Esperanto is not my favourite artificial language (2002) | Hacker News Reader

Esperanto is not my favourite artificial language (2002)

miresperanto.com

80 pointsadamrezich97 comments

If you want to pick up a new language purely for fun, I wholeheartedly recommend toki pona[1]. Its small vocabulary (<200 words) and simple grammar force you to break down complex thoughts into simple ideas. This feels limiting at first, but because there isn't a prescribed way to express certain ideas, it can provide an interesting look at how you see the world compared to others.

As a quick example, there is no word for "child" in toki pona, many speakers say "jan lili" (small person), while others choose to say "jan sin" (new person).

It's shockingly easy to learn and one can become pretty proficient in very little time. There's a great community to learn from on Discord[2], along with plenty of great online resources like /dev/urandom's toki pona page[3].

[1]: https://tokipona.org [2]: https://discord.gg/nbgXCC6 [3]: https://devurandom.xyz/tokipona/

I don't like toki pona. if you're going to have 200 words, why make them long or random? hell, you could have a consonant and a vowel or 2 and you'd have have all of your words and now they could easily be composable together. votgil[1] is basically this but with a single vowel, I haven't dived too deep but it gets close to what it should be. There's also the fact that if you're going to go the toki pona way, you need to go even deeper into the fundamentals of life in order to achieve the right composability. (Think math but for objects). Blyss symbolics also seems to be a better option. Nevertheless it was fun to learn toki pona. and I probably would not have heard of interlingua,blyss, votgil, lojban without it.

[1] http://www.ostracodfiles.com/votgil/main.html

A language should be complicated but only complicate enough to convey the concept in mind.

A 200 word limit will render a language too shallow to even discuss the concept of shallow.

Assertion: no proof. [soz, lol etc]

Yes, I guess (never heard about toki pona before) you would have to rethink how to communicate. Maybe shallow could be presented as "lukin ala" (nothing to examine), "taso sele" (only a surface) or "kote muli" (hearing something dead)

I am more surprised that you would spend one of your 137 words on "sex, to have sexual relations" as almost every language has countless ways of referring to sex without using the word itself

Interesting, I thought of something more of like incompleteness, something like sona wile pini.

It also went ahead and put non binary... which is like, right there "non" and "bi" and "nary" (which could be sex).

> It also went ahead and put non binary... which is like, right there

I think the general idea of adding a word for non-binary was that there were already words for man and woman, but if you didn't identify with either of those, you'd have to come up with your own phrase as opposed to other people who got their own dedicated word. A surprisingly high number of non-binary people speak toki pona (or at least that's my impression from the Discord server), so I think it made sense to add.

I don't think you need man/woman in a language like toki pona either, it's supposed to be this deep philosophical language. If you want to talk philosophy, reproduction is the least of your problems. There's plenty of ways to indicate the sex as well. so long as you have the word for animal/organism, you can say animal + hole/stick, who nurses for the young, hunter, whatever. Its even better because its not exact. the more exact you get, the more words you'll need.

Shallow indicates incompleteness, it lacks a whole, it does not achieve fullness. "wile pini sona" maybe? my toki pona is beyond rusty, I've forgotten all of it.

I think it should be possible but few have attempted to do so. even in programming languages, the goal is to overflow not underflow. Forth-ian languages seem to be the only ones that try to build down not up.

And doing so is difficult because it would require simplification, and simplification is difficult when you lack completeness. English itself is infinite and symbolic. A new concept was identified? lets just make up a word/symbol for it. It does not matter if it breaks structure/grammar.

> why make them long or random?

To decrease their similarity, make it more difficult to confuse/conflate one word with another.

But then use the theory of Shannon and Hamming distance to make the words as different as they can be

Word cadence is also important. We tend to add length to short words and shorten long words naturally until we reach a certain cadence. Longer words can sometimes be easier to say than short ones. If you look at tongue twisters they tend to use short, similar words. By making every word short, you could make an entire language of tongue twisters!

Edit: Also I think Toki Pona uses a reduced character set

https://jansa-tp.github.io/tpcheatsheet/Toki%20Pona%20Cheat%...

Toki Pona definitely has fewer phonemes than English (which IIRC has an above-average number of phonemes compared to other natural languages). Having fewer phonemes is important for a language like Toki Pona because it makes pronunciation much more forgiving.

> Vötgil may also be written using the non-roman alphabet. Every character is a column of 3 cells. Each cell may be white, light gray, dark gray, or black.

Weird and delightful.

Except when drawn by hand with a pencil or pen ... seems like it would inspire a squinty guess-a-rune swearfest.

Is are these triple dots of grayscale unique to Vötgil? Inspired by another alphabet?

I like the idea of a limited vocabulary and word building from there, but I think TP is way too extreme. Randall Munroe wrote a book called "Thing Explainer: Complicated Stuff in Simple Words". In the back of the book, he lists the 1000 words he used to write it. Something like that would be suitable IMHO.

But a child is neither a small person nor a new person...

Child is used in many contexts, and not all of them are covered by "small person" or "new person", true. But I think the point is that you can come up with a precise alternative term for"child" in most contexts.

For those interested in languages... After studying Chinese (Mandarin), Esperanto has always struck me as being not all that great for a constructed language. Chinese is somewhat remarkable in that if you ignore the logographic writing system and tone-based pronunciation, the grammar is remarkably simple and clear. No grammatical gender, no conjugation, particles instead of prefixes/suffixes for basically everything important - you can pick up the basic rules in a couple days. Interlingua is (by design) much closer to this simplicity in a Western-based language.

Some Chinese examples for fun (hoping HN supports Unicode)...

To make something plural, you just add 们. I/me is 我, so we/us is 我们. He/him is 他, so they/them is 他们. No need for separate base words - learn the normal nouns/pronouns and one particle and you're done.

Same concept for possessive, you just add 的. Using the above, my is 我的. His is 他的. Their is 他们的. Easy.

Want past tense? Don't mess with the base words; again just add a particle, in this case 了. Go is 去 so went is 去了. Eat is 吃 so ate is 吃了. Easy. Learn the verb once and it never changes.

Forms of "to be" are similarly simple - the base words again don't change, you just add the "to be" word you want and get on with your life. Want to say "I am American"? America = 美国 and person is 人, so American is literally "America person" 美国人. Am for non-adjectives is 是. "I am American" = 我是美国人. "We are American" = 我们是美国人. Easy. Compare this to something like Italian, where "I am American" = Sono americano, but "We are American" = Noi siamo americani. Absurd in comparison.

Similar to Thai, a lot of particles.

Composition over inheritance, not only useful in CS but also in linguistics. It makes sense not because both use languages but because both manipulate non-physical abstraction and thus come up with similar paradigms.

Composition is akin to agglutinized languages where there is a base and affixes are added to convey concepts like subject, object, plurality etc, such as Chinese or Thai.

Inheritance is akin to fusional languages, like Latin, where each word is a root and changes (or declines) based on the aforementioned concepts by for example changing the word itself.

I've always thought that English is like the PHP* of natural languages: ubiquitous, borrows from every other language, and wildly inconsistent.

*I cut my teeth on PHP3 and haven't used it in over a decade, and I hear it's gotten a lot better.

Vietnamese is similar. Aside from the pronunciation it’s a fairly easy language to learn.

My Achilles heel for learning Chinese is typing and reading the characters, and I suspect they are two sides of the same coin - if I had to make the characters I'd be much better at memorizing them. All the learn-Chinese apps appear to gloss over this by letting you drag and drop the letters when writing.

How do you type Simplified Chinese on a phone? On a Linux/Mac desktop? I know it has something to do with an IME but this is one area of i18n that has never been explained to me.

There are a few input methods, but as a native English speaker and complete novice, I found that pinyin input was the easiest for me. Basically you type the pinyin and you get suggestions for characters. Super easy if you know the pronunciation.

Hmm. On one hand I like dodging having to learn to write it - I can memorize wǒ and ren. On the other hand I'm still stuck having to learn the characters, ha. What do native speakers prefer typing with?

I'm never going to be a Chinese author, I just need to navigate chip documentation and message boards.

pinyin because remembering the exact stroke is a pain, plus pinyin is meant for rapid conversation.

Imo Skritter is the best iOS app for this.

There's a lot of complexity to Mandarin grammar and syntax under the surface though.

1. A rather common grammatical gotcha in everyday Mandarin is what is the grammatical function of 把 and when do you use it? Native speakers are able to easily tell when it's correctly and incorrectly used, but for learners it's completely inscrutable: https://en.wikipedia.org/wiki/B%C7%8E_construction

2. Syntax is constrained by prosody, which is also extremely unintuitive to Mandarin learners. For instance, you can say 我在走路 ("I am walking [the road]") but 我在走 ("I am walking") sounds syntactically incomplete, even though it has all the semantics of 我在走路 and you have to insert the meaningless 路 for the prosody to sound correct (you would say "我在走路" even if you weren't walking on a road at all!).

3. 了 is not a past tense marker at all, you can also use it in certain non-past tense constructions, e.g. 他要吃了吗 for "Is he going to eat?" and you must omit it in certain past tense constructions, e.g. 我还没去过北京 for "I haven't been to Beijing yet."

4. 们 is not always a pluralizer: sometimes it is e.g. 同学 (classmate) -> 同学们 (classmates) is fine, but 牛 (cow) -> 牛们 (cows) sounds a little weird. 们 is much more similar in function to the Japanese 達 and means something more like "and company", e.g. you could say 李先生们 to mean "Mr. Li and company" and definitely not "Mr. Lis". In fact, most of the time, plurals in Mandarin are actually completely implicit and you have to figure it out from context!

My understanding of Mandarin is not strong, so these questions may sound terribly silly:

Is “我还没去过北京” considered past tense? It’s something that hasn’t been done, so I’m confused as to how that could be in the past.

For things like “我在走路”, is it not just needing a thing to receive the action, even if a filler? Or are there cases where you don’t need a filler?

Also, for “他要吃了吗”, is that not a statement from the past perspective of a (desired) completed action? Kind of like “我卖了这些车以后”?

I would characterize 我还没去过北京 as past perfect tense but in the negative: if "I have been to Beijing" is past perfect in the affirmative, then "I haven't been to Beijing" is past perfect in the negative. Maybe a more straightforward example I should've gone with is "我去过北京" for "I've been to Beijing".

Regarding 我在走路, yes, 路 acts as a filler for prosody purposes. If the prosody is fine, then you don't need it, e.g. 他走的很快 (swapping 我 for 他 here because it sounds kind of weird to say such a sentence about yourself) takes no noun just like the fragment 他在走, but mustn't use the filler noun 路 (他走路的很快 is incorrect).

"我卖了这些车以后" I think is more like "After I sold these cars", where "sold" is definitely past tense here. "他要吃了吗？" expresses "Is he going to eat?", where 吃 hasn't taken place in the past.

1. any example on how this is confusing? 把 is "to do to". like 你吃了饭 you ate rice vs 你把饭吃了you did eat the rice etc

2. Its grammatically correct to say "我在走" but it sounds better when its a "pair" so colloquially it uses a 词, which is a pair of characters which may just say the same thing. you could also say 我在走着 which adds no extra meaning but pads it to be a pair

3. 了 is meant for something in the past, past perfect tense is not the same as past tense, past perfect tense is about ambiguity not time.

4. 们 has the 人 radical, so its mean for humans. Otherwise you're to use counting particles instead, plus 很多 or 几 or some other particle to emphasise it.

1. The interpretation is straightforward but the construction is not: why is 他把饭吃了 grammatical but 他把我帮了 not, despite both being SOV constructions you can make SVO (他饭吃，他帮我）?

2. 走路 I agree colloquially you can say 我在走 even if it feels a bit fragmenty. I do think there’s a lot of difficulty again in the construction:

He’s walking: 他在走（路）

He walks quickly: 他走路很快/他走得很快 but not 他走很快

He walks to my home: 他走到我家 but not 他走路到我家

3. I guess this one depends on your view of past tense and if past perfect you consider a subset of past tense or separate: in any case 了 is not a past tense marker, since you can use it to talk about the future (e.g. 他要来了 for “he is about to arrive”)

4. I just wanted to point out that 们 is not a generic pluralizer and the generic pluralizer is usually implicit: besides, particles lie all the time in Chinese: half the words with the 纟 radical have nothing to do with silk!

Mandarin beginner here. On #2, my pet theory is that it's easier to disambiguate zǒulù than just zǒu in the spoken language. My teacher often says "word's character" (e.g. yī de yīshēng, 醫的醫生, probably bad example and/or just wrong, to distinguish from other characters that sound like 醫). I once calculated that there are only about 600 different sounds in Mandarin -- not many more than toki pona! -- so if you tend to pair sounds into concepts, that's a much richer pool to draw from.

> Syntax is constrained by prosody, which is also extremely unintuitive to Mandarin learners.

Definitely true, but I don't think your analysis holds. Prosody is an important part of determining when you need bisyllabic or monosyllabic words in Chinese which doesn't have an equivalent in English, but you don't fill it in with what are effectively nonsense words.

> you have to insert the meaningless 路 for the prosody to sound correct (you would say "我在走路" even if you weren't walking on a road at all!)

Not really. 路 still has some semantic content. E.g. if you're walking about indoors in your house it sounds weird to say "走路." I do agree that a bisyllabic word sounds better there. Other context and emphasis could also make 我在走 work (I am walking for example), but often times that ties back to prosody as you say.

> 们 is not always a pluralizer

It is generally a pluralizer with the constraint that it must be applied for people. That's why it sounds weird applied to cows. 李先生们 could actually mean "Mr. Lis" (i.e. a group of men all with that surname, or some slang usage of "Mr. Li" in the same way you might say "they're a bunch of Benedict Arnolds"). Indeed that was my first instinct when you wrote that phrase. I don't understand what you mean by "and company." That's not how I would understand 李先生们.

Also historically, 们 used to be an even more general people pluralizer that could be applied to demonstratives. E.g. in Song Dynasty Mandarin 那们 is perfectly acceptable and means "those people." It has become more restricted in modern usage.

> In fact, most of the time, plurals in Mandarin are actually completely implicit and you have to figure it out from context!

This is true.

> Not really. 路 still has some semantic content. E.g. if you're walking about indoors in your house it sounds weird to say "走路." I do agree that a bisyllabic word sounds better there. Other context and emphasis could also make 我在走 work (I am walking for example), but often times that ties back to prosody as you say.

That’s a good point, and I think if you wanted to talk about walking indoors the prosody requirements make it even more difficult (我在我家里走来走去？)

> I don't understand what you mean by "and company." That's not how I would understand 李先生们.

I’ve definitely seen writings like “毛泽东先生们” but perhaps the usage is a little more archaic than I thought! I do think you’re right in that the usage really is just a pluralizer for people rather than my wackier definition.

> 毛泽东先生们

This means something closer to "Mao Zedongists" rather than "Mao and company." For example "毛泽东本人的看法与其他毛泽东们略有不同" works, which doesn't quite work if 毛泽东们 is "Mao and company." It's still a pluralizer, it just so happens that proper names can be modifiers in Chinese.

> Interlingua is (by design) much closer to this simplicity in a Western-based language.

If you like Interlingua for its design, have you heard about the Interlingue? It's a lot like Interlingua, but will less exceptions. Interlingue is also called Occidental.

Nice to see a comment on Occidental already. That's the one I've used and supported for 10+ years now.

modern written chinese was designed and constructed as a lingua franca for hundreds of different cultures and languages within china, so the grammar is as minimal as it could be without being ambiguous. It is, in my opinion, the most ambitious conlang in history.

I believe esperanto to be a similar exercise but english is a far easier and more flexible language

> modern written chinese was designed and constructed as a lingua franca for hundreds of different cultures and languages within china, so the grammar is as minimal as it could be without being ambiguous.

No written Mandarin was not invented in the 20th century. There was some pronunciation and orthographic standardization, but written Mandarin has been the lingua franca of at least the literate class since at least the Song Dynasty, maybe even earlier, and written Mandarin itself extends back to the Tang Dynasty.

Here's for example here's the Southern Song philosopher Zhu Xi complaining about "kids these days" in written Mandarin (these are his original words, not translated into modern Mandarin), with modern Mandarin spelling marked in square brackets.

> 今学者不见有奋发底[的]意思，只是如此悠悠地过；今日见他是如此，明日见他亦是如此。《朱子语类·朱子十八》

And here's the Northern Song Cheng brothers complaining about Buddhism (again in the original)

> 佛学只是以生死恐动人，可怪二千年来无一人觉此，是被他恐动也[this 也 is purely for effect, same as it is occasionally done in modern Mandarin]！《二程遗书·卷一》

The article is well-written and makes some good arguments about things that could be better, but like all languages (including conlangs), they're never going to be perfect and they're always going to have biases and rough bits.

I started studying Esperanto 4 years ago and it has become a fluent second language for me at this point. I now almost have more friends who speak the language than don't. While the language could be improved (by fixing word order and dropping the accusative, among other things mentioned here), it's still clear to me that Esperanto is a very easy language to learn. My comparative experience with Spanish has been much harder going.

Thanks to Esperanto I've spent time living abroad with Esperanto-speaking friends from different countries, travelled to local and international hangouts, and met some of the most open-minded and lovely people.

Outside of the hobbyist and academic contexts, there are a surprising number of people who actively use Esperanto (as well as create things: books, music, poetry, apps) and a cool international community to explore.

Just wanted to add a positive take.

P.S. There are a few articles on my blog about my own personal experience with the language and some projects I've worked on, in case anyone's curious: https://martinrue.com/articles

Another reason no language is going to be perfect is that people will disagree on what constitutes perfection. For example, I have always found the flexible word order and the accusative to be a very beautiful trait of Esperanto, and would consider it a downgrade if they were removed from the language. :-)

Personally I like Interlingua which is basically a cleaned-up latin. If you've gotten a good dose of exposure to Romance languages you might be able to read it already:

https://en.wikipedia.org/wiki/Interlingua#Samples

I find Lingua Franca Nova very readable (I speak French decently as a second language), plus it has a very nice flag!

https://lfn.wikipedia.org/wiki/Lingua_franca_nova

Both Interlingua and Ido are improvements over Esperanto.

But as it is with any language, programming language, or ecosystem in general: There is strength in numbers. Esperanto is great because it's easy to learn (Europeans have an advantage, ok) and it's the most popular constructed language, meaning there are good resources and actual people (few, ok) to talk to.

what's the general feeling on Lojban these days? used to know a guy who was a unix command line wizard and lojban enthusiast

From what I can see the language is in constant development. there is a telegram group called ckule. but its a damn tough language to get through.

Lojban's selling point is that it's easy to computationally parse, but not necessarily that easy to learn to speak.

I'd want to try to learn it again eventually, but I failed miserably the first time.

As an academic curiosity I think it's doing great.

I kinda wonder if you can encode knowledge in Lojban and have a more efficient language model using it then only translate to English at the output (if needed). Is there any study on the informational density of Lojban/loglan?

It's conceptually interesting, but aesthetically it sounds horrible.

I like it, but couldn't get past having to learn the argument order of each predicate.

There's strength in numbers, as the author points out, if your goal is to convince a few billion people to speak the language, it doesn't matter much if you're starting with 5,000 speakers, or 5,000,000.

If your goal is to hang out with other language geeks, the latter one sounds preferable, but at that point, you may as well be learning Quenya or Klingon.

> > it's the most popular constructed language, meaning there are good resources and actual people (few, ok) to talk to.

> if your goal is to convince a few billion people to speak the language

I'd go with English or Chinese.

> If your goal is to hang out with other language geeks, the latter one sounds preferable, but at that point, you may as well be learning Quenya or Klingon.

Esperanto is popular enough that I'll randomly run into Esperantists on a street corner in my city, and I'll get newsletters about local Esperanto events. (And yes, there's a lot of retired people who are into Esperanto because it used to be more popular.) That is literally not happening for Quenya, Klingon, Lojban, Ido, or Interlingua. It may be happening for Toki Pona, but then again, tomo tawa supa mi pi lon sewi li jo e kala linja mute mute.

jan utala li toki e ni: ni!

well, for klingon at least there are star trek conventions where you can go and impress people with your language skills.

Interlingua isn't an improvement over esperanto, Interlingua is what esperanto tried to be and actually succeeds in it. Esperanto tried to be auxiliary and fails at it. instead of becoming an international language it becomes another language to learn. while interlingua is a language you already know but have never learned. you just have to learn to speak.

Esperanto was always another language to learn. The idea was never that you people would be able to read Esperanto without learning it; that's pretty much impossible, considering the diversity between languages (even looking just at Europe). The idea was to have a language that would be easy to learn and speak, owing to a simple, regular grammar, and using internationally recognisable words where possible.

Interlingua may be readily intelligible to speakers of Romance languages, but is its grammar any simpler than Esperanto's? Or is comparatively more complex and difficult to learn? Because that's what ultimately matters for the rest of the world's population who do not speak a Romance language.

English has a lot of Latin in it, plus if you are a cosmopolitan English speaker you probably get exposed to some French, Spanish (if only when you read the wrong side of a brochure at Home Depot and realize you understood it anyway), Latin, Italian, Portuguese, etc.

Interlingua doesn’t claim to serve any community outside European language users, but it does succeed at that, I think.

Overall interlingua is about as simple and regular as can be. A Chinese speaker might still be better off learning English but I can’t imagine they’d find interlingua any harder than any natural language.

FWIW As a non native English speaker, it is a terrible language to learn and understand. I have been using it for 2 decades and I still need subtitles for movies, not to mention all the irregularity and exceptions to rules (Honestly, when you have more exceptions than rules, can you even say you have rules at all).

>Overall interlingua is about as simple and regular as can be.

Interlingua never cared much about simplicity and regularity and it shows. Occidental before it has something called De Wahl's Rule that is focused on achieving as much regularity as possible while being easy to read at first sight. Here's how it works: https://en.wikipedia.org/wiki/De_Wahl%27s_rule#Rule

There's a great comic that characterises this https://pbs.twimg.com/media/FeC8E8RaEAAxMxi?format=jpg&name=...

Well no, Interlingua maybe tried to be a better Esperanto, but if so it completely failed. Esperanto is the only constructed language than became a real language and not a nice thought experiment

I fail to see how something "completely fails" and is being talked about here...

>Personally I like Interlingua which is basically a cleaned-up latin.

I on the other hand like it messy and stick with good old fashioned Latin (as an added bonus, the language of my ancient and noble church). Et numquam mutabo, irrumo Interlinguam!

Church Latin is quite different from Classical Latin, though - and surely only the latter truly counts as "old fashioned"? ~

Medieval Latin isn’t really that much different than latin from the classical period.

Pronunciation-wise it sounds pretty different to me.

Wow, I can basically understand every word. It must be even easier for speakers of a more Romantic language (French, Spanish, Romanian, etc).

Yeah, I've been needling people in Discord servers by writing up Interlingua and most of the time most people are able to read it and respond back in English or Romance languages

Si solmente nos ha un sorta de interlingua traducentes del linguas major, assi le personas pote scriber in interlingua sin que non necessariamente apprender lo.

Looks (too) close to Spanish. (Also, I’ve been under impression for a while now that Spanish is the closest thing to a “modernized” version of Latin that we already have.)

There is also another Interlingua which is also called as Latina-Sine-Flexione.

> Myth: Esperanto is internationally neutral,

Somewhat. It is a matter of degree, relative to what came before.

> and is the best (some say the only) candidate for an eventual world-wide auxiliary language.

I'm not in a position to say what is best for such a purpose.

> Reality: Esperanto is overwhelmingly European in design, content and aspiration, and has nothing in common with most non-European languages - such as Arabic and Swahili, two other languages which are widely used as linguae francae in other parts of the world.

Sounds plausible, but I'll need to verify.

> It is thus woefully deficient for world-wide auxiliary use - ...

This does not follow. What are some convincing supporting arguments?

>... a role which English, for all its faults, has been fulfilling perfectly well for decades.

Perfect? This kind of reasoning is laughable.

English is the IAL as far as air traffic control is concerned.

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

> Aviation English is the de facto international language of civil aviation. With the expansion of air travel in the 20th century, there were safety concerns about the ability of pilots and air traffic controllers to communicate. In 1951, the International Civil Aviation Organization (ICAO) recommended in "ICAO Annex 10 ICAO (Vol I, 5.2.1.1.2) to the International Chicago Convention" that English be universally used for "international aeronautical radiotelephony communications."[1] Despite being a recommendation only, ICAO aviation English was widely accepted.

To be fair, the greatest aviation disaster of all time was caused by a mutual misunderstanding from English's complexity and diversity, causing two 747s to believe they both had right of way on an air strip and then they crashed into each other.

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

This is true, but the counterpoint is that we'll never know the converse - we'll never know how many accidents didn't happen thanks to using a single language universally.

I guess the point I'm making isn't against using a single language, but that English isn't a great choice for that language since it's not simple nor standardized. Aviation's lesson from this was basically to invent their own dialect, that unfortunately isn't general purpose enough to be used much outside of aviation.

> To be fair

you're not being fair, though. A common language was chosen for aviation, and nothing in that article suggests that a different language choice would have been better. In this case you had Dutch speakers communicating with Spanish speakers. English does offer a blend between romance and germanic languages (which of course could lead to "false friends", but no language will be perfect)

> Perfect? This kind of reasoning is laughable.

That's not what "perfectly well" means. It doesn't mean perfect, it means doing an acceptable job. Honestly you could just remove the adverb "perfectly" and the meaning of the sentence there doesn't really change.

Hard disagree. I haven't found any resources online to support your claim. I found several that support mine.

Also, contrast "perfectly well" with "well enough".

I'm curious why you see "perfect" as being demoted to "acceptable" in this phrase? Source?

Because that's how English works? "Perfectly well" has never meant literally perfect.

You might want to read Justin B. Rye's "Learn Not to Speak Esperanto" for way more detail on all of this: http://jbr.me.uk/ranto/

Edit: also, as to this

> > Reality: Esperanto is overwhelmingly European in design, content and aspiration, and has nothing in common with most non-European languages - such as Arabic and Swahili, two other languages which are widely used as linguae francae in other parts of the world.

> Sounds plausible, but I'll need to verify.

Like, just given the history of Esperanto's construction -- it was built from European languages -- this should be no surprise. How is it going to end up reflecting non-European languages, which vary heavily and include all sorts of features that are entirely absent in European languages, when it was built exclusively from European languages? I mean I suppose notionally it could happen, but it would be very surprising.

I don't see why an auxiliary language needs to reflect all languages. How could it?

Are you suggesting that an auxiliary language should have elements from as many languages as possible? That such a "balanced" or "blended" approach is better b/c it allows people from more linguistic traditions to feel a little more comfortable?

Maybe a little, but such complexity would sacrifice learnability.

English has been adopted as a common lingua franca though very different from Arabic and Swahili. The driving factors have more to do with geopolitics and economic power than linguistics or conceptual familiarity.

> I don't see why an auxiliary language needs to reflect all languages. How could it?

In general, it doesn't. The claim is not that being Eurocentric makes Esperanto a bad auxlang. The claim is that being Eurocentric means that Esperanto is failing at its design goals; Esperanto is a bad Esperanto. The myth this point is responding to begins with "Esperanto is internationally neutral". That's the problem -- a general auxlang doesn't need to be neutral (English isn't!), but Esperanto claims to be (remember, Zamenhof just called it "the international language") and fails.

You're also correct that a blending approach is probably untenable. However, looking outside European sources gives one a wider range of options with which to optimize over. Some of those options people may realize without exposure to the particular languages, but others -- such as, say, ergative-absolutive alignment (not that I'm suggesting that's a good option to use) -- they won't.

E.g., why isn't Esperanto an analytic language? That would probably make it the easiest. Zamenhof correctly recognized (if not in these terms) that making it agglutinative was better than making it fusional. Why not go further and make it analytic? Likely because he hadn't studied any analytic languages...

Or look at the phonetic inventory; you're not going to get something good by somehow blending the phonetic inventories of the world's languages, but what you could do is look for something simple and minimal that few people will have trouble with. Instead, we get rare distinctions that will only be familiar to Polish speakers. Is having such distinctions in general bad for an auxlang? Of course not! But it's bad for one that claims to be the international language.

Again, I suggest reading through JBR's "Learn Not to Speak Esperanto" and its appendices.

See also, Justin Rye's "Learn Not to Speak Esperanto": http://jbr.me.uk/ranto/

I think the clear winner of functionality, form and easy of finding fellow speakers definitely goes to Ithkuil https://en.wikipedia.org/wiki/Ithkuil

Which, slightly less sarcastically, is probably a great runic language for us to start using for posterity on long lived markers (like at radioactive waste sites) since any Ithkuil manual or reference is so incredibly dense that it's rot rate can probably be measured on the scale of centuries.

Wouldn't information density be a bad thing for such uses? If one part of the rune is damaged, you lose a ton of information. Contrast with e.g. Latin, Greek, Sumerian, etc. in which inscriptions are heavily damaged, but the meaning is still deducible.

Consider this fragmented sentence:

    T___reside_t_of__he_Unit___tates__s_Joe_B__en

A sufficiently determined future archaeologist will be able to figure out what it originally said with some degree of accuracy. The lack of density is essentially functioning as parity bits.

If you think information preservation is important enough to warrant repetition wouldn't it be more efficient to repeat the few highly significant characters several times rather than hope that the deterioration of characters happens to allow significance to still be gleaned? If the sentiment takes 1/6th of the characters to express you could just replicate it six times which would probably have a better chance of surviving wear than a single sentiment written once.

Also - we, living humans, are very good at context based inference due to our high familiarity with the language - but someone who didn't know that "President" was a political position or had never heard of Joe Biden or the US would probably struggle to piece together even what that sentence was trying to convey. "They reside at of the Unit Potates. S'Cuppa Joe By Den"

Re: context, a future archaeologist would have more examples of the language, just like we have lots of examples of Latin, Greek, and Sumerian which allow us to learn the language with enough context to decipher the more damaged inscriptions. The trickiest part might be the name Joe Biden which isn't a part of English, per se. Correspondingly, there are ancient kings whose names we only have fragmented versions of for the same reason.

Ithkuil would obviously also need a large corpus for future archaeologists to decipher, otherwise the runes are meaningless.

Re: repetition - erasure coding (parity bits, basically) is far more efficient than duplication for data durability. How much this applies to natural language (where extraneous characters act sort of like parity bits) I am unsure, but there is nothing stopping you from just mapping your natural language to a binary code and using actual parity bits.

Obviously, a future archaeologist will need to know what kind of erasure coding scheme you are using, but that is actually much simpler to convey than the corpus of the language itself. It's pretty simple math.

What's a Potate?

That's a really interesting question - we're not really sure what Potate or Taters are, we've seen quite a few references to them across a few decades and our best bet is that it was connected to a briefly lived religious cult or or belief system as they are often referenced in connection to "heavenly" or sacred. A fragmentary surviving work by an author only known as STP describes in detail a "Mr. Tulip" who provides our most reliable source on this matter

> 'I thought maybe a maypole,' said Mr Tulip reflectively. 'An' then a display of country dancing, land tillage under the three-filed system, several plagues and, if my --ing hand ain't too tired, the invention of the --ing horse collar.

this whole fragment of a fictional character describing some of the rituals related to the Potate can be found near the quote "Everything will be alright as long as you have your -ing Potate" as expressed near the end of their life.

As a random tangent, looking at that wikipedia page sent me wandering across this example of english text as boustrophedon, and it made me wonder if it might actually be less visially/mentally taxing to read a book typeset that way, since you don't have to keep making big jumps from one side of the page to the other where you might lose your place:

https://en.wikipedia.org/wiki/Boustrophedon#Reverse_boustrop...

Pretty telling that they felt the need to make such a long prologue about why not to be offended

Is the Esperanto community better now?

I think it's more telling about the rant and the ranter than the Esperanto community. Although the Esperanto community is pretty used to non-Esperantists or Esperanto beginners denouncing the language without trying to understand it first, and so there are going to be objections to a rant like this, for sure. See my commment above.

I've always wondered if machine learning would help here - we already have language agnostic embeddings (LaBSE, LASER) - now all we need is a way to pronounce them. Perhaps you could learn one by training an autoencoder that is constrained somehow just like the human vocal chords are

Speaking as an Esperantist, I've heard arguments like this over and over, from disgruntled Esperanto beginners, who half-heartedly learned portions of the language, motivated by fallacious utopian impulses, and then abandoned it completely, after deciding that it isn't "perfect" or "international enough," or "easy enough," or something else. That's naive and short-sighted.

The fact of the matter remains that Esperanto is the most successful, most widely spoken international auxiliary language. It has the largest community, probably the friendliest community of any international language, and has the largest body of literature, with over 30K books published in the language, and with hundreds of new books published each year. It remains the easiest to learn, even for speakers of non-European languages. I know this because I've met tons of Chinese and Japanese Esperantists, who've all learned it after learning English, or Spanish, or Korean, and conclude that Esperanto is orders of magnitude easier and more intuitive than those languages.

It's not completely logical, obviously (Lojban is probably more logical) but is incredibly logical and intuitive, compared to any natural language. It's maybe not as easy as Toki Pona, with its vocabulary of 100 or so words, but it's still pretty damn easy. I learned Esperanto in a matter of weeks, rather than years. Esperanto is not as concise as a language like Volapuk, either, but that concision comes with too many trade-offs.

Sure, it's all a matter of personal taste, but if you learn Lojban or Toki Pona, or even Ido, you'll be able to speak with, at most, tens of other speakers, instead of millions, and read maybe ten books total, rather than tens of thousands in Esperanto. With Esperanto you'll be able to use operating systems like GNOME and websites like DuckDuckGo, Google, Facebook, and Mastodon, all using Esperanto. You'll be able to watch thousands of videos on sites like Tubaro and Youtube. You'll be able to follow hundreds of Twitter and Mastodon users. You'll be able to listen to thousands of Esperanto-language songs and podcast episodes. You'll be able to play games like The Expression Amrilato, and read comic books like Saga, which have Esperanto-language components. Best of all, you'll be able to travel the world with Pasporta Servo, staying for free with Esperantists along the way. There are also conferences attended by thousands of people, summer-long residential courses, and even universities which use the language. Virtually none of that stuff is available in other constructed languages.

As the writer of this article himself admits, Esperanto isn't perfect. Obviously. It was never intended to be. Nor was it intended to be perfectly international, with word roots derived from all languages throughout the world, equally. Nor was it intended to be perfectly easy, or perfectly well-distributed among easy phonemes. But Esperanto is an excellent compromise between all these factors, and is a hugely enjoyable, expressive language with a large and welcoming community. The proof, for me, of its expressivity, is that so many classics of western literature, and of world literature, including much of Shakespeare's works, have been very eloquently and poetically translated into Esperanto. That's not true of any other constructed language.

All natural languages evolve, and so does Esperanto. Esperanto reform and evolution has been happening since the beginning. What's absurd is the beginners, or even total non-speakers who insist on reform before even learning the language to a competent level. I think this has to do with utopian feelings. They think Esperanto should be perfect, and so when they learn enough of it to realize that there are faults, they give up on it altogether. But would anyone do that with say, German? French? It's a language. It has faults. Get over it. The difference is that Esperanto has probably 1% of the faults, irregularities, and idiosyncrasies of any other language.

I'd argue that everyone should learn Esperanto. It takes virtually no time or effort, compared to learning any other language, so it's basically a language you get for free. Like pseudo-code. And the worlds it opens up for you are so immense. Rather than debate about whether Esperanto is perfect or not, just learn it. You won't regret it.

We changed the URL from https://web.archive.org/web/20021210061212/http://www.cix.co... to what appears to be the article's canonical location (since the page says it's the author's website).

It's fine to post archive.org links if there's really no other way, but please search first. I found this one by simply googling the title.

when I initially made this submission 41 days ago the un-archived page looked like this: https://web.archive.org/web/20220212091844/http://www.cix.co...

but now that I've resubmitted it from the Second Chance Pool, you're right, it does seem to be back up. I had assumed the page had been taken down permanently.

The URL that works is not on that domain but http://miresperanto.com/konkurentoj/not_my_favourite.htm

It was a good submission in any case!

Reply on news.ycombinator.com