Year 2020, the United States of Japan sends colonists to the outer space to colonize new planets under the name of the Emperor. Long story short, they finally colonized Planet VII of the Lem solar system. Planet VII is some hundred light years away from Earth, which is pretty much preventing communication. Even prior to colonization they have pretty much stopped contacting Earth, since the communication delays were just too much.

Both, the colonist and the Earthlings, have yet to discover warp technology. So they can't send their message through a warp gate.

Question: how long does it need before the Japanese that the colonists use and the Japanese that the Earthlings use differs so much that Earthlings no longer recognize the Colonist's language as Japanese?

Edit: all colonists are Japanese in both nationality and race, and speak only Japanese. They know no other language except a little English, just like today's Japanese.

  • $\begingroup$ Will the Earthlings at least recognize its base as Japanese or not at all? For example High Tider with 250 years of isolation is considered a dialect of English: en.wikipedia.org/wiki/High_Tider youtu.be/jXs9cf2YWwg $\endgroup$ Commented Dec 19, 2016 at 5:39
  • $\begingroup$ They might recognize its base as Japanese if they went to study it further. What I mean by people not recognizing it, is the general people, not the scientists and linguists. $\endgroup$ Commented Dec 19, 2016 at 5:40
  • $\begingroup$ Do all of the colonists speak Japanese exclusively? In a small population, even a couple of speakers of another language could encourage the development of a hybrid by injecting terms that get adopted throughout the group. $\endgroup$
    – SRM
    Commented Dec 19, 2016 at 6:52
  • 1
    $\begingroup$ I'm sure there is a library inside the interstellar ship, even on a completely alien environment the colonist are still stuck with the same operating sys unless you want to revamp everything... but why reinvent wheels? $\endgroup$
    – user6760
    Commented Dec 19, 2016 at 7:54

6 Answers 6


Well-known historical examples:

  • From Latin to the Romance languages: about three to five centuries (in the 5th century the language was clearly Latin, in the 9th the language was clearly something different). For example, see the Oaths of Strasburg (842 CE); the text in the "romana lingua" is clearly no longer Latin but a very primitive form of French: "Pro Deo amur et pro christian poblo et nostro commun saluament, d'ist di in auant, in quant Deus sauir et podir me dunat, si saluarai eo cist meon fradre Karlo, et in adiudha et in cadhuna cosa si cum om per dreit son fradra saluar dift, in o quid il mi altresi fazet. Et ab Ludher nul plaid nunquam prindrai qui meon uol cist meon fradre Karle in damno sit".

  • From Old English to Middle English: about two or three centuries (at the beginning of the 11th century the language was Old English, by the end of the 13th it was Middle English). Old English is incomprehensible: "Hwæt! wē Gār-Dena in ġeār-dagum, þēod-cyninga, þrym ġefrūnon, hū ðā æþelingas ellen fremedon" ("ġ" is pronounced like "j" in "jar"; "ð" and "þ" correspond to modern "th"). Middle English is much closer to the Modern language: "Whan that Aprill, with his shoures soote the droghte of March hath perced to the roote and bathed every veyne in swich licour, of which vertu engendred is the flour". (Examples from the Wikipedia articles.)

Schools and newspapers and mass-media tend to slow down the evolution of a language; isolation and illiteracy tend to speed it up. It is possible to maintain comprehensibility over a millennium or so -- Greeks in the 2nd century could still understand the Homeric poems, although they were quite far from day-to-day speech; it is also possible to lose comprehensibility in a very short time -- the Turks managed to lose contact with Ottoman Turkish is one generation, see The Turkish Language Reform: A Catastrophic Success by Geoffrey Lewis (1999) (the link goes to Amazon).

Remember that in the meantime Japanse on Earth will continue to evolve. For a language to split into two daughter languages which become so much changed that their genetic relationship is no longer recognizable you need many millennia, and you absolutely need at least several centuries of complete illiteracy, so that the chain of written texts is broken. We can still recognize the genetic relationship between English and Hindi (they are distant cousins) although their most recent common ancestor language was spoken at least nine thousand years ago. (Of course, it helps that we have access to excellent attestations of very old Indo-European languages so that we can form a good impression of their common ancestor.)

Edit: User Simba (in the comment thread) makes the pertinent observation that the history of English may be too unusual -- English went through a close contact with Old Norse which may have influenced the evolution of its grammar; something must have happened to induce the astonishing loss of inflection.

In about the same time span which saw Old English "ġeār-dagum" become Modern English "yore days", Common Slavic (attested as Old Church Slavonic) differentiated into the modern Slavic languages (Russian, Polish, Czech, Bulgarian, Serbo-Croatian-Bosnian-Montenegrin and so on).

The important point is that even after 15 to 20 centuries of divergence, the three main linguistic families of Europe (Romance, Germanic and Slavic) are still obvious for speakers without any special linguistic education. All speakers of a Germanic language will recognize other Germanic languages as closely related, and the same for speakers of Romance languages and Slavic languages.

  • 1
    $\begingroup$ It is interesting that you stated that technology slows down evolution. That was true in the past but not true currently. Tech, such as texting and social media has accelarated evolution. What a twist $\endgroup$ Commented Dec 19, 2016 at 14:30
  • $\begingroup$ @FrankCedeno: We do not have at present a very good understanding on the effect of modern communications technology on language evolution. The only well-established effect is that modern technology tends to put a lot of pressure on local dialects which are put under a lot of pressure from the standard national languages. $\endgroup$
    – AlexP
    Commented Dec 19, 2016 at 14:34
  • $\begingroup$ Note that both of the examples given are complicated by the fact of other languages existing and mixing in with the main language in question. In particular the evolution of Old English to Middle English was heavily influenced by French after 1066. $\endgroup$
    – Simba
    Commented Dec 19, 2016 at 15:38
  • $\begingroup$ @Simba: The French influence on English is mainly in the vocabulary, which, for some reason, is considered by linguists to be of little importance. The French influence on English grammar and phonology was minimal at best. Old Norse had a much stronger influence than French. But in general you are right -- we have few historical examples of languages evolving without any significant contact with other languages. $\endgroup$
    – AlexP
    Commented Dec 19, 2016 at 16:18

tl;dr A long time.

Long version: I am a native german speaker and I can read the texts from Walther von der Vogelweide without many problems and he lived roundabout 900 years ago. The text may sound odd to my ears, but I'll understand it. The bigger problem in written texts is, that the fonts used for writing change over the centuries

You have a shorter timespan for spoken language because the brain has less time to identify the syllables. The problem is, that there are no real recordings of medieval European languages, due to the lack of technology. But one can easily recognise other languages of the same origin although they separated hundreds of years ago.

It may(!) be similar with the Japanese language, I'd expect the timeframe even longer, because there were not so many foreign influences in the Japanese language over the centuries.

But you can find out pretty easy for yourself: Take a trip to the oldest library and read old (and I mean old, not just 50 years) books. How far can you go back in time and still read them? If there are old history re-enactment groups in Japan, visit them and ask them to speak to you in the old tongue and compare it to your modern language.

  • $\begingroup$ There are many more words of foreign origin in Japanese than in German. Chinese-derived words alone make up something like 50% of the Japanese vocabulary. And "gairaigo" (loan words) from other languages are estimated at about 9%: en.wikipedia.org/wiki/Japanese_language#Vocabulary . Since Japanese doesn't use an alphabet, I'd expect phonetic changes to be faster than in alphabetic languages, too. $\endgroup$
    – user2727
    Commented Dec 19, 2016 at 9:48
  • $\begingroup$ +1 The Primary Chronicle / Tale of Bygone Years also dating back 900+ years is still somewhat readable for Slavic people, albeit with some difficulty, and yet it is the oldest Slavic text in existence. $\endgroup$ Commented Dec 19, 2016 at 10:42
  • 1
    $\begingroup$ @KWeiss How do you figure Japanese doesn't use an alphabet? They have three: kanji, the borrowed Chinese characters; hiragana, the syllabary derived from kanji and used for grammatical particles and words of japanese origin; and katakana, the more modern syllabary used for foreign loan words. $\endgroup$
    – senschen
    Commented Dec 19, 2016 at 13:03
  • 1
    $\begingroup$ @senschen None of the three Japanese scripts are alphabetic. Katakana is not newer than hiragana, it's just used most often for recent loan words. $\endgroup$
    – user2727
    Commented Dec 19, 2016 at 13:40
  • 1
    $\begingroup$ Also, surely if (like in many alphabetic languages) the script is very divorced from the pronunciation that would accelerate change. But for Japanese that's only the case with Kanji; Katakana and Hiragana are both very strongly-tied with pronunciation. So I'd expect phonetic changes to be slower in Japanese. IANA linguist though! $\endgroup$
    – Muzer
    Commented Dec 19, 2016 at 14:53

Depends on the level of isolation and population involved, but the divergence of the insular and continental north germanic languages might be of interested. As a small, isolated population that were not exposed to many outside language influences, they pretty much match the profile of your colonists.

Interestingly enough the insular languages (like Icelandic) are a lot more like Old Norse than the others. It seems likely that it will be the home language that diverges rather than that of the colonists.

Mutually intelligibility also doesn't follow the actual family tree. Icelandic and Norwegian both descend from Old West Norse, but Norwegian is intelligible with Swedish (east norse) but not with Icelandic. Perhaps at some point in the future a powerful space-fairing Japanese state will exert a large cultural influence Korea and find itself in the position of being able to converse with them but not with it's own colonists.

So how long did it take until Icelandic diverged from the continental Nordic languages? I think I might ask that question somewhere else on this network, but let's trust wikipedia for now.

"Between 1050 and 1350 Icelandic began to develop independently from other Scandinavian and Germanic languages"

"1300, the Danish language saw a very rapid evolution in both its phonology and its morphology"

"Norwegian and Swedish developed more slowly, but show equally notable differences from Icelandic"

Interesting, so the answer is different depending on which Continental language you compare it to. As I said earlier, it looks like your language change will probably come from the people left in Japan. So it depends on what happens to Japan in your story's world.

Rising Chinese influence might substantially shift what happens to the language. Perhaps Anglophone influence will continue via the internet (which the colonists probably can't get). Etc etc. Fairly huge linguistic (+ writing system) changes came to Scandinavia during this period, so you'll probably need something drastic if you want a short timescale.

But a period of a few hundred years seems to be enough.


A couple examples come to mind.

In the example of the Tower of Babel, God was able to make languages unrecognizable immediately. Maybe that is a replicable phenomena. And not explicitly or necessarily by the hand of God. An effect on people's brains during space travel could confound their language.

Another example is the Mayan language. It is still spoken in regions of Mexico, and various dialects are spoken in Guatemala. Sorry I have to dodge the question a little. I can't say whether the ancient Mayans would recognize any forms of modern Mayan. But here's my take on it. Introducing a new language (Spanish) didn't break the Mayan language. And the Guatemalans handled the changes far differently than the Mexicans. But even Mayan has a pretty well theorized etymology tracing back to Proto-Mayan from around 5000 BC. The thing is, I think it's safe to say the origin is unrecognizable because, well, who knows what Proto-Mayan is? Especially if you're talking about the average layperson?

That took a thousand years or so. But I think you are pretty free to pick almost any length of time.

Factors that would speed up the process would include:

  • losing records of the language,
  • losing the Japanese culture (through blending of cultures. The older generations might hold on for a long time),
  • special effects comparable to the Tower of Babel,
  • regional differences between colonists (and a region with a new dialect becoming dominant),
  • losing the language of origin
  • lots of people who love to make up words and disregard old words.

Factors that slow down the process:

  • Stable, powerful Japanese culture on Earth,
  • Well documented Japanese history and language on Earth,
  • But to recognize colonist language, Earthlings have to receive some communication from colonists, which is apparently impossible??

To make my answer slightly less vague, I would say a minimum of more than two generations for colonists to stop recognizing Japanese. It usually takes immigrants more than two generations to fully adopt a new language and lose their old language. It should also take more than two generations for Earthlings to not recognize colonists, assuming colonists change their language immediately. As a maximum, I would say 2000 years for either culture's language to become unrecognizable because I think the laypeople of most languages don't recognize the 2000 year old version of their language; the 2000 year old version was probably not the language but its likely dead parent languages. I think Japanese could become a dead language that laypeople don't recognize in 2000 years maximum, probably much less.

We seem to be able to trace just about any word etymologies back hundreds and sometimes thousands of years. To lose all trace of that, the Earthlings would probably have to lose all trace that the Japanese sent the colonists out in the first place, and they would have to lose all traces of Japanese. Again, you have a lot of freedom with the time frame for that.


As I see it, there are two parts to this question.

1) The base of the language - the general rules, language construction, common words, etc. These might never change significantly; the major languages are all heavily codified in dictionaries, computers, school textbooks, etc and there's little advantage in changing these.

2) There are however always new words entering languages. Especially with your space-faring civilisation, where with a vastly different environment in space and on the new planet there would be a need for lots of new words to describe things. This could, over time and when combined with new colloquialisms, create a dialect in which the language was fairly recognisable, but where a lot of what was spoken didn't make much sense. Accents may also change; until eventually you're in a situation much like a Glaswegian trying to speak to a Jamaican.

  • $\begingroup$ Codified languages change nevertheless, even Arabic (where the codified classical language has religious significance) has changed a lot from the Classical period to the modern Arabic dialects that are not even mutually intellegible. $\endgroup$ Commented Dec 19, 2016 at 16:45

There are several degrees of "recognisability":

  • Mutual intelligibility: The speakers of the two languages can make sense of a significant amount of the other's speech or writing. This type of recognisability last for a time depth of a few hundred years to about one thousand or one and half thousand years.
  • Genetic relation: With current methods of historical linguistics, Genetic relationship can be demonstrated for languages with a common predecessor about ten thousand years ago. There is not much hope to improve on this time limit.

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