A Few Observations on Minimalist Languages

Tinkerers Anonymous: Some people can't help making changes to "fix" Toki Pona. This is a playground for their ideas.
Tokiponidistoj: Iuj homoj nepre volas fari ŝanĝojn por "ripari" Tokiponon. Jen ludejo por iliaj ideoj.
Mako
Posts: 184
Joined: Fri Jul 16, 2010 11:32 pm
Location: San Francisco

A Few Observations on Minimalist Languages

Post by Mako »

I've been looking at minimalist and auxiliary languages for a few months now, and am mostly pleased. These are a few observations more in service of my hypothetical minilang rather than expressing any desire to tinker with Toki Pona. Toki Pona has flaws, but so do all languages, and Toki Pona should not trigger Ido wars.

1) The base words should have a varied syllable lengths, preferably between 1 and 3 syllables, and varied syllable weights.
2) The final syllable of any base word should not be identical to the initial syllable of any base word (e.g., no 'meli/seli/suli lili).
3) Exact reduplication of a syllable within a base word should not be allowed (no 'lili')
4) Easily confused words should have maximally distinct phonologies (e.g., the mi/ni confusion in Esperanto)
5) Any artistic form, such as poetry, should develop from the language itself rather than an a posteriori decision (suli/seli/selo)
6) All personal pronouns take the verb marker, if there is one (no 'mi kute', 'sina kute', 'ona li kute')
janKipo
Posts: 3064
Joined: Fri Oct 09, 2009 2:20 pm

Re: A Few Observations on Minimalist Languages

Post by janKipo »

A good list to compare tp with, noting the most frequent complaints about it (after the lack of numbers -- but that is another area altogether). Depending on how minimal you want your language (e,g, 3-vowel tp), some of these desiderata become increasingly difficult to achieve for purely combinatorial reasons, but all of them seem reasonable as idealized goals. The other factor in minimal languages is coverage, of course, how much it can say in how much space. This the problem of basic word choice and of the rules for building compounds (and of exactly what you mean by coverage), Again, tp is probably incomplete at the moment, although it is hard to say exactly where, given its very moderate definition of coverage; later languages should look more closely at this problem in advance. The simplest language phonologically that I know of has 4 consonants and no vowels (the consonants are all continuants -- maybe there need to be one vowel symbol to mark which one is the syllabic peak), it turns out to be a mess for constructing words and its semantics is not well thought out yet.
janMato
Posts: 1545
Joined: Wed Dec 02, 2009 12:21 pm
Location: Takoma Park, MD
Contact:

Re: A Few Observations on Minimalist Languages

Post by janMato »

Mako wrote:1) The base words should have a varied syllable lengths, preferably between 1 and 3 syllables, and varied syllable weights.
2) The final syllable of any base word should not be identical to the initial syllable of any base word (e.g., no 'meli/seli/suli lili).
3) Exact reduplication of a syllable within a base word should not be allowed (no 'lili')
4) Easily confused words should have maximally distinct phonologies (e.g., the mi/ni confusion in Esperanto)
5) Any artistic form, such as poetry, should develop from the language itself rather than an a posteriori decision (suli/seli/selo)
These 5 are minimal pair related. I concur, but if you grind out the minimal pairs, you start to lose things that rhyme in any sense of the word. Personally, would sacrifice poetry to gain ease of learning.

Another thing that happens as you get rid of minimal pairs (in all the sense of the word), you start to need either longer words or more phonemes.
Mako wrote:6) All personal pronouns take the verb marker, if there is one (no 'mi kute', 'sina kute', 'ona li kute')
A more general form of this dictum is that "If your language uses particles to separate strings of phrases, they should not be optional" The verb segment also could use some more separators, eg.

mi wile ken tawa noka wawa mute.
I (verb phrase) (modal chain) (core verb) (adverb chain).

Since tp has both implied separators (which means you separate based on if something is a prototypical modal, adverb, etc), and explicit ones, I sometimes make mistakes where I drop particles when an implicit one would do.

* moku ni li moku mi mute.
This is pretty obvious without the separator, but implicit separators aren't allowed here.

Re: jan Kipo
What language has only 4 consonants and no vowels?
janKipo
Posts: 3064
Joined: Fri Oct 09, 2009 2:20 pm

Re: A Few Observations on Minimalist Languages

Post by janKipo »

One of mine. Not published for the reasons mentioned. The C are f,s,n,h (i..e., German ch) -- fun to speak and unpleasant to listen to, but very hard to make work semantically. I call it, for the nonce, fsnx - 2 syllable, both two sounds long, the second vocalic in each case.
Mako
Posts: 184
Joined: Fri Jul 16, 2010 11:32 pm
Location: San Francisco

Re: A Few Observations on Minimalist Languages

Post by Mako »

janMato wrote:
Mako wrote:1) The base words should have a varied syllable lengths, preferably between 1 and 3 syllables, and varied syllable weights.
2) The final syllable of any base word should not be identical to the initial syllable of any base word (e.g., no 'meli/seli/suli lili).
3) Exact reduplication of a syllable within a base word should not be allowed (no 'lili')
4) Easily confused words should have maximally distinct phonologies (e.g., the mi/ni confusion in Esperanto)
5) Any artistic form, such as poetry, should develop from the language itself rather than an a posteriori decision (suli/seli/selo)
These 5 are minimal pair related. I concur, but if you grind out the minimal pairs, you start to lose things that rhyme in any sense of the word. Personally, would sacrifice poetry to gain ease of learning.

Another thing that happens as you get rid of minimal pairs (in all the sense of the word), you start to need either longer words or more phonemes.
Mako wrote:6) All personal pronouns take the verb marker, if there is one (no 'mi kute', 'sina kute', 'ona li kute')
A more general form of this dictum is that "If your language uses particles to separate strings of phrases, they should not be optional" The verb segment also could use some more separators, eg.

mi wile ken tawa noka wawa mute.
I (verb phrase) (modal chain) (core verb) (adverb chain).

Since tp has both implied separators (which means you separate based on if something is a prototypical modal, adverb, etc), and explicit ones, I sometimes make mistakes where I drop particles when an implicit one would do.

* moku ni li moku mi mute.
This is pretty obvious without the separator, but implicit separators aren't allowed here.

Re: jan Kipo
What language has only 4 consonants and no vowels?
I'll address each point
1) For me, at least, successful styles of poetry for a particular language, considered in isolation, emerges from prosodic play within the language. I would prefer a native development before experimenting with foreign form. Rhyme is not the only form of poetry, nor does it seem a suitable form for TP.
2) I would not be opposed to a greater percentage of longer words compared to TP - remember, my Objection 2 above applies to initial and final consonants, not medial ones.
3) Particles are important - but too many particles could over-grammaticalize the language. Topic markers can become subject markers, after all.
4) Extra phonemes? Maybe.
oligo
Posts: 11
Joined: Tue Feb 22, 2011 11:02 am

Re: A Few Observations on Minimalist Languages

Post by oligo »

Nearly all "minimalist" languages are oligosynthetic languages. The best known oligosynthetic languages are: Ygyde, Kali-sise, Sona, and aUI. Ygyde is the most mature of these languages. Its dictionary (posted at: http://www.ygyde.neostrada.pl/ygyded.htm) has over 4000 words. Its description is posted at: http://www.ygyde.neostrada.pl/index.htm. Nearly all Ygyde words are compound words. These compound words are about twice as short as similar Toki Pona phrases. I would describe Ygyde as a grown-up version of Toki Pona. Examples of the Ygyde compound words:
aniga (corrupt) = a (adjective) + ni (secret) + ga (money)
ofyby (leavened bread) = o (noun) + fy (foam) + by (food)
igugo (to vaporize) = i (verb) + gu (liquid) + go (gas).
janMato
Posts: 1545
Joined: Wed Dec 02, 2009 12:21 pm
Location: Takoma Park, MD
Contact:

Re: A Few Observations on Minimalist Languages

Post by janMato »

oligo wrote:Nearly all "minimalist" languages are oligosynthetic languages. The best known oligosynthetic languages are: Ygyde, Kali-sise, Sona, and aUI. Ygyde is the most mature of these languages. Its dictionary (posted at: http://www.ygyde.neostrada.pl/ygyded.htm) has over 4000 words. Its description is posted at: http://www.ygyde.neostrada.pl/index.htm. Nearly all Ygyde words are compound words. These compound words are about twice as short as similar Toki Pona phrases.
Depends on what style you're speaking.
"mi wile e telo" does mean, among other things, "I want some Spanish Red wine, from six to eight years ago", in the simple style.

in toki nasa, the style where you try to say exactly what you mean, then it is good style to be verbose.
mi wile e telo nasa loje e lili tan ma Ispanija tan tenpo sike suno pi LW anu LWWW

If I really wanted to be concise, there are several conlangs optimized for that, including short hand systems, which are really condialects of English.
oligo wrote:I would describe Ygyde as a grown-up version of Toki Pona. Examples of the Ygyde compound words:
aniga (corrupt) = a (adjective) + ni (secret) + ga (money)
ofyby (leavened bread) = o (noun) + fy (foam) + by (food)
igugo (to vaporize) = i (verb) + gu (liquid) + go (gas).
That's agglutinating. toki pona is aggressively isolating. An illustration:
tomo pi telo nasa => becomes plural as => tomo mute pi telo nasa

As opposed to a more synthetic language like English (on the spectrum of course!)
jack-in-the-box -> jack-in-the-boxes (not jacks-in-the-box)

So in tp, everything that looks like a compound word, generally can still be treated as a phrase.

There is one area where this isn't a clear is modifier order in set phrases, eg. plural of friend
? jan mute pona, rare and feels wrong
jan pona mute - reads just fine.
janKipo
Posts: 3064
Joined: Fri Oct 09, 2009 2:20 pm

Re: A Few Observations on Minimalist Languages

Post by janKipo »

For some reason, I don't think of Ygydygy orWahooey as minimalist, since they have indefinitely expandable vocabularies and are not obviously restricted in either phonology or grammar. a UI. Is a straightforward philosophical language, old school. Ygyde is newer in the sense of making its atomic polysemantic under controlled conditions. But this is equivalent to having several times as many atoms, so the apparent small Initial size disappears.
In any case they are after quite different things from tp, where the minimalityaltogether is the goal.
oligo
Posts: 11
Joined: Tue Feb 22, 2011 11:02 am

Re: A Few Observations on Minimalist Languages

Post by oligo »

janKipo wrote:For some reason, I don't think of Ygydygy orWahooey as minimalist, since they have indefinitely expandable vocabularies and are not obviously restricted in either phonology or grammar. a UI. Is a straightforward philosophical language, old school. Ygyde is newer in the sense of making its atomic polysemantic under controlled conditions. But this is equivalent to having several times as many atoms, so the apparent small Initial size disappears.
In any case they are after quite different things from tp, where the minimalityaltogether is the goal.
Ygyde has 90 morphemes/roots which have double meaning. The first letter of the word, called prefix, explains which of the two meanings is valid. This system enables Ygyde to cram 180 different meanings into 90 morphemes. Toki Pona has 123 words, so it is in the same ballpark. In my opinion, Ygyde is as minimalistic as Toki Pona, but it is more precise because the meaning of every morpheme is clear and because the first letter of the word and the length of the word convey lots of information (verb, noun, adjective, time-word, number, color, preposition... and much more). Some of these ideas could be used in Toki Pona.
janMato
Posts: 1545
Joined: Wed Dec 02, 2009 12:21 pm
Location: Takoma Park, MD
Contact:

Re: A Few Observations on Minimalist Languages

Post by janMato »

oligo wrote:Ygyde has 90 morphemes/roots which have double meaning. The first letter of the word, called prefix, explains which of the two meanings is valid. This system enables Ygyde to cram 180 different meanings into 90 morphemes. Toki Pona has 123 words, so it is in the same ballpark. In my opinion, Ygyde is as minimalistic as Toki Pona, but it is more precise because the meaning of every morpheme is clear and because the first letter of the word and the length of the word convey lots of information (verb, noun, adjective, time-word, number, color, preposition... and much more). Some of these ideas could be used in Toki Pona.
How transparent are these compounds? Does mouth-water mean delicious (as it obviously does in English) or does it mean spit (and I can't remember what conlang uses that). Anyhow, *all* conlangs have to build words by various techniques, if only because most conlangers never follow through with a 100,000 word dictionary that can cover everything that people might want to talk about. Compounding is just one of derivational strategies and compounding isn't all that much of a panacea, nor is POS markers.

When I was reading the Klingon dictionary, I thought, man, this reminds me of toki pona because the word set is *closed*. Everything has to be build out of phrases the derivational strategies available.

Anyhow, for me, what makes a language interesting is if it has a community, if it has some sort of 'prestige' factor (which is the linguists term for it and not really a good one). A language or conlang's prestige factor is what ever warm fuzzy feeling you get from using it, either because it's the language that will get you the big house (English) and the pretty girl (Italian), God speaks that language (uh, various), or it's the language that will bring world peace (esperanto), or the language that will make your thought so clear that problems will just wave the white flag upon hearing your logical loglan arguments. And prestige is really to strong of a word because being really good at Klingon or Na'vi is probably not going to help you get into a prestigious university or fraternity and I don't think that is the sense that is meant when field linguists talk about a language's prestige.

What isn't so exciting is that language a or b merely is structurally better than language c, else Ido or Interlingua would have swept away Esperanto and Na'vi would have swept away English, at least in places like Alabama where they weren't very good at it anyhow.

toki pona has been stagnant since about 2002-2003, which is a good thing for a language. The toki pona that people use is much the same sort of stuff, except with a larger phrasal vocabulary, fewer mistakes, newer texts are more likely to use the tools that are available, etc.
Post Reply