o utala ala e jan kepeken lipu pu!
1 Like
Context is everything, I would expect a real conversation to proceed just about as I wrote. I always like to link lipu tenpo for more examples as they do a really good job of making sure their text is fully contextualized + readable for a beginner.
Feels a little strange to me but I am not sure why. To many people, prepositions in the subject is nasin ike but personally I think it's OK as it's disambiguated by li. (Or, I suppose you could interpret that as lon as a content word instead of a preposition in which case the meaning of the sentence is a little off: it's something more like "food existing is great". I think I'd avoid this construct because of that.)
1 Like
Very useful feedback, thanks :D for the record I was going for "eating (existing / being A Thing) is good", with the "noun referring to act of a verb" derivation and "lon" as a content word. With this as context, do you still think it sounds off? How about "kama moku li pona" with intended meaning "the arrival of eating is good"/"thank goodness I finally got to eat"? Being able to refer to events makes a lot of sense in my head, so I would be willing to accept at least a little bit of sounds-weird-ness, but I want to know exactly how much I am causing so I can judge the trade-off. From the perspective of "how weird will it be", the fact that your initial reaction was to see it as a preposition worries me, especially if people for whom that is ike will see it as ungrammatical before they see it the way I intended.
EDIT:
mi lukin 
1 Like
It [lon moku as a subject] would definitely not be ungrammatical. All prepositions can and do work as content words regularly. Personally would interpret lon moku as "existence/realness of (the) food".
I don't think there's an in-sentence way of creating event-abstractions. An idiomatic approach I'd suggest is
ni li pona tawa mi · mi moku ~ This is good to me / I like this: I eat/ate/...
(for reference, I am using middot as a sentence separator, where others might use any of dot, colon, comma, or whatever else)
another option is using la (but of course not an unambiguous one):
mi moku la pilin li kama pona ~ Due to my eating, [my] feeling became good.
1 Like
mi pilin e ike :/
Thanks for the strategies :D
New question: if you want to add more "li" to a sentence that began with a pronoun, can you do it like "ona li X li Y"? to split it apart from "ona X li Y". I know one can repeat it, but imagine the desired set of sentences is, say, {"ona li X e A", "ona li Y e A", "ona li Z e A"} where A has a big stack of modifiers. For example "ona li esun li wan li tu e kulupu esun pi esun mani". Would it be better to rephrase to not use a pronoun in that case? Can you move "kulupu esun pi esun mani" to the topic, pro-drop style? Or maybe say "ona esun e kulupu esun pi esun mani. ona wan e kulupu. ona tu e kulupu"? Are there idiomatic strategies I am overlooking?
"li" is only dropped if the subject is exactly "mi" or "sina". "ona", "mi tu", "sina kin" etc. all take li.
I think "mi X li Y" / "sina X li Y" are pretty commonly accepted as grammatical, but not universally. I seem to remember pu suggesting to split the sentence into "mi X. mi Y."
In my personal nasin li applies to the entire VP, so {"ona li X e Z", "ona li Y e Z"} can't be shortened to "ona li X li Y e Z"; it would have to be "ona li X e Z li Y e Z". Though others may follow different conventions, I'm not sure.
So putting all of this together, to avoid repeating A three times I would probably say:
"ona li esun e kulupu esun pi esun mani li wan e ona li tu e ona"
Moving A into the topic would also work for me, though it'd be confusing in this context because ona has two different referents. Maybe works if you change the subject ona to "(jan) ni" or something.
(Note: I have not been active in the community, so this may differ from community usage)
1 Like
Oops, yes, I forgor, I had "mi/sina/ona" as "pronouns" in my head but that is syntactically wrong yes.
That's also my impression but I want to avoid them due to the ambiguity with modified mi/sina.
Hmm, I'll add this to my mental list of "other grammars than the one I'm currently working with" and be on the lookout for situations where it would make a difference to the meaning so I can determine to what degree people in general do or don't do that; thanks :D
That reassignment of "ona" within a single sentence is boggling my brain! How common is that, if you know?
Hmm so that means the pro-drop part is acceptable, right? And if you say that "ona has two different referents" that implies that you are not required to then drop the pronoun, right? IOW that you can put something in the topic and then use "ona" to refer to it? I'm asking because there's a background/foreground theory of topichood that I admittedly don't understand very well but that I think would imply that if you wanted to refer directly to something, you wouldn't have put it in the topic, and if that theory does not apply to toki pona, then I can save myself from having to actually learn how it works 
1 Like
How common is that, if you know?
Sadly I don't, it honestly might just be me being weird haha.
IOW that you can put something in the topic and then use "ona" to refer to it?
Yeah, I would repeat the objects in this case (either as "ona" or "kulupu") to force the interpretation of esun/wan/tu as transitive verbs.
Though these are exactly the parts where I'm not sure how the community feels on this. Among the options available, I would definitely prefer the first one I listed, or using multiple sentences. I don't think using la to front core arguments that way is very common, though I would understand it.
1 Like
Ahh ofc
"This grammar works for me in practice but is not necessarily universal" is very valuable information, thank you :D I gotta keep in mind and keep reminding myself that I need to make two grammars, one for my own nasin, and one that allows me to (ideally) understand everyone else
Another question: my questions to help me decide what my phonology should be are mostly answered, but there is one un-remarked-upon aspect I would like some feedback on: /o/ as [ɒ̝] before coda /n/. Think of the vowel RP/GA has for "oral" and "forum" (but actually a little more open in my case, I think, but it's hard to distinguish). I have it for (English) "Sonja" and "long", and I wonder if it's worth the effort to eradicate from "jan Sonja", "lon", "kon", or "monsuta".
They are all pronouns!
My opinion is that this works like my example with propositions from nasin toki. Basically it is ambiguous which verbs the direct object belongs to. I would say at least the last verb immediately preceding e, at most every single one in the chain of li.
Seems absolutely standard to me.
Just adding on, you don't need the forced transitivity for it to be able to mean what you want.
As long as it's consistently different from how you say <a>, all should be fine.
1 Like
Yeah but "mi" and "sina" have different syntax, so they have something in common with each other that they don't have in common with "ona"!
My current nasin is based on yours in that respect but with one difference (adding your nasin to my "other people's grammar" model!): if you consider your example of disambiguating "jan li pali li pakala e tomo e ilo lon ma kepeken luka" as "jan li pali e tomo lon ma li pakala e ilo kepeken luka", I think that disambiguation works better if the "e ilo" can only be of "li pakala" and not of "li pali". But I don't have enough experience yet to figure out if that sometimes does need to happen, so my nasin's grammar could go either way for now.
nooo that means it's even harder than I thought to track reference/referent relationships 
Well, it shouldn't just depend on me, but also on how other people will hear it. Like, I can distinguish 6 different vowels between [i] and [a], but to pick which ones to use I need to know how likely it will be for toki pona listeners to map them to toki pona /i/, /e/, or /a/. As another example, my european-portuguese-speaking boyfriend has a vowel somewhere around [ɜ] which I can't distinuish[1] from danish /e/ nor /ɛ/, with my brain mapping it without any pattern I have been able to identify. And I don't want to cause that for toki pona listeners but I don't know enough about cross-linguistic phonology and/or the 5 vowel system to know how high the risk is of other people hearing [ɒ̝] as toki pona /a/.
EDIT: speaking of english phonology, I think the existence of the cot-caught merger is actually all the evidence I need that [ɒ̝] is not a safe realization of toki pona /o/! Wow, this pushed me all the way from "maybe I'll go to the effort of avoiding this, maybe not" to not just saying "I will not do this", but even having an opinion of whether that's a safe thing to do in general (if this is considered to cross the boundary into rudely judging other people's nasin, I will delete that part, but for the record my opinion is not that it's bad, only that it is risky)
danish doesn't really have any central vowels and I've been able to get by without it in english ↩︎
How common are syllabic readings vs nasin sitelen kalama style moraic readings of otherwise unpronouncable (that is, clearly not single-phoneme spelling) name cartouches in sitelen pona? Specifically, if I write myself as "jan [mama nanpa]", how likely is that to be understood as "jan Manan" vs "jan Mana"? I am still very much undecided on how I want to tokiponize my nickname, but I am leaning heavily towards "jan Mana", and while I generally would love to be written as "jan [mama nasa]", would I be able to, say, refer to my mind as "kon [mama nanpa]" on occasion without it being misread?
EDIT: also, something I don't understand about nasin sitelen kalama, from the wiki page:
If the word starts in a valid mora (i.e. a single vowel)
But isn't a CV pair also a valid mora? The wiki page says that "jan [pona]" isn't valid, but shouldn't that be "jan Po"? Why would you need to spell "jan Pona" as "jan [pona · ·]"?
I'm not sure if the standard developed since I last interacted regularly with the toki-pona-sphere, but I would expect an unmarked cartouche to always refer to the first letter of the word only. (Like, maybe the person wants to go by initials or something?)
nasin sitelen kalama was specifically made to be unambiguous about what exactly is spoken, so (at least for the form I know it in) you would have:
- unmarked sitelen: first letter
- any amount of middle dots: that many initial morae
- colon: the entire word
Yes, but the quote is about initial vowels only, so I'm unsure what is unclear to you.
Edit: I seem to have misremembered how middle dots interact with words starting in a vowel. And I think the Wiki article is worded wrongly. It should read "If the initial letter is a valid mora, ..."
1 Like
"i.e." is restrictive, short for latin "id est" aka "that is"; so the quote says "if the word starts in a valid mora (that is, a single vowel)". And in the context of the page, it is contrasting that to a word like "pona".
That's the other part of the reason for my confusion, the first part being that the wiki contrasts "[anpa]" as beginning with a valid mora in contrast to "[pona]". Your analysis does not do so, but still mixes letters and mora.
EDIT: and in doing so, contradicts the wiki page, which says that "[anpa ·]" is two mora "a-n".
Doesn't that rule out being able to include individual letters? Since under that interpretation, you can't really speak "jan [kili pan] li pona", as it would be /jan kp li pona/. If we can only include morae, and only in the sequence they occur in toki pona words (which prevents us from having two coda-n morae in a row), we can at least only break pu phonotactics (IIRC from when I read the names section yesterday, pu does not ban "nn" and "nm") by having multiple vowels in a row.
(Do you see my edits?)
1 Like
Why would it? The system does specifically not make it impossible to spell phonotactically invalid words.
1 Like
ah, didn't see the very last paragraph
That... works, and I can understand it like this. I think a system like that is good for spelling, but [→]
[←] it means the system can do more than write down valid toki pona words, but at the same time cannot write down everything that is spoken (which I don't disagree with per se; an alphabet does the same and I am fine with being able to spell "dshfsfsdfsd" in english)
After all, least under a somewhat rigid idea of phonotactics, "wuwojiti" is no more or less a valid toki pona word than "lknmlknmnkln" or even "kablæpø".
I suppose it's a phonemic writing system for toki pona, then 
EDIT: in retrospect, the difference between "write down what is spoken" meaning "write down what was / has been spoken" (whether or not it conforms to toki pona phonotactics) and "write down what is to be spoken" (can you expect a toki pona speaker to speak words that break its phonotactics?) is what threw me for a loop.
This design is also so it's backwards-compatible with the pu system, which also allows for phonotactically invalid spellings.
1 Like
Ahhh, that makes total sense. Thanks for helping me through my confusion about this :D
1 Like
toki pona is HEAVILY based on contextual intuition and seemingly prefers to be simple at face value but more complex upon further inspection. however, as i like to sometimes say to people: sina pilin nasa tan toki la sina wile sona e kon toki ≈ “if you’re confused, then you’ll ask what it means.” tokiponists are very forgiving about misunderstandings and are more than willing to explain at length what they mean when the need arises
however, i will say ona changing referents if it does usually only happens like at most once in the average sentence? most times ona refers to just one referent. unfortunately, toki pona does not invest in variables ahah, but again: you can always ask for clarification, just like with any other language
1 Like