Speaking of Tones in Chinese

Just as we were talking about Chinese being a tonal language, here is a nice illustration of it's potential pitfalls.

It's Two Days before Our Classes

Finally, our home schooling will officially start this Sunday. It will be exciting, or interesting, or both. Before we start, I talked with my daughter on a few issues, mostly our goals and ground rules.

Goals

I would like to set a few goals that we can check our progress against in a later time. These will be goals for both of us. It is not a trivial task since I have no reference to base upon and language skills are usually hard to measure. To start it off, I decided just to have a near-term, namely six-month goals:

1. Learning Characters: She should be able to recognize at least 200 characters out of the most-frequently-used 500.
2. Pronunciation: She should have a marked, recognizable improvement in her pronunciations.
3. Talking: She should start to speak Chinese in some simply, daily conversations. She may be able to speak comprehensible Chinese mixed with English words where her Chinese vocabulary is lacking.

The first goal is the most straightforward and also quantitatively measurable. Our current baseline is that she recognizes 136. This will be a reasonable improvement in the time period. Of course she would also have learned characters that are not on the list also. The other two goals are what we try to achieve. They are not as easy to measure, but I think we will be able to give our fair evaluations too.

Start talking in Chinese may be the hardest of them all, as it involves both language skills and mental confidence, as well as willingness.

What are not in the goals are other basic language skills such as writing of characters, constructing sentences, and phrases. These will of course remain the essential parts of our lessons and exercises. But I decided to leave them out of goals for now so we have a better focus.

Rules

We agreed on two basic rules:

1. We will have a two-hour class session (with breaks) every week, mostly on Sundays. During the class period, all her regular school rules apply. She should behave just as she does in her school classes.
2. We will have weekly homework assignments and she is expected to finish them before the next class. She probably will be doing her homework with her Mom so that she has a different learning environment than our classes.

Schedule

I had thought of making a calendar for our school year or semester or quarter. But I am still not sure what our weekly pace should be. We will be mainly using the Standford textbook, which covers a lesson every two weeks. I thought two weeks may not be enough from where we are starting. But I am not sure whether to spend three or four weeks in one lesson.

So, we don't have a school calendar yet. We will reevaluate that after we go through a couple of lessons and see how we are doing by that point. The biggest advantage of home-schooling is the flexibility, right?

Are We Good or Are We Weird?

In the Peter Rabbit play, Cottontail has a very simple line "We are always good", which I translated as "我们总是乖的". Whenever my daughter is reading this line, she would sound a lot more like "我们总是怪的". I told her that would mean "We are always weird". We had a great laugh about it.

The characters 乖 and 怪 sound basically the same, except for their tones. The first is a flat (first) tone and the other is a dropping (fourth) tone in Pinyin. When I teach my daughter how to pronounce the words individually, she does it correctly. But when we read that sentence, she just could not get it right.

Then I realized that this is a common pronunciation problem for people who speak English, which is an intonation language. In English, words do not have tones on their own and their tones can change according to their placement in a sentence and the sentence itself. Typically, the tone is rising toward the end of a question ("Are we good?") but falling toward the end of a statement ("We are good."). When an English speaker reads a sentence in Chinese, a tonal language, he/she would still intuitively change the tone in the same way. Therefore we have the incorrect pronunciations.

To confirm this, I had my daughter reading a few different sentences in Chinese:

• "乖孩子."("Good kid."). This is a statement, but the word 乖 is at the beginning of the sentence where the tone tends to be flat. She indeed did well with it.
• "我们乖吗?"("Are we good?"). This is a question, with 乖 near the end. Again, she did very well with it.
• "我们总是乖的". Back to our original statement sentence. She reverted back to mispronounce it!

So, that appears to be the problem. As we proceeded to do more of the Peter Rabbit script, I noticed a few other occasions where she mispronounced tones based on sentence structure.

Once we identified the source of the problem, I explained to her what was going on. It made sense to her too. She would pay more attention to this later. But It will take quite a while to correct, as habits carried over from native tongue are always hard to fight against. But it's okay, we are still having great laughs with her proclamation that we are always weird!

Peter Rabbit for the Holiday

I have been thinking a lot about how to get my daughter to work on some materials that are close to her own life in Chinese. The Stanford Chinese text book, although very good, also lacks in this aspect. I think kids need to learn dialogs in life-like context before or along with studying story-like textbooks.

During the holidays, I did one experiment. Before the winter break, my daughter had participated in a KidStage production of Peter Rabbit show in her school. She was very enthusiastic about the whole project. Within weeks of receiving the script, she had memorized all her own (Cottontail) lines and most of other characters' lines. The show, although a very small production, was a fun experience for all the kids involved.

I found the dialogs in this script to be simple, concise, and cheerful. So I spent some time to translate them into Chinese and see if my daughter would also enjoy learning them in Chinese.

It did not go well at the beginning. Although she was very curious and enthusiastic about it, she was quickly discouraged by most of the characters she could not recognize, pronounce, or remember. So she decided not to do it again. "It is just not helping!" she said.

After a couple of days, I suddenly realized what the problem might be. After translating the script, I simply typed up the Chinese characters and printed them out. I did not include any Pinyin for the characters she did not know. Even though the lines are very simple, she could not remember the pronunciation of all characters and therefore she could not continue to read her lines.

So I went back to the script and asked her to identify the characters she did not know. Then I hand-marked the Pinyin for them. Hooray! Now she can read her lines. She is happy again. We did a few scenes together with her reading her (Cottontail's) lines and me reading everybody else's. Although she can not memorize her lines and have to rely on Pinyin frequently, she is getting used to the pronunciation of a lot of words she hasn't learned yet.

The best thing yet, I didn't have to always ask her to do this. She was asking me to do it with her! Because it is fun.

We will continue to use this script and have her learn more characters' lines. I believe she can learn a lot of new characters and dialogs this way.

This experiment also cast more doubt I had with the no-Pinyin philosophy for earlier grades of the Stanford Chinese textbook. While we will use the textbook when we officially start our home schooling this weekend, I think I will have to supply some Pinyin to help my daughter with the text.

Something is Wrong with SATII-Chinese

One of the driving forces of the rising popularity in learning Chinese in the US is its recent inclusion as a subject test in SAT II. Our local Chinese school has boasted that three of their students have received perfect scores with others doing pretty well scores-wise. Sounds great, right?

I was skimming through some Chinese BBS today and happened upon someone complaining that their child has also scored perfect on the Chinese SAT II but is in the 57 percentile. What? 43% of the test-takers are receiving perfect scores?!

A quick Google search confirmed this. There are much talk in online forums about this skewed score distribution and its being worthless: here is one example.

Obviously, it's not a good thing to have nearly half of participants scoring perfect in a test of SAT caliber. Something is wrong here. It's a new test, so there must be a lot of kinks to be worked out. But another reason may be that most people who (bother to) take this test are the ones who are or were native Chinese speakers, at least when they were kids. Or they have substantial upbringing in a Chinese-oriented family environment. They dominated the test and they dominated the scores.

It does cheapen significantly the aura of the "perfect score", whatever it's worth. It leaves others embarrassed for making one or two silly mistakes to be dropped way off in the scoring scale. If it continues to be this way, then the question would really be: why bother?

It will be interesting to see if anything is going to be done about this new test.

Our Baseline: 136 Out of 500

So, we have a list of 500 most-frequently-used Chinese characters, it's time to do another test. I copied the characters directly off this page, and pasted them into Excel. Then I stripped off the Pinyin and rearranged into five columns, each for 100 characters. This prints out into three pages.

Last time when I did a character recognizing test with my daughter, I asked her to mark the characters with "+", "check", and "-", for words she definitely knew, she thought she knew but couldn't tell, and not knowing at all. That method was not entirely effective and she got tired and it went on. So this time we tried something simpler: she just checks off any characters she knew.

Here is the result:

1. From 1-100: known 46
2. 101-200: 27
3. 201-300: 15
4. 301-400: 19
5. 401-500: 29

The total is 136 out of the 500.

As noted in the previous post, this kind of tests has several weaknesses, mostly because it asks her to recognize characters without pronunciation help or context. But nonetheless, I think this is as good a baseline measurement as we could get, as far as learning characters go. She certainly knows some more characters that are not in this list, but probably not much. (In that other test, she recognized 100 characters out of about 300, which was a more limited sample.)

We will come back to this later on as one measure of our progresses.

Word Count, the English Style

Just as I have been obsessing with the usage frequency of Chinese characters, this comes along from the Language Log.

But we are getting off the topic here.