6

I'm coming at this with a limited grasp of some European languages and for my world I'm faced with the daunting task of creating anywhere between 9 and 2000 languages (we never pick easy projects do we!?). They're nearly all for humans or at least humanoids.

So far most of the characters speak "common" with the odd phrase thrown around in the local tongue. Typically I sort of feel what a language should sound like (sing song, rough, playful, eloquent) type some random words (roughly matching the number of words of what I want them to say) and move on.

This isn't sustainable.

For starters I know I have no sort of dictionary for these languages, secondly I know I'm missing all sorts of rather important things like tenses.

Is there an established technique or procedure for creating, maintaining and developing a new language from scratch? What bullet point steps do the experts follow?

Liath
  • 18,600
  • 11
  • 96
  • 135
  • At the moment this feels far too broad, I think there is a core of a good question here though. Maybe try and narrow down the requirements a bit or split it into multiple smaller questions? – Tim B Sep 25 '14 at 11:09
  • 1
    Unfortunately, this is just too broad. It completely fails the "book test" and needs to be narrowed down a lot to fit within the Stack Exchange framework. – user Sep 25 '14 at 11:09
  • Fair enough - newbie question not quite realising the scale. Unless someone can recommend a way to bring down the scope a lot then let's close it as "Too Broad" – Liath Sep 25 '14 at 11:12
  • Given the history and breadth of constructed languages, this is both too broad and too opinion-based (every answer presenting an existing framework would be equally valid) for Stack Exchange to encompass. – BESW Sep 25 '14 at 11:15
  • Perhaps break the question down into steps and narrow in on those steps. For example "how do I design the grammar for a constructed language" is smaller in scope than this question but still too broad. Really you need to think about just what you are trying to achieve and the exact problem you are facing. For example if you just want a few flavour words to throw into conversation (substituting a few words from the imaginary language from english) or want to mess with grammar a bit or add an accent those are all different requirements from creating a full conlang. – Tim B Sep 25 '14 at 11:21
  • 1
    A type of question that would be more likely to fit within the Stack Exchange format is "I'm designing a grammar for a constructed language; what word classes are generally necessary to convey meaning and why is each required?". However, that wouldn't give the overall picture which I feel is what you seek. This is perhaps the type of question you would want to bring up in chat to try to break down into managable chunks. We also have some precedent here (might be a duplicate of this) and here. – user Sep 25 '14 at 11:28
  • @James Please take that discussion to [meta]. In general, Stack Exchange frowns on questions specifically seeking off-site resources. – user Sep 25 '14 at 16:59
  • During the definition phase we specifically discussed book/reference recommendations being on topic. Thats all I was suggesting, I should have been more clear. – James Sep 25 '14 at 17:59
  • Well before this gets deleted, I want to point out that if you actually try to make working languages you are looking at a lot more effort than you have time for. Seriously; you're basically going to have to go get a PhD in Linguistics and you'll probably be about halfway there. Even Tolkien's languages were largely incomplete. It may help to ask a question about your specific use-case because in a lot of cases it isn't terribly necessary (nor good) to actually write in another language. Depending on your case, maybe start on writers asking when its good to use other languages and when not? – Wesley Obenshain Sep 26 '14 at 02:28

0 Answers0