I read in the Linguistics section on the Wikipedia page for American Sign Language that ASL was "proven [to be a natural language] to the satisfaction of the linguistic community by William Stokoe, and contains phonology, morphology, semantics, syntax and pragmatics just like spoken languages" (emphasis mine).

Are these elements, of phonology, morphology, semantics, syntax, and pragmatics, generally accepted as the necessary and sufficient components of a natural language? Would a constructed language which developed such features be accepted as "natural"? (Note that American Sign Language itself would be an ambiguous case, being a systematically modified version of Old French Sign Language, a signing language which developed "in the wild", so to speak, in the deaf community in Paris well before the French Revolution.)

To consider a somewhat flippant example: what would be the necessary conditions, linguistically (as opposed to sociologically) speaking, for either Esperanto or Klingon to develop dialects which would be considered natural?

Edited to add: I am accepting Alennano's answer as being the most factually correct with respect to what it means for a language to be "natural"; but I also think that, from the answers and comments, the question of whether a language is "natural" is completely beside the point on a matter of linguistics, except inasmuch as the vast majority of "vernacular languages" happened to evolve more-or-less organically most of the time from a prehistoric origin.

  • 1
    I don't think this is how a SE is supposed to work. You could add that part to the previous wording, but rewording completely the question is not the solution. What do we do with the existing answers?
    – Alenanno
    Sep 20, 2011 at 17:44
  • Any edit can invalidate previous answers. In this case, it looks like he simply narrowed the scope of the question.
    – Alek Storm
    Sep 20, 2011 at 17:50
  • @NieldeBeaudrap: It looks like you're redefining some entrenched terms, which is likely to cause confusion. I would replace your usage of "vernacular" with "language", and "language" with "system of communication" (especially in the title).
    – Alek Storm
    Sep 20, 2011 at 17:52
  • 1
    @NieldeBeaudrap: I don't see how it's trivial. Birdsong is a system of communication. Esperanto is a constructed language, and English is a natural language.
    – Alek Storm
    Sep 20, 2011 at 18:02
  • 2
    If you have a separate question, you should post it separately. Severely revising a question invalidates any answers that people spent time on to answer your original question. Sep 21, 2011 at 5:55

4 Answers 4


I'd say a natural language has (or had) native speakers. Natural languages don't cease to be natural when their last native speaker dies. A native speaker is one that learnt it as a child. This means that it had at one time been learned as the first language of children (one child is not enough). So if adults got together to learn a language, it couldn't be a natural language until their children could use it as their sole and first language. A language that is no longer learnt by children will go extinct.

The test of a definition of natural language is what to do with Esperanto. It now has at least three generations of native speakers (though they are all multilingual). Can a constructed language become a natural language by being the first language of children or will a constructed language always be a mere constructed language no matter what. It should be noted that the Esperanto these people speak is different and more complex than the Esperanto of the parents, who learnt it as adults.

  • Given your remarks about Esperanto, are you saying that a natural language is one that is or was the native tongue of a substantial number of children, except when it's not? Sep 19, 2011 at 14:13
  • A minor point: the phrase "sole and first language" muddies the waters a bit with respect to multilingualism, a situation in which children can have more than one "first" language. Regardless of whether they speak another language, whether the Esperanto-speaking children speak Esperanto as a first language (learned fully during the critical period) is precisely what is at issue in deciding whether it has become a natural language. The structural complication you point to argues that the answer is "yes."
    – Aaron
    Sep 19, 2011 at 14:33
  • @Niel: I think what kaleissin is trying to say is that a -first- language cannot be taught academically, and to be a language at all for a first learner, it must be part of a community (more than 1). Any artificial language can become native by the first first-language learner. Modern Hebrew could be considered this way (it was an academic/sacred language, like Latin, for centuries until resurrected). I would expect large changes/adaptations of the rules of the constructed language into the natively spoken one, just like the change from pidgin to creole.
    – Mitch
    Sep 19, 2011 at 14:34
  • @Mitch: that makes plenty of sense to me; my concern was basically whether kaleissin was laying forth criteria for whether a language is natural, and then saying that "we still have to find out" whether a language which fits those very criteria is natural. I was concerned whether the goalposts were being moved (by the academic community), essentially. Sep 19, 2011 at 14:39
  • 2
    Saying that a "natural language" is a language that has native speakers is an incorrect usage of the term in the field of linguistics. Esperanto, though totally awesome, is not a natural language, though it has native speakers.
    – mollyocr
    Sep 19, 2011 at 17:48

A natural language has developed spontaneously over time and it has done it gradually. While an artificial language has been created in a relatively short time, usually for a definite aim. Now that other one is the main difference: it has been created. Also natural languages are man-made but the genesis is totally different, in terms of procedures and time.

No-one thought "let's create a language called English", over time from Old English to the English we know, it developed "by itself", through people's usage. While Esperanto, for example was created in more or less 15 years.

Unlike pidgins, an artificial language has a fairly complex grammar in addition to a good phonology system that is usually as complex as the creator's language, a good furnished lexicon and it's provided with syntax too.

Natural languages have all these things as well, but like I said, in contrast to the artificial languages, they were acquired over time. There were influences, development. In other words: an evolution.

EDIT: I read your question again and I wanted to add that those parameters (syntax, phonology, semantics, syntax, etc) are not parameters to distinguish a natural language from the rest, since also artificial ones have them.

I'm not sure how it works with Sign Languages (about being a natural language or not), but since you mentioned spoken languages too, I answered to that.

EDIT 2: Answering to your comment, the point was to prove that ASL was not a "bad gesturing imitation" of the English language, but rather a true language on its own; this was done by Stokoe, who also encountered many obstacles doing it. The American Sign Language is complex enough under all those points, so it can't just be a simplified version of English; not to mention that it didn't even derive from English. Stokoe wrote some texts about this in case you're interested to go more in depth.

  • 1
    This does not preclude, however, an evolutionary trail of language starting at an artificial language. The differences you are describing between an artificial language and a natural one are not really linguistic in nature; I thought there might have been such a distinction, as discussion of "constructed languages" were proposed to be excluded from this site entirely. Sep 19, 2011 at 10:34
  • It doesn't preclude it, sure. But contrasting an artificial language and a natural one, that one is the main distinction: a spontaneous language vs a purposely created one.
    – Alenanno
    Sep 19, 2011 at 10:38
  • Why the artificial languages have proposed to be excluded? Anyway, I mentioned them just to make the "natural language definition" clearer. I figured a distinction would help to visualize things better. :)
    – Alenanno
    Sep 19, 2011 at 10:46
  • Why it was proposed (by others who were defining the scope earlier) that constructed/artificial languages be excluded would be a question for the meta subsite. Sep 19, 2011 at 11:21
  • So, do you think that in the case of ASL, the point was to prove that it was a language, as opposed to a less rich or systematic mode of communication? Sep 19, 2011 at 12:43

Phonology, morphology and etc are not elements or components that can be developed, they are levels of linguistic analysis. At the level of the morpheme, rules tell how the morphemes fit together into a structure. If they are just jumbled together with no structure, there is no language. That was thought to be the case with ASL, until Stokoe proved otherwise.

If someone invented a language it would by definition have such structures at each level of analysis. Later, if users of the language developed different rules and structures, one could ask if it was now sufficiently different to be a new language, and if so, if that language was 'natural'.
ASL was proven to be a natural language by showing that it was not invented, it is not a modified version of OFLS or anything else; it was not planned, it arose spontaneously, with genetic roots in at least three existing signed languages, and none in any spoken language.

Not all sign language is ASL. Not all “sign language” is a natural signed language. Some is just gestures; some is Signed English or a manual code for some other spoken language. Signed English is an artificial language which was planned and so forth; ASL is a natural language.

enter image description here

Being a constructed language, Signed English has structure at all the above named levels. At the level of phonology its structure is partly that of ASL and partly invented. At the levels of syntax and morphology it has the structure of English. Syntactically, as shown here, the two languages have different word order. Morphologically, Signed English uses one sign for each English word, ASL uses fewer words and more affixes.

It's worth noting that there are children whose first and only language is Signed English, but that doesn't make SE a natural language.


To answer the question: IMO, Esperanto and Klingon (and other constructed/synthetic languages) are not and therefore never will be natural languages. Among native speakers, the languages will begin to act like natural languages because there will always be language change with time, but I think we'll need to coin a new phrase for these types of languages. (The language formerly known as constructed? Bam! Prince joke.)

I'm still not sure about calling ASL a natural language. It is definitely a language of its own right, but it did not happen entirely spontaneously. I guess I wonder about calling any sign language a natural language when it develops in the context of a spoken language society. This is why Nicaraguan Sign Language is so interesting, which developed (among children!) like a pidgin>creole language in an almost entirely deaf community, which may be the closest a sign language could get to being natural since it comes from necessity instead of planning. By the time researchers came in to observe, it already had a basic grammar.

  • This no longer answers my question as it is worded, although thank you for taking the time to answer it as it was originally worded. Sep 20, 2011 at 17:42

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.