How Can We Make RPG Languages Work Better?

questions and answer talk bubbles

Can you write a column about language proficiencies as a game mechanic? Are there any games that do it right?

Language proficiencies (esp in DND5e) often feel like they suffer from the problem of being too situational and thus expensive for their cost, and it being DM dependent makes it both meta-gamey as well as potentially worthless. Is there a way to make picking up language proficiencies as a character worthwhile mechanically? Can it also be anything other than binary states of “Yes, I speak Elvish” or “No, I have no idea what anyone is saying.”

Even worse, in the case of DND5e, spells like Comprehend Languages and Tongues essentially throw the entire system out the window.

Thanks for your time.


Hey Van, thanks for writing in!

You’ve hit on a major difficulty with RPG design, and to be honest, I haven’t written an article about languages yet because I have very little idea how to solve it.

There are two main issues with languages in RPGs. First, even in a fantasy setting, taking an extra language is a major shot in the dark. Unless that specific language comes up, you’ve just wasted your points. The only way around that is for the GM to just tell you that Infernal is going to be useful and someone should pick it up, which isn’t great for immersion.

The second issue is that language barriers are only fun for a very limited period of time. In my experience, it’s about one session. After that, players will get frustrated. They just want to talk to the NPCs and get the story moving! Since it’s rarely feasible to learn a language in that time, the GM always needs to create some contrived way for the PCs to speak the local lingo, further reducing the benefits of spending points on extra languages.

As a result, most games make languages extremely cheap, but even then, they feel like a waste of points. In a lot of d20 games, you can learn a language by spending a single skill point, making it super easy to start with a character that speaks upwards of a dozen languages with total fluency.  And yet, even that small investment feels like a waste compared to more universally useful skills, like stealth or lock picking.

Games that try for a more realistic bent are even worse. In Call of Cthulhu, you’re expected to spend 50 skill points to attain fluency in a language. No thanks, I think I’ll invest in Spot Hidden rather than hoping this will be the game where Mandarin is useful.

In most cases, the best solution is just for the GM to house rule languages as flavor. Maybe give each PC a number of bonus languages for free based on their Intelligence score or its equivalent. They won’t be important to the game, but that’s fine, they can be mentioned once in a while. “Oh hey, you speak Goblin, you can understand this scroll, neat!”

To go more in depth than that requires an extremely robust framework with an eye toward simulating which languages are important and the effects of different fluency levels. I’ve yet to see anything like this, but that doesn’t mean it’s impossible. Maybe there’s some designer hard at work right now on Languages: The Speaking.

Hope that helps!

Keep the answer engine fueled by becoming a patron today. Want to ask something? Submit your question here.



  1. Jenn H

    I’ve seen systems that treat linguistics as a skill. You get specific languages, but can also use the skill to identify and decipher things written in unfamiliar languages (including dead languages). Skill in a language could also imply a greater level of familiarity with the culture that language is from.

    In DnD, the DM could figure out which languages are likely to come up in the campaign and suggest those to players. So if they’re going to be dealing with an orc invasion, knowing how to speak orcish will help. If they’re visiting a dwarven empire, knowing the dwarven language will give players bonuses to persuasion when talking to dwarves, though most locals know common. Knowing how to speak elven may allow the character to see the hidden meaning in a piece of elven poetry, something a person just using comprehend languages would miss.

    • Cay Reet

      It would make sense, I think. If you use a spell or a skill to translate something in your own language, all the nuances in the original text are gone. If you speak that language, you get the nuances as well.

  2. Devor

    It shouldn’t be too hard to develop a working language system. The first step is as mentioned in the article, base it flatly on Intelligence, without competing for any kind of other points.

    The DM should then offer a list of languages that might come up in the setting, and the players could add their own to the list for flavor (there’s no other halflings here, but my character isn’t from around here).

    Then you offer three levels of fluency. Basic, Fluent, and Read/Write. Basic means you can communicate in simple common phrases if you and the other person work at it (not in the middle of combat, for instance). The other two are obvious. If you have three language points you can take basic in three separate languages or put all three points in one. If you really want a robust system, you can group the languages by type, and offer crossover points (if you’re fluent in elven you’re also basic in dark elven… if you can read/write dwarven runes, you can also read gnomish if you’re fluent).

    Maybe you can even have a daft barbarian-type character who’s only got a basic understanding of one language.

    The real key, though, is how the DM uses it in his characters. It’s more useful in a city adventure than a dungeon. You want characters to drop in and out speaking in different languages. It may be the kind of system that’s really only practical in a video game where you can computerize it and it’s more sensible to the time to work it in the story.

  3. Michael Campbell

    One of the oldest rules of being a good Game Master.
    Read the players’ character sheets and see where they intend for the game to go.

    If someone wants to be an dwarfish C3PO; run with it.

  4. Kenneth Mackay

    The Basic Roleplaying (BRP) system has each language as a separate skill. Your percentage in the skill is your chance to decipher a written document or overheard conversation in that language, but if trying to converse in the language both speakers add their skill percentage to give the chance of communicating accurately – which makes it worthwhile to have a smattering of several languages.

    There are options for separating spoken languages and written ones (so a character could decipher Egyptian hieroglyphs but not be able to speak Ancient Egyptian, or speak Cantonese without being able to read or write it) and of giving some chance (a fifth of your percentage, I think) of understanding a related language (so your 50% skill in Orcish gives you 10% in Goblin).

    Like any other system it needs input from the GM, who has to add or subtract modifiers to reflect the difficulty of the document or speech being translated, and decide just what you and the other party think they heard if a conversation roll is failed…!

  5. Wil

    The easiest solution is that languages only become relevant when it’s interesting and the fiction dictates. If both conditions aren’t meet, than it’s unimportant.

    As for how, you just have to look to games like Fate. There’s no language skill… But if my character has the aspect “Served in the French Foreign Legion” it’s a good chance my character knows French. If they have another aspect, “Can’t Leave the Streets of Shanghai Far Enough Behind” it’s likely they know Mandarin and Shanghainese. If they wind up in South America, they can spend a Fate Point to invoke their Foreign Legion aspect and declare the detail, “Remember that Brazilian Legionnaire? I picked up some Spanish from them.” Conversely, the character who Grew Up In A Bomb Shelter wouldn’t be able to do this. Or, the GM can invoke compel the aspect instead for the character’s grasp of the language to lead to some misunderstanding or faux pas.

    So this takes care of the mundane – and frankly boring – part of languages. It’s a temporary or interesting setback at best, with no need for skills and rolling to understand one another. People always find a way to communicate, so it’s pointless to bog things down with it.

    The real question becomes – when does it become interesting? That’s when the PCs need to figure out how to shut down the self-destruct sequence on an alien ship. Or decipher the coded message sent in a near-dead language. Basically, when there’s actual stakes involved that don’t involve just a language barrier.

    • Cay Reet

      Good point – although they wouldn’t have picked up Spanish from a Brazillian legionaire – Brazil is the only country in South America where the native language is a form of Portuguese.

  6. Overprepared GM

    Another way to handle it is assume a world where most people (especially the adventurous) speak a trade pidgin language, but there are a bazillion different languages and dialects. Trade pidgin lets you slowly get basic meaning across, but loses all nuance and social context. How many languages you speak and how well you understand them and their slang and such is based on a skill like streetwise in dnd (depending on your game system. In mine, the skill is “society”). So if you’re trying to overhear the guards talking or bargain for armor or convince a priest to noble to sponsor your quest, you first roll to see how well you can communicate. High roll means you get a bonus to the next interaction you’re eloquent and insightful in a language they speak. Low roll means you get a penalty. You get the gist, but are clumsy about it. Medium means neither. Most players have backgrounds where they understand a couple specific languages, and the GM can grant the bonus automatically if they come up.

  7. Archon

    GURPS 4th Ed. has what I consider a good solution. Languages are cheap, variable-cost non-skill advantages. They come in three levels: Broken, Fluent and Native. They are also split into Spoken and Literacy. So you could have Native Spoken fluency and Broken (or no!) Literacy fluency in the same language, or vice versa. You get Native Spoken fluency in your native tongue for free; if your character is at a time when literacy is common, you get Native Literacy in that language for free as well. Additional languages cost character points (and if you have less command of your native tongue than you should that’s a disadvantage). Individual languages are inexpensive, but knowing a large number of them can get spendy; ditto for reading them.

    Broken lets you get along in the language; Fluent gives you full command though you have a strong accent and may not understand slang, metaphors and so on; Native is what it says though you still might have a slight accent.

    You can then augment this with skills to create poems, stories and so forth. Or to find the allusions, hints and hidden messages in them. Virtually everyone can write a story or compose an poem but it’s likely to be dry, dull and boring if you don’t have some skill at it.

    • Cay Reet

      I like that system, because it makes a lot of sense. You can speak a language good enough for daily life, but be horribly bad at reading it. Or you can read a language, but not speak it well – letters on a page might be easier to interpret than spoken words, especially as dialects and accents are a thing. For a language like Ancient Egyptian, for instance, reading it might be a lot easier than knowing how to pronounce it, since it’s not spoken any longer.

  8. Erynus

    I always took the approach of letting the Npcs do the work, there is no need to know a language if you can hire a translator. Also that gives them someone important to care of. The problem i see is that an specialized linguistic character is a niche character, she just can shine in situations where everyone else cant do anything.

  9. Deana

    Back when I played AD&D or 3.5, I occasionally played with a 10 point gift to players with a 2-to-5 point spend/ language house rule, rather than their excessive language gifts.

    2 points will get you basic spoken language skills (where’s the bathroom? can I have a beer? etc.)

    2 more points will get you the equivalent of a 200-level language course (loosely defined as the ability to speak about past actions and future actions) (spoken only).

    1 point for writing the language.

    This keeps players from playing a ridiculously multi-lingual character but lowers the language cost, and also replicates “real life.” Fluency comes in stages, and it would be fully possible in a multi-lingual culture for you to know enough to negotiate a sale or rent a room, but get bogged down in the details of a story told by the bard at the local pub and not be able to read the written language at all.

    Some people would need full fluency in some languages, while others would want to concentrate on getting by in multiple languages.

    Magic users, clerics, and elves got 20 points and everyone started with just their native language, spoken only.

  10. Erich Flack

    Interesting article and comments. Love this site.

    What about tying language proficiency to specific cultural lore? Someone who has spent time learning about a culture is more likely to try and pick up some of the language. If your native culture, or at least your character, is literate the GM could grant that proficiency.

    It wouldn’t even have to be the full skill amount. Maybe putting three dots/points in “Culture: Plotopia” gives you two points in “Plotopian: Verbal” and one point in “Plotopian: Written”. If the studied culture is dead or far away, you could switch it so the higher language ability is written, not spoken. Maybe a merit “Flair for languages” would give you your full skill in the language.

  11. Rose Embolism

    Campions 2nd Edition had an amazing language system created for it’s spy game Danger International. It grouped languages by similarity, and then had five levels of proficiency, going from a smattering of words, to idiomatic native knowledge with a grasp of dialects. Each cost 1 character point (out of a standard character budget of around 100) But wait, there’s more- depending on how closely two languages were related, one got a reduced cost for learning the second language, or even get a basic proficiency.

    So for example, someone who knows Latin would get a -1 character point cost to learn Spanish and French, while a Russian Speaker would get level 2 in Ukrainian, for free. On the other hand, an English speaker tryiignt o learn say, Mandarin, would have to spend an extra character point. And then there’s the Linguist trait wohhich would reduce the cost of learning languages by one point. Handy if your agent is a polyglot.

    So for an example, a Mandarin native speaker wanting fluent (level 4) Russian, would pay 5 points. But to learn Ukrainian to level 4 they would only spend 2 points.

    As I said it sounds complicated, but with the provided chart, in practice it was easy. But it was too realistic for later versions of Champions, unfortunately. Something similar could be done for a fantasy novel or game as well, for somebody wanting to do LOTS of worldbuilding.

    • Cay Reet

      That system makes a lot of sense. I especially like the idea of grouping the languages and give you the chance to learn some more easily or get a little grasp on, if you know another language of that group. I know that after learning French in school, I could at least successfully guess quite some Italian and Spanish words, because they were similar enough. Didn’t make me fluent in Spanish or Italian, but I still could make an educated guess at simple sentences in those languages. Had I gone for one of those languages afterwards, I probably would have had it easier to learn it.

  12. Adam

    Shadowrun does a pretty good job with languages. Your character gets a native language for free, and extra language points based on your Intelligence. You can spend all those points on one language (so you can speak Mandarin really well, for example) or divide them up so you can speak Mandarin pretty good, and a little bit of Japanese, too. From the rulebook:

    “There are few situations where language skills should
    require a dice roll. Characters with a language skill don’t
    need to make tests to understand one another in every
    day situations. The character’s skill level serves as a
    benchmark for how well they can communicate in a foreign
    language over time. However, in critical situations
    where precise translation is important, a gamemaster
    may elect to require a Language skill test.

    A gamemaster may elect to call for a Language Test
    anytime information needs to be translated hurriedly or
    in a tense situation.
    Failing a Language Test means the parties are unable
    to understand each other. If a glitch is rolled, some
    meaningful portion of the information is presumed to
    be understood but is actually misunderstood. The gamemaster
    may want to make the Language Test secretly,
    in order to maintain the illusion that the characters understand
    what is going on.”

Leave a Comment

Please see our comments policy (updated 03/28/20) and our privacy policy for details on how we moderate comments and who receives your information.