Noodling about 'pi'

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
Post Reply
janKipo
Posts: 3064
Joined: Fri Oct 09, 2009 2:20 pm

Noodling about 'pi'

Post by janKipo »

Some thoughts arising from trying to deal with 'pi'.
An axiom of analysis is that NPs arise from descriptions, that Noun + modifier comes from Noun li modifier, in short. The tidy way of doing this is just
'x li y, A x B' => Ax{[y]}B, where the braces mean "add 'pi' if the content is more than one word long" and the brackets sum up the nominalization of a predicate. The standard NPs arise from a series starting with 'ni li y' and proceeding with cases where A is empty and B is just 'li z', adding another modifier to the chain. The initial 'ni' (or maybe just a dummy materia prima term) is dropped at some point, probably after the first step.

The fun comes in expounding the brackets, which may require several steps but can be summed up roughly all at once using the broad general description of predication:
'(li verb (e NP (PP)0)0 (PP)0)1 (e NP (PP)0)0 (PP)0'
(items in parentheses are optional above the subscripted number of occurrences) (The complexity and repetitiveness here come from transformational considerations, but, at least at first guess, building NPs happens at the descriptive level or, at least, something in which collapse, at least, has already taken place. The short story here is that all the 'e DO's and PPs become modifications and and that all the occurrences of components beyond the first are joined by 'en'. That is, the bracketed form of the above formula becomes
'verb ({DO {(PP (en PP)0)0} (en DO {(PP (en PP)0)0})0 (en verb ({DO {(PP (en PP)0)0} ({DO {(PP (en PP)0)0})0})0 {DO {(PP (en PP)0)0} (en DO {(PP (en PP)0)0})0}.
("but counting the digits gave him such fidgets he dropped Maths and took up Divinity").

Some of the steps here can take place even before nominalization: 'verb e DO' (and presumably the cases of iterated DOs) go over optionally to verb +{DO} : 'ona li alasa e waso telo -> 'ona li alasa pi waso telo' "He hunts ducks" to "He duck-hunts" (or "is a duck hunter"). And similarly with 'verb + PP': 'ona li alasa lon tenpo lete' to 'ona li alasa pi lon tenpo lete' and even (maybe with restrictions on what prepositions are involved) 'ona li alasa pi tenpo lete'. The formula could be simplified by assuming that these optional moved obligatorily occurred before the final nominalization.

While this accounts for most occurrences of 'pi', some cases remain that need separate discussion at least. One is the verb 'p* NP', which becomes just NP in the nominalization formula, but which gives rise, in Pije's dialect and earlier tp, to 'li pi NP' in ordinary contexts. The modern reflex is more nebulous: an "appropriate noun" + {NP}, where the appropriate noun is left up to the speaker, although the head of the subject NP and 'ijo' are usually safe (the latter so long as jan are not involved, perhaps). So, Pije's iconic 'ni li pi mi' appears as 'ni li ijo mi' and 'kili li pi jan San' as 'kili li kili pi jan San' or 'kili li ijo pi jan San'.

Another is the matter of degrees of ads(-jective/-verb). Almost all of these has the potential to be taken in degrees, ranging from "not at all' ('ala') through "a little" ('lili') to "a lot" ('mute') and maybe to "totally" ('ali') and "more than before" ('sin') and "really" ('kin'). At least 'mute' and 'lili' (and probably 'sin') are themselves among the ads that can come in degrees, so the possibility of degree markers of at least two words (and then two more and so on ad infin) has to be allowed. Presumably, 'ali' and 'ala' (and 'kin'?) are absolute and stop the chain, and 'sin' doesn't occur beyond the first step. But at least 'ala' can turn up at any point in different roles: denying the whole thing or denying the latest degree: 'pona mute ala' v. 'pona pi mute ala' "not very good" versus "slightly good" (and can there be a 'pona ala mute'? Yes, a polite 'ike mute', but not 'pona pi ala mute' ). Getting the rules for all of these, plus the dialect that allows dittos like 'pona pona' or the various ways of saying "enough" (and "insufficient" and "excessive") seems to push recursion too far for the moment and no transformational rules seem to work either. But a list of pretty clearly acceptable cases may give some guidelines: (using 'pona' as a paradigm, treating 'mute' and 'lili' as the same, here covered by 'mute'))
pona
pona kin
pona ala
pona ali
pona sin
pona mute
pona pi sin ala
pona pi sin mute
pona pi mute ala
pona pi mute mute (not ditto)
pona pi sin mute ala ('ala' modifies 'sin mute', so different from -- tentatively -- 'pona pi sin mute, ala', which is a negation)
pona pi sin pi mute ala
pona pi mute mute ala (as above)
pona pi mute pi mute ala
And then the flights of exaggeration
Post Reply