Hacker News new | ask | show | jobs
by kindawinda 1047 days ago
I've been delving deep into the structure of toki pona and noticed some intriguing gaps:

Abstract concepts - The language seems to skate over complex philosophical or scientific ideas. You won't find an easy way to express ideas like "justice" or "quantum". It's a testament to its minimalist design, but it also means some concepts are lost in translation.

Specific flora/fauna - While you can use terms like "kili", "kala", or "soweli" to describe plants, fish, or mammals, respectively, don't expect to convey specific species such as "maple tree" or "cat". It's a broad brush approach that lacks granularity.

Intensive modifiers - Toki pona leans hard on the word "mute" for intensifiers. This simplification might hinder nuanced expressions, as there's a big gap between "mute" and a neutral descriptor.

Collective nouns - The language only really has "kulupu" to describe collectives. So, distinguishing between, say, a "flock" and a "crowd" might get tricky.

Idiomatic phrases - A repertoire of idioms or set phrases for common topics would make conversations feel more culturally rich and engaging. It's one thing to communicate efficiently, but another to do so with character and flair.

This isn't criticism but an observation of the trade-offs made for simplicity. Every language has its quirks, and understanding these nuances can make for a richer linguistic experience.

7 comments

> Abstract concepts - The language seems to skate over complex philosophical or scientific ideas. You won't find an easy way to express ideas like "justice" or "quantum". It's a testament to its minimalist design, but it also means some concepts are lost in translation.

Justice could be translated as “sama pona” (literally “good sameness”). Quantum could be translated as “wan” (literally “unit”); so a quantum leap in the sense of a non-gradual physical movement would be “tawa wan” (“single movement”), whereas a quantum leap in the sense of a large change would be “ante suli” (“large difference”).

You're right that it is very easy to lose nuance in translation. I guess that's true of translating between any two languages that are quite different.

> Specific flora/fauna - While you can use terms like "kili", "kala", or "soweli" to describe plants, fish, or mammals, respectively, don't expect to convey specific species such as "maple tree" or "cat". It's a broad brush approach that lacks granularity.

The idea is that rather than naming things, you describe them. A maple tree could be “kasi suwi” (“sweet tree”) or “kasi pi ma Kanata” (“Canadian tree”), depending on context. A cat could be “soweli pona” (“friendly animal”) or “soweli utala” (“fighty animal”) or “soweli utala pi luka kiwen” (“hard-clawed fighty animal”) or “soweli kalama” (“noisy animal”) or “soweli pona utala kalama” (“noisy fighty friend animal”), depending on your relationship to the individual.

Toki pona forces you to think about which aspects are relevant to your meaning, and state the context explicitly. This is one of the stated intentions of the language, and therefore a feature rather than a bug :)

> Idiomatic phrases - A repertoire of idioms or set phrases for common topics would make conversations feel more culturally rich and engaging.

There are some of these, for example “jan pona” (“good person”) very commonly means “friend”.

> “jan pona” (“good person”) very commonly means “friend”.

But “good person” and “friend” are two very different things. I might describe a well-known charitable figure as a good person, but that wouldn’t mean I would be calling them my friend. One is a value judgement and the other is a personal relationship.

...in English

But English has similar ambiguities. For example, a "girlfriend" is usually not just a girl friend. Other expressions are used when there is no romantic relationship.

I don't know toki pona but there is probably another expression for actual good people. Maybe something like "person of high value".

The language is also geared towards expressing ideas in general much more than expressing your own feelings. In this way, a "good person" is someone you would consider a friend even if you've never met, which while not exactly a concept that's entirely sane on the internet, makes a lot of sense in a more thighs knit community (eg: a good person is a friend to the community and therefore a friend to me).
I have a direct answer: once you get good enough at toki pona to want to distinguish, you do some personal reflection about what a friend is to you and find more specific ways to say that, such as jan poka (accompanying person) or jan mi (my person). You can also specify whether jan li pona tawa mi (good to me) or li pona tawa kulupu (good to the broader community).
"Arm" and "hand" are two very different things (I guess?), but they are the same word (рука) in Russian.
That's likely no different than a homonym in English or a word with multiple meanings. Almost always context makes the meaning clear.
Even in English the term "friend" is ambiguous though, some people refer to someone they've only met once as a friend already, whereas others really need to think about who they consider a friend.
So call them "my good person" giving the implication of familiarity. Similarly my best person could be your best friend.
If I ask you "do you have a pet?", and you respond "yes, I have a cat", I can easily imagine the pet you have and be generally right (maybe except the colour). How would you respond in Toki Pona so that an average person would understand you have a cat, not a dog or a guinea pig or fish or a rabbit?
From other comments, you would probably describe the pet’s features (likely with a fair amount of confusion / fuzziness), or some feature-descriptions might be stand-ins for specific animals.

For instance imagine that rather than a cat your pet is a chuckwalla, while that name is quite precise, it is almost certainly useless to the average conversation partner because most people have no idea whatsoever what a chuckwalla is (or that there’s something called a chuckwalla in the first place).

If you have an unusual pet, the name is useless, agreed. But everyone knows what an average pet, like a "cat" or a "dog" or a "guinea pig", looks like, and it would be weird to describe what a cat is in response to this question in most non-conlang languages.
> Justice could be translated as “sama pona” (literally “good sameness”)

But how is that not just a compound word that needs to be learnt? Because I would not guess justice from "good sameness", it might as well mean morality.

So now you have to learn these compound words and the idea of a 130 word vocab is a bit dubious.

Yeah I seem to be noticing a common theme here of claiming that new words are easy to create by putting these simple words together without realising that this doesn't really help anything without people agreeing on which combinations of words mean which things.

If everyone starts to use "good sameness" for "justice", you now have a 131 word vocab. Repeat for how many combinations become common. Eventually you will end up with as many words as any other language. It's almost like the proponents of the small vocab of this language don't actually understand what a vocabulary is

The idea of the language is that you describe what you want to say in order for yourself to understand what you mean by each word

The commenter thinks justice is good and about equality, do you? It's mostly a challenge in self-discovery, to know how you would describe different things with the limited dictionary

It's not really made to communicate, although it is of course possible

I've studied three foreign languages (four if you count a failed Toki Pona attempt a while ago), and I find it neat to discover a new grammatical concept, such as subjunctive in Spanish, which exists but only barely in English, or to realize that a language is missing an English construct, such as Mandarin, which doesn't have anywhere near as many gendered words as Germanic or Romance languages. As with computer languages, it's rarely the case that something is inexpressible, but it might have to be done differently compared to another language. I don't completely buy into the Sapir-Whorf hypothesis, but I do believe that cultural differences are partly attributable to grammar and vocabulary.
As I understand it, neither Mandarin nor Irish have words that directly mean "yes" and "no."

Then again, half a millennium ago, English had 4 such words: 2 for yes, 2 for no. (Sir Thomas More tried to explain this circa 400 years ago, and even he got it wrong.)

Norwegian and Czech, both of which I speak very badly, preserve two forms of "yes" and I cannot keep them straight.

Learning Mandarin does make me yearn for the simplicity of English "yes" and "no." But I have to admit, each word in Mandarin works a lot harder than it does in English, so you can often say a heck of a lot in just one or two syllables.
In Mandarin and Vietnamese, you can build new words by concatenation of words with broad meanings. Say, a plane is "flying machine", a mobile phone is a "hand machine", and a fridge is a "cold closet". In practice, it's very effective and it allows you to either guess the commonly used concatenations, or make up your own and you might be understood. I assume you can do that in Toki Pona too ?
Most languages are like this. That English words are so opaque is quite a historically contingent thing, not the baseline of how languages usually work. English retains a lot of Latin and Greek and French compounds, which are usually translated more literally in many other languages.

In German, plane is Flugzeug (flight tool), fridge is Kühlschrank (cooling closet). In Hungarian, they are repülőgép (flying machine) and hűtőszekrény (cooling closet) as well, calqued from German.

English speakers often perceive this as some kind of endearing primitive Tarzan-like communication without all those classy prestigious words, but English also has plenty examples. How about "washing machine" for example?

Calques are great, as are loanwords.

In English, "calque" is a loanword, but "loanword" is a calque.

That is simply fun.
Right? :-D
In English, we make words the same way - it's just that we construct them out of latin or greek roots.

"Television" is tele (Greek, 'far') + vision (Latin, 'seeing'). German for Television is "Fernsehen" - fern (German, 'far') + sehen (German, 'seeing').

seatbelt pineapple
Even English kind of does this, just playing parcour through multiple languages, and with time and use shortening the compound words to shorter alternatives. A plane is an aeroplane, or aeróplanos, an air wanderer. A mobile phone is a mobile telefone, a movable far-voice.
And of course in many languages more related to English as well, such as (a bit infamously) German. And obviously there are compound words in English as well, either concatenated ("snowman", "airport") or not ("cell phone", "compound word").
My favorite is "electric brain," for "computer."
Well, the kind of computer you have in mind used to be electronic computer to distinguish it from a computer or calculator which described a human being who computed. See, for instance, Nevil Shute's biography Slide Rule in which he describes being in charge of a department of calculators when working on the R100.

It's a fascinating read. You can get it from Anna's Archive.

“Don't you see that the whole aim of Newspeak is to narrow the range of thought? In the end we shall make thoughtcrime literally impossible, because there will be no words in which to express it."
There are well over a hundred species of maple. It’s an entire genus.
Relevantly, many of them have no single word name in English, or even any specific name in English at all.
Pragmatically, in English we deal with this by briefly dropping into Latin to talk about acer caesium, for example, and this is fine because we’re generally not aiming for any kind of linguistic purity - it’s not so easy if we try to use Toki Pona as a highly puristic, controlled vocabulary language.
This is suggestive that ~Hard Whorfism may have some truth to it does it not? Expert consensus seems to be that it is a "fact" that Hard Whorfism is false, though I strongly disagree, wonder what your take is on it?

https://en.m.wikipedia.org/wiki/Linguistic_relativity

this was circulating on usenet in early 2000s for a while, id love to see it official:

# Toki Pona: Snoufax II Advanced

## Basic Particles - *a* - particle - *e* - object marker - *en* - and - *la* - used for conditions and time - *li* - links subject and predicate - *o* - vocative particle - *pi* - of

## Tense and Aspect Modifiers - *sili* - Continuous or ongoing. "mi sili toki" – I am speaking.

## Moods and Modality - *kin* - Potentiality (can/might). "mi kin toki" – I might speak.

## Technical and Scientific Terms - *ilo sona* - Computer (knowledge tool). "mi kepeken ilo sona" – I use a computer.

## Abstract Concepts - *pona nasin* - Justice (right way/path). "pona nasin li tawa mi" – Justice is important to me.

## Idiomatic Expressions - *pakala luka* - Break a limb (good luck). "pakala luka tawa sina!" – Break a leg (Good luck)!

## Politeness Levels - *sewi* - To elevate the status of the person being spoken to. "sewi jan Lisa, mi toki tawa sina" – Honorable Lisa, I speak to you.

## Detailed Descriptors - *jelo pimeja* - Dark yellow (like mustard). "mi jo e tomo jelo pimeja" – I have a mustard-colored house.

## Complex Syntax - *seme* - Universally before a statement to make it interrogative. "seme sina toki?" – Are you speaking?

## Pronouns - *ono* - Neutral pronoun (they/them). "ono li toki tawa mi" – They are speaking to me.

## Numerical System - *ten* - Ten. "mi jo e ten tomo" – I have ten houses.

## Prefixes/Suffixes for Intensifiers - *suli-* - A prefix to mean "very/big". "suli-pona" – Very good.

## Passive Voice - *palisa* - Passive. "mi palisa toki" – I am spoken to.

## Directional Particles - *noka* - Toward. "mi tawa noka tomo" – I am going towards the house. - *monsi* - Away from. "mi tawa monsi tomo" – I am moving away from the house.

## Negation of Modality - *ala kin* - Cannot, impossible. "mi ala kin toki" – I can't speak.

## Temporal Phrases - *tenpo pini* - Previously, before. "tenpo pini la, mi toki" – I spoke before. - *tenpo kama* - Soon, later. "tenpo kama la, mi toki" – I will speak later.

## Comparison and Superlatives - *sama lili* - Less like, lesser. "mi sama lili jan Lisa" – I am less like Lisa. - *sama suli* - More like, greater. "mi sama suli jan Lisa" – I am more like Lisa.

## Conditional Statements - *sama... la* - If... then... "sama toki pona li pona tawa sina la, sina toki kepeken ona" – If you like Toki Pona, then you speak using it.

## Reflexive Actions - *sama mi* - Myself. "mi toki tawa sama mi" – I talk to myself.

## Possession - *pi mi* - Mine. "tomo ni li pi mi" – This house is mine.

## Emphasis Particle - *kin li* - Emphasizes the action or state. "mi kin li toki!" – I really am speaking!

## Uncertainty or Hesitation - *...anu...* - Either... or... "mi toki anu toki ala" – I either speak or don't.

## Verb Structures - *li... e* - Direct object indicator. "mi li toki e toki" – I speak the language. - *tan... tawa* - Cause and effect. "mi tan tomo tawa tomo sona" – I go from home to school.

## Adjectives and Adverbs - *sama sona* - Intelligently, knowledgeably. "ona li toki sama sona" – He speaks intelligently. - *pona mute* - Very well. "mi toki pona mute" – I speak very well.

As an experienced toki pona speaker, this list is amusing because most of this is either a pointless terrible idea or already valid and accepted - somehow with very little in between. I'll leave it to the next commenter to go point by point.
That makes sense as to why it's from the 2000s prior to your modern tongue.