Noun classes, olin and "pona tawa"

Mind and thought: Wisdom, mental health, cognition, self-talk, consciousness, philosophy, psychology, optimizing your thinking, productivity hacks
Menso kaj penso: Saĝaĵoj, psiĥa sano, kogno, memparolado, psiĥa stato, filozofio, psikologio, rearanĝi sian pensadon, plibonigi sian produktokapablon
janMato
Posts: 1545
Joined: Wed Dec 02, 2009 12:21 pm
Location: Takoma Park, MD
Contact:

Re: Noun classes, olin and "pona tawa"

Post by janMato »

jan Ote wrote: I prefer to not repeat 'jan' in the next sentence than to have one syllabe less. Avoiding monotony, boredom in talk/text is natural. More than counting syllabes.
...
Some conlangs (e.g. Interlingua) aren't economical by design and their users like it, because they feel this more natural and usefull.
I'm confused. So when we have a partially designed language, do you expect the more verbose or less verbose variant to be the expected variant? I'm saying that in partially defined languages, we have to look for rules of thumb to decide how the resolve the rest of the language. So in absence of the things that normally cause people to be prolix (giving them selves time to think, sounding more formal or polite, trying to paint a picture (like the way tolkien describes trees), people would be expected to choose what ever is shorter to say. If a speaker has an unacceptable pronoun, they'll have to look for something else, and if it isn't explicitly illegal, it might serve the role of a pronoun. Imho, the hard thing here is conclusively coming up with the diagnostic tests for what is a pronoun.

Re: Using "on" in Slavic languages-- "on" is masculine-animate. Slavic languages do have grammaticalized noun-categories, so not a very good parallel.
Re: Slavic pronoun dropping and using only the bound morpheme i.e. 3rd person singular verb suffix, (which covers much the same ground as ona in toki pona). Still not sure if this illuminates toki pona, because it isn't controversial that Slavic languages have grammaticalized noun-categories, even if they don't *always* use them.

If anything, this helps my case because if Slavic languages have noun categories even when they don't always use noun categories, then toki pona might have noun categories that sometimes are ignored.

I'm trying to assert that toki pona does have noun classes, and can be inferred by looking at it's pronoun system. The language definition originally made all sorts of assertions ("toki pona has no plurals", well, actually plurals exist, they're just optional, "it has no tense", well, actually it has tense, it is just optional) And exactly how optional? If one wants to be understood, some of these optional markers will appear 50%, 80%, 90% of the time. Somewhere along that cline the construction becomes an obligatory grammaticalization. The day that toki pona readers see "mi en jan Foo en jan Bar li musi. tempo ni la mi li pali ala" and think it's unacceptably ambiguous and instead consistently say "mi mute li pali ala", then "mi mute" might as well be obligatory.

Any how, I'm starting to think that if I tried to make my case based on modified pronouns (which have official precedent) I might have an easier time making my point.

ilo li olin e ilo. is wrong because if we used pronouns we'd get ona ilo li olin e ona ilo. And "ilo" class pronouns can't love or be loved.
jan ilo li olin e jan ilo is okay because if we use pronouns we'd get ona jan li olin e ona jan. And "jan" class pronouns can love and be loved.

Therefore toki pona has noun classes which explain the difference between "olin" and "pona tawa". And it looks like there are ~100+ noun classes (one per modifier in the tp lexicon). As a conclusion, it's probably not really provable or disprovable, so keep in mind here, this isn't exactly science.
janKipo
Posts: 3064
Joined: Fri Oct 09, 2009 2:20 pm

Re: Noun classes, olin and "pona tawa"

Post by janKipo »

We seem to be comparing apples and schist. The fact that a language has a means of saying that more than one thing is involved does not mean it has plurals, nor does the fact that it can place an event in time mean it has tenses. Plurals (number) and tense are grammatical categories; telling how many or when are semantic/pragmatic functions, very different things. to be sure, the grammatical categories are sometimes used for the semantic functions in some languages. But there are other grammatical devices which are used for these purposes in other languages (and, indeed, in those that have number or tense). To have number or tense requires that there are regular ways to do these things and that their use is obligatory (with certain exceptions, defined for each language). tp has nothing like this for either category -- nor for other similar categories that come to mind: there are no obligatory connections with any content word (including exclusions).
To, be sure, each word has a net of connections of various sorts with other words and this net is different for each word. But that does not mean that each word is a different word class in a grammatical sense; it just means each word has a different meaning. Given the meaning of a word, it will go naturally with some things and not with others, 'sona' doesn't fit with 'kiwen' well, because stones tend not to be sentient, let alone wise (but, of course, we can tell a story about a wise old mountain , ...).
So, the most you can say is that 'ilo li olin e ilo ante' looks odd at first glance, but not that is is grammatically forbidden -- or even semantically -- but that it needs a story. And then we get on to the fuzzy story-classes, which are less help than just saying that certain things are assumed to have certain properties until otherwise revealed.
janMato
Posts: 1545
Joined: Wed Dec 02, 2009 12:21 pm
Location: Takoma Park, MD
Contact:

Re: Noun classes, olin and "pona tawa"

Post by janMato »

janKipo wrote:We seem to be comparing apples and schist. The fact that a language has a means of saying that more than one thing is involved does not mean it has plurals, nor does the fact that it can place an event in time mean it has tenses. Plurals (number) and tense are grammatical categories; telling how many or when are semantic/pragmatic functions, very different things.
I'll have to concede on that one. I'll come back to that someday when I have a chance to write about toki pona's dynamic stability. A rational, work minimizing, communication-success-maximizing agent would probably never choose to leave so much ambiguity when relatively cheap constructs are available, so I suspect quite quickly some constructs (plurals and tense) would be come so common as to be easier to mentally track as rules instead of words.
janKipo wrote:To, be sure, each word has a net of connections of various sorts with other words and this net is different for each word.
So this would just move my question from what are the grammatical noun types of toki pona to what are toki pona's "semantic categories." Brings to mind the problem of the lack of a defined metaphor system. Both questions could be resolved with either a conculture or a real culture.
janKipo wrote:So, the most you can say is that 'ilo li olin e ilo ante' looks odd at first glance, but not that is is grammatically forbidden.
The rules of jan Pije don't distinguish among points of conculture, style, grammar, or semantic modeling. It's obvious when Strunk and White are talking about optional style, in language definitions, I think it is inevitable that people like myself will at first interpret the language spec to be talking about grammar and not style recommendations.
janKipo wrote:And then we get on to the fuzzy story-classes, which are less help than just saying that certain things are assumed to have certain properties until otherwise revealed.
Interesting, I'll have to read up and see if anyone has tried to apply fuzzy set theory to grammatical noun-classes or semantic classes. I guess it would be something along the line of saying that it is 80% true that a robot is in the class of entities that can love and robots in love is 80% valid as a sentence.

Likewise reminds me of a issue in Slavic languages between going by wheeled conveyance and just going (different verbs for each). Airplanes must "go", but they go by wheeled conveyance on the tarmac. If one of the landing gear should jam and the airplane taxing along on on wheel, but not yet crashed from lift, is it on wheeled conveyance or flying?
janKipo
Posts: 3064
Joined: Fri Oct 09, 2009 2:20 pm

Re: Noun classes, olin and "pona tawa"

Post by janKipo »

Yes, it is all fuzzy logic from the get-go in the real world: grammatical classes are fuzzy (with very steep membership functions, to be sure), semantic webs are fuzzy, even physical characteristics (and indeed physical bodies) are fuzzy. But, happily, fuzzy things can be dealt with relatively unfuzzily at various practical levels. So, here, we can make specifications for 'olin', say, which work fine and also tell what sort of story is required to dodge the bullet. That doesn't make 'olin' a class, just a separate word, as it already was (tp is blessedly short of synonyms).
This has been a useful discussion, precisely because it brings these questions up and they need to be discussed at some point (maybe later, when there is a bit more culture to mess with). there are more pressing issues at the moment -- of the how to say correctly sort, for example.
By the way, don't confuse the value of the membership function with the the truth value of the claim that something is in the class. That is only one possible tv assignment, and often not the most useful one. For most practical cases, something about an acceptance function also plays a role, where acceptance presumably varies with membership but not necessarily directly. Of course, the strongest acceptance functions are those that give 0 up to a certain membership value and 1 thereafter, i.e., the preciding functions.
User avatar
jan Ote
Posts: 424
Joined: Thu Oct 08, 2009 1:15 am
Location: ma Posuka
Contact:

Re: Noun classes, olin and "pona tawa"

Post by jan Ote »

janMato wrote:Re: Using "on" in Slavic languages-- "on" is masculine-animate. Slavic languages do have grammaticalized noun-categories, so not a very good parallel.
Your source is not good. Pronoun "on" is NOT "masculine-animate", but for ALL nouns of grammatical masculine gender. In Slavic languages, just like in French or German, nouns for unanimated things do have grammatical gender:
 EN: a rose; a garden
 DE: die Rose (feminine); der Garten (masculine)
 FR: une rose (feminine); un jardin (masculine)
 IT: la rosa (feminine); il giardino (masculine)
 PL: róża (feminine); ogród (masculine)
 RU: роза (feminine); сад (masculine)
Now a small example:
tp:
mi lukin e jan e tomo.
 ona li lukin pilin ike.

EN:
 I see a man and a house.
 He/it looks (like) feeling bad.
PL:
 Widzę człowieka(m) i dom(m).
On wygląda na czującego się źle.
The pronoun 'on' is for masculine, but BOTH, 'man' and 'house' are masculine in Polish. So, the grammar of the second sentence doesn't indicate to which noun from the first sentences the pronoun 'on' refers to. But we don't see any problem here, do we? We do not need any additional grammar means to avoid confusion. Even if sometimes there are some ambiguity, we solve it without additions to the grammar. If this works for any natural language, why cannot for tp? This is my point.
janMato wrote:Likewise reminds me of a issue in Slavic languages between going by wheeled conveyance and just going (different verbs for each). Airplanes must "go", but they go by wheeled conveyance on the tarmac. If one of the landing gear should jam and the airplane taxing along on on wheel, but not yet crashed from lift, is it on wheeled conveyance or flying?
An old riddle. What answer do you give for English? Is it going on the ground or flying?
janMato
Posts: 1545
Joined: Wed Dec 02, 2009 12:21 pm
Location: Takoma Park, MD
Contact:

Re: Noun classes, olin and "pona tawa"

Post by janMato »

jan Ote wrote:
janMato wrote:Re: Using "on" in Slavic languages-- "on" is masculine-animate. Slavic languages do have grammaticalized noun-categories, so not a very good parallel.
Your source is not good. Pronoun "on" is NOT "masculine-animate", but for ALL nouns of grammatical masculine gender.
D'oh! My bad. Sady my source was my brain. But half of my sentence is still half true and in the domain I had in mind it is still sorta true (pronouns involving people). The prounouns in IE are unpredictable and have no semantic content except to indicate that when you say "oh" it means one of the bazillion words with masculine gender and not the feminine ones. It's still a category, albeit an unpredictable and partially phonetics based one, it still is used to resolved anaphora (deciding what one was referring back to)

Pronouns: I entirely forgot that in (at least some) IE languages grammatical gender trumps animacy. Only example I can think of is Iclandic barnið. So an Icelander would say(I think), the child is sick, it is sick. I haven't had a chance to check with an Icelandic speakers about edge cases where there is a Schrödinger's box with either a child(n), woman(f) or man(m), car(m) or bus(f) in the the box which hasn't been revealed yet, what pronoun would be used? I know in English, if we don't know the biological gender, masculine trumps feminine, i.e. if in doubt, use "he" or sometimes "it". If they were told it was either a female mouse, female human or female elf, would they pick "on" or "ona"? If a Russian guy's nick name is Vovja, would people switch to using OHa to refer to him in a previous sentence just because his name ends in "a"? I will let someone else see how Vovja reacts when we say "Vovja has been drinking all day. She should go home already" in Russian.(Or maybe it's perfectly natural, I haven't spent enough time in Russia to know for sure)
jan Ote wrote:The pronoun 'on' is for masculine, but BOTH, 'man' and 'house' are masculine in Polish. So, the grammar of the second sentence doesn't indicate to which noun from the first sentences the pronoun 'on' refers to. But we don't see any problem here, do we? We do not need any additional grammar means to avoid confusion. Even if sometimes there are some ambiguity, we solve it without additions to the grammar. If this works for any natural language, why cannot for tp? This is my point.
Now getting back to tp. Tp has "ona" which can stand in for *any* word. This is a more confusing situation than even strictly grammatical gender because Polish *oh* can only stand in for the 100,000s of words that are grammatically masculine and not the 100,000s of grammatically feminine ones. So that useful information for deciding what that pronoun was referring to. If I get you, "on" can't refer to roses. That is a valuable clue that some toki pona sentences lack, and imho, unnecessarily when other constructions do exist that imply and induce a noun-class way of thinking.

I see the man and roses. He doesn't look so good. (Sorry, can't say this is polish, but I suspect the he could only refer to the guy)
mi lukin e jan pali e kule kasi. ona li ike. <-- This option loses a lot of information. What is looking bad, they worker or the rose?
kule li ike. <-- clear and same length, legal to intepret kule as repeated noun, I'm no thin ice to call kule a pronoun
ona kule li ike <--- legal, and imho implies toki pona has tons (at least 120+) of phrases that quack like pronouns. ona kule also guarantees the sense that I'm refering back to something and not talking about a new kule.

quote="jan Ote"]
janMato wrote:Likewise reminds me of a issue in Slavic languages between going by wheeled conveyance and just going (different verbs for each). Airplanes must "go", but they go by wheeled conveyance on the tarmac. If one of the landing gear should jam and the airplane taxing along on on wheel, but not yet crashed from lift, is it on wheeled conveyance or flying?
An old riddle. What answer do you give for English? Is it going on the ground or flying?[/quote]
I'd probably say, 'The airplane is half-flying, half taxi-ing down the runway.' The example doesn't work as well in English because going by wheeled conveyance vs going by foot, etc is an important distinction in Russian, but not in English anymore. When I asked a Russian once, they said the difference was if they thing has wheels. Hence the case of things with wheels that are going, but not "driving". I can drive a car and a football player can drive down the football field (motion on foot with unusual vigor). Although If I drive to school, I'm almost certainly using a car. If I'm flying down the hallway, I'm also probably on foot.

Anyhow, even if classes (semantic or grammatical) don't help me decide if robots can love each other or merely find each other pleasing in the toki pona mind, noun classes do reveal what distinctions the language thinks are culturally important. For example, the drive/walk distinction comes from PIE which was one of the first societies to rely on pastoralism with wagons.
User avatar
jan Ote
Posts: 424
Joined: Thu Oct 08, 2009 1:15 am
Location: ma Posuka
Contact:

Re: Noun classes, olin and "pona tawa"

Post by jan Ote »

janMato wrote:Pronouns: I entirely forgot that in (at least some) IE languages grammatical gender trumps animacy. Only example I can think of is Iclandic barnið. So an Icelander would say(I think), the child is sick, it is sick.
Surprise!
PL:
 Dziecko jest chore. Ono jest chore.
 dziecko (neuter) - "a child";  ono - pronoun for singular neuter, "it"
janMato wrote:If a Russian guy's nick name is Vovja, would people switch to using OHa to refer to him in a previous sentence just because his name ends in "a"?
Vova? No. Because they know this is a popular male name and he's a male. But otherwise... If a name ends in "a" and there is no way to say is the person a woman or a man, then grammar will be used. E.g. some young people don't know that "Barnaba" is a male name, then young Poles tends to think it's "she". And vice-versa: "Mercedes" doesn't end in "a", moreover, 'a car', 'samochód' is masculine, then many can think that if it is a name then it's a male name.
janMato wrote:I see the man and roses. He doesn't look so good. (Sorry, can't say this is polish, but I suspect the he could only refer to the guy)
Right. Even in Hungarian, which hasn't got a grammatical category of gender ("ő" means "he/she", but pronouns for things are "az" sing. / "azok" pl.).
janMato wrote:mi lukin e jan pali e kule kasi. ona li ike. <-- This option loses a lot of information. What is looking bad, they worker or the rose?
In such cases I use:
mi lukin e jan pali e kasi kule. jan ni li ike.
mi lukin e jan pali e kasi kule. kasi ni li ike.
(By the way: noun - first, adjective - second; so: plant-colourfull)
janMato wrote:kule li ike. <-- clear and same length, legal to intepret kule as repeated noun, I'm no thin ice to call kule a pronoun
It means 'color is bad'. Strange.
janMato wrote:I'd probably say, 'The airplane is half-flying, half taxi-ing down the runway.'
Then the same for Slavic languages. 'W połowie leci, a w połowie jedzie po pasie'.
janMato wrote:When I asked a Russian once, they said the difference was if they thing has wheels.
Not only.
The verb 'ехать' (PL: jechać) can be used for a general travel (especially travelling abroad: 'I go to Himalayas'), for riding a horse or other animal (на лошади, верхом), sledging (на санях) or skiing (на лыжах) . Or even using a ski-lift (на подъемнике).
janMato
Posts: 1545
Joined: Wed Dec 02, 2009 12:21 pm
Location: Takoma Park, MD
Contact:

Re: Noun classes, olin and "pona tawa"

Post by janMato »

jan Ote wrote:
janMato wrote:Pronouns: I entirely forgot that in (at least some) IE languages grammatical gender trumps animacy. Only example I can think of is Iclandic barnið. So an Icelander would say(I think), the child is sick, it is sick.
Surprise! Dziecko jest chore. Ono jest chore.
Interesting, but not surprising. they're all descendents of PIE, which was crazy enough to pick such a noun-class system. The Australians have a better system where noun classes are grammatical and do have semantic content. The PIE noun class system has the benefit of helping with some cases of anaphora (which is good!) and a serious cost of learnability (which for native speakers only shows up in delayed speaking age, and even that is only noticeable for some of the most difficult languages like Danish)

I think your point is that when a speaker of a modern IE language says he, she or it, they really have no idea what it might mean (it could be biologically male, inanimate, etc) and aren't worse of for it, therefore toki pona doesn't need any fancified pronoun constructions--do correct me if I'm wrong.

All I'm saying is that in fact IE languages are better off (at least with respect to anaphora) than toki pona, and toki pona sentences can already avail themselves of constructions that have a whiff of noun-classes-ness (err, semantic-classes), it would be useful for reducing ambiguity and maybe useful for exploring what categories the tp noun currently covers, what noun classes go well with certain verbs.

Re: jan ni vs ona jan vs jan as tools for expressing anaphora
I would guess that "jan ni" give me a sense that I'm not talking about what came before.

me lukin e jan pali. jan ni li pali ala. I see a working man. (on the other hand) This (other) man is not working.

And as usual, this TP sentence has mutliple readings and one of them is, "I see the worker, he (the specific person, or nearby person) is not working."

mi sama li jan utala li ni: jan utala li toki e "Ni!
As for myself, I am a knight, I am this, a Knight who says NI!
User avatar
jan Ote
Posts: 424
Joined: Thu Oct 08, 2009 1:15 am
Location: ma Posuka
Contact:

Re: Noun classes, olin and "pona tawa"

Post by jan Ote »

janMato wrote:Interesting, but not surprising.
It was meant to be a joke exclamation.
janMato wrote:therefore toki pona doesn't need any fancified pronoun constructions
Right.
janMato wrote:All I'm saying is that in fact IE languages are better off (at least with respect to anaphora) than toki pona
...and they have plural, tenses and "gradable" adjectives, while toki pona doesn't have them. But it works.
janMato wrote:Re: jan ni vs ona jan vs jan as tools for expressing anaphora
I would guess that "jan ni" give me a sense that I'm not talking about what came before.
No, you are completely, totally wrong. The demonstrative pronoun 'ni' is used like English pronoun/article 'this'/'the' or Deutch 'dieser'/'der', i.e. for an object that have been previously mentioned:
mi lukin e jan pali. jan ni li suli.
I see a working man. This man is fat = The man is fat.
I see a working man, who is fat.
janMato wrote:me lukin e jan pali. jan ni li pali ala. I see a working man. (on the other hand) This (other) man is not working.
This doesn't make sense. Until you (by non-verbal means!) point at some other man, between saying the two sentences. If you do not do it, 'ni' refers to the man mentioned in the first sentence.
janMato
Posts: 1545
Joined: Wed Dec 02, 2009 12:21 pm
Location: Takoma Park, MD
Contact:

Re: Noun classes, olin and "pona tawa"

Post by janMato »

jan Ote wrote:No, you are completely, totally wrong. The demonstrative pronoun 'ni' is used like English pronoun/article 'this'/'the' or Deutch 'dieser'/'der', i.e. for an object that have been previously mentioned:...
No, at most, I'm partially wrong. (And I'm not trying to imply you are wrong, because I long ago lost track of what I was arguing for or against. It was something about using the word olin in the sentence, jan pi kili upa li olin e kili", I'm no longer sure how I got from there to here)

This little piggy went to market,
This little piggy stayed at home,
This little piggy had roast beef,
This little piggy had none.
And this little piggy went...
"Wee wee wee" all the way home...

"this" in each case introduces a *new* entity, usually the next toe on a the foot of a child.
jan Ote wrote:This doesn't make sense. Until you (by non-verbal means) point at some other man, between saying the two sentences. If you do not do it, 'ni' refers to the man mentioned in the first sentence.
re: finger pointing + this
Yes, paralinguistic evidence is important, more so in toki pona and probably worth of more formal attention than the current language definition gives it. Everyone talks about context, but is anyone doing anything about it?

Well, determining if a sentence has been produced correctly is just a matter at looking at toki pona's noun-phrase-grammar rules (I suppose the ones on Wikipedia). That can objectively resolve if a sentence is legally produced. On the other hand, the issue you're drive at (and that jan Kipo often drives at) is a matter of how to take a well formed toki pona sentence and transform it into meaningful ideas in the brain. That process is generally invisible, and so harder to prove true or false, harder to study compared to spotting a misspelling ore a missing "li", "e", or "la"

So you might be right about "ni" being strictly for referring to previous antecedents (I don't think you are) but I can't prove it-- the language isn't defined enough to do so. Maybe someone will write the flip side of the noun-phrase-grammar and explain how to turn a valid sentence back into "ideas." I suppose in a natural language one could utter a sentence and record how peoples behaviors altered and then infer how they must have parsed the sentence back into ideas.

Re: IE languages having grammaticalized, obligatory plurals, tenses, etc, vs toki pona.
We'll, I'll have to concede that nothing that serves the same role of the grammaticalized plural, tense, etc in toki pona is obligatory. But toki pona does have means for expressing plural, tense, etc and one would be hard to understand if one *never* used plural, tense, etc constructions. In fact, I suspect that plural, tense, etc are so useful that people trying to maximize the odds of being understood will start using these constructs so often, that the community would grammaticalize these currently optional ways of expressing plural, tense, etc. It would be a community innovation and cause grief when deciding if its part of toki pona, just like the Esperantists worry what maltrinki means and if it is a good community innovation. But that is a topic for another day.
Post Reply