Tabletop RPGs address languages in the game world in various ways. Written languages, henceforth called Scripts, are sometimes an afterthought, and the results can be either too simplistic, too fiddly, or just not well thought out.
Most RPGs I’ve seen take two basic approaches: either a literate character can read every language they speak, or every written language is as much a skill as every spoken language. In this article I’ll present a “Hybrid Approach”, and then offer some options for more unusual situations.
The “You Speak It, You Read It (If You Read)” Approach
The simplest assumption is that if a character can read, they can read every spoken language they know. The criteria for determining whether a character can read at all are one or more of the following:
- The character is a player character.
- The character’s class or profession requires literacy (e.g. wizard, cleric, scholar, scribe).
- The character has an Advantage, Feat, Edge, etc. that says they can read.
- The character comes from a time or culture where everyone can read.
This sidesteps the complexity of many actual writing systems, but GMs and players may simply not care about such things.
Extinct Languages: If a spoken language is extinct, its Language covers mainly its written form plus a reconstructed pronunciation.
The Literacy Skill Approach
Systems that treat spoken languages as ordinary skills more often than not treat written languages as ordinary skills. In the author’s opinion this adds complexity, especially compared to the previous approach.
If two written languages use the same alphabet, the GM may rule that they’re similar enough to be Related or even Close, and use the same rules as for spoken languages. Just because two spoken languages are Related or Close doesn’t mean the written languages are Related or Close, and vice versa. In the real world, some Turkic languages use the Latin alphabet, and others use the Cyrillic alphabet. Conversely, both English and Turkish use the Latin alphabet, but in completely different ways.
On the other hand, Dialects of a parent language (almost?) always use the same writing system – and the same skill – as the parent language. Generally writing systems use a standard dialect for “educated people”, and native speakers of a Dialect learn to read this standard form. British and American English may spell some words a bit differently (because English spelling is a nightmare) but a Briton can easily read American text and vice-versa.
Extinct Languages: If the spoken Language is extinct, having two skills for it seems even more like overkill. The author recommends having only a “Literacy” or “Read/Write” skill for that language, which includes a reconstructed pronunciation.
Since this article argues for a more “realistic” approach to languages in RPGs, the author will outline a hybrid of the previous two approaches.
Essentially, scripts are either Normal, Archaic, Adapted, or Complex.
A Normal Script is ideally suited to a single Language. Anyone who can read learns that Script when they learn the Language. This is the default.
An Adapted Script writes multiple Languages using conventions and notations specific to that Language. (e.g. Latin, Cyrillic, and Arabic).
If the system uses Literacy Skills define only one skill for all uses of the Script, no matter how many Languages it transcribes.
Otherwise, use the rule for a Normal Script: anyone who can read and speaks a Language learns how to write it with that Script.
An Archaic Script is like a Normal Script except that it’s little used.
Anyone who can read and speaks a Language with an Archaic script doesn’t learn the Archaic Script unless the language is extinct or no other scripts are available. Treat the Script as a distinct Language, albeit one that’s easier to learn if someone knows its spoken counterpart. (See language groups for possible mechanics.)
A Complex Script is virtually a Language unto itself, either because it contains a large number of characters (like Chinese) or because the characters used to represent a word have little relationship to the sound of a word (like English), or because it’s just convoluted (like Japanese with kanji).
Treatment of Complex Scripts depends partly on the rules for handling Languages:
All or Nothing: If the rules assume that a character either speaks a language or doesn’t, Treat the Script like a Language distinct from its spoken form.
Fluency Levels: If the rules define multiple levels of fluency in a spoken language, the easiest approach is to treat the Script as another language.
Language Skill: If the rules treat each spoken languages as a separate skill or ability, define a Literacy Skill for the Script, or if the Script transcribes more than one Language a distinct Literacy Skill for each pairing of Language and Script.
If the language is extinct, define a Language only for the written form.
Usually one can only write a language in a script designed for that language. If someone transcribes that language in any other script, e.g. transcribing Japanese into Latin letters1, the writer must pass a skill test to write it correctly, and all readers must pass their language tests to read it correctly. Any failure causes a Mistranslation.
Some ancient scripts are hard to read, or the languages themselves haven’t been fully translated. Each time a character reads an inscription in one of these scripts, there is a chance that he misreads it. To determine whether a Mistranslation occurs, the GM will secretly roll against that script’s Mistranslation chance.
If a Mistranslation occurs, the Referee will obscure or even invert the meaning of the original text. This should only involve 1d6 words of the translated text.
While the exact details of a particular script works may have few if any in-game ramifications, knowing how writing systems work can suggest ideas for creative mistranslations, ambiguities, or other difficulties reading an old text.
Nearly every writing system yet devised breaks into these categories:
Pictograms or ideograms convey meaning by directly resembling an object or indirectly suggesting an idea through an associated object. They’re among the earliest forms of writing on Earth, including cave paintings, and many Mesoamerican writing systems. Modern examples include hobo signs, computer icons, and the many shapes and symbols used in traffic signs. However, they lack grammatical structures and levels of abstraction that characterize spoken and sign languages. They’re most effective as mnemonic devices for storytellers, warnings for travellers, and religious or secular art.
Game Effects: The GM has three options:
- Let players interpret them, e.g. “You see the picture of a skull.”
- Treat them as a Written Language, e.g. “You read Hobbit Signs, so you see the symbol for danger.”
- Devise a system of signs and show them to players. This might be the most fun, but also the most work.
Logographic scripts evolved from ideograms; they became simpler and more abstract, and eventually a grammar and syntax. The best known example is written Chinese.
The large number of symbols makes logographic scripts harder to learn; each logographic script is a language unto itself. While a logographic script (mostly) eliminates differences in pronunciation and dialect2, writers can make mistakes drawing a character, or accidentally substitute one character for a related one. Older texts may use characters that are either obscure or wholly obsolete, and without a reference it’s virtually impossible to determine the meaning.
Game Effects: These Scripts are necessarily Complex. Characters falling out of common use may give rise to mistranslations of centuries-old documents.
A logophonetic script combines logograms and symbols describing sounds. Sometimes the same symbols are used for both sounds and words, and the reader has to figure out which from context. In other cases phonetic elements evolved from abstracted logograms, or came from other writing systems. Examples include Egyptian hieroglyphics, several cuneiform writing systems, and written Japanese which combines Chinese-derived kanji and phonetic kana (hiragana and katakana) for grammatical endings and particles.
Game Effects: These Scripts are necessarily Complex. Sometimes, however, the phonetic elements form a complete subset, e.g. the kana characters of Japanese. PCs and NPCs could treat them as “Normal” scripts, although writing in that subset might mark one as uneducated.
A majority of modern writing systems transcribe the sounds of a spoken language, although not always consistently. How they do so varies considerably, but these are the most common techniques:
A syllabary theoretically has a unique character for each syllable in the language. As this rapidly becomes unwieldy, most take shortcuts. For example, each Japanese kana represents only a [mora](https://en.wikipedia.org/wiki/Mora_(linguistics)) or “light syllable”, either a short vowel, consonant and short vowel, or N. Long vowels are represented as two characters which contain the same short vowel, for example.
An abjad transcribes the consonants only, although modern versions add notations, usually dots and angled lines, above the letters to fill in vowels. Examples include Arabic script and Hebrew.
An abugida transcribes consonants with a default vowel, usually A; notations above each character alter the default vowel. Examples include Devangari and most other writing systems of India.
An alphabet uses one letter each for a vowel, a consonant (or consonant cluster), or letters that alter the sound of preceding or following letters. The most prominent examples are the Latin and Cyrillic Alphabets.
A featural writing system encodes fine phonetic details in the shape of each character. For example, Canadian Aboriginal Syllabics represents each syllable with a shape that denotes a consonant or consonant cluster, an orientation (up, down, left, right) that indicates the vowel, and dots to indicate long vowels or a “W” sound between consonant and vowel. The Korean Hangul alphabet is sometimes cited as another example. Except for Hangul, most came into existence in the last two centuries.
In the past century, linguists have created systems for transcribing the sounds of all languages, notably the International Phonetic Alphabet. While in theory this sounds like the ultimate phonetic system, in practice it can capture every nuance of dialect and pronunciation. If it somehow replaced English, for example, not only would we have to learn over a hundred new notations, we would have to sound out the equivalent of, for example, “faar”, to realize it was the Southern U.S. pronunciation of “fire”.
Few Phonetic scripts are “Complex”, unless they’re truly huge syllabaries or drifted very far from the actual pronunciation of the language. (As English sometimes does.)
Phonetic scripts, particularly alphabets, are often “Adapted” and/or used as foreign scripts. Sometimes sounds change from one language to the next or one era to the next. Scribes often repurpose old letters or invent new letters for sounds specific to their language.
The quirks of each phonetic writing system might inspire creative mistranslations, e.g. a Script with missing or unclear vowel markers might lead to ambiguous verb moods and tenses, which would leave the reader wondering whether something had happened, is happening, might have happened, or will happen in the future.
Depending on the GM’s preference and the approach taken, they can add a Script column to a Languages table, create a Script Table to map Scripts to Languages, or both.
Fields in a Script Table might include:
- the name of the script
- Type (optional)
- see Script Type, above.
- one of Normal, Adapted, Archaic, or Complex. If left blank, the default is Normal.
- Mistranslation (optional)
- If using the Mistranslation rules, the chance of mistranslating text in the script. Usually this is a percentile value, but the table could specify a value rolled on some other dice (given in the header).
- Languages (optional)
- If not given in another table, the list of spoken languages written in this Script.
If the resulting tables are too busy or too sparse, the GM may wish to break them up by Category. Archaic and Complex Scripts, in particular, may have attributes of spoken languages like Language Groups and Rarity levels.
Returning to Erebus, here are the Script Types and Mistranslation Chances for the scripts used in Erebus.
Normal Scripts in Erebus
All Normal Scripts have a Mistranslation chance of 0%
|Common Runes||alphabet||Dorlandish, Jotlandish, Nordlandish, Westlandish|
|Dwarf Runes||abugida||Old Dwarfish|
|Vanir Runes||alphabet||Dwarfish, Elfin, High Vanir|
|Zarnic Letters||alphabet||Thalestrian, Zarnic|
Archaic Scripts in Erebus
All Archaic Scripts have Rarity Extinct.
|Ophidian, High||logograms||Aklo Logograms, Ophidian||15%|
Complex Scripts in Erebus
|Tianxi Logograms||logograms||Tianxi||-||(age in centuries)%4|
Pictographic Writing in Erebus
In addition, Erebus has the following pictographic forms of writing.
|Pictograms||Purpose||Approx. Number of Signs|
|Barbarian Runes||identity, boundaries of tribal territory||1-5 per tribe|
|Halfling Signs||indicates traps, entrances to tunnels, caches of food and weapons, etc.||several dozen|
|Shop Signs||indicates name and/or type of business||a dozen or so|
|Traveler Signs||indicates places to camp, friendly or unfriendly residents, etc. similar to hobo signs.||12-20|
Using the Hybrid Approach, here are the main Scripts used in Telluria:
|Druidic Script||abugida||-||Vethic (Druidic)|
|Elissan Alphabet5||alphabet||Archaic||Vethic (all Classical)|
|Keshitic Script||abjad||-||Old Keshitic|
|Kymric Runes||alphabet||-||Kymric (all)|
|Theran Alphabet||alphabet||Adapted||Keshite, Kymric (Common), Northlandish, Sarkennian (all), Theran, Thervingian|
|Verdan Alphabet8||alphabet||Adapted||Kymric (Verdan), Vethic (all)|
Note that we’re not using the Mistranslation rules.
Under an “All or Nothing” or “Fluency Level” system, we’d define the following new Written Languages:
- Aklo Logograms
- Elissan Alphabet
- Naunetic Hieroglyphs
- Thervingian Runes
Under a “Literacy Skill” system, we’d define the following Literacy specialties:
- Aklo Logograms
- Elissan Alphabet
- Naunetic Hieroglyphs
- Theran Alphabet
- Thervingian Runes
- Verdan Alphabet
The remaining scripts would be folded into their spoken languages.
Under a strict “Literacy Skill” approach, all eleven Scripts would have at least one skill. In the worst case, every language would have its own Literacy skill plus one each for the Archaic scripts.
If all you have are d6’s, here’s a rough translation of percentiles to a “d66”, i.e. two distinct six-sided dice, one of them designated “tens”.
Find the closest value and roll; what’s +/-2.78% between friends?
|03||1,1 - 1,1||36||1,1 - 3,1||69||1,1 - 5,1|
|06||1,1 - 1,2||39||1,1 - 3,2||72||1,1 - 5,2|
|08||1,1 - 1,3||42||1,1 - 3,3||75||1,1 - 5,3|
|11||1,1 - 1,4||44||1,1 - 3,4||78||1,1 - 5,4|
|14||1,1 - 1,5||47||1,1 - 3,5||81||1,1 - 5,5|
|17||1,1 - 1,6||50||1,1 - 3,6||83||1,1 - 5,6|
|19||1,1 - 2,1||53||1,1 - 4,1||86||1,1 - 6,1|
|22||1,1 - 2,2||56||1,1 - 4,2||89||1,1 - 6,2|
|25||1,1 - 2,3||58||1,1 - 4,3||92||1,1 - 6,3|
|28||1,1 - 2,4||61||1,1 - 4,4||94||1,1 - 6,4|
|31||1,1 - 2,5||64||1,1 - 4,5||97||1,1 - 6,5|
|33||1,1 - 2,6||67||1,1 - 4,6||00||1,1 - 6,6|
In the real world there are conventions for writing Japanese in Latin letters. Unfortunately there are multiple conventions, and they don’t always capture nuances of pronunciation. Here we assume the writer is making up their own convention based on how Japanese sounds to them. ↩︎
One myth about written Chinese is that speakers of any dialect can read it, because it abstracts out the sounds of the words. However, it’s more correct to say that all people who read and write Chinese write it the same way, with the same grammar (based on Mandarin), regardless of what “dialect” of Chinese they speak. ↩︎
Pelganic replaced Kharat and can represent a wide range of sounds. Pelganic can trancribe the following languages without incurring the foreign script penalty: Darthani, Dorlandish, Kulhani, Plains Ologan, Thalestrian, and Zarnic. ↩︎
Older manuscripts may use logograms that dropped out of the modern language, or that changed meanings over time. ↩︎
Derived from the Theran Alphabet. ↩︎
Derived from the Theran Alphabet. ↩︎
Derived from Kymric Runes. ↩︎
Derived from the Theran Alphabet. ↩︎