On an Internet teachers discussion list, the comment was made with regards to the targeting/non-targeting question in TPRS/CI teaching:

like say “here are my 3 phrases that my kids will acquire today.” No one says that.

No one…except a CI teacher working in Chinese or Japanese. And some other languages. But trust me, we say that all the time. Because we have to say it.

Why? Let’s back up. First, we cannot assume that all languages work like Spanish and French. Spanish is a WYSIWYG language. If you have even a rough idea of what a word sounds like, you can recognize it when you see it in print. Because the spelling system is reliable. You can even accurately sound out words you’ve never heard of.

Russian is a semi-opaque language (for English native speakers). It uses an unfamiliar alphabet. It’s phonetic, once you figure out which of those letters represents which sound. So there is a period of time when, traditionally, learners were taught the alphabet letter by letter, and then and only then would they read meaningful language in context. Until then, it was just “what would this word sound like, based on the rules you just memorized”. Doesn’t sound like something a CI teacher would be pushing, does it? Uncontextualized memorization of rules? Uh-uh.

Chinese and Japanese (Chinese most of all) are opaque languages. No alphabet. Yes, Japanese has two systems of kana that represent the sounds of syllables, but they are not used to represent every word in standard writing. So part (in Japanese) or all (in Chinese) of a text is written in a way that you must know each individual word. Despite all the “easy Chinese characters” hype out there on the internet, there are VIRTUALLY NO parts of characters that are reliable enough in meaning and sound to be of much use to a non-native speaker (beginner OR advanced!) who sees a character he doesn’t already know.

Why is that? Imagine a system in English where we wrote words using a prefix showing what category they belonged to, then wrote a suffix showing APPROXIMATELY what the sound of the word was. So a word like “oatmeal” would be written food-oml; steak would be food-stk; “bitter butter” would be sense-btr food-utr. How could you correctly read the word for “butter” if you didn’t already know — really know — what “butter” sounded like? (And this system does have phonetic clues. Chinese can often be more like saying “mayonnaise” is written food-qrk. Good luck with that one.)

If we want to avoid rote memorization as the basis for reading in Chinese or Japanese, we MUST produce students who are micro-fluent (have an automatic accurate receptive ability of the new language items) so that they have a “voice” to push the recognition of totally non-phonetic text. When the students know the sound of the words and their meaning very, very well (well enough to immediately and accurately recognize them and know what they mean, in any combination, but not necessarily well enough to output perfectly, as would be the case when those words were acquired) they have this “voice”. It makes language sound “right” or “wrong”.

The proponents of “non-targeted input” (and there seem to be more and more) are, as far as I can tell, teaching WYSIWYG languages. I see people pushing teachers of Chinese to do this, but I do not see the “pushers” teaching Chinese themselves, or really knowing how it works. All the CI Chinese teachers I have ever talked to about non-targeted input just shake their heads and laugh. It is simply not possible, unless we want to take a step backwards into the way WE learned to read Chinese — “Here’s a list of 30 words written in the 36 characters that make them up. Memorize them.” Individually. As single words. I challenge you to find a non-native-speaker who went through a typical Chinese program who did not have to learn to read this way. And if you find one, I doubt he would want his students to learn to read the same way, if he really “gets” what Comprehended Input is all about.

And yet, when I say “non-targeted input will not work for Chinese” or “non-targeted input is not the next natural evolution of all of TPRS/CI teaching” or even “circling is not evil congealed into a popsicle”, people accuse me of being divisive. We should not potentially “put off” those new to TPRS/CI by arguing about things that “don’t matter” (they say).

Well, I would alter that statement. There are some things that really don’t matter. Whether or not you choose to have desks in your classroom doesn’t matter at all. It’s a personal preference thing. It has nothing to do with acquisition or literacy in any language. But the choice of targeted (planned language with massive repetition) or non-targeted (unplanned language, no massive repetition, only “natural” repetition over a long period of time) DOES matter if you are not teaching a WYSIWYG language.

If you’re teaching Spanish, it doesn’t matter if your students can come up with the sound of a Spanish word out of thin air. The text is there to provide the sound for them. Heck, Spanish students can even read cognates in text and realize what they mean, because the sound is given to them on the page. Students of non-WYSIWYG langugaes can’t do that. For

Now, maybe folks would be more comfortable “healing this rift” within TPRS/CI by having Chinese and Japanese secede? IMO it’s not a bad thing to admit that we must change and expand “standard practices” for opaque-script languages (Chinese and Japanese), and for some period of time in the beginning for semi-opaque languages (using alphabets other than the Roman alphabet, or very non-phonetic or phonetics very different from English — think Welsh) if we don’t want our students to have to memorize the rules of a system or individual words, and then painfully apply those rules or single words to decoding a line of text. That’s not reading. The results we get in cold character reading relying on the voice that microfluency provides are night and day compared to the memorization route of literacy.

I personally would be delighted to have people recognize that Chinese and Japanese ARE different and must be taught differently than WYSIWYG languages, even when both classes are being done using CI. That’s not true if you only want the spoken language, of course, but if you want to have speech AND literacy. It’s the literacy piece that makes the difference, and IMO if we’re going to include it but do not integrate it into our teaching, we are not serving our students.

So, to the people who are trying to “heal the rift” — let’s just widen it. Let’s break off the non-WYSIWYG languages from the motherland. Amputation for the good of the patient, assuming the patient is something like an earthworm, where both halves will then grow into something complete that suits it completely.