Loading ...
Sorry, an error occurred while loading the content.
 

How to evaluate a conlang

Expand Messages
  • João Ricardo de Mendonça
    How do you evaluate a conlang? How do you define a good or a bad conlang? I understand this is a personal criteria, so I m not looking for a definitive
    Message 1 of 13 , Sep 27 8:15 PM
      How do you evaluate a conlang? How do you define a "good" or a "bad"
      conlang? I understand this is a personal criteria, so I'm not looking
      for a definitive answer. I just wanted to know other people's opinions
      on this.

      João Ricardo de Mendonça
    • Peter Bleackley
      ... I personally would start by reviewing the artistic aims of the creator. Are his or her aims interesting, or coherent? Are they flawed in some way (e.g.
      Message 2 of 13 , Sep 28 1:39 AM
        Staving João Ricardo de Mendonça:
        >How do you evaluate a conlang? How do you define a "good" or a "bad"
        >conlang? I understand this is a personal criteria, so I'm not looking
        >for a definitive answer. I just wanted to know other people's opinions
        >on this.

        I personally would start by reviewing the artistic aims of the creator. Are
        his or her aims interesting, or coherent? Are they flawed in some way (e.g.
        setting a goal that cannot realistically be achieved)? Does the conlang
        achieve its stated aims, and how does it go about doing so? e.g. Kélen aims
        to be a convincingly alien language, and does this by abolishing the large
        open class of verbs, found in all human languages, with a small, closed
        class of relationals. I find it one of the most convincing alien languages
        I've seen. Is the language intelligible? Try translating a passage - relays
        are good for this. Finally, how do you find the language from an aesthetic
        point of view? This is the most subjective bit, of course, and one where
        you might find yourself at odds with the creator's design goals.

        Pete
      • Henrik Theiling
        Hi! ... I d say in *most* languages. The counterexample I will cite without knowing much about the language is Basque, which was claimed here to have only
        Message 3 of 13 , Sep 28 6:12 AM
          Hi!

          Peter Bleackley <Peter.Bleackley@...> writes:
          > Staving João Ricardo de Mendonça:
          > >How do you evaluate a conlang? How do you define a "good" or a "bad"
          > >conlang? I understand this is a personal criteria, so I'm not looking
          > >for a definitive answer. I just wanted to know other people's opinions
          > >on this.
          >
          > I personally would start by reviewing the artistic aims of the
          > creator. Are his or her aims interesting, or coherent? Are they flawed
          > in some way (e.g. setting a goal that cannot realistically be
          > achieved)? Does the conlang achieve its stated aims, and how does it
          > go about doing so? e.g. Kélen aims to be a convincingly alien
          > language, and does this by abolishing the large open class of verbs,
          > found in all human languages, with a small, closed class of
          > relationals. ...

          I'd say in *most* languages. The counterexample I will cite without
          knowing much about the language is Basque, which was claimed here to
          have only very few real verbs.

          Wrt. the original question: I do essentially the same: look for what
          the author's goals were and see how nicely they were achieved.
          Additionally, I can't help but also apply my own ideals to a conlang I
          read about. I usually find languages nice that are original -- they
          don't need to be very alien. This is very hard to formalise, of
          course, and just pure personal taste.

          When designing a conlang myself, I start making a list of goals.
          Usually, those goals have developed over a certain time, then trigger
          the start of a new conlang, and only need to be formulated.
          Sometimes, however, this process leads to unsolvable problems because
          the goals are contradictory, and then it either becomes a funny
          conlang, or none at all. :-)

          **Henrik
        • Carsten Becker
          On Wed, 28 Sep 2005, 05:15 CEST, João Ricardo de Mendonça ... Clones of your L1 or any tolkienean language are usually disregarded. And there are some other
          Message 4 of 13 , Sep 28 10:56 AM
            On Wed, 28 Sep 2005, 05:15 CEST, João Ricardo de Mendonça
            wrote:

            > How do you evaluate a conlang? How do you define a "good"
            > or a "bad"
            > conlang? I understand this is a personal criteria, so I'm
            > not looking
            > for a definitive answer. I just wanted to know other
            > people's opinions
            > on this.

            Clones of your L1 or any tolkienean language are usually
            disregarded. And there are some other typical newbie errors
            to avoid, like assigning unsensibly complex meanings like "a
            gummy bear that accidentally fell under the sofa pillow and
            was found by the dog of good friends of our family ten years
            later when it was hard like stone" to a word like "at".
            That's simply hilarious. I think there's once been a thread
            where we discussed about the features of an "übernoob"
            parody jokelang. Just keep your stuff consistent and try to
            avoid to re-invent the wheel called "your native language".
            FWIW, I must admit my first conlang was heavily based on my
            L1 German, too, and wasn't consistent at all as for its
            sound for example. Oh, and it also helps to first collect
            general ideas about the sound of your language and some
            interesting grammatical featuers (e.g. I wanted Ayeri sound
            at least a bit like Austronesian langs like Indonesian
            originally, though it didn't completely work out as I
            wanted). Actually, conlanging is much like model building.
            First, you have the idea. Then you see how you could realize
            things. Then you go to a hardware shop or something and buy
            the materials you need. Then you sit down and build the
            model, and if you don't like something, you need to refine
            it.

            Carsten

            --
            "Miranayam cepauarà naranoaris."
            (Calvin nay Hobbes)

            Current projects:
            www.beckerscarsten.de/?conlang=ayeri
            www.beckerscarsten.de/?conlang=tarsyanian
          • Carsten Becker
            Oh, and this might also interest y all (found today on the ZBB): http://www.angelfire.com/psy/platinumkow/descrimorpho.pdf -- it s the TOC of Describing
            Message 5 of 13 , Sep 28 11:07 AM
              Oh, and this might also interest y'all (found today on the ZBB):
              http://www.angelfire.com/psy/platinumkow/descrimorpho.pdf -- it's the TOC of
              Describing Morphosyntax inclusively the questions at the end of each chapter
              which point out things to consider in order to write a quite complete grammar.

              c.
            • Jörg Rhiemeier
              Hallo! ... This is indeed an interesting question that has been discussed here several times. In my opinion, there is no absolute criterion what makes a good
              Message 6 of 13 , Sep 28 12:43 PM
                Hallo!

                João Ricardo de Mendonça wrote:

                > How do you evaluate a conlang? How do you define a "good" or a "bad"
                > conlang? I understand this is a personal criteria, so I'm not looking
                > for a definitive answer. I just wanted to know other people's opinions
                > on this.

                This is indeed an interesting question that has been discussed here
                several times.

                In my opinion, there is no absolute criterion what makes a "good"
                conlang. Any conlang ought to be gauged against its design goals.

                An international auxiliary language, for example, is to be evaluated
                by such criteria as international neutrality, easy learnability and
                full expressive power (a jargon where one can merely express such
                simple concepts as "How much is the fish?" is not enough - an IAL
                ought to be capable of handling scientific and legal matters as well).

                With artlangs, things are different, and it again depends on the
                "school" of artlang the creator adheres to. The greatest "school"
                is probably the "naturalist" school, who seek to create languages
                that resemble natural languages. Such a naturalist conlang ought
                to mimic the complexity found in natural languages, and should
                by the same token not be a relexification of any other language.
                If someone says, "This is the language of a tiny minority in the
                Austrian Alps that survived the spread of Indo-European in Europe",
                then the language should be a plausible pre-Indo-European language
                of the Austrian Alps. It should not have features that give away
                its artificiality (such as a self-segregating morphology); it also
                should not show features known to be restricted to an area which
                it doesn't belong to (e.g., clicks). Or to take another example:
                a Romance conlang ought to be derived from Latin by a series of
                *plausible* sound changes - not some bizarre and arbitrary
                permutations of phonemes such as /n/ > /b/, but sound changes
                observed in natlangs, such as /t/ > /d/ in intervocalic position.

                But not all artlangs are naturalist artlangs. There are those
                among us who wish to explore the vast possibilities of languages
                that lie beyond the limits of what is found in human languages,
                either under the rationale of designing languages of alien
                sapients, or as a piece of linguistic abstract art. These people
                set themselves a particular design goal, e.g., "a language that
                doesn't involve any lip movement", or "a language that uses a
                stack-based syntax and has nouns as the only open lexical class",
                and proceed making a language that meets this goal. You may like
                the resulting language or not, it is a perfectly legitimate kind
                of conlanging, and the resulting languages are often highly original.

                And after all, beauty lies in the eye of the beholder.

                Greetings,

                Jörg.
              • Remi Villatel
                ... I don t think there s any objective way to evaluate a conlang, that would be like trying to use a ruler to measure poetry. So, it s only my very subjective
                Message 7 of 13 , Sep 28 4:15 PM
                  João Ricardo de Mendonça wrote:

                  > How do you evaluate a conlang? How do you define a "good" or a "bad"
                  > conlang? I understand this is a personal criteria, so I'm not looking
                  > for a definitive answer. I just wanted to know other people's opinions
                  > on this.

                  I don't think there's any objective way to evaluate a conlang, that
                  would be like trying to use a ruler to measure poetry. So, it's only my
                  very subjective opinion.

                  First, to be "good", a conlang must be original. I find re-working
                  somebody else's conlang very insulting for the original author.

                  Second, I look at the sounds. I always give bad grades to conlang which
                  mimic english soundings and orthography. There are so many of them that
                  it's getting boring. Apart from that, I don't care much about the
                  phonology. There's always something you would change. I however always
                  give a good grade for the unexpected, like one or two letters with
                  unexpected sounds assigned to them. If the orthography is too complex, I
                  totally disregard it to go the next step.

                  Third, the grammar. I dislike Euro-clones, natlang clones and elfic
                  clones. I prefer more exotic features. I give my better grades to
                  conlangs totally devoided of one common class of words --any kind-- or,
                  at least, which have strange ways around one class. Of course, some
                  coherency makes a "good" conlang so I prefer regular conlangs i.e.
                  without too many grammatical exceptions.

                  Fourth, the grammar again. I qualify as "good" a conlang that is hard to
                  translate without loosing half of its subtleties. That's what happens
                  with natlangs. A translator doesn't translate, he/she re-writes with
                  words from L2 what he/she reads in L1... if you see what I mean.

                  Well, I think that your question will teach you more about the
                  psychology of the conlangers of the list than about what is a "good" or
                  a "bad" conlang. ;-)

                  Hope it helps but don't think so,

                  --
                  ==================
                  Remi Villatel
                  maxilys_@_tele2.fr
                  ==================
                • João Ricardo de Mendonça
                  ... What makes you think this was not my intention? :-) João Ricardo de Mendonça
                  Message 8 of 13 , Sep 28 4:23 PM
                    On 9/28/05, Remi Villatel <maxilys@...> wrote:
                    >
                    > Well, I think that your question will teach you more about the
                    > psychology of the conlangers of the list than about what is a "good" or
                    > a "bad" conlang. ;-)
                    >

                    What makes you think this was not my intention? :-)

                    João Ricardo de Mendonça
                  • Isaac Penzev
                    ... of ... chapter ... grammar. Archi-useful thingie! -- Yitzik, from behind the Long Buck Curtain
                    Message 9 of 13 , Sep 29 2:30 AM
                      Ktebe Carsten Becker:

                      > Oh, and this might also interest y'all (found today on the ZBB):
                      > http://www.angelfire.com/psy/platinumkow/descrimorpho.pdf -- it's the TOC
                      of
                      > Describing Morphosyntax inclusively the questions at the end of each
                      chapter
                      > which point out things to consider in order to write a quite complete
                      grammar.

                      Archi-useful thingie!

                      -- Yitzik, from behind the Long Buck Curtain
                    • caeruleancentaur
                      ... Wow! What a great tool for analyzing one s conlang!! Now, when I have about a month free I ll start working with it, although I still don t understand
                      Message 10 of 13 , Sep 29 4:29 AM
                        --- In conlang@yahoogroups.com, Carsten Becker <naranoieati@B...>
                        wrote:
                        >Oh, and this might also interest y'all (found today on the ZBB):
                        >http://www.angelfire.com/psy/platinumkow/descrimorpho.pdf -- it's the
                        >TOC of Describing Morphosyntax inclusively the questions at the end
                        >of each chapter which point out things to consider in order to write
                        >a quite complete grammar.

                        Wow! What a great tool for analyzing one's conlang!! Now, when I
                        have about a month free I'll start working with it, although I still
                        don't understand all the terminology. Schönen Dank, Carsten!

                        Charlie
                        http://wiki.frath.net//user:caeruleancentaur
                      • Mia Soderquist
                        ... A good conlang makes its creator happy. A bad one makes its creator vaguely depressed. (Bad conlang! Bad! No biscuit! ) I m not willing to judge other
                        Message 11 of 13 , Sep 29 8:24 AM
                          On 9/27/05, João Ricardo de Mendonça <somnicorvus@...> wrote:
                          How do you evaluate a conlang? How do you define a "good" or a "bad"
                          conlang? I understand this is a personal criteria, so I'm not looking
                          for a definitive answer. I just wanted to know other people's opinions
                          on this.

                          A good conlang makes its creator happy. A bad one makes its creator vaguely depressed. (Bad conlang! Bad! No biscuit! )

                          I'm not willing to judge other people's work as "good" or "bad", but more in terms of whether or not it achieves the explicit goals of the creator and/or happens to have bits and pieces that appeal to me personally. There are some conlangs I like better than others, but I can't think of any that I especially *dislike*.

                          I can be pretty hard on my own work, on the other hand. I love ea-luna, in many ways, but I see so many shortcomings that it keeps spawning whole new projects every time I go back to it. I love the sound of Muhilamanyani, my very first conlang, but it had some really ridiculous points too. Heck, I woke up today and completely revised the existing grammar of my current project, which I am creating myself and for the amusement of friends on another, non-linguistic list. What do I judge my own language on? It's a subjective sense of how well I can express my thoughts without stumbling on bad grammar and vocabulary choices. If I were to start evaluating other people's languages, it might be in those terms as well... except, of course, for those languages designed for alien minds.

                          M.S. Soderquist
                          Alas! Sigless!
                        • 轡虫 (kutsuwamushi)
                          ... The first thing that I look for is originality. I don t want to spend much time reading about a conlang that s basically a natural langauge with a few
                          Message 12 of 13 , Sep 29 9:31 AM
                            2005/9/27, João Ricardo de Mendonça <somnicorvus@...>:
                            > How do you evaluate a conlang? How do you define a "good" or a "bad"
                            > conlang? I understand this is a personal criteria, so I'm not looking
                            > for a definitive answer. I just wanted to know other people's opinions
                            > on this.
                            >
                            > João Ricardo de Mendonça

                            The first thing that I look for is originality. I don't want to spend
                            much time reading about a conlang that's basically a natural langauge
                            with a few tweaks, or is too similar to other conlangs. And more
                            subjectively, I want "interesting" features - features that make me
                            think outside of the Indo-European/Japanese/Chinese box.

                            I also like completeness. Really good conlangs could be used to
                            translate an article or story without coming across too many gaps in
                            the grammar. (Naming languages and basic sketches of conlangs-to-be
                            aren't /bad/, if that's what the creator wanted. There's nothing wrong
                            with them. They're just not as interesting /as conlangs/, in my
                            opinion.)

                            Beyond that, it depends on the author's goals. A naturalistic conlang
                            should be plausible, a loglang should be logical, and so on.

                            --
                            Kate
                          • João Ricardo de Mendonça
                            Thanks to everybody who replied. As I hinted on my response to Remy Villatel, I wanted to know how you people relate to this hobby, how you perceive it.
                            Message 13 of 13 , Sep 30 6:46 PM
                              Thanks to everybody who replied. As I hinted on my response to Remy
                              Villatel, I wanted to know how you people relate to this hobby, how
                              you perceive it. Perhaps I should have written "How to APPRECIATE a
                              conlang" instead of "evaluate".

                              As for the newbie mistakes, I think I won't fall for the "cloning your
                              mother-tongue" one, but I'm prone to go for the "add to it every
                              little detail from every language you know" kind of thing. So much
                              stuff, so little time... :-)
                            Your message has been successfully submitted and would be delivered to recipients shortly.