Posted by & filed under Podcast.

This week John Ericson joins us to talk about his wacky and wonderful FairyLang.

Top of Show Greeting: Russian (translation by Boris Listunov)

Featured Conlang: FairyLang



Is it possible (or does there exist) languages in which the lexicon can be automatically generated (or at least guided) by a heuristic employing a pre-existing ontological framework?

I am new to this hobby (it was your podcast that really got me interested), and am trying to design a language where the semantics and lexicon are somewhat coupled. That a person might automatically generate a word from the very meaning he wishes to convey.

For instance, I have been toying around with a language whose lexicology is based upon an infix system combined with single prefix and suffix slots. Each word can possesses two to four consonants separated by vowels. An optional vowel can come before the first consonant and also trailing the final consonant. Each vowel slot represents some feature in the language. The pre and postfixes add to it more subtle meanings.

I have worked a simple naming language like this where the consonants are chosen arbitrarily. I like where I was going with it. But it occurred to me that, if I could assign to each consonant position a meaning in a similar fashion to the vowels, then I could build a framework where at least the meaning of a word is somewhat self-evident.

I am unsure if you could reasonably do this where you necessarily get a one-to-one correspondence between specific meanings and a single word. In any case, I am curious if any conlang has attempted to build a lexicology that fuses with semantics in this way.  If so, it would greatly help me figure out the most complete and least ambiguous ontological for such an endeavor.



(some resources we found for this question listed below:)

3 Responses to “Conlangery #77: FairyLang”

  1. Rhamos Vhailejh

    If you guys ever decide to change the format of the website… please don’t let Mike have any say in the changes, whatsoever. xD

  2. Qwynegold

    I know the OP probably doesn’t care anymore or won’t see this, but anyway… I think it’s impossible to construct an oligosynthetic language where the meaning of a given word is always completely predictable, or where a given meaning should only have one possible word for it. Because you will run into meanings where you could choose between several different ways to construct the word for it, and it will be subjective which method you choose. I have an oligosynthetic conlang myself, btw:


Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.