Here’s an interesting article in the New Yorker about conlangs and specifically about Ithkuil, which, according to its creator, John Quijada, is “an idealized language whose aim is the highest possible degree of logic, efficiency, detail, and accuracy in cognitive expression via spoken human language, while minimizing the ambiguity, vagueness, illogic, redundancy, polysemy (multiple meanings) and overall arbitrariness that is seemingly ubiquitous in natural human language.”
For me ambiguity, vagueness, illogic, redundancy, polysemy and overall arbitrariness are some of the things that make languages so interesting, and I suspect that in some ways languages work better because of them. This certainly seems to be true of redundancy, which can help get the message across in less than ideal conditions, i.e. noisy environments, etc, and without ambiguity and polysemy puns and similar word play would not be possible, and poetry would difficult.
I have considered adding details about the Ithkuil script to Omniglot, but decided not to when I saw its complexity.
I’ve studied Lojban for a while. Some of its grammar constructs are precisely defined, while others are intentionally vague. I’ve found that sometimes expressing a statement in a precise way is way too lengthy, while a vague formulation of the same statement is concise enough and sufficiently clear.
On the other hand, some features that make languages interesting also make them inefficient.
I once realized (not necessarily correctly) that when a language is designed (like Lojban or Ithkuil) it is done so by language geeks, resulting in an efficient language lacking emotion and all the things you need to make a language beautiful. When a languauge would be designed by people loving language, I believe it would become very inefficient. So my conclusion was that letting a language evolve would make a nice balance between efficiency and beauty. Although the Ithkuil script looks anything but efficient to me…
I agree with the above. I’ve checked the writing system page and it was way too lengthy. Might be useful for encoding though!
I think the concept of a “logical” language is vague by its own. Take for example the words “nap” and “sleep” in English. I don’t recall there are other words to describe the status of sleeping except of these two maybe. In classical Arabic, it is said there are 7 words to describe 7 different levels of sleeping. In modern dialects there might be at least 2-4 words for sleeping as well.
What sounds “logical” in one language or culture doesn’t necessarily mean it is logical in another. Hence, we should ask, on what bases or logic was Ithkuil built? In any possible way, the logical base (at most) would be that of the creator and his or her own environmental influence – and that might not sound logical in other cultures for other individuals.
Just another example of how logic can be different, Take the word “carbon” which translates in German as “Kohle” or “Kohlenstuff”. If you happen to have some snacks or chocolate just read the ingredients in German, and where it says “Carbohydrates” you will find “Kohlenhydrat”. The thing is, “Kohle” also means “coal”. Here, we see that the English language differentiates between the two statuses of Carbon (Arabic does as well) – but in German, the two substances are conceived to be the same aspect. Which logical rendering is right here? It is both, as I believe.
I definitely agree that ambiguity is far from a negative aspect of language. As a musicologist I would say that music’s ambiguity allows it to communicate even more, but that’s a big, sticky process. I would liken a “completely logical” conlang like Ithkuil to the most abstruse 12-tone composition—interesting, but completely “heady” and a little off-putting in its exclusivity. You have to have that element of messiness or humanness to even invite interpersonal communication. Anyway, a very interesting thought. It led me to wax eloquently for a while in this post:
http://mjroy.blogspot.com/2012/12/ithkuil-and-ambiguity.html
One thing that bothers me is that structural and semantic ambiguity is rarely distinguished in this type of discussions. By structural ambiguity I mean such cases where the structure of the sentence isn’t uniquely related to the sound, as in English “the little sister’s house”, where it is not clear whether it’s a house belonging to the little sister, or the little house belonging to a sister. It isn’t that difficult to remove this sort of ambiguity from a conlang with little costs. (And no, Jerry, it wouldn’t deprive the language of its beauty. Neither is beauty connected anyhow to efficiency, efficiency doesn’t mean omitting all emotional expression as unnecessary, or at least it shouldn’t mean that. An efficient language shoul allow one to express one’s emotions, well, efficiently.)
By semantic ambiguity, on the other hand, I mean all things somehow related to meaning, ranging from the question whether the boundary between “green” and “blue” light is at wavelength 500 nm or 515 nm, to the issue whether “the sky is blue” refers to all instances of sky everywhere at all times, or to the sky visible at one moment from one place, or… This sort of ambiguity is inversely related to precision, lack of ambiguity is then equal to infinite precision. Since there is no way how to achieve infinite precision with finite number of words, people often put an artificial boundary on “ambiguity” and declare that everything on one side of the boundary is ambiguous. Where the boundary lies depends one one’s intuitions and is, of course, rather arbitrary.
For example, sentiments I have encoountered several times went along the lines “the language shouldn’t allow passives or other constructions lacking the subject” (because they are ambiguous and therefore bad) or “how great are the languages whose grammar forces the speaker to inform about the source of information they provide” (because that removes a lot of “ambiguity”). Some people seem to think that grammar should make inclusion of some information mandatory, even if it is irrelevant to the communicated statement. (Incidentally, sometimes the same people try to invent ways to overcome such grammar rules which already exist in their language, as when they use Spivak pronouns to get rid of the clumsy phrase “he or she”, when grammar just doesn’t want us to be ambiguous about gender.
And, by the way, Ithkuil looks ugly, its phonology, morphology and the script all the same. And the grammar seems quite arbitrary, as grammars normally are, but with additional complications; I am not sure why people tend to ascribe the attribute “perfectly logical” to such things, but it certainly gives logic undeservedly bad reputation.