Is there any evidence that a ‘natural’ language that exists today was created by somebody a thousand (or two) years ago. And that it evolved till today?

  • 3
    I'd like to point out that the definition of natural language is (retronym, linguistics) Any human language that has evolved naturally in a community, usually in contrast to computer programming languages or to artificially constructed languages such as Esperanto. Could you clarify what you mean by the quotes around 'natural'? Otherwise, this question has by definition no answer (if natural language's definition is held) or is too broad (if you don't define the scope of ''natural' language'). – Duncan May 2 '18 at 19:06

Possibility? Of course, there is always a possibility, but rather remote one, if you mean a full featured well-known modern language.

The best candidates are "taboo languages", where the lexical items are replaced because of taboo/religious pressure. Perhaps the best known is Dyirbal, where a special version of the language is to be used in front of "tabooed" persons.

Though it probably does not go millennia into the past, and it's questionable if it should be considered a language, since it is just a vocabulary encoding.

  • These sound like registers to me. – jambrothers May 2 '18 at 20:49
  • Were some cants and argots constructed? The established ones have probably become 'natural'. – Qsigma May 9 '18 at 9:02

Yes, the Damin language (Wikipedia article), a ceremonial language and only natural click language outside Africa, was probably constructed.


People learn languages from other people who speak them. If someone ever got others to adopt his language as their "natural language," they would have had to learn it from him. Without a community of speakers, it would have been impossible for this to happen. At some point we have to account for the fact that someone would have had to design the language and actually teach it to others.

I believe that this has never happened in recorded history. We do indeed find that an early Greek philosopher named Pherecydes invented an "artificial" word for "table":

He maintained that the divine name for "table" is θυωρός, or that which takes care of offerings. (Diogenes Laertius 1.119)

There is no evidence for anyone ever having invented an entire language, which would certainly be much more notable.

This leaves thousands of years of prehistory (before the invention of writing) when this could have happened, but it seems much more difficult to design a language without writing, or, for that matter, to do it without the linguistic knowledge that exists today.

Coining of new words has always happened, but that is language change, which develops from an existing language, and not the invention of an entire language from scratch.

  • I disagree with your claim that no evidence anyone invented an entire language. People these days do it all the time and my answer provides some tentative claims (depending on definition of entire language). – qwr May 2 '18 at 18:53
  • Of course there is evidence for people having invented entire languages. This has happened with some frequency throughout history. The questionable thing is more a matter or if there was one that was constructed several thousand years in the past that's still in use. – Cubic May 3 '18 at 12:21

You could argue that modern Hebrew is at least partly constructed.

Hebrew fell into disuse and only survived as a 'sacred' language in a religious context. Then, with the rise of the Zionist movement towards the end of the 19th Century, Hebrew was modernised and used as a lingua franca in the Palestine region. In the process, aspects of Hebrew variants were combined, and the vocabulary was extended to allow the use of Hebrew in modern day life.

See the Wikipedia article on the Revival of Hebrew.

  • I know the intricacies of modern Hebrew, I wouldn’t argue that. :-) – Dr. Shmuel May 3 '18 at 11:38
  • 3
    On re-reading the question, this is actually kind of the other way round: a natural language from more than 2000 years ago died out and got revived through constructing a modern version of it. – Oliver Mason May 3 '18 at 11:38

Cryptolects (aka Cants or Secret languages) come to my mind as candidates. While they are typically not fully constructed, they often contain a constructed core vocabulary making the language unintelligible to outsiders. When they exist long enough, they are learned as first language by the group members.

Examples of such cryptolects include Verlan (French based) and Läppe Tellep (German dialect based).

EDIT: It took me some time of searching for this one because I forgot the name: Eskayan spoken on Bohul island in The Phillipines.

EDIT2: Another example where it is probably no longer decidable whether it is an isolate language or a secret language is the Nihali language in central India.

  • 2
    Eskayan is a bit dubious of an example because it's a) only ~100 years old, not 1,000, b) I can't find any strong evidence of people actually speaking this outside of very specific contexts (i.e. is not how I interpret OP's meaning of "natural") and c) it's more or less just Boholano with a largely artificial lexicon: If only a (partially-)constructed lexicon satisfies criteria, e.g. Turkish would be a much better candidate. – errantlinguist May 2 '18 at 13:34
  • Indeed, time depth is a problem—it will be very hard to find something fitting that is a millennium old or older. – jk - Reinstate Monica May 2 '18 at 13:39
  • 2
    I'd be willing to bet that with enough digging one might find a secret "language" with only an artificial lexicon especially among clergy that may have lasted in some form for a while, but, again, it would only be used for very specific purposes and only by a very few people. – errantlinguist May 2 '18 at 14:06

Depending if you include scripts in the definition of "natural language", Hangul is a famous example of an alphabet that was created in the 15th century and has evolved since then.

In addition, the notoriously difficult Tangut script was created "in a very short time" around 1036 and was widely used in books and inscriptions by the Western Xia for about 500 years.

  • Yeah, I was going to mention Hangul as example of successful man made script, but anyways it’s not a language – Dr. Shmuel May 2 '18 at 19:35
  • 1
    Are there any scripts that weren't constructed by someone? At least Glagolitic and Cyrillic definitely were. – Vladimir F May 2 '18 at 19:37
  • @VladimirF my examples are scripts we know were created in a short amount of time – qwr May 2 '18 at 19:38
  • OK, but so are mine. – Vladimir F May 2 '18 at 19:49

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