I know linguists like to say "no languages are older or younger than other languages" because they all evolved from ancient roots. With exceptions such as Nicaraguan Sign Language.

So let me explain that I'm specifically talking about cases where one settled language splits and one branch gains a degree of non-mutual-comprehensibility with the trunk. I'm interested only in the cases where few linguists would disagree that the result is two languages.

The recent cases I can think of are:

  • Yiddish from High German
  • Afrikaans from Dutch
  • Maltese from Arabic

Without getting too deeply into controversial cases, would one of these or some other I've overlooked be the most recent?

  • 2
    Note that these are all cases of isolation; speciation through geographical separation. Given that nobody was keeping statistics about mutual intelligibility until quite recently, I'd guess that some Asian language (possibly Hakka, Hmong, Hiligaynon, or some other minority language with a large overseas contingent) would have split into more-or-less mutually unintelligible groups (like Cantonese and Mandarin are in China today, for instance). Then there are innovations like Light Warlbari. "Splitting" is not quite as monolithic as meiosis appears.
    – jlawler
    Commented Nov 24, 2014 at 5:24
  • Of course but I tried to explain what I'm looking for. I would assume that there's a degree of mutual comprehensibility between Light Warlpiri and Warlpiri or at least that deciding that is controversial. Not having an ISO 639 code would make it and its ilk not uncontroversial anyway. I really don't think it's hard to see what kind of languages I'm asking about here. Commented Nov 24, 2014 at 7:27
  • What do you mean by "youngest"? The split was the most recent in history? The split happened the shortest time after the parent language split from its parent?
    – curiousdannii
    Commented Nov 24, 2014 at 7:39
  • The most recent in history, as in the examples given \-: Commented Nov 24, 2014 at 7:40
  • 1
    I would suggest that the place to look would be the many Creoles around the world today. But they may not meet the criterion of a 'settled language diverging'. It would be helpful if you specified why you want to know. The key problem with the question is that the universe of language change and mutual comprehensibility is too complex to be easily captured in just this one dimension. Commented Nov 26, 2014 at 7:53

3 Answers 3


A recent example from Southern Africa:

The Mfecane in South Africa in the early 1800's, caused by Shaka's military campaign, created a whole lot of social upheaval, the repercussions of which were felt as far away as Tanzania.

In the midst of all the attacks and counter-attacks, invasions &c, a powerful group of Sesotho speakers (the Makololo), having adopted many of the techniques used by Shaka, went on the rampage, ended up in Southern Zambia, and subjugated a community of Luyana speakers. The original conqueror died and his daughter took the reins, his daughter gave up the throne to her brother, and her brother was defeated by the Luyana some 30 years after the initial invasion, chasing most of their invaders out of their homeland.

During those 30 years the Makololo had forced the Luyana to speak Sesotho -- a rather distantly related language with clicks, a large number of vowels, a distinction between aspirated and unaspirated consonants, and many other features which made the language difficult for Luyana speakers -- and the result is that the Luyana speakers developed a new language derived almost entirely from Sesotho (essentially, a creole).

That is, we have

over a period of a few decades, as recently as 170 years ago.


We do not (I believe) want to get into the tedious discussion of what is a “language” as opposed to a “dialect”. I think most of the instances of what you are asking about involve situations where an already written language is provided, for political reasons, with a new system of writing as a way of distancing itself from what had until then been regarded as a form of the same language. For instance:

Portuguese > Galician

Romanian > Moldovan

Persian > Tajik

and probably many others.

  • Actually my understanding is that Portuguese and Galician share a common ancestor rather than the latter being an offshoot of the former, Moldovan is uncontroversially an orthography of Romanian and there is mutual comprehension, Persian and Tajik and Ossetian and others all share a common ancestor with the orthographies not part of the linguistic question. But I may be wrong about some of these if supporting evidence is shown. Commented Nov 25, 2014 at 2:30
  • Standard Tajik is simply classical Persian, written in Cyrillic script, reflecting how Persian was traditionally read by the literati in Bukhara. The spoken dialects of Tajik stand in the same relationship to Classical Persian as the Persian spoken dialects in Iran and Afghanistan. At present there is actually a debate in Tajikistan about whether or not to reintroduce Arabic/Persian script. Persian is a South-West Iranian language. Ossetic is an Eastern Iranian language and is completely different from Persian/Tajik. Galician is debatable, but it is in any case very close to Portuguese,
    – fdb
    Commented Nov 25, 2014 at 9:27
  • Interesting regarding Tajik and Persian. Agreed that Galician is very close to Portuguese or vice versa, but closeness isn't what the question is about. It's about most recent branching, not even how common branching is. Commented Nov 25, 2014 at 12:29

In some ways there are a huge number of dialects which are no longer mutually intelligible. How well would someone with a broad Texan accent understand someone with a broad Scots accent? How well would someone speaking African American Vernacular English understand someone speaking Australian Aboriginal English? But these are usually considered to be all one language because even though the extremes may not understand each other, there is a chain of dialects between them which can all be understood. (And most people can speak at different registers.)

Perhaps an example might be the Melanesian creoles of Tok Pisin, Bislama and Torres Strait Creole. They were developed in Queensland in the 1870s onwards, but the first two were then taken back to PNG and Vanuatu. Some people say that they are still mutually intelligible, but after spending some time with one of them I'm sceptical.

  • The examples in your first paragraph were the things I was trying to filter out by asking only for instances that were accepted without controversy. While Kriol is accepted as a language none of the examples listed are. I'm not ignorant of them. They're just not what I was looking for when I posed this question. Commented Nov 25, 2014 at 2:32
  • The creoles are much better examples and for some reason I only thought of them after I'd already asked the question. I wondered about creoles having more than one parent making them other than offshoots but in my examples Yiddish and Maltese clearly have several main "parents" even though they're not usually labelled "creoles". Commented Nov 25, 2014 at 2:34

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.