In my medieval world, population A and population B have not had any contacts for thousands of years. But pressing events will make them look for each other.

However when they meet... how are they going to understand each other?

I assume that:

  • Population A mainly speaks language 1.
  • Population B speaks several languages including language 2.
  • Language 1 and language 2 are derivated from the same language 0 that they used to talk to each other thousands of years ago.

Because languages evolve with time, they are now quite different. What kind of difference happen in a thousand years? Can I show them in the dialogues?

Another important fact: NOBODY speaks both language 1 and 2, and there is no magic or language book to make it easy.

I have to do this because several cultures separated for such a long time simply cannot have the same language. It would feel unrealistic and break the immersion. Suspension of disbelief is a serious issue.

The problem is that makes it much more difficult for me to explain the communication between characters who don't speak the same language. Would characters understand each other if languages 1 and 2 are not too distinct? As I said, there is an urgent threat, and the story/heroes certainly do not have a few weeks to focus on language learning.

How to make sure that the characters understand each other without troubling the reader?

  • 8
    $\begingroup$ If they have been completed separated for thousands of years, how are you bringing them together now? $\endgroup$
    – Kyyshak
    Commented Jul 24, 2018 at 11:17
  • 2
    $\begingroup$ A long and perilous trip from a group of Population 1 to Population 2. But I am afraid that the group will simply not understand the locals of Population 2 when they find them. $\endgroup$
    – Dreamk33
    Commented Jul 24, 2018 at 11:19
  • 3
    $\begingroup$ Worldbuilding.SE is about helping you with specific problems about the setting, whereas Writing.SE is about helping you with the writing process and the craft of writing in general. I removed that part because it's basically up to you where to post as long as it's on-topic on the site you post and you are aware that answers will have a different focus. If you would like feedback, for example on the on-topicness of a question you want to ask, feel free to use the Sandbox for your draft. $\endgroup$
    – Secespitus
    Commented Jul 24, 2018 at 11:20
  • 4
    $\begingroup$ Confirming AlexP: I speak Spanish about level B2, i.e., far from fluent. But that was enough to pass an A1 test for Italian. Ten hours of lessons allowed me to pass the listening and speaking parts for A2. See “CEFR” for a description of those levels. $\endgroup$
    – WGroleau
    Commented Jul 24, 2018 at 14:57
  • 3
    $\begingroup$ @AlexP you should write that as an actual answer, even though you already wrote a good answer. There probably aren't going to be interpreters or many scholars --- but there are going to be young & hormonal unmarried folks meeting new folks for the first time and very keen to learn the new language! You can bet that within three months almost everyone of the newcomers will be more or less fluent in the new language. Within a generation or two, the migrating language will be all but forgotten. $\endgroup$
    – elemtilas
    Commented Jul 24, 2018 at 15:25

11 Answers 11


Let's consider a real historical example

Let's fix the historical period to the 14th century, and let's say that "language 1" is French and "language 2" is Romanian. Both French and Romanian are descended from Latin (which is thus "language 0"), and in the 14th century, they had had no contact for about 1000 years. No 14th century Frenchman spoke Romanian, and no 14th century Romanian spoke French.

But it came to pass that in 1396 a French army and a Romanian army had to join forces with a Hungarian army at Nicopolis (in the north of modern Bulgaria), to fight against the Ottomans. So how could the French and Romanians communicate?

  • First of all, it's absolutely clear that the top level leaders, for example Jean Le Maingre (marshal of France) and Mircea the Elder (Prince of Wallachia) would speak through interpreters. A great lord does not speak an approximation of a foreign language except in case of dire necessity.

  • Second, there was no time for them to learn the other language, although it was clear that the languages were related.

  • So, how did the interpreters communicate? Well, they had four possibilities:

    1. Speak Latin ("language 0"). Both the French and Romanian armies had clerks who knew Latin, because Latin, although long dead, was used as a written language in diplomacy and administration. Of course, the French clerk's Latin would sound completely different from the Romanian clerk's Latin, but since the differences would be perfectly systematic they would very quickly learn to get over them.

    2. Speak Italian, or actually, at that time, Genoese. The French had bona fide Genoese crossbowmen in their army, and there were Genoese colonies in Wallachia so that there were bound to be several Romanians who understood Genoese. (And anyway, the medieval opinion of Romanian peasants was that Italians spoke Romanian too, but quite badly.)

    3. Speak Greek. Greek was a language of culture. On the Romanian side, there must have been traders or churchmen who spoke Byzantine Greek; the French must have had one or two priests who understood Common Greek. The differences in pronunciation would be even more marked than the differences between the pronunciations of Latin, but they could still be overcome.

    4. Use Hungarian interpreters. The bridge power at that time was Hungary, whose king, Sigismund, would eventually become Holy Roman Emperor. The Hungarians maintained diplomatic relationships with both France and Wallachia, and thus had people who could speak French, and people who could speak Romanian, and, with a bit of luck, maybe they had some clerk who could speak both French and Romanian.


The technical question is how to make sure that the characters understand each other without troubling the reader. This is quite interesting because the story, as far as I understand, is to be written in English for an English-speaking audience.

Let's assume that the story is to be written from the point of view of a character from population A, who speaks language 1, which will be conventionally represented by English. So the question becomes: how to represent a language similar enough to English and still different enough, so that the readers will perceive it as a foreign yet closely related and understandable language.

Now, if this were a story written in Romanian, it would be easy: let language 2, the language of population B, be represented by Italian; if the author chooses the Italian phrases carefully Romanians will understand them, and yet they will clearly be in a foreign language. Or if the story were written in French, language 2 could be represented by Middle French (the language of Rabelais); with a bit of effort a French reader could understand Middle French -- after all, Balzac wrote an entire book, Les contes drolatiques, in mock-Middle French, and it is still in print (and hilarious). Or if the story were written in Russian, language 2 could be represented by a Russian recension of Church Slavonic.

But where can one find a language clearly related to English, foreign enough to confer an air of difficulty, but close enough to remain intelligible? My suggestion is to represent language 2 with a slightly modernized Middle English; it may be that my view is prejudiced (after all, English is so much unlike my mother tongue that for me the differences between Middle English and Modern English seem quite small), but I believe that slightly modernized Middle English would be intelligible, with a little effort:

'Wepyng and waylyng, care and oother sorwe I knowe ynogh, on even and a-morwe,' quod the Marchant, 'and so doon oother mo that wedded been.'

which in Modern English would be

'Weeping and wailing, care and other sorrow I know enough, in the evening and in the morning,' said the Merchant, 'and so do many others who have been married.'

(Chaucer, The Canterbury Tales, as quoted and translated by Wikipedia.)

Of course, it should not be overdone. A phrase here, a sentence there, a few words from place to place ought to be enough to convey the color. The story should remain a story written in English; use the stand-in for language 2 sparingly.

  • 10
    $\begingroup$ +1 for mentioning Middle English and not making references to J R R Tolkien $\endgroup$ Commented Jul 24, 2018 at 17:52
  • 1
    $\begingroup$ +1 for the middle English, but as a native English speaker, let me tell you, the divergences been missile English and modern English are huge. Itb is understandable but barely. Is better to use something like kings James English, or cockneese? (English with a cockney accent) and both are more easily understood, by modern English speakers. $\endgroup$
    – Pliny
    Commented Jul 25, 2018 at 12:59
  • 1
    $\begingroup$ @AmiralPatate: Of course the majority of the dialogue would be rendered in English. As I said, a word here, a phrase there, for color: but it should not be overdone. $\endgroup$
    – AlexP
    Commented Jul 25, 2018 at 13:02
  • 1
    $\begingroup$ @AlexP Tangential pro tip: you write "Suspension of disbelief is a serious issue". It's a poor start to have the reader laugh out loud at the author's naming scheme, or force them into syntactic memory games of remembering which meaningless M-word is which. Look how name languages, and places, and people. They tend to be short, simple, based on real root words. We abbreviate if they hit three syllables, let alone 5-6. Even then, with very few exceptions, names containing X, Z, Q, apostrophes or unusual doubled letters don't feel foreign, they feel like keyboard-mashing names from bad fanfics. $\endgroup$ Commented Jul 25, 2018 at 20:32
  • 3
    $\begingroup$ @DewiMorgan: I am not the original poster. The fictitious names in the comment are fictitious, intended to be humorous, and almost certainly do not represent the original poster's actual intentions. And the phonotactics of the automatically generated names used in the example are actually not that far from Greek, a bit Latinized and somewhat Englished; the script which generated them intended to represent barbarian names as heard by Greeks. $\endgroup$
    – AlexP
    Commented Jul 25, 2018 at 21:14

The simplest solution is that in both population you have some rare few scholars who speak language 0 and can communicate in this language, even though they do not actually speak each others' modern day language 1 or 2. This is basically how political and scientific discourse was kept alive in the middle ages in most parts of Europe where Latin became the language for discourse of the educated few. Even though the Italian, German and Normanic Scholar or politician spoke completely different languages which they did not understand mostly, they could all converse in Latin.

Your party of delegates from Population 1 should take at least one priest/scholar/ you name it who still knows language 0 well enough to speak it fluently with them. When arriving in the territory of Population 2, they should avoid the lowly educated locals and try to reach a city/monastery/university where old knowledge and texts written in language 0 are still available and thus also people who understand and even speak this language. The scholars would then work as interpreters for the nobles, politicians, military men who wish to forge an alliance and derive a joint military plan of action against your external threat.

If you do not wish to exclude your heroes too much from being able to act directly, make it a part of compulsory education for nobles, politicians, military men and scholars alike to learn language 0 in both countries. Then they can communicate directly, but possibly less eloquently than the learned scholars who spent more times reading texts and improving their use of written rhetorics and grammar in language 0.

  • 5
    $\begingroup$ This was a plot device in Heinlein's Have Space Suit, Will Travel. An American astronaut gets involved in time travel mayhem and must find a way to communicate with a Vulgar Latin-speaking Roman Soldier. Fortunately, the astronaut speaks Spanish and has also taken Classical Latin courses. Neither of those is exactly the same, but they are "close enough" for a basic level of communication and they both gradually converge on a pidgin that they both can speak to each other. $\endgroup$ Commented Jul 24, 2018 at 15:39
  • 8
    $\begingroup$ @RobertColumbia Am I reading the wrong plot synopsis or did you get story titles mixed up? Have Space Suit, Will Travel doesn't concern that at all. $\endgroup$
    – Dispenser
    Commented Jul 24, 2018 at 17:41
  • 1
    $\begingroup$ Or maybe they don't know how to speak Language 0, but the scholars can read and write in it. $\endgroup$ Commented Jul 24, 2018 at 22:14
  • 2
    $\begingroup$ @Dispenser This was a relatively small part of the story, but allowing the two groups to speak meant getting another point of view that they needed. But yet, I remember this as part of Have Space Suit, Will Travel. $\endgroup$
    – NomadMaker
    Commented Jul 24, 2018 at 22:45
  • 2
    $\begingroup$ @Dispenser: It does, though it's not time travel. Essentially the teenage would-be astronaut gets picked up by a member of an advanced race, who (after various adventures) use him as one of three samples to determine whether or not to destroy Earth. The other samples are a IIRC a Neanderthal and a Roman soldier, who were picked up earlier and kept in stasis. $\endgroup$
    – jamesqf
    Commented Jul 25, 2018 at 4:08

Lewis and Clark, when they explored the American west, consistently had language barriers as they communicated with the natives. When they needed to use the spoken word they went through a chain of interpreters, at one time it was English (Lewish and Clark)->French (Labiche)->Hidatsa(Charbonneau)->Shoshone(Sacagawea). Source

Typically they just relied on sign language for direct communication. I believe it is very easy for a system of rudimentary gestures to quickly adapt into a functional language for most tasks, especially if the cultures are enough aligned that each party can have a good guess of what type of responses to expect.

Plains Indian Sign Language

  • 1
    $\begingroup$ Welcome to Worldbuilding, Bob! If you have a moment, please take the tour and visit the help center to learn more about the site. You may also find Worldbuilding Meta and The Sandbox (both of which require 5 rep to post on) useful. Here is a meta post on the culture and style of Worldbuilding.SE, just to help you understand our scope and methods. Have fun! $\endgroup$ Commented Jul 24, 2018 at 17:03

One possibility would be a common written language. In China there are many major and minor dialects that sound very different, and cannot be understood by speakers of other dialects, but they all have the same written characters. Even the Japanese and Koreans use many of the same characters, although they have very different sounds for them. So you could have a situation in which your two populations can communicate by writing in the dirt or on chalkboards.

  • 4
    $\begingroup$ Chinese characters are only standardized because they were explicitly made that way. This is essentially suggesting that Languages 1 and 2 have changed from Language 0 in speech but not in writing, which is probably more plausible than them not changing at all but still not really reasonable. Also, contrary to popular belief, it's not at all true that one e.g. write a sentence in Mandarin and have it be perfectly grammatical Wu pronounced differently; the grammars still differ. $\endgroup$
    – Danica
    Commented Jul 24, 2018 at 16:34
  • 4
    $\begingroup$ It doesn't have to be grammatically perfect, though. Nor do they have to have exactly the same set of characters -- just as long as their two character sets have a large overlap. By the way, I think something similar is true of Latin in the middle ages. The way it was pronounced and used in daily life was probably different in different times and places, but educated people would have been able to understand the ancient usage of Cicero, Plutarch, etc. $\endgroup$
    – workerjoe
    Commented Jul 24, 2018 at 17:43

Here in my real world we have something called the "Ohne Wörter Buch" Which would translate to "Dictionary without words". (It's actually a pun in german language so you cannot really translate this 1:1)

It simply is some kind of picturebook. There is no reason why this sould not work in your world, too.

Communication does not always mean talking.


enter image description here


For most people, usage of a foreign language ranks from most difficult to easiest:

  1. Hearing the foreign language is often the most difficult task. You are at the mercy of the speed, accent, and articulation of the speaker. If you missed or did not understand something, it can considerably impair communication.

  2. Speaking the foreign language is also a difficult task. You control the pace of the information flow. However, if you cannot recall the word you are looking for, you are in trouble. You might be able to look up the word, but that becomes extremely awkward as the other person waits for you to express yourself.

In many situations, the order of these first two may be reversed; however, that does not affect my final argument.

  1. Writing the foreign language comes next. Because parties expect writing to be slower than speaking, you can take more time. If you don't know a word, you can look it up or interpolate it. However, sometimes the way you think something should be expressed isn't the way it is actually expressed, particularly when a language evolves.

  2. Reading the foreign language is easiest. You have plenty of time to decode what you see. If you don't understand a word, you can put it together from pieces that you do recognize. The one caveat is that when languages evolve, they often retain similar pronunciations of words but the spelling can change dramatically. For example, Scandinavian languages sound familiar to a German when spoken, but their spelling is quite different.

  3. Reading and hearing a foreign language together is easiest, as it eliminates all of the problems described above.

Therefore -- if a common language is not feasible or not yet found -- the best way to initiate communication is the following: 1. One party writes in their native language. 2. If the second party does not understand, the first party repeats in spoken form what was said. 3. The second party responds in their written language. 4. If the first party does not understand the response, the second party repeats their response in spoken form.


So, lots of good answers here, but there are a few things I haven't seen pointed out yet that I feel is worth mentioning.

Just because two languages are 'quite different' doesn't mean that it's nontrivial to learn to understand one when you speak the other.

No, seriously. Portuguese and Spanish are 'quite different'. They're also close enough that someone who is fluent in Spanish can (if they're smart) learn to understand Portuguese at a really basic level pretty quickly with some reasonable exposure to the language. Similarly, Latin is 'quite different' from modern members of the Italic language family, but it's also an insanely good starting point for learning things like Spanish, French, or Italian (or Portuguese, or Romanian, or any other romance language), because they're still similar enough that knowledge of one transfers reasonably well to the others.

When you really start having issues with mutual intelligibility is cases where you're looking at far removed language families. Using the example of Spanish and Portuguese, they're quite different if that's the only reference you have, but are somewhat similar to each other compared to German, and damn near identical when you compare them to things not even in the Indo-European family like Xhosa or Japanese.

This statement: '...several cultures separated for such a long time simply cannot have the same language.' is actually just as unbelievable as them having identical languages.

Let's look at well established facts first:

  • Their phonetics will drift (real world example, US and UK English). However, learning new sounds is damn near impossible for most people once they get a few years past puberty, which means that 'new' sounds don't happen overnight. They don't even happen much over centuries. Realistically, both languages will almost certainly have almost the same set of phonemic consonants (most likely one or two different), and will probably have very similar collections of phonetic vowels (more likely variance here, as vowel sounds are inherently a bit more variable). Overall, it's very likely that it won't be too hard for someone who speaks Language 1 to learn the phonetics of Language 2, even though they may 'sound' very different.
  • Their syntax is not very likely to drift by much. Syntax is one of the big things that makes it reliably possible to trace a language's history because it changes very slowly, even in isolated populations. Look for example at English and German. They diverged over a thousand years ago, but they still have rather similar syntax. Similarities in syntax greatly simplify understanding of another language.

OK, so those two are in favor of mutual intelligibility. Now for the bits that are less certain:

  • Vocabulary will probably drift some, but drift in vocabulary is most often in the form of superficial changes or loanwords. Looking again at Germanic languages for an example of superficial changes, 'mit' (from German and some other Germanic languages) and 'with' (from English) are two different forms, but have the same meaning, and the basic structure is close enough that they're recognizable as the same word. Loanwords, and new words for new things, are where things are likely to diverge in an unintelligible manner. Using another example with German and English, 'luft' and 'air' mean the same thing, 'luft' is from the original Germanic root, and 'air' originated as a loanword from Greek (by way of Latin and Anglo-French). If both populations are isolated from other languages as well, you're very likely to still have two mutually intelligible languages even after a thousand years. Even if they are exposed to other languages, it's still likely that a decent percentage of core vocabulary is likely to be similar enough to be mutually intelligible.
  • Idioms might drift a lot, or they might not. The idioms in a language are quite often a byproduct of the shared cultural experiences of the people who speak it. If your two populations lived in very similar conditions for those thousand years, they're likely to have (mostly) similar sets of idioms. The really interesting part is that this is often true of two completely different languages (for example, 'when <farm animal> files' is a remarkably common idiom in a lot of Indo-European languages for indicating something is impossible). Idioms also tend to stick around if they are in widespread usage (mimetics at work), so well established ones from before the split will likely still be used a thousand years later. New idioms though, are where things may get very confusing, but only if it's something really out there (the equivalent of the above mentioned idiom in some Balto-Slavic languages (like Russian) is 'when the crawfish whistles on the mountain', which still conveys the meaning well enough to anyone who knows what a crawfish is). The neat thing about idioms though is that it's easy to ask for an explanation, which means that while they may slow down learning a new language fluently, they generally aren't a huge issue for people who are just trying to understand the language.

So, overall, I seriously doubt that the languages would be different enough that it would be a serious issue. I mean, they're going to be different, but it's more likely that they'll be English and German different, not English and Mandarin different.

Any sign language that existed before the split is likely to still be very similar.

Same arguments as above, except sign languages change far slower than spoken languages (they need more consistency, for multiple reasons). You can see this easily in real life. There are six main sign language families consisting of a total of about a hundred known languages, with more than half in one family and that one family constituting what's used throughout most of the world. In comparison, there are over 50 spoken language families (the number is much higher, I'm just too lazy to count them right now) consisting of over 7472 languages (the two biggest families both have more than a thousand languages), and the only one that doesn't cover a mostly contiguous geographic area is the Indo-European family (and that's mostly because of English, French, and Spanish).

I also find this 'NOBODY speaks both language 1 and 2, and there is no magic or language book to make it easy.' rather unlikely.

The populations may be isolated, but even if the journey is very perilous, it's highly unlikely that nobody has ever in those thousand years crossed between them. If there is any third population involved who has had contact with both, then your chances of this being true go down even further. This may be a viable writing option to allow for a good story (your MC fumbling around trying to learn the local language can make for a very entertaining aside from the main story), but it's not a very realistic one even for a medieval setting (keep in mind that even in medieval times, Europeans did make it all the way to both China and North America).


Body Language for humans is almost a universal language. I would say they could communicate through lots of pantomiming and pointing at things. Get some base movements down, like trying to eat something inedible and making a no sign, or something edible and a yes sign (and reaffirming through other actions) and go from there.

I'm sure that each society has had mute/deaf people before, it might even be easier to try and get a person from each society with that disability to be the "interpreter". They would probably be more perceptive of body language and the emotion/meanings that people are trying to get across.

  • 3
    $\begingroup$ Body Language as you describe it is less universal than you might think. Some cultures shake head for yes, and nod for no, and others nod for yes, and shake for no. Some indicate numbers on there fingers via number of fingers held-up, others by number of fingers not held up. Even things like facial expressions can vary in meaning dramatically between cultures. $\endgroup$ Commented Jul 25, 2018 at 2:06
  • $\begingroup$ @LyndonWhite Yeah, that's why I brought up the reaffirming, doing good things and making a sign for yes, bad things and then a no sign. Until that meaning is understood by both parties. Once you get the yes/no down, it becomes a lot easier. They could count out things and hold up a finger for each item added to a basket, or something similar. There's many ways to emphasize a meaning. $\endgroup$
    – Sensoray
    Commented Jul 25, 2018 at 12:24

To my experience, some people have more talent than others in picking up languages and communicating with foreign people. So the first step is to identify such people in your crew and let them act as interpreters afterward.

When the two languages A and B are both derived from the same old language (whether it is yet known or not) there is certainly some mutual intelligibility left over, and the talented people can bridge the gap.

The process of establishing reliable communication is not immediate, but within a few days it should be set up.


I don't know how true it is but I've read that several languages, Basque and Lithuanian among them, are considered to be "extremely conservative" being stable in many aspects over many thousands of years. If you posit Language 0 as such a language then Language 1 and 2 while different should still remain similar enough in terms of proper nouns, honourifics, and numbers to allow some communication even if it is a bit stilted.


Generally, high level, formal language, academic language and generally written language is subject to less change over time. While peasants might find it impossible to comunicate because spoken, vernacular language mutates fast, academics, elites and (especially with a common religion) priests/monks might understand each other.

You should also consider that some languages of common decent are similar in writing, but not in sound, while others are more similar in sound than in writing.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .