Hacker News Reader

Comment by strogonoff | Hacker News Reader

Invented languages are overwhelmingly boring in their likeness to English, Spanish and other Western languages.

What if we tried to create, say, a language with a logographic written system that is pure WYSIWYM (as opposed to “what you see is how you pronounce”) and synthetic to boot?

Make it use vocal cords differently.

Instead of borrowing around, use a random seed in generating a minimum set of unique basic “native” words according to language rules and build on top of that (borrowing for meanings outside of that set).

This could be so much more fun!

View on HN

> Invented languages are overwhelmingly boring in their likeness to English, Spanish and other Western languages.

Toki Pona is not like English, Spanish, or other Western languages.

It has no singular/plural distinction. It has no past/present/future tense. Its pronouns have no gender. All of its phonemes are present in almost all languages (this is on purpose). The way it forms questions is not like Enlgish (I don't know of any language that it's similar to). Its word order is subject-verb-object, like most languages. [EDIT: not most, only 42%]

The only thing its taken from English, as far as I've seen, is a bunch of vocabulary. Though honestly its sounds are so limited that sometimes you can't recognize which English word a Toki Pona word came from.

> What if we tried to create, say, a language with a logographic written system that is pure WYSIWYM (as opposed to “what you see is how you pronounce”) and synthetic to boot?

I'm not sure exactly what you mean, but I'll just leave this link here...

https://omniglot.com/conscripts/conlangs.htm

> Instead of borrowing around, use a random seed in generating a minimum set of unique basic “native” words according to language rules and build on top of that

Lojban does this.

> Its word order is subject-verb-object, like most languages.

Most languages use subject-object-verb.

https://en.wikipedia.org/wiki/Subject–verb–object

> I'm not sure exactly what you mean

One way of classifying languages is as “analytic” (where unit’s role in a sentence, its relationship to other units, is denoted by auxiliary units/particles or unit order) or “synthetic” (where unit inflects, its form changes to reflect its role).

Most languages are somewhere in between, but languages with logographic written systems tend to be more on analytic side.

What could a pure attempt at a synthetic grammar that also has semantics-focused writing system (one that tries to maximize difference in visual representation appropriately to distance in meaning) look like?

> The way it forms questions is not like Enlgish (I don't know of any language that it's similar to).

The "x ala x" pattern is directly modeled on the Chinese "x不x" (and "有没有") pattern, including the answer ("x" / "x ala" in toki pona, "x" / "不x" in Chinese). I think Sonja has mentioned this explicitly somewhere.

For example, in Chinese I think you can ask "你可不可" 'you can not can?' with the possible answers "可" 'can' and "不可" 'cannot'. This corresponds directly to toki pona's "sina ken ala ken?" 'you can not can?' with the answers "ken" 'can' and "ken ala" 'cannot'.

There's also the "anu seme?" pattern which is similar to the https://en.wikipedia.org/wiki/Tag_question phenomenon in a number of languages; the one that I find it most similar to is German, with the "oder?" tags.

https://en.wiktionary.org/wiki/oder#Particle

I understand the "oder?" to have a connotation of 'or what?' (like "are you coming or what?"), in which case "kommst du, oder?" should correspond literally to toki pona's "sina kama anu seme?" 'you come or what?'.

I agree with you that toki pona doesn't feel very similar to English overall.

Other similarities to English: use of prepositions, something like an accusative case (though marked differently from English). Frequently being willing to use the same word as both a verb and a noun (without any morphological changes).

Other differences: adjectives follow the nouns they modify (like in Romance): jan lili, soweli suli, tenpo seli, tenpo pimeja, telo kasi. English would say small person, large animal, warm time, dark time, herbal beverage. No participles, no relative clauses, no general word for "and", no comparatives. Particles are super-weird from an English point of view, especially li and la.

I get the impression that you are not very deeply connected with the general conlanging community. Eurolangs are only a very small part of the vast diversity!

> What if we tried to create, say, a language with a logographic written system that is pure WYSIWYM (as opposed to “what you see is how you pronounce”) and synthetic to boot?

UNLWS.

> Make it use vocal cords differently.

Dritok. Or, if you just mean "differently from Europe", any conlang which makes use of creaky voice.

> Instead of borrowing around, use a random seed in generating a minimum set of unique basic “native” words according to language rules and build on top of that (borrowing for meanings outside of that set).

So freaking common I can't even begin to list a representative set. The whole category of oligosynthetic languages does this (although the initial generation may not be completely random). There are multiple software packages for assisting with the generation of this kind of language.

You’re right, I had no idea about this diversity. UNLWS looks really cool!

I should have said “popular invented languages”, those that tend to regularly pop up on HN. Which, admittedly, says more about the audience using invented languages than people who do the invention.

There is a written-only language that tries to use only meanings and no sounds:

https://en.wikipedia.org/wiki/Blissymbols

Its history is interesting—its creator grew up in an English-speaking family initially living in China, and he learned to read Chinese but not speak it, in distinction to a huge number of people who learn to speak Chinese but not read it. It seems that that experience of thinking of the Chinese hanzi characters as meanings (initially in his case English meanings) rather than sounds eventually led him to try to make a written conlang with a similar characteristic.

Edit: this is also interesting to me because I myself know a few Chinese characters whose pronunciation I have no idea of (the ones that came to mind for me right away were 一，二，三，上，下，女, 日, 本, 天, and 月, although I'm sure there are several others). Maybe I have a little tiny mini-Charles Bliss experience when I see those characters, like "oh yeah, I know what that means, no idea how you would say it though!".

Edit: according to Wikipedia Bliss was actually a German-speaking refugee in China, so I misremembered the circumstances of his exposure to Chinese!

Chinese in practice is a rebus for a language that was spoken 3,000 years ago. There are a few hundred basic characters and most other ones have a meaning half and a phonetic half shoves together. E.g. 論 lun (discourse) = 言 speech plus sounds like 侖 lun.

> What if we tried to create, say, a language with a logographic written system that is pure WYSIWYM (as opposed to “what you see is how you pronounce”) and synthetic to boot?

So Hangul?

Hangul characters carry no meaning, it is in fact purely pronunciation information.

(Also, Korean isn’t a synthetic language, I should’ve been clearer that synthesis I have in mind is regarding language grammar not [just] the writing system.)

Sorry about that. Just noticed that I misread your comment. It took me a while to realize that you wrote WYSIWYM to mean "what you see is what you mean".

I don't get the appeal of logographic scripts though. I personally think an alphabet with strict pronunciation rules is the way to go. It's easy to learn and more predictable. And psychologically speaking it doesn't even matter all that much whether meaning is contained in words or characters. We read scripts by meaning anyways. In reading English for example, we read words, not characters.

"It deosn't mttaer in waht oredr the ltteers in a wrod are, the olny iprmoetnt tihng is taht the frist and lsat ltteer be at the rghit pclae. The rset can be a toatl mses and you can sitll raed it wouthit porbelm. Tihs is bcuseae the huamn mnid deos not raed ervey lteter by istlef, but the wrod as a wlohe."

Most of those words are short though, which means the displacement of a letter from its expected location is not very large. And if you combined typical spelling errors with the scrambled order, it might be too much to parse. I think it's going to far to suggest letter order has no useful information.

I just copied and pasted an example, I didn't write it myself. It's true that order of letters do matter more as words get longer. The point is that we don't really read letters. We read words.

So I would even argue that written English is basically like Chinese since in both cases you have to recognize the word/character and memorize its pronunciation and meaning. In both scripts you get hints of pronunciation, but in both cases you kind of have to do a lot of memorizing. In the case of English it's because it has no consistent rules or predictability. But then the words aren't really logographic, so I guess we're stuck with the worst of both worlds. No meaning, and very unreliable pronunciation hints.

The example I usually come up with is like “bad” and “bed” in English. Two basic units with entirely orthogonal meanings, but difference in written representation is minimal and “a” vs. “e” do not have a semantic component (e.g., “a” doesn’t mean “bad”, and “e” doesn’t mean “rest”).

To be clear, it does happen in, say, Chinese too, especially with complex characters that have a phonetic component. That’s why it could be an interesting experiment to design a written system that is “pure” in such a way that increasing distance in meaning always results in increasing difference in written representation.

(And considering that most languages with logographic writing tend to be analytic, what if this language was actually very synthetic?)

> I don't get the appeal of logographic scripts though. I personally think an alphabet with strict pronunciation rules is the way to go. It's easy to learn and more predictable.

I think it’s subjective. With Hangul as an example, to me it was indeed easy to “learn” the writing, but is much harder to actually comprehend and generally use Korean language.

> I think it’s subjective. With Hangul as an example, to me it was indeed easy to “learn” the writing, but is much harder to actually comprehend and generally use Korean language.

You're right, but that doesn't really explain why you want logographic writing. The script can be independent of the spoken language. Latin is used in various languages, and even Hangul isn't exclusively used by Korean. I think the point you made is not really relevant.

> You're right, but that doesn't really explain why you want logographic writing.

I thought radically novel approaches to language were underrepresented in invented languages, and gave an example of something I [subjectively] consider to be an obviously interesting area for exploration.

(1) I think a language in which written forms of units differ appropriately to the distance in their meaning could be pretty cool, and (2) a fully phonetic system would thus not be unfeasible, as it would require pronunciation to always differ proportionately to semantics, which—since we are physically limited in the variety of sounds we can produce—could mean very long written forms.

That leaves logography of some kind. (I added “synthetic” as an additional challenge—logographic systems seem to be used by mostly analytic languages, so what if the opposite is attempted?)

As it turned out thanks to elsewhere in the thread, there do in fact exist constructed languages that address some of those points, although they are not very popular (and don’t tend to come up on HN) compared to Esperanto or Toki Pona.

Reply on news.ycombinator.com