Toki Pona Forums

I suppose we might distinguish 'len pi tawa lete' (for cooling hot hands) and 'len pi tan lete' (for toting dry ice, say)

It is new for me. So, I need to think about templets:

• "X pi tawa Y"
• "X pi tan Y"

Maybe, we should add still one templet in this set:

• "X pi lon Y"

Well, that already is common. ''tawa' and 'tan' in this way are open to some questions, but do come out of the common store of lore (just not the motion part, only the directional).

Well, that already is common. ''tawa' and 'tan' in this way are open to some questions

Without key words it is diffucult to understand what is what. You say:

'len pi tawa lete' (for cooling hot hands) and 'len pi tan lete' (for toting dry ice, say)

I see here a word "for", which is used in both cases:

• tawa (for...)
• tan (for...)

I have no answer how to distinguish them. In my understanding, "for" describes "finish, direct object", "tan" is related with "start, subject". May I ask you to give some examples with "pi tawa" and "pi tan", to catch a difference. (maybe, tomo, moku, ilo or something like this).

'for" here is about purpose, ordinarily 'tawa' but now buried in 'pi'. 'tawa' now takes on the sense of "goal", something toward which the object is directed, in this case cold. So, clothing moving toward bringing cold" or some such reading. "tan" takes the opposite reading, that from which we move away, so "clothing for removing/getting away from cold". Of course, the shorter "clothing for cold (as a goal)" and "clothing from/because of cold" (as a point of departure) makes sense, too. As I said, these are not likely to become idioms, but they are principled suggestions.

buried in 'pi

While it is difficult for me. Though, I have got a little hint from your comment in another post about "x la S" and "S lon x". It inspire me to draw a scheme with "S pi x" and "x tan S". I do not say that this is a correct scheme, only a draft for analysis.For example, a relation between "lon" and "tan". First one is about "to put in". Second one is about "to put out". Also, a relation

• "x la" (situation)
• "pi x" (base, source)

are equal, because "situation = base, source". But, to give a formal decription for a relation "pi -- tan", is hard. What do you think about the "pi -- tan" tie? I will try to find an answer. Hope, it will help a tp-community to operate with these four word in the best way.

Additionally, I see, these four words are static, immobile, in comparison with a word "tawa".

Well, just for starters, 'pi' covers just about every conceivable relation, not specifically (or even majorly) 'tan'. If there is any dominant relation, it is probably possession in the very broad sense already of classical genitives.
'lon' is not ablut putting in but about being in (or at) already, putting in is more 'tawa'. 'tan' is also less about pulling out than about being out (the motion possibly implicit in 'tan' is not exploited in tp, so your remark about staticness, at least, is to the point.)

janpona o,

according to the Official Book, "pi" links two noun phrases. Therefore "tomo pi lon lete" would read like "house of cold location" (which coincidentally seems to make sense, but you have to read "lon" as a noun and "lete" as an adjective, which may be not what you intended).

Why not simply "tomo lon lete", "house in the cold"? Nothing in the Official Book seems to prohibit this.

Maybe you want to say "house, which is in the cold", "tomo pi (ona li) lon lete", but the Official Book doesn't have such complex examples. (Even though nothing seems to prohibit these either.)

The official pu says some strange things about 'pi' of which this is the strangest (usually accompanied by a counter example on the same or the next page). 'pi' simply groups together a complex modifier (more than one word); what the structure of the modifier is is not relevant, nor, for that matter, need what goes before the 'pi' be a noun (phrase), since 'pi' is used with verb heads as well as nouns (and, indeed, adjective heads). We did try to get Sonja to modify her description to fit her examples, but without success (she -- like you -- calls whatever comes after 'pi' a noun, all evidence to the contrary notwithstanding). Fortunately, it almost never makes a difference in practice. The occasional occurrence of a universally ungrammatical item like 'tomo lon lete li' is usaully caught and corrected (prepositional phrases occur only at the end of sentences).

I very like a perfectness of digits's order (0, 1, 2... 9). In this order, I always know where I am: left-right, more-less... etc. When I am trying to define where I am in tp, I am confused: "li pi", "pi X". But, the most hard for me is: to divide a semantic field on two parts between "pi" and "tan".Though, guess, I have some progress in key words for lon: "is placed (located, planted)" or "have placed (located, planted)". For example, "sewi li lon e jan". Here "lon" means: Lord have placed Adam (in the Garden of Eden). Contrary, "sewi li tan e jan" means: Lord have expulsed Adam from Eden. I know a variant where "lon" means "create". But, for me "lon"-"tan" is an undivided pair. In my member "Lord was working ("pali", creating the World) six days and one day he was "lape". So, "sewi li pali (have created) e jan". Here, I was talking about "lon" in a verb slot. "li lon" (be located), "li lon e" (have located).

If to say about "lon" in a noun slot. "Tomo pi lon lete" -- House was located on a frozen place. Though, in the cartoon "Frozen", the house was grown from the frozen place -- "tomo pi tan lete". Of course "X pi Y" is "noun pi noun". Therefore, "pi Y" (pi lon, pi tan) make substantivation (nominalization). That is, a verb "to place" (to locate, to plant) becomes substantivated -- a noun "a placed" (a located, a planted), as the word "Frozen" is a substantivated name from a verb "to freeze".

Why not simply "tomo lon lete", "house in the cold"? Nothing in the Official Book seems to prohibit this.

Here, we have a templet "X lon Y". Now we may insert different words instead X, Y, and to compare them:

• "jan lon telo" (man at-in water) -- I imagine a situation where someone is jumping into water
• "ma Atalantis lon telo" (Atlantis at-in water) -- I see how the Atlantis goes under water
• "kasi lon ma" (a plant at-in soil) -- I have planted a flower (I have a planted flower)

And, the most significant example, for me, is "jan lon len" (a man located at-in-inside the clothes). I see in the "lon" only the substantivated (verb-adjective) name. Of course, we should analyze this templet more deep. A perfect solution I see in that: to find out key words for all "mediators": la, lon, pi, tan. If you have some ideas about keys for these words, guess, it willl be very interesting for all to know them.

'li pi' is automatically ungrammatical because 'pi' has to come between content words (the ancient use of 'pi' for possessives -- the source of the pu formulation but long out of date as the principal use -- did allow that form, but not since about 2003 (in Pije's first draft).
Your ;lon' is right, your 'tan' is probably OK but would probably be 'weka' in most idiolects. 'pali' tends to ask teh question "made out of what", so it works for the Adam story (tan ma) but not Gen 1, where the implication is that they were simply called into being, hence 'lon e'.
'lon' does mean "place" (among other things) as a noun, but 'tomo pi lon lete' would normally be read as Noun + prep phrase, and similarly with 'tan'. Well, 'lon; and 'tan' are prepositions basically, not verbs at all (they take objects without 'e') but do get substatnivized as the genus of their objects (place, source)
pu says that prepositional phrases come after the direct object, so not after the subject (and not, grammatically, after the direct object either).
'lon' and 'tan' are prepositions, 'pi' is a particle for grouping modifiers, 'la' is a sentential connective. they are very different in every respect, so a way of dealing with all of them at once seems inherently unlike to be useful or meaningful.