Draft one of toki pona lesson for twitter

Language learning: How to speak Toki Pona, translation problems, advice, memory aids, tools and methods to learn Toki Pona and other languages faster
Lingva lernado: Kiel paroli Tokiponon, tradukproblemoj, konsiloj, memoraj helpiloj, iloj kaj metodoj por pli rapide lerni Tokiponon kaj aliajn lingvojn
janMato
Posts: 1545
Joined: Wed Dec 02, 2009 12:21 pm
Location: Takoma Park, MD
Contact:

Draft one of toki pona lesson for twitter

Post by janMato »

I was aiming for 365 lessons of 140 character or less, I'm close.

Any comments before I schedule these for daily release?

1234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890
#tokipona Grammar. li means the verb phrase is starting. Can also mean predicat is starting. jan li suli. The man is big.
#tokipona Grammar. e means the direct object phrase is starting. Dropping "e" is a common beginner's mistake.
#tokipona Grammar. It isn't clear if you can omit an "e phrase" for transitive verbs. People do it anyhow. mi moku (e moku).
#tokipona Grammar. Modifiers can modify modifiers. ilo musi sike li laso mute.
#tokipona Grammar. kin and taso can be injected almost anywhere. Only these words can do that.
#tokipona Grammar. toki pona is an analytic, isolating language. Its all phrases and 1 morpheme per word.
#tokipona Grammar. li, e, pi, la, en, anu, taso, kin(?) are strictly function words. ss
#tokipona Grammar. lon, sama, poka, kepeken, tawa, tan are function words and prepositions.
#tokipona Grammar. All content words generally can, and are used as adjectives, adverbs, nouns or verbs.
#tokipona Grammar. Modal verbs precede the head verb. jan li ken pali. The man can work.
#tokipona Grammar. ken, kama, tawa, awen, lukin can be used as modal verbs.
#tokipona Grammar. kama takes unmarked complements as if it were a preposition, but it goes in the verb slot.
#tokipona Grammar. Inherent possession rules aer unclear. mi pona e luka mi vs mi pona e luka.
#tokipona Grammar. Multiple direct objects are joined by "e". mi moku e moku e telo. Not mi moku e telo en moku.
#tokipona Grammar. en in a li or e phrase implies mixture. mi moku e telo en moku. I ate food mixed with water.
#tokipona Grammar. en in a li or e phrase implies mixture. jan li lape en kalama. He talked in his sleep.
#tokipona Grammar. pi means many things. Sometimes it means "of", sometimes "about"
#tokipona Grammar. pi can mean possesion but is omited for 1 word posessors. tomo mi. my house. tomo pi jan suli. big guy's house.
#tokipona Grammar. pi can act like a generic or univeral relationship operator. P1 pi P2 = P1 is *somehow* related to P2.
#tokipona Grammar. Despite the rule that pi must be followed by 2 words, this is legal: N1 pi N2 pi N3 N4

#tokipona Grammar. pi can act like parenthesis, ie. group. mi moku e (moku pi (seli mute.))
#tokipona Grammar. pi can act like genative, ie. group. jan pakala e ilo sona pi jan Mato. Matt's computer.
#tokipona Grammar. A kitten dies every time you use "jan li pi ma lete" to mean "jan li tan ma lete" He's from the N/S pole.
#tokipona Grammar. mi and sina are different because they refer to the speaker and listener.
#tokipona Grammar. mi, sina, ona can safely take wan, tu, mute, meli, mije as modifiers. mi tu li pilin pona.
#tokipona Grammar. mi, sina, ona can take other modifiers, but is less common. jan li moku mute. tenpo kama la ona jan li suli.
#tokipona Grammar. "transitive nouns" mean transformation. moku mute li jan suli e mi. Feasts make me fat.
#tokipona Grammar. X la Y means "Y if X", and sometimes it means simutaneous events
#tokipona Grammar. anu, en, taso, la, ante are logic words, but toki pona isn't strict about logic.
#tokipona Grammar. bare mi or sina, must not be followed by li. BUT mi mute li lukin. sina mute li lukin.
#tokipona Grammar. The last pi must be followed by two words.
#tokipona Grammar. There are only six prepositions, lon, sama, poka, kepeken, tawa, tan
#tokipona Grammar. kepeken is strongly related to "instrumentality" and doesn't always match up with English "with"
#tokipona Grammar. tan indicates causation, and is more commonly found in la phrases, tan ni la mi wile lape.
#tokipona Grammar. Modifiers (adverbs and adjectives) follow the modified. jan suli li tawa wawa e sike laso.
#tokipona Grammar. In a predicate sentence, tan, tawa can indicate motion. mi tawa tomo pali. mi tan tomo pali.
#tokipona Grammar. sike, sewi, sipin, monsi, etc are not prepositions, so must follow lon. kasi laso li lon sike tomo mi.
#tokipona Grammar. Few people rigorously follow the part-of-speech rules listed in the classic word list.
#tokipona Grammar. There are three or four formal grammars of toki pona, none written by jan Sonja.
#tokipona Grammar. The complement of prepositonal phrases can be joined by en. Or maybe not. The rules aren't clear.
#tokipona Grammar. o is a vocative after a name. jan Mato o!
#tokipona Grammar. o is imperative before verb phrase. o moku! o moku o pona e sike sina!
#tokipona Grammar. o hortantive/optative before a complete sentence. o mi mute li lape. Let's sleep!
#tokipona Grammar. noun + verb can behave like participles. waso tawa = running bird, ostrich.
#tokipona Grammar. verb + noun can behave like participles. kama sona = getting knowlege, learning
#tokipona Grammar. Possessors follow the possessed. Any third word after pi might be a possesor or adjective.
#tokipona Grammar. Possessors follow the possessed. mi tawa tomo mi. soweli li lape lon supa ona.
#tokipona Grammar. ona can co-ordinate with anything but an entire sentence. Use sparingly.
#tokipona Grammar. X li pona tawa Y means Y likes X. olin means romantic attraction and the like.
#tokipona Grammar. X li lon Y e Z menas X puts Z on Y. Not everyone uses or likes this construction.
#tokipona Grammar. X li tawa Y e Z means X moves Z to Y. Not everyone uses or likes this construction.
#tokipona Grammar. X li tawa Y. This is a predicate with implied motion. English doesn't really work like this.
#tokipona Grammar. X li Y. This is a simple predicate meaning equivallents or subclass. X is Y. or X is a Y.
#tokipona Grammar. X li Y. In simple predicates, Y can be a noun phrase or an adjective. jan li suli.
#tokipona Grammar. X li M e Y. This means X makes Y have the qualities of M (a modifier). jan li laso e tomo.
#tokipona Grammar. X li N e Y. This means X makes Y into N (a noun). jan li soweli tomo e soweli. Man domesticates it.
#tokipona Grammar. ala negates the preceding. There is no clear guidance on double negation, most people single negate.
#tokipona Grammar. ala negates the preceding. jan li pala ala. jan li ken ala pali.
#tokipona Grammar. Subjects are always required. ijo/jan/ona are sometimes used as placholders for unimportant subjects.
#tokipona Grammar. Verb phrases additional verb phrases start with li. Not en
#tokipona Grammar. Proper modifiers must follow a noun. nimi mi li nimi Mato.
#tokipona Grammar. Most words have a "home" part of speech, but move around in with predictable meaning shifts.
#tokipona Grammar. Mutliple subjects can be joined with en, anu, taso, ante. meli en mije li musi.
#tokipona Grammar. Time is expressed in tenpo X la phrases, e.g. tenpo ni la, tenpo kama la, tenpo pini la

#tokipona Grammar. Tense is always expressed by phrases. Past: tenpo pini la jan li pali. He worked.
#tokipona Grammar. Tense is always expressed by phrases. Future: tenpo kama la jan li pali. He will work.
#tokipona Grammar. Tense is always expressed by phrases. Present: tenpo ni la jan li pali. He is working.
#tokipona Grammar. Without time phrases, a verb can mean any tense. jan li pali = He (is/will) work(ed)
#tokipona Grammar. Without number modifiers, pronouns can mean any number. sina = you, you two, y'all

#tokipona Grammar. Some adverbs are expressed in la phrases. ken la = maybe. ike la = unfortunately.
#tokipona Grammar. la fragments, e.g. kin la, mute la, open la, have been controversial, but are used everywhere.
#tokipona Grammar. Manner is expressed in prepositional phrases at the end of li phrases. mi tawa sama waso.
#tokipona Grammar. Manner is also expressed in adverbs after verbs. jan li lape mute.
#tokipona Grammar. Place is expressed in a lon phrase at the end of a li-phrase. jan li pali lon tomo.
#tokipona Grammar. Reflexives and recipricals use "sama". meli en mije li lukin e sama.
#tokipona Grammar. Questions must have an X ala X, seme or "anu seme?" construction to be a question.

#tokipona Grammar. Causitives usually involve kama. jan lawa li kama ni: jan pali li pali mute.
#tokipona Grammar. Some verbs take unmarked complements. mi kama sona. Parsing these is controversial.
#tokipona Grammar. Prepositional phrases can't go in subject slot.
#tokipona Grammar. The subject stays the same for all li phrases in a sentence. jan li tawa li pali li pini li lape.
#tokipona Grammar. Yes/No questions must have an X ala X. sina wile ala wile e suwi? OR they use "anu seme?"
#tokipona Grammar. Yes/No questions must have an S anu seme?. sina wile anu seme? OR they use "X ala X?"

#tokipona Grammar. Yes/No questions are answered "Yes" by repeating the sentence. sina wile ala wile? mi wile.
#tokipona Grammar. Yes/No questions are answered "No" by wile ala. sina wile ala wile? ala.
#tokipona Grammar. Other quesions replace a word with seme. sina wile e seme? What do you want?
#tokipona Grammar. Alternatives require "anu seme?" sina wile e telo e moku anu seme?

#tokipona Grammar. Join the arguments of a prep. phrase like this: tomo li lon nena en nasin.

#tokipona Punctuation. The "e ni:" phrase indicates two linked sentences. The sentences are always independent.
#tokipona Punctuation. Only proper modifiers are capitalized. Start of sentence is never capitalized.
#tokipona Punctuation. Sentences must end in a period or colon.
#tokipona Punctuation. Titles can be sentence fragments. Otherwise, you must avoid sentence fragments. No rules for them.
#tokipona Punctuation. Direct speech goes in quotes. jan li toki e ni: "ale li pona"

#tokipona Vocab. Toki pona has somewhere between 118 and 140 words depending on which words you count. 124/125 are in common use.

#tokipona Vocab suwi, pan, kili, moku, telo, namako are foods. Candy, bread, food, drink, spice.
#tokipona Vocab ni means this and that, can refer to a thing or sentence.
#tokipona Vocab jan means any person or human-like creature. jan can be of either gender, maybe includes homonids & neanderthals.

#tokipona Vocab meli means woman
#tokipona Vocab mama means parent. It can refer to the father or mother.
#tokipona Vocab palisa is something like a stick.
#tokipona Vocab lipu is someting like a sheet.
#tokipona Vocab supa is a flat horizontal surface
#tokipona Vocab linja is someting like a wavy line, but not necessarily straight!
#tokipona Vocab supa, lipu, palisa, linja, sike are generic shape words.
#tokipona Vocab jaki means dirty. ike is similar but doesn't automatically mean dirty or disgusting.
#tokipona Vocab kute means "to hear", but people use it to mean "ear" anyhow.
#tokipona Vocab ike means bad, and sometimes "complex" Sometimes the meaning depends on who's talking.
#tokipona Vocab ilo means tool.
#tokipona Vocab jelo means yellow, loje means red, laso is blue/green. Each color covers a broad band of colors.
#tokipona Vocab pimeja and walo mean black and white. kule means color. Not sure if pimeje or walo are kule.
#tokipona Vocab sike is a circle, but also a sphere or a round point.
#tokipona Vocab lon can be true, or exists. It does not work like the English, "is" or "to be"
#tokipona Vocab kala means fish, but also anything living in the water.
#tokipona Vocab soweli means animal, but particularly furry, pleasant ones.
#tokipona Vocab waso means bird, but also anything living thing that flies.
#tokipona Vocab akesi means animal, but paricularly lizards and unpleasant ones.
#tokipona Vocab moli means kill, dead, death, but has no simple opposite word.
#tokipona Vocab sinpin means face. But also front. meli pi lukin pona li jo e sinpin pona.
#tokipona Vocab lawa means head. But also leadership. tenpo ale la jan li pana e mani tawa jan lawa.
#tokipona Vocab noka means foot and/or leg. It may become obsolete, but people still use it.
#tokipona Vocab anpa means lower part, down but also defeat. It isn't a prep. soweli li lon anpa supa.
#tokipona vocab sewi means upper part, up, but also holy/religious. It isn't a prep. waso li tawa lon sewi kon walo.
#tokipona Vocab weka means to be missing or to get rid of, or drop. jan li weka e linja lawa tan lawa.
#tokipona Vocab wile means want and need. Must is difficult to express.
#tokipona Vocab pona and ike mean good/bad. Sometimes this depends on whose talking. Sometimes not.
#tokipona Vocab pona and ike mean good/bad. Sometimes the meaning is by convention. jan pona=friend, even henchmen.
#tokipona Vocab jan sama, jan lili, mama mije, mama meli are siblings, children, dad and mom.
#tokipona Vocab unpa means sex, but also fuck. The community really does treat it as taboo.
#tokipona Vocab pali means work, but also is a generic verb for "to do (something)"
#tokipona Vocab pana menas give (gifts), but also is a generic verb for "to give (something)"
#tokipona Vocab kon means air, but also magic.
#tokipoan Vocab kon, ko, kiwen make up a set classifiers, gassy, gooey, rocky.
#tokipona Vocab tawa means go, move, to. It is a noun, verb, and preposition.
#tokipona Vocab kama means come, arrive, but also "to become". It's complement is unmarked. mi kama jan sona.
#tokipona Vocab nimi means word and name. nimi mi li nimi jan Mato. mi jan Mato. But not "nimi mi li jan Mato."
#tokipona Vocab kulupu is a group, collection or set of things.
#tokipona Vocab jan, ma, toki, kulupu, esun, ilo are the most common words to take proper modifiers.
#tokipona Vocab wawa means power and fast. waso li tawa wawa. tomo tawa mi li wile e wawa.
#tokipona Vocab lupa means hole- used for windows, mouths, holes in the ground
#tokipona Vocab ante means different, change, other. mi kama tan ma ante.
#tokipona Vocab ante is also a conjunction, "otherwise, else" mi wile e moku ante lape.
#tokipona Vocab anu means or. It overlays the li and e. mi pali anu lape. mi moku e kili anu kasi.
#tokipona Vocab en means en. en (usually) can't overly li or e. mi pali li lape. meli en mije li pali.
#tokipona Vocab esun means a store. It also gets used for all sorts of commerce related phrases.
#tokipona Vocab ijo means "thing", and maybe living thing, too.
#tokipona Grammar ijo might be used in a passive construction. ijo li moli e mije. The man was killed.
#tokipona Vocab awen means keep, stay, but also continue(!) as a model. tomo mani li awen e kiwen suno.
#tokipona Vocab lete means cold and seli means hot. lete also means polar and seli, equatorial. e.g. ma seli, ma lete.
#tokipona Vocab You can eat kili (fruit), pan (bread), kasi (plants), moku. & Carnevores eat soweli, akesi, waso, kala
#tokipona Vocab mani is money. mani, esun are the commerce words. li jo is to own, to have.
#tokipona Vocab suno and mun are sun and moon. laso suli means blue sky. stars are mun lili.
#tokipona Vocab telo is water. telo suli is the ocean. telo nasa is booze.
#tokipona Vocab nasa is crazy, but also medicinal or narcotic. ijo nasa = drugs
#tokipona Vocab utala is fighitng and moli is death. They lack single word antonyms in toki pona.
#tokipona Vocab toki is talking. nimi are words. kalama is just noise. sitelen are pictures, but also writing
#tokipona Vocab tenpo is time. Be careful to not spell it with a "m" tenpo kama la ale li pona.
#tokopona Vocab sin is new or news. It lacks a single word antonym.
#tokipona Vocab when lili, suli, pona, ike are used as modifiers, the meaning often can't be guessed. waso pona = chicken(!)
#tokipona Vocab mi-me, sina- you, ona-he/she/it, ni-this/that, jan-one is the basic pronoun system.
#tokipona Vocab tomo is a people-sized container(car or building), poki is a smaller container

#tokipona Tools Toki pona can be colorized using this webpage. (TODO: add bit.ly link)
#tokipona Tools You can get a reasonable approximation of your name in toki pona here. (TODO: add bit.ly link)
#tokipona Tools Toki pona can be converted to and from a variety of scripts on this webpage. (TODO: add bit.ly link)
#tokipona Tools A primative gloss can be generated from any text on these pages. (TODO: add bit.ly link)
#tokipona Tools You can search the entire tp corpus for usages of words or phrases on this page. (TODO: add bit.ly link)
#tokipona Tools To find tp users to interact with, you can use chat, forums, blogs listed here (TODO: add bit.ly link)
#tokipona Tools jan Pije's lessons have been formated in book format. Other books are listed here. (TODO: add bit.ly link)

#tokipona Vocab jelo, laso, loje, pimeja, walo are yellow, blue, red, black and white respectively.
#tokipona Vocab ala, wan, tu are exactly numbers, 0, 1, 2. suli, lili, mute are aprox. measures.
#tokipona Vocab ala means not, nothing, no.
#tokipona Vocab ale and ali mean everything. Both spellings are valid and in use.
#tokipona Vocab Do not confuse ala, ale, ali! ala is "no", "nothing". ale/ali is everything!
#tokipona Vocab mu and a are words describing noises made by animals and humans respectively
#tokipona Vocab pu, as of 2010, is a reserved, undefined word.

#tokipona Vocab monsuta means fear. mi monsuta e akesi. mi mute li wile monsuta e monsuta taso.
#tokipona vocab namako means spice and extra. pona kin li namako lon moku. nimi mute li namako.
#tokipona Vocab kipisi means cut. sina wile e pali la o kipisi e linja lawa.
#tokipona Vocab alasa means hunt or gather. jan pi tenpo kiwen li alasa tawa moku.

#tokipona Vocab There are 100s of canonical country and language names, generally requiring memorization.

#tokipona Semantics. If you create a bahuvrihi in toki pona, often it isn't clear which word goes first.
#tokipona Semantics. You can't create noun-noun compound words in toki pona. The 2nd word must be a modifier.
#tokipona Semantics. Toki pona splits up the world coarsely.
#tokipona Semantics. In the right context, the entire lexicon shifts. In a coffee shop, telo is coffee.
#tokipona Semantics. Use kepeken to describe the language you speak. mi toki kepeken toki Inli.
#tokipona Semantics. There isn't a clear culture or conculture in tp. Meaning depends on *your* situation.
#tokipona Semantics. pakala! is the cuss word. Use unpa! or nena suli! or jan sewi! if pakala doesn't work for you.
#tokipona Semantics. Phrases must be grammatical. You can't build arbitrary compound words.
#tokipona Semantics. Proper modifiers should only refer to people, places, brands, companies. Not loan words.
#tokipona Semantics. Reduplication has no established meaning. jan jan li pali pali mute mute = ? ? ? ? ?
#tokipona Semantics. Some modifiers act like derivative affixes. jan -> jan lili, child, diminutive, but not always.
#tokipona Semantics. Some modifiers act like derivative affixes. jan -> jan suli, adult, augmentive, but not always.
#tokipona Semantics. Headless noun phrases are formally illegal. Modifiers must describe something.
#tokipona Semantics. ala officially can behave like "un-" and mean polar opposite. It's still uncommon though.
#tokipona Semantics. Almost everyone chooses a name headed by jan, meli, or mije.
#tokipona Semantics. There are 3 official number systems and many more proposals. None are used frequently.
#tokipona Semantics. There are several proposed spelling systems. None are used frequently.
#tokipona Semantics. There are dozens of writing systems. Outside latin only abbreviation systems get much usage.
#tokipona Semantics. Ordinary modifiers can be shuffled. n m1 m2 = n m2 m1 Sometimes shuffling makes a phrase look wierd.
#tokipona Semantics. Sometimes moving mute closer to the head phrase is clearer. eg. n1 pi n2 m mute vs n1 mute pi n2 m

#tokipona Mistakes. Some mistakes make tokipona unreadable (reversing modifier order)
#tokipona Mistakes. Some mistakes don't affect readability. e.g. proper modifiers missing their head.

#tokipona Phonetics. There's no guidance on intonation or sandhi. Maybe anything goes.
janKipo
Posts: 3064
Joined: Fri Oct 09, 2009 2:20 pm

Re: Draft one of toki pona lesson for twitter

Post by janKipo »

Nice to see this getting done so fast. The longer chapters on pckipo.blogspot.com are taking longer than I hoped.
Then predicate/verb phrase distinction doesn't make sense -- they are the same thing, unless you get uptight about where the words come from, which is unimportant in tp -- at the grammatical level.
Of course you can omit the 'e DO' for a transitive verb -- it merely increases the ambiguity slightly.
Modifiers can modify modifiers, but the example isn't a case of that happening. you want something like kili pi loje suno li suli pini, where 'suno' modifies 'loje' which modifies 'kili'.
Can we really talk about morphemes on twitter?
'en, anu, taso, kin' are not strictly functional -- they all have significant content. They do not, however, move around to a lot of the possible roles (yet).
The next six aren't function words at all, but they are prepositions occasionally at least (lon, tan and tawa almost always).
Generally can go anywhere, but most don't go everywhere,
Modals take a whole verb phrase as complement, what follows a modal (with its frills) is indeed the head verb of that phrase.
'wile, open and pini' at least are also modals.
'kama'' is -- as you noted -- a modal, it takes a VP complement, not an NP one, unlike prepositions
Possession is not assumed grammatically; a floating 'lukin' can be anyone's (context decides, as the saying goes)
'pi' is merely a mark that the following two (at leaast) words are to be taken together as a unit modifier of whatever is to the left. What to make of that depends upon the words and the context. Personal words in that spot tend to be possessive, other to be just complex modifiers. I suppose there are cases where "about " is a good reading.
'pi' requires at least two words after it, since otherwise there is not distinctive grouping.
N1 pi N2 pi N3 N4 means that the block N3N4 modifies N2 before the whole modifies N1, that is, it is a straightforward case of the right grouping rule.
Grouping is all 'pi' does grammatically; what it means is something else again and depends on context.
Nice threat for kitten lovers. 'pi' has to have a head of some sort to its left.
I suppose that 'mi, sina' are different, but because they don't take 'li', not because they happen to refer to speaker and hearer (although the two may be related).
Any word serving as a head noun can take the number words as modifiers, nothing special about the pronouns.
Ditto for other modifiers. I don't get the point here.
What is a transitive noun? Your example is an ordinary causative derived regularly from an ordinary noun.
'la' means "under the precedingly mentioned condition" which translates sometimes as "when", sometimes as "lf" and sometimes something else altogether.
Not sure what a logic word is, These are connectives, though not sentential ones, and they are some of them at least used in a variety of ways (not enoiugh data in the case of 'ante').
Don't make final lists of any role here, since expansion is always possible (and taking place with the 'lon x' forms) and kepeken is for now somewhat different from the others as a verb.
'kepeken' means "with the help of, as a tool", not "accompanying", a very different sense of "with" (poka)
'tan' means "from a source: and so get used for where you are coming from, what your reasons for doing something are, what the cause of something is, and other analogous concepts.
add 'anpa, poka' to the list of things that need 'lon'. 'sinpin'
The POS in the "official list" (there are at least three claiming to be this) are meant to be suggestive and are regularly violated on the same page.
Any structure can be conjoined to a similar one by 'en' (etc.), the only interesting question is what it might mean.
Possession is a type of modification.
'ona' coordinates with a previous noun phrase (which one depends on context). It is not used to repeat sentences.
Out of time. More tomorrow. BTW most of this list is semantics, not grammar.
janKipo
Posts: 3064
Joined: Fri Oct 09, 2009 2:20 pm

Re: Draft one of toki pona lesson for twitter

Post by janKipo »

Numbers would have helped, even if they use up space.
'olin' is only between jan
don't forget X li tan Y e Z, X takes Z away from Y.
X li Y is a simple predication, but it means anything that that can mean, which is much more than equivalence and subclasses -- differentiation, for example.
X li N e Y can also mean X applies N to Y (with a rather broad interpretation of "apply")
on 'ala', it is not clear what the scope of the negation is 'jan li ken pali ala' can't work ? or can loaf?
using 'ona' for a dummy subject is frowned upon, even though it is etymologically justified.
You always need to say "in complete sentences' (where the predicate is also always required), but a lot of sentences in conversation are not complete.
'lape mute' is quantity, not manner
Yes-no questions use the 'ala' construction, wh-questions use 'seme'; 'anu seme' falls in between: you can say yes or no or supply a correction.
causative: kama E ni (note, there is nothing special here, it is a causative from an intransitive.
'lama sona' is just a modal with a VP complement; what is controversial?
PP in S slot" What does this mean" Since a PP is by definition not an NP and S by definition is, why mention. Or do you mean that an NP in the S slot cannot have a PP modifier? That is just not true.
Make up your mind about what a y-n q must have. can be either. The answer to an X ala X question is X for yes, ala for no -- or as much more of the sentence as you want.
seme questions are answered by an appropriate replacement for 'seme' and as much more of the sentence as you want.
'anu seme' is tricky, Strictly, if you want what is offered, a yes or some sort will do, otherwise you should offer an alternative -- including no for nothing.
independent in the sense of being separate sentences, but, as you note, linked in some way.
Or a question mark or an exclamation point.
Sentence fragments are OK if thee is enough context to fill in the gaps -- answering questions, for example, or picking up a pattern mi tan ma tomo SanLuwi. sina?
which reminds me that you can ask questions just by an upward tilt of the voice at the end of the sentence.
Indirect speech uses the 'ni:" trick.
But suwi is an adjective, moku a verb, pan, kili and namako nouns (though I am not sure about the last one) and, of course, mean a whole lot of other things/
can refer when use as a noun, but it is also an adjective (and mainly that). It can maybe be modified to get a better range of deixis 'ni weka' "that yonder"
jan definitely includes all hominids, indeed all great apes. It also includes other things if they are dramatis personae in a tale.
meli means female, species indifferent.
mama means parent in any sense of the word and moves on to familial and beyond and ancestral as well.
the stickiness of palisa is long, thin, pointed.
say how like in each case : lipu flat and floppy, supa flat, hard and horizontal, linja - one dimensional
jaki means dirty, disjgusting, ike means bad, evil, complex, unhappy. pona is the opposite of both.
Who uses 'kute for ear? 'nena kute' or 'lupa kute'
jelo also contains a lot of greens.
soweli officially land mammals except herd domestic animals, often used without the quantifier or, as you say, for cute fury ones. also used for animals in general (or in kindomal).
akesi are reptiles and amphibians and any kind of yucky, spooky thing like that
sinpin is also wall
weka means afar, basically, the rest is derivative
tp does NOT treat 'unpa' as tabu and does not use it in the colloqial sense of "fuck" (see 'pakala')
pana refers to any action outward from the subject, shooting for example.
Quotes help when talkinga bout words and being careful (as, I notice, I am not).
'en' means "and"
esun really means exchange, but that's pedantry since the rest is right.
ijo li moli e jan. Is not passive, merely indefinite. and not a good idea generally
mani is also those herd animals, technically
but do remember to *pronounce* 'tenpo' with an /m/.
meanings of modified-modifier structures are contextual - at lunch, waso pona may be chicken at the hunt a pheasant and gambling a pigeon
jan and ni are not strictly pronouns, but can be pressed into service as needed (like any thing else, come to that).
toki is a constructed object, defined by its being built, poki is defined by its holding things. A poki can be as big as a house (the pokey, perhaps), a tomo can be a nanobot.
Thanks for the tools. Have I mentioned thanks for all these notes, too. They really help.
You're more confident about what monsuta means than I am, though I guess we have the right area.
ditto namako
in your tools, you might think to keep lists of names, places, etc. and reference them I think my list was up to date when the old list closed, but there have been lots since. maybe a wiki, since it would be work to keep up to date.
You basically can't create a bahuvrihi (do you really want to bring that up?) in tp len lawa loje is going to be a hat, not a person with one on.
But in noun-noun compounds, the second word IS a modifier, by definition. What did you have in mind?
Bad advice on swearing. To be sure, other expressions than 'pakala' will come along, but hopefully not just calques. Big bump?
How do you do something ungrammatical in tp? It is easy to be meaningless, but the only ungrammatical things are shifting the slots around.
Whatever comes first is the head, so headless noun phrases (and verb phrases) are impossible -- though you may not get what you want. And don't forget short answers.
Well,. the Latin alphabet is pretty generally used.
And sometimes the shuffling is just wrong.
aikidave
Posts: 106
Joined: Thu Oct 01, 2009 12:07 pm
Location: Boston, MA
Contact:

Re: Draft one of toki pona lesson for twitter

Post by aikidave »

Great idea. Great start.
It is going to take me some time to wade thru this, but here are my comments so far:

> #tokipona Grammar. toki pona is an analytic, isolating language. Its all phrases and 1 morpheme per word.

Will the general reader know what these words mean? What do you mean by all phrases? Should you introduce the sentence template: condition la subject li verb e direct object optional prepositional phrase ?

> #tokipona Grammar. li, e, pi, la, en, anu, taso, kin(?) are strictly function words. ss

I'm confused as to what you mean by function words. I think of li, e, pi, la as separators

> #tokipona Grammar. kama takes unmarked complements as if it were a preposition, but it goes in the verb slot.

What are unmarked complements? Will typical Twitter reader understand this?

> #tokipona Grammar. en in a li or e phrase implies mixture. jan li lape en kalama. He talked in his sleep.

I never knew this about en. Thanks.

> #tokipona Grammar. mi, sina, ona can take other modifiers, but is less common. jan li moku mute. tenpo kama la ona jan li suli.

Why would I say ona jan - human person?

> #tokipona Grammar. In a predicate sentence, tan, tawa can indicate motion. mi tawa tomo pali. mi tan tomo pali.

What do you mean by predicate sentence? Maybe when tan or tawa are used as a verb?

> #tokipona Grammar. Few people rigorously follow the part-of-speech rules listed in the classic word list.

maybe add that most TP words can be used in any POS?

> #tokipona Grammar. There are three or four formal grammars of toki pona, none written by jan Sonja.

What are the 4 formal grammars? May be too much info for Twitter?

#tokipona Grammar. o hortantive/optative before a complete sentence. o mi mute li lape. Let's sleep!

Making Twitter users use the dictionary a lot, aren't we? Sorry, I'm an engineer not a linguist!

> #tokipona Grammar. noun + verb can behave like participles. waso tawa = running bird, ostrich.

I know tawa defaults to a preposition, but when I see it after a noun, I think of tawa as an adjective: waso tawa - moving bird

> #tokipona Grammar. X li Y. This is a simple predicate meaning equivallents or subclass. X is Y. or X is a Y.

Maybe mention zero copula here?

More when I get some more time.
janMato
Posts: 1545
Joined: Wed Dec 02, 2009 12:21 pm
Location: Takoma Park, MD
Contact:

Re: Draft one of toki pona lesson for twitter

Post by janMato »

Re: templates
Yes, it's in there. Good idea to provide multiple versions of it.

Re: linguistic jargon
I'm pro-conventional linguistics jargon. But I agree, I should include some tweets defining the key definitions. (For example, I think the word "describer" found in the wiki is less useful than "adjective")

By toki pona is all phrases, I mean, to contrast with languages that inflect, e.g. run, ran, running, amo, amas, amat, amorum, etc. I'm not even sure what language that amo stuff is. Spanish I think. The alternate to inflections is phrases. To oversimplify, languages have 3 ways to express an idea-- make up a word for it, glue a chunk of sounds to another word, string a bunch of words together. So thats lexical, morphological and periphrastic solutions to problems. toki pona more radically than any other language relies on the last solution for solving all challenges.

Re: intro with template
In twitter world, there really isn't a first message, just tweets close to each other. I can expect a twitter reader to have read the last 1 or 2 messages, but not necessarily the whole stream. Frankly, it is a bizarre medium to write for. After I finish proofing, then I wire these tweets up to a scheduler and they get tweeted once or twice a day until they've all been broadcast.

Re: function words.
Function words are semantically bleached, tend to be a closed lexical class, tend to not stand on their own, e.g. "the", "in", "for", "how", "do" in English questions, etc. They contrast with content words, which are semantically rich, e.g. "fish", "equivocate", "fragile". UPDATE: D'oh I wrote this late at night and wrote the 2nd part backwards

li, e, pi and la, in normal linguistic jargon are either particles or clitics. The decision about if it is one or the other usually hinges on diagnostic tests, like can you insert a word in between the potential case ending/particle thingy or not. I haven't been clever enough to come up with many diagnostic tests. I know li is not a verbal affix because modals can be inserted in between li and the verb. e and pi always attach to the head of the noun phrase and you can't insert anything in between these and the noun they attach to. Prove me wrong! :-) Then I can call them particles. Otherwise, e and pi look like clitics.

mi lukin e jan. -> mi lukin e .... jan. No matter what I put in the blank, it's just not the same sentence, not in the way that the following is the the same thing.
mi lukin e jan -> mi lukin e jan lili. <-- The head of the DO stays the same.

re: What are unmarked complements?
mi tawa tomo ante.
mi kama tomo mi.

jan Kipo is analyzing these as modals + intransitive verbs. I'm not entirely convinced, I think either tawa/kama are verbs that are taking an unmarked complement OR they are predicates of motion (to use Paynes jargon). And from time to time we see people write stuff that sort of looks okay with out the "e", such as

? mi toki ilo.
I'm talking about computers

If this was a ordinary human language, it would have moving parts and jan Kipo (or me) would be able to say, "hey, here are the six characteristics of intransitives and this set of phrases meets all of them" or "rephrase this as a question and it become obvious that x, y, and z", but toki pona is so limited, there's not a lot of obvious diagnostic tests.

>Will typical Twitter reader understand this?
Dunno the underlying problem is actually hard. One the harder chapters in the linguistics text book I'm reading is on valence, i.e. how do languages cope with verbs that semantically seem to need more than just two participants (subject and object). In toki pona, we have managed, correctly or incorrectly to put tertiary things in all these slots with X's:

X1 la S (en X2) li V X3 (pi X4) e DO Prep X5 (another Prop X6 ... etc)

Why things end up in the X slots as opposed to the S or DO slots, is a hard question.

>> #tokipona Grammar. en in a li or e phrase implies mixture. jan li lape en kalama. He talked in his sleep.
>I never knew this about en. Thanks.

I don't know how canonical it is, but it makes sense. It makes even more sense with adjectives. moku pi loje en laso.

>Why would I say ona jan - human person?
Exactly, it means something like s/he, but not "he/she/it (that animal over there)". I would use something like that, but thats just because I like modified pronouns. I can't just drop them altogether like they do in Chinese, so might as well make them useful.

> #tokipona Grammar. In a predicate sentence, tan, tawa can indicate motion. mi tawa tomo pali. mi tan tomo pali.
What do you mean by predicate sentence? Maybe when tan or tawa are used as a verb?

"My father is a dentist" i.e. my father is an example of the class of people known as dentists.
"Pork is the other white meat" A=B. These can be reversed without changing meaning.
"The other white meat is pork" B=A.
There are more subcategories. Like I said earlier, jan Kipo prefers to analyze these as intransitive verbs, I don't buy it yet. For example, in a equative predicate sentence, the arguments can be reversed and it means the same thing. But this can't happen in intransitives

mi kon. I breathe.
? kon mi. Air is me.

>maybe add that most TP words can be used in any POS?
Good point.

>What are the 4 formal grammars? May be too much info for Twitter?
jan Kipo's, morpheme addict's, the one on wikipedia which has been through multiple iterations. I wanted people to know they exists. I probably should try to work in what a formal grammar is.

>>#tokipona Grammar. o hortantive/optative
>Making Twitter users use the dictionary a lot, aren't we? Sorry, I'm an engineer not a linguist!

Yeah, you're right, that is one is a bit over the line. Function vs content is a valuable framework, imho, but commands and demands are just commands and demands.

>> #tokipona Grammar. noun + verb can behave like participles. waso tawa = running bird, ostrich.
>I know tawa defaults to a preposition, but when I see it after a noun, I think of tawa as an adjective: waso tawa - moving bird

Participles are adjectives made out of verbs. I'll rewrite it to say adjective since the distinction between an ordinary adjective and a participle isn't important.

> #tokipona Grammar. X li Y. This is a simple predicate meaning equivallents or subclass. X is Y. or X is a Y.
Maybe mention zero copula here?

On this issue, I agree with jan Kipo. toki pona doesn't have a zero copula. It's a piece of misinformation that is in the wikipedia article. Russian has a zero copula. (And if you don't like imaginary things, the other word for it is juxtaposition)

ya pilot (russian). I'm a pilot. The phrase works by setting the two next to each other.

toki pona on the other hand, has an invariant particle. (invariant to time, person, number and all other know semantic categories and anyone could imagine)

Thanks for the proof reading!
janKipo
Posts: 3064
Joined: Fri Oct 09, 2009 2:20 pm

Re: Draft one of toki pona lesson for twitter

Post by janKipo »

janMato wrote:Re: templates
By toki pona is all phrases, I mean, to contrast with languages that inflect, e.g. run, ran, running, amo, amas, amat, amorum, etc. I'm not even sure what language that amo stuff is. Spanish I think. The alternate to inflections is phrases. To oversimplify, languages have 3 ways to express an idea-- make up a word for it, glue a chunk of sounds to another word, string a bunch of words together. So thats lexical, morphological and periphrastic solutions to problems. toki pona more radically than any other language relies on the last solution for solving all challenges.
Latin not Spanish and what is 'amorum'?
Re: function words.
Function words are semantically bleached, tend to be a closed lexical class, tend to not stand on their own, e.g. "the", "in", "for", "how", "do" in English questions, etc. They contrast with function words, which are semantically rich, e.g. "fish", "equivocate", "fragile".
Contrast with NON-function (alias content) words. And, with the possible exception of "do", all the words you list for English are content. I sometimes think you think "function" means "prepositions and conjunctions" (which is largely NOT what it means).
li, e, pi and la, in normal linguistic jargon are either particles or clitics. The decision about if it is one or the other usually hinges on diagnostic tests, like can you insert a word in between the potential case ending/particle thingy or not. I haven't been clever enough to come up with many diagnostic tests. I know li is not a verbal affix because modals can be inserted in between li and the verb. e and pi always attach to the head of the noun phrase and you can't insert anything in between these and the noun they attach to. Prove me wrong! :-) Then I can call them particles. Otherwise, e and pi look like clitics.

mi lukin e jan. -> mi lukin e .... jan. No matter what I put in the blank, it's just not the same sentence, not in the way that the following is the the same thing.
mi lukin e jan -> mi lukin e jan lili. <-- The head of the DO stays the same.
Clitics are not separate words and 'li' and 'e' and 'pi' are. If you insert anything anywhere, you have a new sentence, so this is not much of a test. It turns out to be subjective on what is a different sentence or some such thing. Note that inserting a modal (which is a verb after all) after 'li' changes the sentence too, so it is not a separate case. Actually, what you call them doesn't matter much, so long as you use them correctly.
re: What are unmarked complements?
mi tawa tomo ante.
mi kama tomo mi.

jan Kipo is analyzing these as modals + intransitive verbs. I'm not entirely convinced, I think either tawa/kama are verbs that are taking an unmarked complement OR they are predicates of motion (to use Paynes jargon). And from time to time we see people write stuff that sort of looks okay with out the "e", such as

? mi toki ilo.
I'm talking about computers

If this was a ordinary human language, it would have moving parts and jan Kipo (or me) would be able to say, "hey, here are the six characteristics of intransitives and this set of phrases meets all of them" or "rephrase this as a question and it become obvious that x, y, and z", but toki pona is so limited, there's not a lot of obvious diagnostic tests.
'kama' and 'tawa' are in different classes. 'tawa' is a Preposition and takes a NP complement for the goal, the place going to. 'kama' does not take such a complement, but needs a 'tawa' PP to express this notion. On the other hand, 'kama' is a Modal and so takes a VP complement, which may also be an NP at another level of analysis, for what the subject becomes. So the example above is most naturally read as "I become my home" and less easily as a verb" adverb combination "I arrive my-homely", whatever that may mean.
As for 'mi toki ilo', that is just one of many attempts to find a way to say "talk about" and probably not the best, since it naturally means "I talk mechanically", with the aid of a voice box, say.
>Will typical Twitter reader understand this?
Dunno the underlying problem is actually hard. One the harder chapters in the linguistics text book I'm reading is on valence, i.e. how do languages cope with verbs that semantically seem to need more than just two participants (subject and object). In toki pona, we have managed, correctly or incorrectly to put tertiary things in all these slots with X's:

X1 la S (en X2) li V X3 (pi X4) e DO Prep X5 (another Prop X6 ... etc)

Why things end up in the X slots as opposed to the S or DO slots, is a hard question.
What do these Xs mean? X1 can be an NP or a sentence or a modifier or Lord knows what, X2 has to be an NP, X3 can be just about anything, depending on what V is, X4 is a modifier, which may be an NP at a different level; X5 has to be a NP and similarly for the rest of the Xs. In short, I don't see a level of analysis where this pattern fits in.
>> #tokipona Grammar. en in a li or e phrase implies mixture. jan li lape en kalama. He talked in his sleep.
>I never knew this about en. Thanks.

I don't know how canonical it is, but it makes sense. It makes even more sense with adjectives. moku pi loje en laso.
The official line is that you can't have 'en' in verbs, but that is clearly wrong, since the added 'li's give the wrong reading. and similarly for DOs. It is too bad we don't have the same distinction for other noun phrases (but English gets by without it).

>Why would I say ona jan - human person?
Exactly, it means something like s/he, but not "he/she/it (that animal over there)". I would use something like that, but thats just because I like modified pronouns. I can't just drop them altogether like they do in Chinese, so might as well make them useful.
But they are rarely needed and generally look odd. Go with 'ona' alone unless confusion is sure to follow (or the wrong reading).
#tokipona Grammar. In a predicate sentence, tan, tawa can indicate motion. mi tawa tomo pali. mi tan tomo pali.
What do you mean by predicate sentence? Maybe when tan or tawa are used as a verb?

"My father is a dentist" i.e. my father is an example of the class of people known as dentists.
"Pork is the other white meat" A=B. These can be reversed without changing meaning.
"The other white meat is pork" B=A.
There are more subcategories. Like I said earlier, jan Kipo prefers to analyze these as intransitive verbs, I don't buy it yet. For example, in a equative predicate sentence, the arguments can be reversed and it means the same thing. But this can't happen in intransitives

mi kon. I breathe.
? kon mi. Air is me.
Well, "A dentist is my father" means something quite different from "My father is a dentist". And, similarly, if 'mi kon' means "I am air" then "Air is me" probably means something different, too. In saying they are intransitive verbs I mean simply that they occupy the verb slot and don't take a direct object. At another level of analysis, some are Nouns, some Adjectives, some verbs and so on. And these facts may underlie further moves that can be made logically with them. But these logical additions are not yet grammatically significant.

>>#tokipona Grammar. o hortantive/optative
>Making Twitter users use the dictionary a lot, aren't we? Sorry, I'm an engineer not a linguist!

Yeah, you're right, that is one is a bit over the line. Function vs content is a valuable framework, imho, but commands and demands are just commands and demands.
And wishes and encouragements and so on.
#tokipona Grammar. X li Y. This is a simple predicate meaning equivallents or subclass. X is Y. or X is a Y.
Maybe mention zero copula here?

On this issue, I agree with jan Kipo. toki pona doesn't have a zero copula. It's a piece of misinformation that is in the wikipedia article. Russian has a zero copula. (And if you don't like imaginary things, the other word for it is juxtaposition)

ya pilot (russian). I'm a pilot. The phrase works by setting the two next to each other.

toki pona on the other hand, has an invariant particle. (invariant to time, person, number and all other know semantic categories and anyone could imagine)
I'm afraid I just don't get the distinction here unless it is a contrastive analysis, that is, Russian has a perfectly good copula but just doesn't have to use it, using juxtaposition instead, whereas tp has no such contrast.
janMato
Posts: 1545
Joined: Wed Dec 02, 2009 12:21 pm
Location: Takoma Park, MD
Contact:

Re: Draft one of toki pona lesson for twitter

Post by janMato »

janKipo wrote:Contrast with NON-function (alias content) words. And, with the possible exception of "do", all the words you list for English are content. I sometimes think you think "function" means "prepositions and conjunctions" (which is largely NOT what it means).
Oops! That was I typo, fixed in original. li, pi, e, la are clearly function words and none of them are prepositions or conjunctions. For the typology to work, only one of the words needs to be non-controversially defined. Function words are everything that isn't a content word. Quantitatively speaking, the function words tend to also be the ones that are the most common and in tp, they are so common they don't match the curve of the rest of the words. There is something going on with function words vs content words, even if I don't have an airtight definition of them yet.
If you insert anything anywhere, you have a new sentence, so this is not much of a test. It turns out to be subjective on what is a different sentence or some such thing. Note that inserting a modal (which is a verb after all) after 'li' changes the sentence too, so it is not a separate case. Actually, what you call them doesn't matter much, so long as you use them correctly.
Well, I'm not arguing in favor of saying "ni li X" and "ni li X Y" are the same strings. I, too, can see that they differ by two characters in the 8th position. If X is a noun and Y is a modifier, then these two sentence are going to be describing the same semantic stage. ni li kili. ni li kili jelo. These describe the same scenario in a way that ni li kili vs ni li jelo kili doesn't.

Re: clitics/cases/particles
I'm just an amateur field linguist, not a theoretician. I'm sure in another 100 years, the taxonomy and diagnostic tests will improve. In a hundred years, the best reference grammars will look as bizarre as the reference grammars from the 1600 which imagined that all languages, including Amerind, were essentially latin and that a reference grammar of a foreign language was a matter of plugging the appropriate foreign phrases into the verb and noun charts for Latin.

X1 la S (en X2) li V X3 (pi X4) e DO Prep X5 (another Prop X6 ... etc)
Why things end up in the X slots as opposed to the S or DO slots, is a hard question.
What do these Xs mean? X1 can be an NP or a sentence or a modifier or Lord knows what, X2 has to be an NP, X3 can be just about anything, depending on what V is, X4 is a modifier, which may be an NP at a different level; X5 has to be a NP and similarly for the rest of the Xs. In short, I don't see a level of analysis where this pattern fits in.[/quote]

The valence chapter of a reference grammar. Particularly when the number of things on the stage goes beyond a subject and a object. A good field linguists is supposed to check for these features, or at least find out how these common scenarios are dealt with.

I was trying to say that in tp, when we have valence above 2, then people, rightly or wrongly start putting the other words into all the places marked X. e.g. mi toki e toki Lasin. moku la mi wile e pan sike. And so on.

Valence of 1
---------------
Intransitives -
Reflexives - subject is the same person/thing as the object
Passives (subject omitted)
Omitted Direct objects

Valence of 2
---------------
Transitives - the subject is doing something to the object
Reciprocals - what the subject is doing to the object, the object is also doing to the subject
Applicatives (oblique promoted to direct object and direct object is omitted)

Valence of 3
---------------
Obliques & prepositional phrases & datives and the world of cases and the like.
Double direct objects. An oblique is promoted to direct object and the other direct object remains. English has these. I gave him the book.

Valence of 3+
---------------
Lojban style valence (e.g. verb foo has 5 arguments)
C# style method names, e.g. writeToScreen(int boxStyle, int Color, int Width, int Height)


>Well, "A dentist is my father" means something quite different from "My father is a dentist".
Yes, that isn't an test of subclass predicates, only equative ones. Here is a test:

1) mama mije li jan pali pi uta pona. <--- possible subclass predicate
2) mama mije en jan ante li jan pali pi uta pona.
3) jan pali pi uta pona li mama mije en jan ante.

If we asked a subject, "do 2 & 3 describe different scenarios?" They'd say no. Obviously we are not blind and don't think that the strings are the same, they differ in the first place. The above operation cannot be performed on an intransitive verb.

1) mi kon. I breathe. <-- possible predicate of some sort.
2) mi en jan ante li kon.
3) kon li mi en jan ante.
Not only is 2 and 3 likely to elicit an "No, these are not describing the same scenario", #3 looks like nonsense, where as #2 is merely describing breathing. And yeah, it can parse as "I am gas", but that is nonsense and can be safely discarded from the available options.

>And, similarly, if 'mi kon' means "I am air" then "Air is me" probably means something different, too.

I don't think so. All I can think of is topic fronting, and there is no particular reason to believe that topicality is indicated by fronting the topic.
I'm afraid I just don't get the distinction here unless it is a contrastive analysis, that is, Russian has a perfectly good copula but just doesn't have to use it, using juxtaposition instead, whereas tp has no such contrast.
Russian doesn't have a present tense copula and saying "yest" or "yavlyaetcsya" (exists, shows itself) means something different.

? ya yest pilot. I'm obviously not fluent in Russian, but this just sounds wrong.
ya pilot.
sir yest? Is there any cheese? Does cheese exist?
? yest sir. This still sounds wrong.
da. vot cir.

In past tense, Russian does have a copula and the copula inflects just like any other past tense verb. So the contrast is that while Russian either has ordinary juxtoposition in present tense, or an verb that varies in past and future, toki pona has an invariant particle. While the taxonomy for clitic/case/particle is fuzzy even in my own mind, the distinctions among the type of predicates is clear to me.

ya bil pilot. I was a pilot.
ona bila pilotka. She was a pilot.
janKipo
Posts: 3064
Joined: Fri Oct 09, 2009 2:20 pm

Re: Draft one of toki pona lesson for twitter

Post by janKipo »

none of the tp words you listed was a conjunction or preposition (though you have listed 'en' sometimes) but the English list was almost entirely of that sort plus the contentful "the," The distinction is fairly easy to work out in toki pona, practically impossible in English (which has been argued to have no function words) the difference between function words and content words is that function words only have functions, content words have meaning (content) as well. And that doesn't get us much forwarder. but then, it is not clear what one is to do with this distinction once one has it.

I wasn't talking about their being different strings, but about their describing different situations: eating a yellow fruit is eating a fruit but eating a fruit is not the same as eating a yellow fruit. And, of course, interchanging modifier and modified makes something new, what is the point of all this? or of distinguishing clitics and particles and whatever else. does it apply to toki pona? If so, does it make a difference?

Do all reference grammars have valence chapters? None of mine do. But, clearing away the verbal miasma here, there is an interesting point -- one we meet with frequently: how do you juggle a number of "objects" when you have only a limited number of slots? How do you say what we are talking about, what language we are talking in (solved finally), who gets the gift (given early on) and so on. So, typically we are looking at NPs and, in toki pona, these can go just about anywhere. So, supposing that the subject and object and the verb are set, we get conditions, modifiers on subject or verb or object, and PPs as possible places -- oh, and NP complements to a verb. The valence article looks pretty superficial to me, being only about the surface appearance of the sentences, not about anything significant outside of pragmatic considerations. Notice that tp treats reflexives as bivalent, not mono and, of course has no passives (which in English are just like omitted direct objects, omitted, not not involved at all). Typically, of course, reciprocals involve a compound subject and a reciprocal pronoun object, so are pretty much like reflexives which are given as monovalent. I'd like to see what might be meant by Applicatives; this seems to be on a whole different grammatical level from the other things and so seems misleading to include here. Obliques we know how to handle and, in English certainly, there are no double direct objects, "him' in the example is an INdirect object and so a case of the first type of trivalence. So, back to the question of what do do with some of these floating problems. I don't see that this discussion has helped.

How do you tell what test is decisive? The bit about my father the dentist fails the equivative test, so? So it is not an equivalence. And the difference that makes is exactly what? You then say that samples 2 and 3, both about my father and another guy and dentists, are the same, whereas they clearly are not. one is about class membership, the other (to be as generous as possible) is about identity: that's all the dentists there are. (And I expect most people would say Yes, they are different, too), But the point seems to be that that there are subclass verbs (alias, for the most part, Nouns) and intransitive verbs and they behave differently. So far, I haven't seen a case that stands up -- nor can I see the significance for anything we are about here. The air case of course is strange because it starts out by reinterpreting a Noun and then expects that reinterpretation to continue throughout, even when it makes very little sense: "Breath is me and other people" is poetic as Hell, but doesn't fit in with the other stuff, since it takes 'kon' back to being a noun again. If we leave 'kon' be, then 'mi kon' and 'kon li mi' just fail like the dentist case.

OK. So Russian never use the present tense copula (so how do we know it is yest?) and so we can say either that Russiona omit the present tense copula or that have a 0 for the present tense copula. But either way we can get this only by contrast with something. In tp there is nothing to contrast as far as copulas go.

If the distinction among the types of predicates is clear, can you explain it better? And also give some motivation for it?
janMato
Posts: 1545
Joined: Wed Dec 02, 2009 12:21 pm
Location: Takoma Park, MD
Contact:

Re: Draft one of toki pona lesson for twitter

Post by janMato »

janKipo wrote:none of the tp words you listed was a conjunction or preposition (though you have listed 'en' sometimes) but the English list was almost entirely of that sort plus the contentful "the," The distinction is fairly easy to work out in toki pona, practically impossible in English (which has been argued to have no function words)
will, have, had, going when used as auxillary verbs in English. Some of these have obvious diagnostic test to decide if they are words separate from "will" (desire) "Will" the person (etc), which are content words. For example, I can't contract either one.

I'll be there.
* I see'll (I see Will)
* I haven't the'll. (I haven't the will)

going, as in, "I'm going do that" is function. The test is that it is also pronounced gonna, which can't be used as content word:

* The cat was irritated by all the comings and gonnas.

The application to tp is that content words aren't as specific and can't really go into the slots for content words. When they have by decree been made content and function words-- i.e. most of the prepositions, it results in really nasty gardenpathing, where you have to backtrack, sometimes several times to cover all the possible ways to read it. The content words generally can go into about any slot-- you may end up with Colorless green dreams doing things, but it subjectively sounds grammatical in a way that "gonna most had will the in" isn't.
janKipo wrote:the difference between function words and content words is that function words only have functions, content words have meaning (content) as well. And that doesn't get us much forwarder. but then, it is not clear what one is to do with this distinction once one has it.
Anyhow, the toki pona scholar Morpheme Addict said much the same thing, so if I got a metal model built on sand then at least I got company. It's like porn, I know when I see it, but the definition eludes me.
janKipo wrote:I wasn't talking about their being different strings, but about their describing different situations: eating a yellow fruit is eating a fruit but eating a fruit is not the same as eating a yellow fruit.
The diagnostic test doesn't work when taking info away, only when adding it. The equivalence operator isn't working like the one in algebra-- I'll have to work out which operator it's like, maybe >= / <=

Also, if you can remove info, in that sense, then "eating fruit" and "eating fruit" aren't necessarily the same. In the former I omitted that it was a mango in Minnesota, and in the latter I omitted that it was a banana in Bombay.

Given that
ni li kili.

What of the following still describes the same stage?
ni li kili jelo - This can be the same situation as above, but doesn't have to be
ni li jelo kili - This is a different situation.
ni li jelo - This is a different situation (Actually, it sort of makes sense as an adnoun, this is a yellow thing, in which case it is equivalent to the above)

Given that
ni li kili jelo

What of the following still describes the same stage?
ni li kili- Yes.
ni li jelo kili - This is a different situation.
ni li jelo - This is the same situation.
And, of course, interchanging modifier and modified makes something new, what is the point of all this? or of distinguishing clitics and particles and whatever else. does it apply to toki pona? If so, does it make a difference?
I can't remember anymore. Ah, that's right, its about diagnostic tests. Without diagnostic tests, we can only go on untested intuition about our mental models, authority and other unsatisfying tests of correctness.
Do all reference grammars have valence chapters? None of mine do.
In the real world, facts tend to fall on a continuum, so sentences with universal qualifiers are (usually) wrong. Some reference grammars valence chapters, some will cover it in the section on verbs.
But, clearing away the verbal miasma here, there is an interesting point -- one we meet with frequently: how do you juggle a number of "objects" when you have only a limited number of slots? How do you say what we are talking about, what language we are talking in (solved finally), who gets the gift (given early on) and so on. So, typically we are looking at NPs and, in toki pona, these can go just about anywhere. So, supposing that the subject and object and the verb are set, we get conditions, modifiers on subject or verb or object, and PPs as possible places -- oh, and NP complements to a verb. The valence article looks pretty superficial to me, being only about the surface appearance of the sentences, not about anything significant outside of pragmatic considerations. Notice that tp treats reflexives as bivalent, not mono and, of course has no passives (which in English are just like omitted direct objects, omitted, not not involved at all).
Assuming that there aren't situations that can't be described in tp, then there has to be a mechanism for describing a stage with an unknown, missing, or unimportant actor. If there really, truly isn't a mechanism for that, i.e. all possible sentences will necessarily get a modestly competent toki pona reader to answer "Who did it?" with a "He/she/it did it." then we could say tp can't do passives. I really wish there were a pair of words for all the language features, one that tied such a feature to the technical details and one that indicated the semantic goal of that feature. It's like saying "Tp lacks past tense because you can't tack an -ed to the end of verb (or any other morpheme), therefore tp speakers are incapable of answering "Has it happened? Is it happening? Will it happen?" and maybe not even capable of understanding the question"
Typically, of course, reciprocals involve a compound subject and a reciprocal pronoun object, so are pretty much like reflexives which are given as monovalent. I'd like to see what might be meant by Applicatives; this seems to be on a whole different grammatical level from the other things and so seems misleading to include here.
Wikipedia discusses applicativesand does mention that English has a similar operation that the wikipedia article decided is a different phenomena, dative shifting. Again, like passives, it doesn't help that applicatives show up in languages where there are 15 different things technical things going on in the verb (e.g. markers for transitivie and intransitive), so the bit about an oblique being promoted to direct object may seem like a small part of the story, when, imho, that is the main part of the semantic thing going on. i.e If the means of an action is important and the target is vague or important, how do we express that in toki pona? Hundreds of people have demonstrated that something in the brain encourages people to promote the oblique to direct object in the classic beginner's sentence:

mi toki e toki pona. (toki pona is the important part of the message. When I use a language, the target is this vague complex of people, keyboards, chalkboards, words, bursts of air from my mouth, and frankly that isn't the part the lister cares about. So people are doing a type of applicative, imho.)
Obliques we know how to handle and, in English certainly, there are no double direct objects, "him' in the example is an INdirect object and so a case of the first type of trivalence.
Pinker disagrees with you with respect to double objects. He takes a chapter to beat the subject to death, I lack time and space to do so. I would guess that Pinker would disagree with the wikipedia article above, too.
So, back to the question of what do do with some of these floating problems. I don't see that this discussion has helped.
We'll if I only had my own mental model, then I would never realize how different it is from other peoples.
How do you tell what test is decisive? The bit about my father the dentist fails the equivative test, so? So it is not an equivalence. And the difference that makes is exactly what? You then say that samples 2 and 3, both about my father and another guy and dentists, are the same, whereas they clearly are not. one is about class membership, the other (to be as generous as possible) is about identity: that's all the dentists there are. (And I expect most people would say Yes, they are different, too),
If I restrict my thoughts only to concept that are easy to define, I'd have to stop thinking. The bit about dentists (dad and and possibly others are dentists. The dentists are my dad and possibly others), they could describe the same stage, where as "I am sleeping." and "Sleeping is me" can't possibly describe the same stage. At the deepest level, I'd guess this has to do with how each of the neurons and dentrites in the brain are wired together, but the explict model of that would be cumbersome to discuss. So I'm left with "mean the same thing" as an approximation.
OK. So Russian never use the present tense copula (so how do we know it is yest?)
yest is the infinitive of "to be" It's merely the closest option and imho, doesn't really work well as a present tense copula-- when one attempts to use it, it means something else, namely it is constrastive with non-existence, where as "I'm a pilot" is just presentation, there isn't an attempt to contrast this with the listeners thoughts about existence-- i.e. "I'm a pilot, and you thought I didn't even exist" vs "A pilot exists". The defective example is like trying to imagine how object incorporation would work in English. After a short description of object incorporation, you can ask someone to do the same in English and the might say:

* I pigsold at the market today.
And then if we said, "Well, how did we know the word pigsold if English lacks (productive) object incorporation"?
and so we can say either that Russiona omit the present tense copula or that have a 0 for the present tense copula.
I'm certain you've said you didn't like 0 copula's. And in this case there is an alternate model-- juxtaposition, which work fine.
But either way we can get this only by contrast with something. In tp there is nothing to contrast as far as copulas go.
Yes, lots of diagnostic test for English rely on contrasting with questions and commands. But in tp, these constructions are so similar to the indicative that not much happens. So diagnostic tests in tp have to rely on what few re-arrangements are possible.
If the distinction among the types of predicates is clear, can you explain it better? And also give some motivation for it?
I'll address the motivation, I may be an amateur, but the gist of predicates has entire chapters in college linguistics textbooks dedicated to it. I as a practical matter, I can't presume they are all fools and crackpots and rebuild the science of linguistics from first principles, much less so for each new language encountered on possibility that something as basic as predication might not apply to a new language.

The largest motivation is that I'm an amateur field linguist. Lemme pick another example that I'm working on. I'm reading up on Algonquian, which doesn't seem to have a Subject-Object system, it has Obviate and Proximate and the Subject/Object are mostly unmarked. Theoreticians can't explain all the cases as to why real people, when given a obviate-proximate system will put the semantic recipient of the action into obviate or proximate (and it's seems to be legal to do both, that is why it is not a true subject-object system)

If someone comes along and says, well, the mental model for subject-object systems are well understood, lets shoe-horn our analysis of obviate-proximate systems into that, then the model is probably less descriptive, even though the model of how a subject-object system is closer to being airtight. (Or taking predicates, which isn't my idea, it's a well attested idea in the linguistics literature, and deciding it's not airtight and choosing to use intransitives, a possibly more airtight mental model, as a way of explaining what usually is explained using predicates)

Languages need to be invented, sooner the the better and if we wait for theoreticians to come up with airtight formal models (or attempt to do it ourselves), then we could be waiting a long time and reference grammars (airtight or otherwise won't get written)
janKipo
Posts: 3064
Joined: Fri Oct 09, 2009 2:20 pm

Re: Draft one of toki pona lesson for twitter

Post by janKipo »

This will take a while. The first problem is, as noted, that several levels of analysis are treated indistinguishably and so these have to be teased apart to even begin.
For now, just this: why in the wide world would you say that 'll, gonna, 'd, 've are not content words? The test seems to be that we can contract them. But by that logic, n't is a mere function word, too. All the contractions show is that they are different (content) words from other more or less similar words, which can't be contracted. But we knew that already from the different function they perform.
Post Reply