Wikipedia:Conlangs/Criteria

These were the original suggestions from Almafeta's userspace:

If at least two of these minor conditions can be met, a conlang is notable:
- Has at least 50 speakers. Few constructed languages get to this mark.
- A book with an ISBN has been published about the language. That prevents many small-press and unsupported languages (other than those who think it's a good idea) from having articles.
- A book with an ISBN has been published in the language about a topic other than the language itself. That shows that it's used.
- Older than usual. According to Langmaker.com, 1950 (post-WWII) was when the personal constructed language really exploded; there are only 78 known conlangs before then, and those should be considered notable for predating the 'conlang explosion.'
- Among the 100 most popular conlangs, as determined in Langmaker.com.
If any of the following major conditions can be met, a conlang is notable:
- Been mistaken for a real language.
- Caused controversy. For example, Adjuvilo was created by an Esperantist to ferment dissent in the Ido movement.
- Has (or had) at least 500 speakers.
- Has an ISO code in any of ISO 639-1, ISO 639-2, or ISO 639-3.
- Inspired a conlang that was notable by one of the major conditions. For example, Occidental directly inspired Interlingua.
- Has a Wikipedia. I mean, really.

-- Almafeta 14:28, 28 July 2005 (UTC)[reply]

Concerning the "been mistaken for a real language" criteria, how would we verify this?Emînough (talk) 23:47, 28 June 2013 (UTC)Emînough[reply]

Other suggested criteria

Feel free to add any criteria you come up with to this list. (Don't forget to sign them.)

Has caused active discussion (at least)
- Has been discussed in a professionally published book or magazine article
- Google search turns up at least N independent discussions/reviews of the language by people other than the creators --Jim Henry | Talk
Was created by J. R. R. Tolkien.
Sufficiently large vocabulary (~1000-2000 words)

2,500 words has been more frequently mentioned and would probably be a good threshold. I tentatively proposed criteria for the quality of the lexicon but IJzeren Jan talked me out of it. --Jim Henry | Talk

Worked by m people for n years. (Carlos Th (talk))
Differential criteria due to purpose (Carlos Th (talk))
Has a corpus of at least 25,000 words publicly available (in books, on the web, or mss. in a library's special collections) --Jim Henry | Talk
Has a reference grammar that covers all or most of the points in the Lingua Questionnaire (I see this as a way of making the "has a thorough grammar" criterion more objective) --Jim Henry | Talk
Has been used in professionally published fiction/media. [Should we quantify "professionally published" in terms of print run, as in the John W. Campbell Award?] --Jim Henry | Talk 20:52, 1 August 2005 (UTC)[reply]
A conlang should be included if and only if it can be the subject of an informative, NPOV, well-sourced article that includes opinions and facts gleaned from verifiable sources other than the creator's own site, provided that the article does not stray into Original Research. This is essentially a restatement of the general criteria for inclusion of any article in Wikipedia. As a practical matter, any conlang that has been sufficiently discussed to make such an article will tend to be significant, interesting and/or have artistic merit, and any conlang worth including will acquire such commentary eventually, at which point it can be added to Wikipedia. Robert A West 16:56, 2 August 2005 (UTC)[reply]

Let's say "...other than the writings of the language's creator(s)". That will make it general enough, not tied to conlangs by a single author published only on the web. --Jim Henry | Talk 17:20, 2 August 2005 (UTC)[reply]

For the "books about a language" criterion, there should also be requirement of either a) having an Amazon book sales rank number lower than 500,000, or b) verifiable evidence of sales of more than 100 copies. Anyone can buy an ISBN number. I personally know of one book with sales of less than 50 copies: it has an Amazon book sales rank number a bit above 700,000. Dpbsmith (talk) 16:30, 5 August 2005 (UTC)[reply]

I posted earlier saying I oppose the "with an ISBN" criterion entirely because it would exclude older professionally published books and include newer vanity press books. Your sales criteria are an improvement, but I think it would be better still to allow in books with lower sales rank if they they are written by someone not involved with the language's creation. Being able to reference independent writing about the language by someone other than its creator means the language is verifiable and we can write an article without doing original resarch. --Jim Henry | Talk 19:51, 5 August 2005 (UTC)[reply]

Has a grammar of itself written in itself. ThomasWinwood 22:24, August 13, 2005 (UTC)

This is certainly a great feat and an interesting criterion, but it shouldn't be a cut off point, as most (and I do mean most) natural languages will fail this criterion. 72.21.131.202 (talk) 21:50, 22 March 2013 (UTC)[reply]

Has established notability outside of the conlanger community. The Literate Engineer 05:38, 30 August 2005 (UTC)[reply]