Expanding on "mi moku e kasi e soweli":
I am trying to test for a distinction between some specific senses of toki pona words, which I will illustrate with “moku”, namely:
the noun “food” (call it moku0)
the intransitive verb “to be food” (which is how I analyze “kili moku”, with an implicit copula; call it moku0c1; this part of the analysis is not super elegant tbh)
the intransitive verb “to eat” (call it moku1; I have not been able to find a way to distinguish “I cause: soweli li moku1” and “I apply moku0 to the soweli” to my satisfaction, and as such will not be trying to elicit moku1 readings in the example below)
the transitive verb “to eat something” (call it moku2)
Imagine the following have been made clear from context:
“pan” refers to a specific slice of bread
“kasi” refers to a part of a plant that needs to be cooked in order to be edible due to its very tough shell
“linja” refers to some cooked spaghetti
Can the following toki pona utterances have the following meanings? Not that they necessarily must, but that they can. If you think that the usage is unlikely but not impossible, that’s also valuable information.
Toki Pona
English translation
How I derived it under my analysis using the constructions in the document; spoilered to avoid affecting judgment
mi moku e pan
I butter the bread
mi moku0 e pan, I apply food to the slice of bread
mi moku e kasi
I cook the plant
mi moku0c1 e pan, I cause: the plant is food
mi moku e linja
I eat the spaghetti
mi moku2 e linja
mi moku e pan e kasi
I butter the bread and cook the plant
mi moku0/moku0c1 e pan e kasi, I apply food to the bread and (I cause: the plant is food)
mi moku e pan e linja
I butter the bread and eat the spaghetti
mi moku0/moku2 e pan e linja, I apply food to the bread and I eat the spaghetti
mi moku e kasi e linja
I cook the plant and eat the spaghetti
mi moku0c1/moku2 e kasi e linja, (I cause: the plant is food) and I eat the spaghetti
mi moku e pan is possible but very unlikely to mean I butter the bread (thought process: pan is already moku, there is no sense in causing the pan to become moku)
mi moku e kasi feels fine to mean I cook the plant under the given context. Without knowing the kasi is not yet edible I would feel the same about this sentence as I do about mi moku e pan: possible, but not a likely interpretation.
mi moku e linja feels fine to mean I eat the spaghetti.
mi moku e pan e kasi I would interpret as I eat the bread and cook the plant because applying moku to pan here is a non-change + it makes no sense to eat the kasi as it stands.
mi moku e pan e linja I would interpret as I eat the bread and the spaghetti for the same reason.
mi moku e kasi e linja I would interpret as I cook the plant and eat the spaghetti, but honestly, I might just fail to interpret that sentence entirely. Same idea as with mi moku e pan e kasi as above.
The existence of other words that I would preferentially use for cooking and buttering (seli and pana respectively) makes me less likely to interpret mokucorrectly as intended by the author across all these phrases, I think.
also, moku is a transitive verb, so it can’t be zero derived to mean “turn X into food” in most speaker’s usage. toki pona has lexical nouns, transitive verbs, and adjectives, and a very clearcut zero derivation system. some words, like lukin, have two lexical entries, in the case of lukin a noun (eye) and a transitive verb (to see).
so to answer your question: I would say none of those sentences can have those meanings.
I don’t think toki pona’s zero derivation system is clearcut. Here is an example from lipu pu that goes the other direction, turning a lexical noun into a transitive verb and introducing new meaning (p61):
jan Sili li pilin pona, li uta e jan Mawijo
Sili is happy and kisses Mawijo
The noun->transitive verb derivation is clearly valid despite the fact that uta’s proposed pu meaning is NOUN: mouth, lips, oral cavity, jaw.
What is stopping us from deriving a new transitive verb from the derived noun? As per the analysis framework we are using, there is no issue. It’s just confusing, not invalid.
The analysis framework I’m using is linguistic descriptivism. Using uta as a verb for “to kiss” is completely in line with the zero derivation framework.
The reason why we can’t just make up things is because we’re describing language usage for a living language. Your new derivation is hypothetical and is not attested among speakers.
“olin” doesn’t have a noun definition, but according to lipu pu, lexical transitive verbs used as nouns take on the meaning of the act itself OR the direct object of the verb, so according to lipu pu, olin means “loved person” or “loved object” as a noun.
So you can use toki pona the way you’re suggesting and nobody will stop you (or they shouldn’t; that would be prescriptivism), but they WILL misunderstand you. Of that I have no doubt
First of all <fanboying> I have been reading your semantic spaces dictionary with great interest and have a very similar perspective on how answers ought to look like for “what does [word] mean?” </fanboying>
Ok, with that out of the way:
That’s exactly the sort of thing I was trying to test for, yes! Because I was building up a theory of polysemy as I was learning, but I wanted to confirm that there wouldn’t be some simpler (smaller lexicon) explanation available.
I wasn’t deriving it in that direction, I am going from the noun “food” to the intransitive verb “to be food” (as in the classic example of how “mi moku” can both mean “I eat” and “I am food”), and then using the derivation from @shamu’s document which says that for an intransitive verb Y1 you can treat X li Y1 e Z as “X causes: Z li Y1”.
Is this something I can read about in pu or ku or some well-accepted community resource? (navigating prescriptivism and descriptivism is always a bit strange for me with others’ conlangs)
Even if not, is there somewhere I can read about it? I am especially interested in the “clearcut zero derivation system”, but also just in a resource that will tell me as a learner which lexical entries are generally considered to exist for which words.
(pu has clearcut descriptions of zero derivation from all three lexical parts of speech to the other two positions in a sentence that matches up with modern usage)
I’ve been meaning to write up about how zero derivation works in toki pona, but pu actually goes over it. I don’t have my copy of pu on hand so I can’t give you pointers yet but I will when I get home.
Also: I do not think that there are ANY intransitive verbs in toki pona. This is kind of related to zero copula in the sense that any part of speech can be an intransitive predicate without its meaning changing at all. This is not an intransitive verb meaning of moku, it’s just the noun meaning of moku. “e” is what makes a word take a transitive verb position.
In some cases, even without e, verbs are transitive. if “mi moku” means “I eat,” it actually IS transitive, semantically and lexically, it just doesn’t have an object right now.
By the way, on the "navigating between descriptivism and prescriptivism" topic, I want to push back a little bit on some of your argumentation.
Any theory of a language must necessarily apply some sort of generalization beyond its corpus, otherwise it will predict that the only grammatical sentences are those which have already been spoken. In other words, we must make things up, because speakers will. Our theories need to include generalizations.
Which is exactly what happens here. Surely a similar analysis applies to any lexical transitive verb used as a noun, whether or not that usage has been seen yet? For example to transitive nimi sin, which a descriptivist theory needs to be able to account for.
In general, unless we are being prescriptivist, we can never speak about how a language can/will/does behave, only about how it has behaved or how it can/will/does behave according to some theory.
And even if we are being prescriptivist (say, if we are the ones defining the language), all we can do is to say that our theory is the proper one, it's not like that will stop people from taking the language and making it their own.
But that gets messy with a language which is both living and constructed. When does the construction end? Who is allowed to prescribe, and to what extent?
(as a side note, it's at least the case that for engelangs with objectively judgeable goals, we can say that if someone's theory, even the creator's, results in a language that doesn't meet the goals, then either the language itself is currently failing to meet its goals or the theory is wrong)
Post it when available and let's apply it to the situation at hand then -- I flipped through lipu pu and the olin/uta derivation examples were the most relevant I could find.
If you can explain the difference between "mi moku" (I am food) and "mi moku" (I eat) then this sounds like a pretty good theory to me, cf. the thing I said about finding the moku0c1 thing messy.
However, I must object to this definition of "transitive" which does not require a verb to take any objects in order for the sentence to be grammatical! I don't necessarily object to the theory, but if you call the verb "transitive" you need to postulate how it gets its object when there is none explicitly provided. This is how transitive verbs behave in loglang spaces, and it leads to the problem of underfilling and I swear, if underfilling is haunting me all the way into toki pona...
More from lipu pu on derivations, straight from the horse's mouth:
[Preposition->Verb derivation] "A sentence can also use a preposition without a regular verb [...]" (p35)
[Verb derivation] "Every sentence has a 'main verb'. This is the verb or another word used in that position, such as an adjective, noun or preposition. The main verb is normally the word that comes after the particle li [...] Some pre-verbs have a slightly different meaning when they are used as an adjective or verb. For example, compare the many meanings of kama and lukin. To know all the meanings of a word, consult the official dictionary [...]" (p48-49)
[Adjective derivation] "When another word is added to [the end of] a noun phrase, it describes the sum of all previous words" (p44)
[Noun derivation] "You can convert any verb into a noun: 'toki: VERB to speak / NOUN something you speak [...] the act of speaking'" (p26)
[Adjective derivation] "You can also use any verb as an adjective. For example, jan sona is a person who knows, a knowledgeable person" (p26)
[Adjective derivation] "A second noun can also fill the role of the adjective. This is similar to using the word 'of' in English: lipu soweli / 'book of animals'" (p26)
[Possessive derivation] "They [mi/sina] can be used as adjectives to give information about a noun" (p23)
The wording on the noun->verb derivations ("... To know all the meanings of a word, consult the official dictionary...") makes it sound like the semantics were left specifically vague in order to construct specific definitions in the future.
Do you know whether, according to pu, deriving a noun from a verb can refer to its subject? "someone who speaks" from "toki". If not, what, if anything, does pu have to say about deriving nouns from intransitive verbs?
the exact wording on p26 suggests "no, you cannot refer to the subject, only the object" to me. See for yourself:
"You can convert any verb into a noun:
toki: VERB to speak / NOUN something that you speak, i.e. a language; the act of speaking, i.e. speech
moku: VERB to eat / NOUN something that you eat, i.e. food; the act of eating"
It looks to me that your choices for verb->noun derivation are "the direct object of the verb" or "the abstraction of the verb".
okay y'all i'm at a bus stop with two giant bags of vegetables, sheet music, and a practice chanter (for bagpipes) so i don't have that many free fingers but i'll try my best to respond to everything lmao lemme back read
i guess in the very strange case that you consider bread indelible without butter, "mi moku e pan" could mean "i butter the bread" (i make the bread into food).. same with the kasi being inedible without being cooked.. and "mi moku e linja" is definitely valid for "i eat the spaghetti", but i definitely would be confused by the others (probably even in context)
