Wednesday, February 10, 2010

cantonese grammar. thoughts. Richard Dawkins.

Cantonese is an analytic language where, in a sentence, the arrangement of words is important to its meaning. A basic sentence is in form of SVO, i.e. a subject is followed by a verb then by an object, though this order is often violated because Cantonese is a Topic-prominent language. Unlike synthetic languages, seldom do words indicate time, gender and plural by inflection. Instead, these concepts are expressed through adverbs, aspect markers, and particles, or are deduced from the context. Different particles are added to a sentence to further specify its status or intonation.

D: topic-prominent...
A topic-prominent language is a language that organizes its syntax to emphasize the topic–comment structure of the sentence. The term is best known in American linguistics from Li (1976), who distinguished topic-prominent languages, like Japanese, from subject-prominent languages, like English.

D: so we do have a clash between East and West in this respect. Sticking to SVO word seems prudent.
We can incorporate a means of emphasis. Alternatively, allow reiteration.
For example,

Mandarin Chinese
張三 我 已經 見過 了。 -> Original order: 我 已經 見過 了 張三。
Transcription: Zhāng Sān wǒ yǐjing jiàn-guò le. Transcription: wǒ yǐjing jiàn-guò le Zhāng Sān.
Gloss: Zhang San I already see-EXP RES Gloss: I already see-EXP RES Zhang San.
Translation: (As for) Zhang San, I've seen (him) already. Translation: I've already seen Zhang San.

D: Strange that ASL sign, derived from an English, would invert the subject-topic prominent aspect.

D: A topic(object)-prominent arrangement would mean that use of a pronoun later would necessarily be of the topic and not subject.
Bill, Ted has seen him.
HIM being Bill.
A nuanced pronoun system like Lojban's can help.
Bill (he1), Ted (he2) has seen him(1) or (2).
Verbal Aspect

In contrast to many European languages, Cantonese verbs are marked for aspect rather than tense - that is, whether an action has begun, is ongoing, or has been completed. Tense - where an action occurs within time, ie past, present, future - is specified through the use of time adverbs. In addition, verbal complements may convey aspectual distinctions, indicating whether an action is just beginning, is continuing, or at completion, and also the effect of the verb on its object(s).
Aspect particles are treated as suffixes bound to the verb.

D: so Chinese is essentially verb-suffixing.
Again, we have a source of clash with English.
English: will have been XYZ-ing. Or -ed.


Cantonese uses the following pronouns, which like in many other Sinitic languages, function as both subjective (English: I, he, we) and objective (me, him, us):

D: our English subject/object pronoun distinction is a BAD idea in an interlang.
Word order takes care of this.
One day, I don't mind some subject/object word particle to differentiate.
Right now, KISS. Keep It Simple, Stupid!

I like me (I). He saw her (she). And so on.

Copula ("to be")

States and qualities are generally expressed using stative verbs that do not require the verb "to be". For example, to say "I am hungry", one would say 我肚餓 (ngoh5 tou3 ngoh4)(literally: I hungry).

D: English sentences get cluttered enough with 'be' as it is. The English mind does not see 'be' as inherent in stative (verbal) adjectives.

A stative verb is one which asserts that one of its arguments has a particular property (possibly in relation to its other arguments). Statives differ from other aspectual classes of verbs in that they are static; they have no duration and no distinguished endpoint. Verbs which are not stative are often called dynamic verbs.
D: so stative verbs do not interact with TIME.

D; a stative verb can be recognized in English. It cannot form part of an imperative command.
John is tall.
You there - be tall! (Huh?!).


Examples of sentences with stative verbs:
I am tired.
I have two children.
I like the color blue.

Static versus dynamic

The same verb may act as stative or dynamic. An English phrase like "he plays the piano" may be either stative or dynamic, according to context.
Some languages use the same verbs for dynamic and stative situations, while other use different (if often etymologically related) verbs with some kind of qualifiers to distinguish between the usages. A stative verb is often intransitive, while a corresponding one would be transitive. Compare, for example, modern English with modern Danish.

D: these sort of implicit qualities kick the butt of ESL speakers, particularly those from linguistic background that use a clear system. In this respect, I do not believe that a pidgin-like omission of this detail is desirable. The sheer ambiguity created is staggering.

Dowty's analysis
Dowty gives some tests to decide whether an English verb is stative. They are as follows:
Statives do not occur in the progressive (the * before a sentence means that it is ungrammatical or absurd to most native English speakers):
John is running. (non-stative)
*John is knowing the answer.
They cannot be complements of "force":
I forced John to run.
*I forced John to know the answer.
They do not occur as imperatives.
*Know the answer!

D: I made John be taller! ....?!


Many negation words start with the sound m- in Cantonese; for example, 唔 m4 (not), 冇 mou5 (not have), 未 mei6 (not yet). Verbs are negated by adding the character 唔 m4 in front of it.
D: this is a verb PREFIX of sorts.

D: Decimese reserves -M in the word-final consonant position. An objection might initially seem that the -M ending of adjectives/adverbs could become confused with such a word-initial negation. The syllable format of various grammatical elements as well as word order ensure this is not the case.
Hmm. I had not planned to use M in a word-initial position for word particles. If I were to do so, I would need to ensure that word boundaries are perfectly clear. I wish to avoid -in rapid speaking - 1 identical phonetic tract that has 2 possible phonemic and thus semantic structures.
OK, and then we have English using N + V (+C?) format for negation.
OK... if M or N + V word particle can only occur in sequence with a HYLR + V, word particle, we could ensure that a sequence of sounds is not ambiguous. A logical construction for Decimese at the advanced optional diphthong stage could be:
1) default to generic M + V
2) MOU - not have - possibly (recall that H can disappear entirely, bequeathing it's following vowel to another word's vowel diphthong) MO_HU.
3) MEI- not yet. ME_HI.
Using rigid word order, let us say that we have noun (subject)
Possible alternative sentence format.

D: I have noted before that we can redesign Decimese to function with only 2 nasal consonant word-finals. For Mandarin, that would omit M, leaving only N and NG.
Well, if it becomes an issue, we *could* reserve all 3 nasal consonants as word-final BUT
1) use -N for subject
2) use- M for object.
This would serve to make topic and subject-prominent layout crystal clear.
It has 2 drawbacks:
1) we lose use of -M word-final consonant to indicate adjective/adverb.
Well, stative verbs almost deserve -NG for verb instead. Of course, we would need a 'to be' substitute to clarify.
Is that a verb or adjective? Keep in mind that word order would place a subject's adjective behind a verb.
I do NOT wish for the whole sentence to be read first, and then the meaning disentangled.
That would serve to place great demands on working memory. It makes complex sentences a bad idea.
It overwhelms new speakers that are still struggling to just find word boundaries and recall vocabulary meaning.
The second drawback of -M for object is:
2) the newly variable word order will become confusing to speakers with linguistic backgrounds that have invariate order.
If (and I say IF) we were to use -M word final consonant to denote object, then pronouns potentially could acquire the matching format to show object status.
He does. To him. (-M!).
Since decimese defaults to gender-neuter, we get:
It does. To it. .... To -it-M?
It will likely require the animated/subset human (for possible motivations) indicators for nuance. At least if context does not already provide it.

I am seriously considering making the entire language extremely humanocentric. Literally even the math and sense of scale is pegged to the human condition.
We are meso-space and meso-time -scale creatures. Atomic scale interactions confuse us as much as the vast scale of black holes. Planck-time events seem impossibly fleeting. The age of the universe- 13.7 BILLION - and geologic time scales (say, 1.37 billion) are not what our brains were made to grasp.
We are the middle. The middle size. The middle perception of time.
The middle. And to our own inflated sense of self-importance, the CENTRE.
Decimese, after all, is made for humans to commincate about human concerns. Humans.
I do wish for our speaking and thinking to be aligned with the nature of the universe, though. Joseph Campbell- my spiritual guru- said one of the purposes of religion is to align humanity with the world he lives in.

Back to Cantonese grammar now.

In contrast to the examples of sentential negation above where the entire sentence is negated, 唔 can also be used lexically to negate a single word. The negated word often differs slightly in meaning from the original word; that is, this lexcial negation is a kind of derivation. Evidence for this is that they can be used with the perfective aspect particle 咗 jo2, which is not possible with sententially negated verbs.

D: Pidgins can accomplish this by simply placing a simple negation such as 'ne' adjacent with the part to be negated.

D: I have considered a similar approach to interrogative. With rigid word order, we an tack a query particle ("chu"?) adjacent to the word in question.

Using the same approach for both is powerful and succint.
And regular!

N.B.: 啲 is a very versatile word in Cantonese, besides pluralizing certain phrases, it can also mean "a little/few", e.g. 一啲 jat1 di1 (a little), or 早啲 earlier (literally: early + (intensifier)).
D: Plural-few-early. ! This would confuse the heck out of English speakers.
However, a carefully planned system could allow something like this to be optionally denoted via a diphthong, perhaps.
D: this could align with my number system. Whole number (invert) - fraction. 2/1. 1/2.
Add a space and time indicator and we arrive at:
1) plural: say 2 or 3. Why? 1-2-3. Single-dual-plural.
2) invert. 1/1 - not sure what that means inverted! 1/2 1/3 .
3) add time... Don't ask me what time 1-2-3 means LOL. We express time as a line.
It could mean farther down the line. Just happened:1. Happened: 2. Happened long ago: 3.
A +/- system, like the whole #/fraction system (possibly one is derived from the other) could express the additional time nuance of earlier/later. Past, future.
So we arrive at 1-2-3 and +/1, with #s/time/space indicators.

Well I'd say that is enough speculation during my first morning coffee!

Aside. Book /author review. Richard Dawkins.

D: deep and profound is my respect for this man.

I read his work from 20 years before that, "The Blind Watchmaker", at work this week.
Since creationism is one of my bugbears, of course I loved it! I learned so much despite the book being so old, and despite having read many books on the subject already. It inspired me to write no less than 3 short stories.

However, I did notice what I consider a gap in his thinking when I compared these 2 books.

His recent position on atheism relies heavily on the behavior of probability at the extreme to function like proof.
He essentially argues that a 'vanishingly small' chance for a deity is comparable functionally to no chance.
This sleight of hand then allows him to swap the concepts of probability and proof.
Otherwise one ends up with a nominally agnostic but not *quite* atheist position.
The "Blind Watchmaker" has a similar fudging of details.
In it, he takes umbrage at both the caricature of the steady-speed gradualist evolutionist, as well as the Gould-ian punctuated evolutionist. This is where he trips up. He mocks that a tiny growth in a horse-ancestor's leg each generation, or that a slight increase in proto-human's brain capacity each generation could matter. He uses the word 'significant' to hedge his bets.
However, earlier in the book he emphasizes how evolution operates on just such vanishingly small- but present- variations for selection to occur. I confess this seems like a case of debating 'silly bugger' to me.

However, I believe that reading his book in high school would have single-handedly swayed me from taking a liberal arts education. I would have chosen biology instead. (My decision was close.)

His distinction between archaelogy/historical time and geology/evolutionary time was illuminating.
His ambiguous use of what constitutes an 'infinitesimal but significant' versus an 'infinitesimal but INsignificant' (???) distinction got me thinking. About infinity. I was planning Decimese #-base naming conventions and concepts.
Geometry: 360 degrees. 60 arc-minutes and 60 arc-seconds.
Time(day): 24 hours. 60 minutes. 60 seconds.
I read a primer on economic indicators this week at work too. ( Heh. Love my job sometimes! <:)
Economists use percent-of-percent. Unit: 1. 100 percent. each percent then contains 100 'number points', resulting in 10,000 number points derived from 1 whole number.
60-60. 100-100. 360-36x10. I found these 2 systems interact.
Anyway, I started applying these subconscious ideas to historical versus geological thought. And 'impossibilities'.
Impossible- in a human lifetime. Or highly improbable, a vanishingly small 'insignificant' chance. But on a geologic scale?
The economic system of 'number points' was useful as mental shorthand. 100x100 is 10,000.
The universe, shall we say, is c. 10 BILLION years old.
(New Earthers would disagree. But relatively speaking, then by their own admission, they were pretty much 'born yesterday'. I tend to agree, pretty much. )
I initially used multiples of 1000 to try to chunk time scales. 10B. 10M. 10,000. 10 years.
10,000 worked well too (economic # points). 10B. 1M. 10 years.
If we need very large #s, then recycling the 1-2-3 dimension system can work. Take the base 10.
A line. 1D. 10 units long. Square -2D- area. 10x10=100. 3D-volume-cube-10x10x10=1000. Hypercube.... 10,000. # points.
This is only for a naming convention. But it allows us to rapidly adjust scale to discuss a concept appropriately.

I have always been intrigued by Eastern philosphical thoughts on infinity.
Is infinity all time, or beyond time? Ditto space and the concept of void.
It touches on math infinity and zero. Fractioned, infinitesimals and the Planck limit.
Discussing probability on scales beyond the age of the universe looks like infinity. At least if we simply consider it 'more than all time'. Which suddenly doesn't seem so long in this light.
I read that the odds of quantum effects allowing a glass of water to fall through the counter it rests on can be calculated.
But the chance of this happening during the age of the entire universe is highly improbable. Perhaps vanishingly small. Maybe even insignificantly so.
The concepts of human and geologic time can be linked to scales linked to the various MELTS acronym principles.
Before the Big Bang, we don't think we had spacetime yet. Math? Then bang. Things happen fast! Time. Things expand. Space.
Things cool off. We get subatomic particles derived from tinier particles. Science: physics. Atoms and molecules form: chemistry. Life emerges: biology. Complex life leads to self-awareness. Logic. Moral considerations. Ethics. Each step is an emergent property of the last.
Time scales. Physics - 10 billion years. Life- more like billions. Millions... thousands - historical time. A 100 - human.
Things that seem 'impossible' or requiring near-infinities to happen, even by our physics? Math.

D: there seems to be natural math and then human math. The universe deals in whole numbers.
Hydrogen: 1. Oxygen: 1. H2. O2. H2O1. Water. Sure we have a 2:1 ratio. 2/3 is a messy .666 repeat.
But that is human math. The universe only deals in whole numbers.
Evolution. DNA is binary- on/off. 4-base. A snake doesn't evolve 1/1000th of an extra vertebra over each of 1000 generations.
About 1 in 1000, it gets another complete vertebra segment. Or maybe 3 or 4 once in 3 or 4000 years. Or 40,000.
This seems like a trivial mental exercise, but it was considered deadly serious by humans in the past.

With the Pythagoreans; mathematics, numerology, and religion, seem to blend in a way difficult for us to understand today. Key to Pythagorean idealism is the idea that number lies at the essence of all things -- abstract as well as concrete. Their study of mathematics was extensive, and the applications to music and astronomy were especially impressive. They believed strongly that all numbers were commensurable; that is to say, that any two numbers could be expressed as integral units of a smaller number. This amounts to a belief that all numbers were rational in modern mathematical terminology. In these ideas alone we see something of Plato's idealism, as well as the ancient notion of the atom.

It is ironic that this discovery lead the Pythagoreans directly into a major crisis: an immediate consequence of the Pythagorean Theorem is that the diagonal of a unit square is Ö2 -- a number which is not commensurable with one (i.e., is irrational). Legend states that this discovery was found by Hippasus, who was drowned at sea for heresy as a consequence. Whether or not the legend is true we can be certain that the existence of irrational numbers was another (and reluctant) contribution to Western knowledge made by the Pythagoreans.
D: unable to tease apart the math of the universe versus the math of man, we see this amounted to a religious crisis.

Joseph Campbell might have suggested that they did not have a belief system that aligned them to be in harmony with the universe they lived in.
Maybe a language can contribute to such harmony. Just a little bit.

POS spell checker still is not working.


Dino Snider said...

Stative verbs are invariate in time. Recall the Decimese acronym MELTS. Math, ethics, logic, time, space. So we could assign verbs descriptors based on these. Static verbs could be 'logic verbs'. For that matter, grammatical elements seem to show affinity for one concept or another. Verb-action - time. Thing -noun-space. Conjunctions-Boolean- logic. And so on. Ethics - tricky. But important.

dino snider said...

Oops. I meant his other book "The God Delusion", not "The Blank Slate" - I read both. Hell I'll read anything Dawkins writes!

Dino Snider said...

Regarding Cantonese being a "topic-prominent" language...

I can babble in ASL sign language. I read a book in the library on the formal linguistics of it. It gave me insights for my other language projects I would not have otherwise gleaned. Anyway, ASL also violates word order to introduce new and salient topics, instead of starting with older facts that have already been established.