Roman Numerals and Christmas Tree

Signs and symbols: Writing systems (hieroglyphs, nail writing) and Signed Toki Pona; unofficial scripts too
Signoj kaj simboloj: Skribsistemoj (hieroglifoj, ungoskribado) kaj la Tokipona Signolingvo; ankaŭ por neoficialaj skribsistemoj
janTe
Posts: 50
Joined: Wed Nov 04, 2009 4:34 pm

Re: Roman Numerals and Christmas Tree

Post by janTe »

If you were going to use colors to represent numbers, there's already an established system you can use: Resistor colour codes (http://en.wikipedia.org/wiki/Electronic_color_code). Every electrical engineer can read this like the back of his or her hand.

But this is getting a bit off topic.

I like the TP numerals and the Christmas tree :-)
janTe
Posts: 50
Joined: Wed Nov 04, 2009 4:34 pm

Re: Roman Numerals and Christmas Tree

Post by janTe »

Oh, you might want to use LLLL for 20, and MMMMM for 100.

In practice, I can see people saying 'luka luka luka luka' instead of 'mute', to avoid confusion with the more general meaning of 'mute'. But 40 would be MM, unambiguously.

Same with 100 and 'ale'.
janMato
Posts: 1545
Joined: Wed Dec 02, 2009 12:21 pm
Location: Takoma Park, MD
Contact:

Re: Roman Numerals and Christmas Tree

Post by janMato »

janTe wrote:Oh, you might want to use LLLL for 20, and MMMMM for 100.

In practice, I can see people saying 'luka luka luka luka' instead of 'mute', to avoid confusion with the more general meaning of 'mute'. But 40 would be MM, unambiguously.

Same with 100 and 'ale'.
The ambiguity problem improved in writing by changing notation to Roman Numerals or in speech by prefixing with nanpa. LLLL and MMMMM exacerbate the verbosity.

mi wile e mute. I want a lot. I want 20. I want to grow.
mi wile e luka luka luka luka. I want 20. I want 15 hands. (When buying illegal gorrilla hands at the bush meat market in Africa)
mi wile e LLLL. I want 20.
mi wile e nanpa luka luka luka luka. I want 20.
mi wile e nanpa mute. I want 20.
mi wile e M. I want 20 (but unambiguous only when written)
mi wile e nanpa M. I want 20.

The official system at the moment encourages following the conventions up to about 150, after that it starts to become laughably cumbersome again. And that is where community innovation kicks in with all it's problems and solutions.

At the moment I figure that a toki pona would deal with modern math the way we deal with unfamiliar number systems we see today, such as hexidecimal.

0xCAFEBABE would make a nice tattoo. => oh, ex, see, aye, ef, e, be, ah, be, e would make a nice tattoo.

It would be descriptive of the notation, not necessarily a grammatical English phrase, not necessarily recognizing the underlying structure-- i.e. reading 2010 as two-zero-one-zero or twenty-ten, instead of Two-Thousand-Ten, which is more clearly a place system.
Last edited by janMato on Tue Jan 26, 2010 3:16 pm, edited 1 time in total.
janKipo
Posts: 3064
Joined: Fri Oct 09, 2009 2:20 pm

Re: Roman Numerals and Christmas Tree

Post by janKipo »

'mi wile E...' although, when what follows is a number or, worse, 'nanpa' it is not clear what you are asking for. Yeah, yeah, "context will decide." Writing is even clearer with Hindu-Arabic notation -- as long as you're gonna break the rules.
janMato
Posts: 1545
Joined: Wed Dec 02, 2009 12:21 pm
Location: Takoma Park, MD
Contact:

Re: Roman Numerals and Christmas Tree

Post by janMato »

janKipo wrote:although, when what follows is a number or, worse, 'nanpa' it is not clear what you are asking for. Yeah, yeah, "context will decide."
Good point. Are numbers necessarily modifiers (like proper names, etc), even when prefixed with "nanpa"? If so, a hyperorthodox solution would be prefixing numbers with "ijo nanpa/jan nanpa/soweli nanap" etc. similar to it looks like we should do with bahuvrihi. Reminds me of Japanese that has different number constructions for different categories of things.

* uta kiwen pi palisa utala => saber tooth
akesi pi uta kiwen pi palisa utala
janKipo wrote: Writing is even clearer with Hindu-Arabic notation -- as long as you're gonna break the rules.
Ah, conculture rules. We know that tp must have a cultural value of not borrowing too much (or it wouldn't be a small language anymore). Has anyone thought about what the implied culture has to say about code switching? Outside of the brief life of tokipinglish, I'ven't seen too much discussion on the matter. Would the imaginary toki pona be expected to be bilingual and code switch or mono-lingual and mostly talk in tp? The closest thing I can think of is the aboriginal mother-in-law avoidance speech, where speakers talk a greatly modified subset of their mother tongue when the mother in law is around-- i.e. at least some people in the world think it is good to code switch back and forth from a simple language to a full featured one.
janKipo
Posts: 3064
Joined: Fri Oct 09, 2009 2:20 pm

Re: Roman Numerals and Christmas Tree

Post by janKipo »

tp number words (even 'luka' [ptui1] and 'ale' and 'mute' [hyuck!ptui!] ) seem to be normal words, functioning in all contexts (except, perhaps, as prepositions). What they mean in these various positions is less clear, however. In particular, as head (and only) noun, they can obviously refer to the number itself (whatever that may be -- and, believe me, we don't want to get into that) and, in context, to a number of things offered -- to indicate how many are accepted. Starting off with 'nanpa' muddles matters further, since now we might mean a number of numbers (which ones unspecified) or the item offered in the ordinal positions specified or, just possibly, the number itself again. I don't see what the point of 'jan/ijo/soweli nanpa...' is nor how it relates either to Japanese or to bahuvrihi (which are usually '... pi jo ...' in tp). But then, I am not sure what the good point was. My point was that 'mi wile e wan' say, is a very peculiar thing to say out side of, say, a sudoku puzzle version of Wheel of Fortune, unless there is a lot of context to fill it in, typically with an NP to be modified.

Are there really saber-toothed toads?

I'm not quite sure where conculture ends and language rules begin in this case. Strictly speaking, we can, with enough patience and skill, use any number we want in tp as a linguistic fact. We are limited in fact by just those factors of patience and skill to a a relatively small number of relatively small numbers. The language offers little in the way of overcoming the problems of patience and skill, like having words for bigger magnitudes. Now, a consequence of this is that a culture using this language will probably not have much to do with big numbers. OR it will change the language in some way to accommodate them. At the moment, we are right at the OR, and a cheating all over the place: using illicit or at least unsightly and ungainly number expressions or codes which either abbreviate the same or rely on different systems altogether, the latter being serious code-shifting. I have -- over the years -- suggested at least three solutions to the problem and others have added a handful more. Hopefully one day we will get the final, definitive answer and place tp either in its proper therapeutic and cultic role or in its role s a utile language in the real world.
janMato
Posts: 1545
Joined: Wed Dec 02, 2009 12:21 pm
Location: Takoma Park, MD
Contact:

Re: Roman Numerals and Christmas Tree

Post by janMato »

janKipo wrote:I don't see what the point of 'jan/ijo/soweli nanpa...' is nor how it relates either to Japanese or to bahuvrihi (which are usually '... pi jo ...' in tp). But then, I am not sure what the good point was.
a a a! sometimes I can't tell if you're saying "meh", or asking for clarification.

tempo mute la mi mute li pakala e soweli tawa moli ni.
For a long time, we've been beating this dead horse.

re: POS of numbers, if they're nouns then no problem, but you seem to be saying "nampa wan" does/can/should mean only 1. To mean one cat, it would need to modify something, right? Else I'd have a words that modify something unspoken. If it must modify something then one will have to pick a category, e.g.
? mi jo lipu kule. mi jo e lipu nanpa MLTT.
I have 29 sheets of construction paper.
Now we've inserted the extra "lipu" and if it were obligatory, it would be just like having to know that hitotsu is one object, hitori is one person (different suffix after the "hi"), ichimai (one flat thing), etc.

re: bahauvirihi, well if a noun phrase must have a head, then pick a generic one, e.g. sabertooth -> sabertooth thing, cat, beast
re: x pi jo y for bahauvirihi
? akesi pi jo nena uta pi jo palisa utala
This may actually be an more important pattern that it seems. In the world of software and databases, we pretty much model the world using 2 relationships, Is-A and Has-A. "pi" is almost the Has-A relationship, but "pi jo" would make it much clearer.
re: toads... akesi is in part a category that depends on who's talking. Cat's are cute soweli, but relative to someone afraid of being eaten by a sabertooth, they're akesi.
janKipo wrote:Hopefully one day we will get the final, definitive answer and place tp either in its proper therapeutic and cultic role or in its role s a utile language in the real world.
I'm working on my list of 100 uses for tp, in the therapeutic section I've got treating depression, a faster to acquire sign language for adult onset deafness, and a backup language for stroke patients. This is on the speculative idea that if one got a stroke in the Broca or Wernicke's, the ordinary machinery for language production would be (partly) gone, but the remaining general purpose parts of the brain could possibly still process toki pona-- being that toki pona doesn't seem to require recursion or the ability to recall one word of 100,0000 in sub-millisecond response times. So if a stroke patient could do computer programming or algebra but not process lanuage, they might be able create and interpret toki pona sentences. Or maybe not, who knows?
janKipo
Posts: 3064
Joined: Fri Oct 09, 2009 2:20 pm

Re: Roman Numerals and Christmas Tree

Post by janKipo »

'nanpa wan,' as a NP alone means "one number" (no telling which) or, just possibly, "the number, one" or, again, with a little stretch, "the first, best, ... thing" (a calque). As a modifier, it means this last primarily, though it might mean either of the others (though I think it needs 'pi' then). In certain contexts, it could be a modifier of an implicit noun (indeed, I think that is what happens in the case of its mean "the first, best, ...").
Your e.g. says "I have colored paper. I have the 29th sheet" (in some unspecified ordering). It does not say "I have 29 sheets" that would b 'mi j e lipu MLTT' or the written form of the spoken equivalent. I don't see why 'lipu' is "inserted" -- but then, I don't see why you don't say 'mi jo e lipu kule MLTT' either.

Do you know what a bahuvrihi is? tp doesn't have any that I know of, but the ones it gets from other languages (which it probably shouldn't all together) tend to reproduce the underlying semantics. It might be possible to do this with the overt form but no one has done so so far so far as I know. You are right however that usually 'pi', without 'jo' is enough for the effect. Calling a smilodon "saber-tooth" is a bahuvrihi, but that gets translated into tp pretty much as you suggested ('akesi' aside). that is, the center becomes the head.

I doubt saber-toothed tigers are ever akesi, maybe mansuto. The basic meanings seem to stick for the head nouns, the derivative features, "ugly" for example, turn up more as modifiers.
janMato
Posts: 1545
Joined: Wed Dec 02, 2009 12:21 pm
Location: Takoma Park, MD
Contact:

Re: Roman Numerals and Christmas Tree

Post by janMato »

janKipo wrote:'nanpa wan,' as a NP alone means "one number" (no telling which) or, just possibly, "the number, one" or, again, with a little stretch, "the first, best, ... thing" (a calque). As a modifier, it means this last primarily, though it might mean either of the others (though I think it needs 'pi' then). In certain contexts, it could be a modifier of an implicit noun (indeed, I think that is what happens in the case of its mean "the first, best, ...").
Ah, my bad. jan Pije does explicitly say that "nanpa #" is ordinal. jan Pije's cardinal examples show them as modifiers, but doesn't say if they can only be modifiers (like proper names)-- I'm starting to think numbers have to modify something.
janKipo wrote:Your e.g. says "I have colored paper. I have the 29th sheet" (in some unspecified ordering). It does not say "I have 29 sheets" that would b 'mi j e lipu MLTT' or the written form of the spoken equivalent. I don't see why 'lipu' is "inserted" -- but then, I don't see why you don't say 'mi jo e lipu kule MLTT' either.
re: inserting lipu. Because something has to go in between "e" and the "MLTT". (Unless I'm wrong about numbers having to modify something)
janKipo wrote:Do you know what a bahuvrihi is? tp doesn't have any that I know of, but the ones it gets from other languages (which it probably shouldn't all together) tend to reproduce the underlying semantics. It might be possible to do this with the overt form but no one has done so so far so far as I know.
If by "know" you mean recently looked up in wikipedia, then yes. Otherwise, not sure. AFAIK, they show up in TP by as mistake-- but not obvious to catch because a [modifier] [modifier] pattern can be read as [noun] [modifier] and probably not with the intended meaning. If they are mistakes (and I'm thinking they are) they need to be guarded against by plugging in a suitable thing to be modified, usually something generic like jan/ijo/etc. Bahuvrihi still remind me of phrases like "I want six" because what ever six is modifying isn't named (it's paralinguistic meaning those donuts, not the plastic tag with the 6 on it).
janKipo wrote: You are right however that usually 'pi', without 'jo' is enough for the effect. Calling a smilodon "saber-tooth" is a bahuvrihi, but that gets translated into tp pretty much as you suggested ('akesi' aside). that is, the center becomes the head. I doubt saber-toothed tigers are ever akesi, maybe monsuto.
Hmm. The classic definitions use the phrases "non-cute" and "lovable animal", which still sound to me like subjective categories that can't easily be resolved until one knows who's talking and what their preferences in animals are.

akesi waso pi kon seli li toki ni. "mi soweli suli pi laso suli. o monsuto e mi."
One would expect (non-clinically depressed) dragons to have a healthy ego and consider themselves lovable animals.
janKipo
Posts: 3064
Joined: Fri Oct 09, 2009 2:20 pm

Re: Roman Numerals and Christmas Tree

Post by janKipo »

I don't see any reason why numbers can't be head nouns grammatically, even if they are semantically (or pragmatically) serving as modifiers of contextually specified class nouns. So, 'mi wile e wan' seems OK, and, in an appropriate setting, almost obligatory ("I would like one of those doughnuts you are now offering for my selection" seems way too much).

It seems that bahuvrihi turn up eventually in most languages by erosion; you can't say "the man with the red cap on" all day without eventually shortening it ("Red Cap") or replacing it ("porter") . It hasn't (so far as I can tell) happened in tp yet, but I won't be too surprised if it does (it is less likely in Lojban because the rules are stricter and the mindset toward literalism has been almost insurmountable). It's a move I didn't think of when discussing the Zipf Wall in my paper, but it now seems a natural step: 'linja uta li mu e kalama "miao"'

Yeah, I don't like the subjective part of the definitions, but I have to admit that, given them, you're probably right about the applications.
Post Reply