mU-notation: a way to simplify tp-grammar

Discuss any other topic in here.
Diskutu ĉiujn aliajn temojn ĉi tie.
janpona120
Posts: 97
Joined: Wed Apr 20, 2016 5:44 am

Re: mU-notation: a way to simplify tp-grammar

Post by janpona120 »

combinations become idioms and what idioms they become is largely accidental from an external view. 'tomo awa' might have come to mean "hotel" but it didn't. Similarly, 'noka' could mean "walk", but we have settled on 'tawa noka' instead (this one subject to change, no doubt).
Phrases become idioms in situation of poly-semantic coding of the words. This way is difficult to memorize. You should keep in mind each meaning. 100 accidental idioms -- 100 hard probes to memorize. More effective approach is to use proto-semantic features of tp. This way allows to build a sense that you want, when you need, and to extract a sense from phrases that you never used. The proto-semantics allows translate automatically, using only one linguistic formula. For example:

-- fish ("kala") has a mission "to swim" (actively)
-- liquid ("telo") has a mission "to flow" (passively)

If you go by the way of proto-semantics, you will never use the word "telo" for "swimming"-idioms.

-- direction ("tawa") has a mission "to direct" (passively -- to keep a direction)
-- head ("lawa") has a mission "to redirect" (actively -- to change a direction)

So, "tomo tawa" (hotel) and "tomo lawa" (headquarter) are strongly mutual related.

-- winter ("tenpo lete") -- season for cold
-- coat ("len lete") -- clothes for cold

These examples apply only one linguistic formula "thing_1 for thing_2".
Evolution is slow but works pretty well.
Agree. The poly-semantic approach is the past of tp, the proto-semantics opens a "pona" future for tp.
janKipo
Posts: 3064
Joined: Fri Oct 09, 2009 2:20 pm

Re: mU-notation: a way to simplify tp-grammar

Post by janKipo »

Well, that is a nice system for some language, which right now happens to look a lot like toki pona but is in fact totally different. None of the "protosemantic" features assigned to what look like tp words are features of the words in tp, so comments about their systematic use are not relevant. It is true, however, that accidental -- as opposed to systematic -- idioms are hard to learn, if you sit down to "pump pred" (old Logjam expression) by memorizing words from a list. On the other hand, idioms, even accidental one, are picked up fairly easily in use when they become common (and are looked up or asked about or puzzled out, when they are less common), so little is lost by developing in this way (not nearly so much as memorizing hundreds of unused forms or even questionable rules). I doubt that protosemantics (certainly this partiuclar version) is anywhere in tp's future. But I do look forward to whatever language you create along this lines (though changing the vocabulary would be a kindness).
janpona120
Posts: 97
Joined: Wed Apr 20, 2016 5:44 am

Re: mU-notation: a way to simplify tp-grammar

Post by janpona120 »

None of the "protosemantic" features assigned to what look like tp words are features of the words in tp, so comments about their systematic use are not relevant.
In my imagination i draw a picture: a tender flower (tp) under a hard freezing wind (dogmatic "pu"). I am sure that protosemantics for tp really is watering and fertilizing. By, protosemantics, the flower of tp will grow as beautiful as a little girl becomes a pretty woman.
I doubt that protosemantics (certainly this partiuclar version) is anywhere in tp's future. But I do look forward to whatever language you create along this lines (though changing the vocabulary would be a kindness).
Any language is based on protosemantics. For example, Hebrew or ancient Russian alphabet is based on protosemantics. The prime language of humanity (pre Babylonian era) was based on it. Even tp has proto-seeds.
I do look forward to whatever language you create along this lines (though changing the vocabulary would be a kindness).
Thanks.
janKipo
Posts: 3064
Joined: Fri Oct 09, 2009 2:20 pm

Re: mU-notation: a way to simplify tp-grammar

Post by janKipo »

Well, the idea that an alphabet -- any alphabet -- is based on semantics just seems weird to me. Phonology, maybe (though very loosely) but not grammar and certainly not semantics (I except philosophical languages like aUI, but that is a very deliberate creation on an existing base, not creating the base itself).
While I don't quite get what you mean by protosemantics (my ideas tend to float around the notion that there is a store of meanings -- a possibly words embodying them -- that every language incorporates and which have been passed down from Noah or the 24 first humans), the way you talk about it sounds fairly unrealistic or, at least, unscientific. On the other hand, of course, as a universal grammarian (in the broad sense that goes up through pragmatics and rhetoric), I have to hypothesize a nearly Platonic conceptual nexus from which any given language draws globs as the meaning of words (complete with preexisting implications, etc.). But -- unlike Plato -- I would take the items in this nexus to be induced, not deduced, and so to be tentative and constantly changing in the light of new data. And so, also, not given and passed down from ages past (the currently trending story about "blue" being a nice case in point). Working with what I have at the moment, I am mapping out tp's vocabulary and finding both new items and new connection. Most of your claims don't fall into these maps or do so only in peripheral ways and I think that shifting the focus to put them at the centers would distort the ongoing realities.
janpona120
Posts: 97
Joined: Wed Apr 20, 2016 5:44 am

Re: mU-notation: a way to simplify tp-grammar

Post by janpona120 »

the idea that an alphabet -- any alphabet -- is based on semantics just seems weird to me.
Compare two letters "b" and "p", from one side, and words "sina" and "sin", from another side. We observe here equal pattern -- small modification. In a wide field of vocabulary it means that each word is a separate letter. In a narrow field of alphabet it means that no any alphabet is semanticated, because most of them are frivolous and tugled (or "nasa" in terms of tp).
While I don't quite get what you mean by protosemantics
There is the table of chemical elements (a set of chemical atoms: Hydogen, Helium, Carbon...). Also, there exists an equivalent set of semantic atoms: space, time, being, matter, etc. If you have a complete knowledge about semantic atoms, you may divide words on separate semantic atoms. With less knowledge, you may divide big semantic molecules on a few simple semantic molecules. For example, a word "toki" is a combination of two simple semantic molecules: "talk" and "think". Or in example proposed by jan Mato six years ago about "telo mani" to code a milk, using for it a simbolic semantic atom "a cow". In some cultures, cows are the means of payment and the milk-givers. So,in jan Mato's proposition I see a protosemantic dividing the word "mani". It is very interesting scientific approach with great future before coming of the Artificial Intelligence. You also may try to use a dividing of words on simple semantic parts or to mix a few simple semantic ingridients into one big semantic molecules. For example, I like a Sonja's solution about the word "toki" -- to joint together "to talk" and "to think". It is an excellent idea to joint these two senses in one word. I say, what I think, and contrary. It means that tp stimulate to be sincere. And tp is rich on such diamonds a lot. Let's try to use protosemantics, and we will have done "pu li sona kule". For a while, tp is based on polysemy, therefore we have "pu li sona pi walo en pimeja".
janKipo
Posts: 3064
Joined: Fri Oct 09, 2009 2:20 pm

Re: mU-notation: a way to simplify tp-grammar

Post by janKipo »

I fear this didn't clarify matters much. The analogy (or identity?) between letters and words just does not resonate with me.
As for protosemantics, I expect that eventually some of the projects that work out the semantic nets among words will 1. get some really useful maps of at least some word classes in some languages and 2. from these extract a set of semantic primes that can be used to explains the maps derived empirically. Nothing like this is even close to realization yet and so appealing to it as being already in play in tp can mean nothing more than that tp might be an useful language to try to make an empirical map of (but we knew that since it has only 120 words or so). As for 'toki' as "talk" and "think", this is an innovation (and an unexplained one) from earlier "say" and the use of 'pilin' for "think" as well as "feel" in both senses. So far, these various currents have not been sorted out, though I suppose the 'pilin' will be more about believing than about thinking things through.
janpona120
Posts: 97
Joined: Wed Apr 20, 2016 5:44 am

Re: mU-notation: a way to simplify tp-grammar

Post by janpona120 »

As for protosemantics, I expect that eventually some of the projects that work out the semantic nets among words will 1. get some really useful maps of at least some word classes in some languages and 2. from these extract a set of semantic primes that can be used to explains the maps derived empirically.
Protosemantics is a pro-active tool. It is fit to extract: knowledges, motivations, inventions. A main idea is next. You study a protosemantic alphabet (like digits from 0 to 9), and from now you can generate knowledges (like numbers of any sort). It is a top goal. A lot of mauntaineers are trying to reach the peak. And I guess, during 10 years this test will be successfully finished.
As for 'toki' as "talk" and "think", this is an innovation (and an unexplained one) from earlier "say"
and the use of 'pilin' for "think" as well as "feel" in both senses
Some people have a logic mind (at first position), so they are thinking through a word "toki". Another ones have intuitive mind, and theirs magic word of thinking is "pilin". In my opinion, a tie between "to talk" and "to think" is a beautiful idea. If to put on first place the word "pilin", then language should be named as "Pilin pona". ;)
janKipo
Posts: 3064
Joined: Fri Oct 09, 2009 2:20 pm

Re: mU-notation: a way to simplify tp-grammar

Post by janKipo »

Well, if you read the literature, the language is almost called 'pilin pona' 'feel good" with all the change-of-life talk. But 'toki' basically means (embraces, whatever) "communication" and thinking is a different activity from that -- more about sorting and connecting and rearranging than about disseminating results. To be sure, there is a sense in which much thought is talking to oneself (even literally, given measures on the glottis and tongue), so I don't object to the shift (although it might mean a lot of corpus revision), I just don't accept it outright yet.

Given that I remember working with protosemantics (it had a different name then but the idea was the same) in 1960 and that the present state, so far as I can tell, is not markedly better than it was back then, I think your ten years is deliriously optimistic. Of course, since most people seem to go about it in rather odd ways, it might be that someone with a non-odd approach could come along and revolutionize the whole study. Most researchers are not even shooting for something like what we have in mind but much more low level issues of clusters of co-occurrences and the like' that is data rather than theory, still.
janpona120
Posts: 97
Joined: Wed Apr 20, 2016 5:44 am

Re: mU-notation: a way to simplify tp-grammar

Post by janpona120 »

But 'toki' basically means (embraces, whatever) "communication" and thinking is a different activity from that -- more about sorting and connecting and rearranging than about disseminating results.
These two word have genetically a common semantic atom -- communication. Speaking participants make communication by words signals. Thinking participants (neurons of the brain, or computer) make communication by electric signals. Proto-semantically they are absolutely identical things.
Given that I remember working with protosemantics (it had a different name then but the idea was the same) in 1960 and that the present state, so far as I can tell, is not markedly better than it was back then, I think your ten years is deliriously optimistic.
Such a situation was in a story of developing of the table of chemical elements. A lot of decades without visible results. And one day it was done. Now, in the protosemantic research we have more than ten atoms with strong description of relation between them, like relations of chemical atoms. During next ten years will be found all needed elements, and the table of semantic atoms will become complete. The same situation we have with a quantum computing or a DNA sequencing. The work goes slowly, but powerful.
it might be that someone with a non-odd approach could come along and revolutionize the whole study
Sure, the person is first.
janKipo
Posts: 3064
Joined: Fri Oct 09, 2009 2:20 pm

Re: mU-notation: a way to simplify tp-grammar

Post by janKipo »

The community -neuron net analogy is not sustainable, at least partly because the connection between thinking and neuron nets is happenstantial so far as we can demonstrate. And, in any case, what is transmitted in communities is meaning, what is transmitted in neuron nets is distinctly not (though, on some standard theories, meaning arises from it). But, in any case, we, as speakers and thinkers, expreience these as rather different things, even if loosely related at some remote level. Still, as I say, I can live with the identification; I just don't yet see it as natural nor accepted.

I'm glad to know there is at least a small net (ten atoms -- nodes?, you say) that is worked out. That is a lot farther than anything I have seen reported (but admittedly, this doesn't get talked about much since it is not high priority -- although it should be). Ten years still seems optimistic but nto, perhaps, so impossible as my rather jaundiced view.
Post Reply