just to give them the info.)

If the information makes the game better, give it to them in Common.

If not, don't.

Languages in the GLOG

Everyone speaks Common.  No one speaks any other language.

Languages are treated like a Skill, and take up a Skill Slot.  The only difference is that they improve twice as fast (gaining +2 per improvement instead of +1).

There are three implementations of languages.

1. The information is available in Common, and is accessible to everyone.  (Good default choice.)

Before you leave the lighted realms of free info, think about what you gain by locking the information away behind a foreign language.  Is the game improved by this sequestration?

2. The language communicates occasional, optional bits of information, and the PCs have time to learn it if they wish.

Learning a language my allow access to a few foreign NPCs, it might allow you to speak to the goblins without an untrustworthy interpreter, or it might allow you to read the dungeon graffiti.  This only works if:
  • The foreign language isn't used to communicate anything essential, and there is a minimal penalty for not knowing the language.
  • The PCs have the opportunity to learn the language.
  • The PCs spend enough time here to learn the language over multiple sessions
Note the third point.  If the party doesn't have opportunities to learn the language, it doesn't matter that the language is learnable.

3. The information is inaccessible and there are no opportunities to learn this language prior to encountering it.  The party will have to (a) go to a Library to research it, (b) hire a linguist, or (c) use magic to decipher it.

I'm less and less a fan of #3 these days.

Not that it isn't fun to drag a complaining linguist through a dungeon, but there's an opportunity cost to be considered.  It takes a lot of time, words, and attention to find a linguist, negotiate the cost, keep the linguist alive in the dungeon, and extinguish the linguist when they catch on fire from reading the forbidden words.  It's not bad, it's just. . . are there better uses for your group's time and attention?

Secondly, what information would be better delivered in town, than in the dungeon?  Because delivery information in the dungeon (point-of-use) is probably preferred.

Maybe for dramatic pacing ("My god!  Do you know whose tomb we were in?")

Maybe for introducing quests ("According to this book, Zharkhoon was buried on his golden barge with his jeweled monkey sewn inside his chest.")

But even then, it's usually more satisfying to let hit points, torches, and spells be the reason that the party returns to town, not an inability to translate an inscription.

A library is a decent compromise, though.  It can be a good way to give them dungeon-related things to do while they are in town.  For example: "I need 10s up front.  I won't have time to translate it until tonight.  Come back tomorrow morning." 

And of course, magic is always a limited-used resource.  If a party wants to use magic to gain an advantage, that's always acceptable.  (If it wasn't, the DM should never have given them the magic in the first place.)


  1. Interesting thoughts. I'm experimenting with languages right now, with a sort of cross between your "EVERYONE speaks Common" and Skerples' notes on how monks aren't just literate (lots of non-monks can read and write), but Literate. The result is probably closest to Scenario #2.

    Basically, "Common" is a trading sign language (inspired by Plains Sign Talk) which pretty much everyone knows and, in most cases, will cover everything you need. Knowing someone's native language is largely unnecessary but sometimes useful: If the only language you and the kobolds have in common is, uh, Common, then you can parley, ask for directions, etc., but complicated subjects will require you to describe things creatively *and* keep them interested (e.g. the kobolds will be more willing to sit through your painful circumlocutions if you're talking about traps, or maybe your plan to kill owlbear that's been eating them).

    (It's a sign language because (1) I think it *feels* different to have your character signing rather than speaking in these situations, (2) it gives the players another tool because they know from the beginning that they have a robust non-verbal method of communication during those times that they have to be quiet, and (3) Plains Sign Talk is great).

    1. The signing is fun, I'll admit. And I like the idea of giving the clerical cleric the superpower of being an effective communicator. But then that necessarily means that everyone else is *not* an effective communicator. It (potentially) locks the other characters out of a roleplayed conversation.

      It reminds me, a little of how, once Diplomacy rolls were codified in 3rd edition, everyone got locked out of social encounters except the bard. Or how everyone got locked out of disarming traps once the Thief was introduced (allegedly--I wasn't there for that one).

    2. I like signings because it also means you have to not be holding a weapon, so that makes putting your weapon away to negotiate a clear way to transition from fighting to negotiating back to fighting.

    3. @Arnold At least in my games, people tend to default to "just one person is the communicator" anyway, though I *am* looking for ways to tinker with that (I'm running a GLOG campaign for the first time and *explicitly* stated that there are no social skills or anything like that, and CHA doesn't necessarily mean you're going to persuade people any better than otherwise, and the party dynamics still shook out how they usually do, lol.

      "Or how everyone got locked out of disarming traps once the Thief was introduced" Huh. Yeah, I can see that. It's something for me to keep in mind as I continue to tinker with skills.

      @Unknown "You have to put your weapon away" is another thing that I like about signing, but I was starting to worry that my comment was turning into its own post. >:P It's good to see that some of the implications are as intuitive as I hoped!

  2. I have an idea. What if we use a Story-Game style approach to languages?

    Say you have 2 Language points at character creation. The first one gets spent automatically and you now know Common, which is what most civilized people from "Around Here" use. It's your local lingua franca. But you keep the second point.

    Then, at any point during your journey, if you encounter, say a group of Orcs who only speak Draconic and Orcish, you can spend that language point and say to the Referee, "I actually speak Draconic." You should, of course, have to justify it. Ex: "I learned it in Wizard College." And of course, some languages will just be impossible to learn because they've been dead for centuries or were never written down or speaking them aloud causes mortals to bleed out every orifice and die, etc.

    1. Satisfying from a story game, perspective, I guess.

      Boring from an efficacy perspective. You have language points that can only be spent on languages--nothing else. Therefore, when there is an opportunity for someone to say "I speak orcish", someone in the party will chime in and say "I speak orcish". There's no interesting decision there.

      Additionally, the DM will probably design encounters for people to spend their language points on. It's not so different from the DM deciding, during character creation, who speaks what language, and then writing the encounters accordingly.

    2. The only difference is that the DM can write "okay, this dungeon has 5 foreign languages in it, and I have 5 characters, so we're good" and the language becomes merely window dressing. It's no different if the DM had just made everyone speak Common and then given every PC a different-colored hat.

    3. I feel like I'm judging too harshly here. My own advice was "why not make everyone speak Common" and then my criticism of your implementation was "it's basically the same as everyone speaking Common".

      Language points are fine. It's very similar in practice to "Everyone speaks Common" which is something that I advocate, and is perhaps a little bit more satisfying on the story-game front, with the slight illusionism of retconning your own abilities.

    4. I'd say this option still provides choices. Players aren't necessarily going to spend a point on every language in the dungeon. Just because you put kobolds in the dungeon doesn't mean that players are even going to try and parley with them.

  3. One of the early house rules we adopted was to pick languages during the game as they cam up (so during char gen the players note how many languages they know, but they may ad hoc declare that one of those languages is the one needed at the time).

    I absolutely agree about the 3rd problem. I think if knowing a language leads to maybe extra information (like Gumshoe's non-essential clues), then it's fine.

    The second problem is not something I can agree with. Party composition may already severely limit the capabilities of the characters (i.e. whether fighting, sneaking, or trickery is viable) - but that's not arbitrary: it's a direct result of the choices made during character generation.

    At any rate, I don't think I've been using languages to their fullest potential. A little bit of extra information here or there, money sink in the terms of sages and translators, and +1 on reaction rolls if the party is able to converse in the creature's native tongue. Surely it could do more.

    1. I think a key difference between our implementations is that you are using languages as something derived from chargen, which is cool.

      "You have 18 Int, therefore you get +3 languages, therefor you get +1 on certain reaction rolls and have a 3/6 chance of understanding dungeon graffiti."

      That's fine. You're making the Int stat more useful.

      I'm interested in languages as a skill. How a character learns a language, the cost of learning that language, and how that language is leveraged once learned. So we are answering almost entirely different questions here.

    2. I feel there's a distinction to be made with your objection to the second point. When a player chooses to play one class or another, they know ahead of time what opportunities they're making more difficult for themselves. With languages, it's often more of a random draw

  4. I think speaking a creature/persons native language should give a bonus on reaction rolls

  5. Languages giving bonuses to reaction rolls is kind of the Tolkien model - everyone can speak to everyone else, but various elves and dwarves are delighted if you can speak even a few phrases in their language, speaking the language of Lothlorien would let the fellowship talk to the rest of the borderguards if they knew it (not just the one translator), the Rohirrim *can* speak Common but their king is suspicious and looks unfavorably on people who speak foreign languages, and knowing a dead/evil secret language gets the party a useful but non-essential clue. I think this strikes a nice middle ground of languages mattering without gating off information for no good purpose.

    I also think this model integrates fairly well with the illusionism of retroactively declaring what languages you speak because 1) the number of languages actually in use is fairly limited, and 2) you're deciding which bonuses and extra clues you want (and you can speak to whatever it is in common first, to suss out whether or not you the bonus is likely to be useful), rather than the non-choice of using a language-shaped key to open a series of language-shaped doors. I don't think the illusion is too bad here either - the justification either being "I already knew this language, but this being the first time we've encountered it in game it obviously never came up" or "having previously encountered the language and not gotten the bonus, I thought it would be useful to brush up on my conjugation". The advantage over learning it as a skill I think is that it's quick and becomes immediately relevant at point-of-use.

    Characters can derive language points from Intelligence (or whatever), and then your wizard/scholar/monk/sage class can get some extra ones - languages become a universal tool, but specialists have more flexibility. Alternatively, everyone gets the same language options, but specialists get a bigger bonus: Frodo knows enough elvish to make a polite introduction, but Aragorn knows enough to carry on a conversation.

    1. Not that simulating a Tolkien aesthetic is a goal for most games, but I think there's a nice flavor to having languages in certain map-fantasy worlds. If you were going for a fairy tale or gonzo science-fantasy vibe, then I think the assumption "everything speaks your language except for maybe a handful of creatures (Martian tripods, maybe) that are totally incomprehensible" is the way to go.

      For Centerra it seems like it'd make sense for everyone to speak the church's language, except maybe for very old people. Their grandchildren could interpret for you, but knowing the language might still be useful (perhaps a local myth doesn't translate into Common well, or maybe the goblin matriarch is a lot smarter than her children are).

  6. I've struggled with this one for a while, and I think my latest approach is the worst possible. I let the players pick their languages out of the book, and the whole campaign takes place in a 'lost world' with strange cultures & humanoids they've never seen before. Nobody speaks any of those languages they picked anyway!

  7. I have a "Languages" skill, and when players encounter a language for the first time then everyone rolls their Language skill, with some + or - depending on the situation. Whoever is successful speaks the language.

    Typically I make some sort of 'Common' that can be understood by most in at least relatively simple context. Speaking to someone in their native tongue usually offers other benefits - they will probably be friendlier.

    Fluency in language often means familiarity with culture, so I can assume that players speaking to the Goblins in Gobblish will be aware of specific cultural faux pas and will be better able to behave themselves in ways that the goblins find acceptable.

    1. It sounds like it's effectively becoming a reaction roll. If no one succeeds on their Language roll, it's harder to get a friendly reaction.

  8. Vayra at The Mad Queen's Court has an interesting approach to languages in that some of them seem to provide a mechanical effect in line with that of the language's flavor, making them more of a skill to acquire. There is, for example, the Trader's Cant (the common language) which imposes a -2 penalty on any interactions other than haggling when using that language or the High Charter, which is so incredibly deep and complex that allows an Intelligence roll instead of a Charisma check to influence other speakers. All the languages are setting specific, as all languages should be, and they contribute a great deal to world-building, but I guess making them a skill to learn with an added benefit/restriction is an interesting approach.

    Maybe that could be one of the solutions. Speaking a language means, after all, looking at the world through a particular lens. Perhaps you cannot be shoved around or knocked prone when grinding your teeth to communicate with stone, but at the same time you become so heavy and thick and incredibly stubborn that you sink to the bottom if you try to swim or must pass a Wisdom/Will/Presence check to follow any plans not suggested by you.

    Or the Inquisitorial language, which makes it unable for you to tell a straight lie but allows you to ask yes/no questions with a Charisma/Language check that the "suspect" must answer truthfully (everyone you talk to is a suspect when speaking this language).

    This would make languages a treasure of its own kind perhaps and getting to learn them could become its own adventure. As per the skill system, I don't know, they just get the bonuses to whatever thing the the language allows them to do or something.

    1. Huh. I really like the idea of a language giving you a benefit beyond just communication. Like, if you speak Neropian, you can read the words on the ancient monuments, but you can also speak to the dead if your skill gets high enough.

      Neropian - speak with dead.

      Elvish - point to point encryption (can shout secrets in a crowded room, no one but the person you are talking to can understand you)

      Orcish - bonus to save vs divine spells

      Dwarven - +1 HP

      Honestly, the idea of "learn dwarvish, you'll get an extra +1 HP" is such a fun one for me. I don't think I'd use that one, exactly (too synergistic) but I still love the idea.

  9. I think a lot of the issue is that languages are generally done in a "everyone speaks common" in one camp and then "only a very specific group speaks this language" in the other camp. What I think would make language more useful and interesting is if there were a group of other "commons" you could learn. Eg. the orcs/goblins/kobolds/trolls etc all speak Outlander which is the common that exists outside civilization so knowing it guarantees parlay (or at least the chance to offer it) instead of hoping they have someone who speaks common. Dungeons tend to have writing in The Old Tongue on them (effectively latin) which is also what dragons and similar ancient creatures speak which lets you try to read the description of the old pedestal with the nondescript ring on it. Animals all have their own simple tongue for those who want to put in the time to learn it. The etc. This way language choice isn't randomly choosing a group of people, it's a set of skills that makes you better at certain specific things like any other in your toolbox. Diplomacy is a lot easier when you aren't both doing the equivilent of "Me" *points to self* "El Looko" *mimes looking around* "magic-o.. anillo? yeah, ANILLO!" * points to hand *

  10. I tend to say everyone _can_ speak common (or at least has an interpreter), but because of politics, fluency, or general distrust of outsiders will speak other languages when they can.

    If you need to speak to them, Common will do, but knowing the right language opens up options like eavesdropping or speaking to the rank-and-file directly.