Monday, 15 March 2010

Conlang Article on the New York Times

I know this article has been up for a few days already, but it's notable enough that I wanted to link to it here. The article is entitled Questions Answered: Invented Languages, and is a lengthy series of questions and answers on the subject of Constructed Languages. The questions were asked by readers of the New York Times, while the answers were eruditely given by Arika Okrent (author of In the Land of Invented Languages, one of the few to have studied the modern conlanging movement without prejudice) and Paul Frommer (the linguist that was hired by James Cameron to develop Na'vi, the language of the blue aliens in Avatar).

The article is remarkable for quite a few reasons:

  • It's not everyday that a mainstream publication devotes some time and space for conlanging (OK, it's a New York Times blog, but still).
  • It's very long and articulate (and not for the faint of heart. I advise everyone to read it in two or more sittings. For those not familiar with this particular hobby and art form, some of the answers will need a while to digest).
  • It goes further than the usual targets of Esperanto and Klingon (thanks to Arika Okrent who really knows her stuff).
  • It's surprisingly positive! Although a few questions betray a somewhat negative bias against the Secret Vice, they are very few, and Arika handles them neatly.

One thing that jumped to my attention is that the general questions about invented languages were all answered by Arika, while Paul Frommer only handled the questions concerning Na'vi itself. I wonder why that is. I can think of a few explanations:

  • The questions themselves were divided in two groups: generic versus Na'vi, and Paul received only the Na'vi themed questions.
  • Paul felt he was not familiar enough with the conlanging world to answer meaningfully. Fair enough, he's quite the newcomer here.
  • Paul doesn't have any actual interest in conlanging itself, and only does those interviews because he is contractually obliged to do so (he was contracted, i.e. paid, to create Na'vi, after all, and there's no evidence that he has done any conlanging before this). Once again, fair enough. Not everyone needs to be bitten by the conlang bug!

Note that I'm not judging here. Whatever reason Paul fοcussed only on the Na'vi questions is fine by me. It was just an obvious pattern that I can't help wondering about.

Of course, this blog post would be a bit pointless if I didn't add my two (Euro)cents to the article. So here are my own answers to a few of the questions asked in the article:

What is the process for “making up” a language? Do you just go on inspiration, or is there some preconceived structure to your work? In what order do you proceed? Where do you start?

There seem to be two main types of conlangers out there. The first type works on language creation in a very systematic way. They have a kind of shopping list of what needs to be designed in a language, and follow it to the letter. That list contains usually phonology, morphology, syntax and vocabulary, and usually in that order. The other type works on their conlangs in a much more organic way. They design features as they need them and seem to let the language grow, rather than go by consciously designing it. They also tend to have an iterative approach, revising older parts of the language as they invent new features. Both approaches have their merits and pitfalls, and in truth most conlangers will fall somewhere between those two extremes.

Personally, I'm rather systematic in my conlanging efforts, although once I've left a conlang lie for a while I may pick it up again and start revising it, in a sort of iterative process not unlike software design.

Do you create an alphabet first? When languages are invented, how do the creators choose the set of vowels and consonants the language will have? Do you utilize the International Phonetic Alphabet?

I normally define the phonology of a language using the IPA (which is a godsend for anyone creating a spoken language!). Creating a writing system (not necessarily an alphabet) is often one of the first things I do after defining the phonology and morphophonemics of the language, but depending on the type of writing system it can take a long time before I'm finished designing it (in which case I will normally work on it in parallel with designing the language itself). I also often create a transliteration of the language, because sometimes the writing system is just impractical for my tools at hand (no computer fonts available), and because writing everything with the IPA gets old very fast.

Why not revive and disseminate an endangered language rather than make up a new one? There are innumerable beautiful languages around the world that will disappear in a generation, and precious few folks recording them and learning them. The notion of a “better” language seems to ignore that language use follows population growth, transportation, and political and economic power. Why would one choose to invent a new language (aside from the whole “wildest academic dreams” thing) rather than revive a “dead” language or a dying one, like Cornish or Manx?

I'm not a field linguist. Actually, I'm not a linguist at all! I have no professional linguistic training, and all I know of linguistics I've learned in my own time. I have also no interest in becoming a professional linguist: my current job is more than enough, thank you very much (and it probably pays better than whatever I could make as a linguist!). Conlanging is just a hobby for me, and doesn't take time away from my saving endangered languages, because even if I had any interest in it I wouldn't have the ability to do it! Besides, I'm a Frenchman. How would my learning Cornish or Manx in any way help save those languages? I would probably never be able to reach a level of command of the language that would allow me to communicate with it natively anyway (and with whom? Do you really think the only reason the Cornish community isn't speaking its language is because I haven't learned it yet?). Wouldn't it make more sense to investigate the reasons why those languages are endangered in the first place, and help the communities speaking them by solving those issues? That's a socio-political problem, not a linguistic one. In any case, that's not the kind of problem you would even want a French hobbyist to tackle! Anyway, Arika summed it up best when she wrote: There is no general pool of “effort” from which all endeavors are drawn. Even if you somehow managed to make me stop conlanging, what makes you think I would use that extra time to study an endangered language? (notwithstanding the fact that doing so would probably be as useless for that language as my conlanging anyway) I would probably spend it sleeping!

OK, enough rambling from my part, just go and read that article. It contains many more questions masterly answered by Arika and Paul, and you'd make yourself a disservice if you missed it. Even if you're not interested in conlanging yourself, it might help you understand why other are, and why it's not a complete waste of time, or at least no more than any other artistic endeavour!

Reblog this post [with Zemanta]


  1. I think that the choice is between English or Esperanto as the future global language rather than an untried project.

    Your readers may be interested in

    Dr Kvasnak teaches English at Florida Atlantic University.

    A glimpse of Esperanto can be seen at

  2. Thanks for posting this link! Tempo ni la mi kama sona e Toki Pona.

  3. Good post.

    That question regarding reviving old languages as opposed to making new ones bugs me to no end. It's as pointless as the question about what the point of conlanging is.
    My answer to both is: To be creative. Period. /opinionated

    I do have an interest in (near-)extinct languages, but only as sources of inspiration for the next conlang.

    So. There's my two cents. :)

    Re: Transliterations - Just out of curiosity and only if you do any worldbuilding for your languages, do your transliterations suffer from the same bugs real world transliterations do?

  4. Indeed. It's like complaining to a painter specialised in landscapes that they should spend their time doing environmental work in the Amazon jungle instead. Where does *that* follow?

    As for transliterations, mine are usually meant only for outside the conculture the conlangs are set in. So I try to prevent bugs as they are supposed to give a good idea of the pronunciation of the language.
    When I make a Roman *orthography* for a language, however, it usually has quirks and bugs like all orthographies.