I’ve been an ESL teacher for over eleven years here in Taiwan. Several years ago, I wrote an article in the Taipei Times titled “Cram schools are bad for English”. That wasn’t the title I gave the piece; the editors replaced my title with that horrible one. (I don’t recall my title exactly anymore, but it was something along the lines of “How Cram Schools Teach Kids to Speak English–Badly”). I usually archive all my articles here on my website (like this other article I wrote published in the Taipei Times), but I see that I never did that one, so I’m putting it up now, for my own purposes. I also now see that a couple of people wrote letters to the Taipei Times criticizing my article, so I’ll put those up, too, and for my own satisfaction comment about them. If you’re interested in languages and teaching, you might get something out of it, too. Heck, even if you’re not interested in those subjects, you might enjoy it. And if not, well, at least I’ll have my own self-satisfaction.

Cram schools are bad for English

September 5, 2010, Taipei Times

There are many problems with private English cram schools, insofar as the quality of education they offer is concerned. Among others, the fact is that many, if not most, teach their students to speak poor English. There are numerous ways in which this can occur, but there is one in particular that warrants special examination, and that is the common school policy of enforcing a “no Chinese” rule.

The conventional wisdom on this policy is that by not allowing kids from speaking Chinese, they will use English more, and if they use it more, their language ability will therefore improve. This logic is flawed because it assumes that students will improve their English by being forced to use what little vocabulary they have to create phrases or sentences. In practice, what this policy does is reinforce bad habits.

Kids are smart. So when you tell them to speak only English, they will use their limited vocabulary and apply those words to the language framework that is already firmly embedded in their minds — Chinese.

Thus, one hears kids making sentences like, “This is what,” “Here have a book” or “I don’t have go.”

In their ingenuity, they have used English words to speak Chinese (not to be confused with speaking English). And what else could they be expected to do?

By forcing kids to speak before they have established a foundational framework from which to properly invent their own phrases and sentences, constructions like these are repeated and reinforced, so that the “Chinglish” itself becomes the structural basis for their understanding of English.

Once such bad habits are developed — which happens very quickly and easily — it is extremely difficult to get students to “unlearn” what is firmly implanted in their brains. To compound the problem, teachers then waste valuable classroom time trying to fix the problems the school itself created in the first place — which often proves to be a futile task.

The way to avoid this problem is to provide students with a proper framework well before they are ever expected to invent their own conversational English.

This is not to say kids should be discouraged from using English. However, the distinction must be made between the expectation that students use specific sentence patterns they are taught in the classroom and the expectation that students create their own sentences in conversation.

The former expectation is reasonable and can reinforce good speaking habits; the latter is illogical and only reinforces bad English.

Students must first have a great many hours of active listening before they should be expected to speak in a creative capacity.

This should make sense if one thinks about how people learn their own mother tongue. Children learn to speak their native language at a very early age not because they are taught phonics and grammar, but because they are constantly exposed to it and figure it out on their own, for the most part, through contextual clues. This is the basic concept behind the principle of language immersion.

By the time children whose first word is “ball” actually manage to utter the word, they must have heard the word numerous times. Moreover, they must have heard the word in a meaningful context to have figured out the meaning on their own. They must have seen that round thing and heard that sound together enough times to be able to deduce that the sound is the name for the thing.

By such means, children build their vocabulary. Similarly, they do not start speaking in sentences because they have magically guessed how to correctly put words together using proper grammar, but because they have repeatedly heard those patterns used before in various contexts.

The only creating a child is doing at this stage is in swapping various vocabulary words to use that pattern in different contexts. Not until the child has firmly established a proper framework can the process of inventing original sentences truly begin.

In the classroom, it is better to allow students to answer properly in Chinese than to force them to answer improperly in English. At this point, what is important is that they heard the question in English, understood it and are able and willing to answer it.

Discarding the “no Chinese” rule — or at least scaling it back to a loose guideline applicable to only certain situations — would not only help prevent the formation of bad habits, but would also prevent students from becoming unnecessarily frustrated and discouraged.

The other reason some schools enforce this rule is that many foreign teachers themselves do not have enough knowledge of Chinese to be able to do their jobs effectively if the children speak in their own native tongue. However, this obstacle is probably not as difficult to overcome as it may seem.

First, it is often easy to comprehend what a child is trying to tell you even if you do not understand the words coming out of his/her mouth (teachers deduce meaning based on nonverbal and contextual clues just as their students do).

Second, teachers would, as a direct result, rapidly acquire enough language ability to be able to understand their students (which is to say that the teachers would also need to be students).

Third, in cases where a student’s English is not good enough and a teacher’s Chinese is not good enough for the two to effectively communicate, schools already have Taiwanese teachers or assistants available to help out — and teachers would probably discover that instances where they deem it necessary to employ translation help would be relatively few.


John Coomber from Richmond, Canada, didn’t like my article much. The Taipei Times published his letter on September 7, 2010. Here it is, with my comments interspersed:

Language lessons

It is difficult to know quite where to begin with Jeremy Hammond’s article, so convoluted is it in its attempts to rationalize why Taiwanese students speak incorrect English (“Cram schools are bad for English,” Sept. 5, page 8).

One would think an experienced English teacher would know better than to so misuse the word “rationalize”, not synonymous with “explain” in this context.

However, first, as a point of departure, let me say that I do agree with the idea of exposing students to lots of listening (and reading later) — comprehensible input is a very powerful learning tool.

Also, I am not a fanatical “English only” enthusiast and I do not subscribe to the view that kids (or any learners for that matter) should be forced to speak. However, forcing learners to speak and insisting they try and use English when they choose to speak are not quite the same thing.

Yes, in the context of the “no Chinese” rule enforced in cram schools, they are the same thing. And if John agrees with me that kids should not be forced to speak English (or to not speak at all), then where is his disagreement?

Beyond these two points, it is not easy for me to find much common ground. I notice Hammond uses terms like “reinforce” and “bad habits,” which hark back to the behaviorist ideas of language learning, the importance of which have somewhat diminished over the past few decades.

If Hammond wishes to cling to the notion of language learning as simply a process of habit formation, he is free to do so, but he risks advocating some ideas that fly in the face of more contemporary views.

I did observe that kids form bad habits, but where did I say that language learning is “simply a process of habit formation”? Is John denying that students form bad habits in their use of English? Does he consider this an impossibility?

Now let’s see if I have his central idea correct. Children who are made to speak only English may make mistakes because they transfer aspects of their first language onto the target language. Well, what a surprise!

Yes, an obvious thing. And yet schools continue to make the mistake of forcing kids to constantly reinforce those mistakes with their “no Chinese” rules.

It has been known for some time that learners have difficulty organizing second language knowledge into coherent structures, and it is also generally accepted that this is a consequence of how people get new languages; it is an inevitable part of the learning process, and as such should be expected in the classroom.

Expected, yes. Reinforcedno.

The kind of utterances that Hammond quotes as examples of bad habits are heard around the world (subject to local variation) and can be frustrating for teachers, but as he rightly notes, they are examples of ingenuity, and in my view should be welcomed.

Ingenuity welcomed, yes. Incorrect English reinforced, no.

However, that does not mean such utterances need to be rewarded, yet that is what Hammond seems to be suggesting with his reference to reinforcing bad habits.

Where did I suggest that? John doesn’t seem to have tried very hard to understand what I wrote. He seems rather to have made quite an effort to deliberately misunderstand it.

Perhaps he is referring to teachers who fail to correct and offer appropriate models to learners.

Nope. I’m referring to schools who punish kids for speaking Chinese, thus forcing them to speak incorrect English — that is, to use English words to speak Chinese, which is the effective result of the rule.

I am not sure, but let’s stay with this idea of bad habits for a moment. Better, says Hammond, to allow learners to answer correctly in Chinese rather than incorrectly in English.


Wait a minute — is responding to English with Chinese appropriate language behavior?

Certainly. Why wouldn’t it be? It’s called communication.

Doesn’t this count as implanting bad habits into their brains?

No, it doesn’t count as bad habit formation. Is John acknowledging that it is possible to form bad habits? Or is he guilty himself of harking back to the “behaviorist” ideas of language learning (whatever that is)?

Hammond is silent on how much valuable classroom time is taken up in unlearning these particular bad habits, but one is led to assume that there comes a time when a magical transition from perfect Chinese answers to flawless English ones takes place.

No, one is not led to assume that there is any kind of “magical transition”. On the other hand, one is forced to assume from the logic of the “no Chinese” rule that kids will just magically be able to actually speak proper English. As I stated in my article, children “do not start speaking in sentences because they have magically guessed how to correctly put words together using proper grammar, but because they have repeatedly heard those patterns used before in various contexts.”

Some people have devised teaching methods based on the belief that learning a second language has similarities with acquiring a first language, but Hammond’s analogy is unfortunate because it glosses over a number of inconvenient facts that fail to support his point.

I agree that young children appear to work out grammar by themselves, but as well as exposure, they need opportunities for interaction and at these times they make mistakes and they continue to do so for a number of years.

In fact, evidence suggests that children master first-language grammar through a process of trial and error, a process that would be proscribed in Hammond’s classroom for fear of ­promoting bad habits.

So advocating that kids learn a second language by trying to figure out how to speak through trial and error is better than them learning by listening to the language first, figuring out meaning through context clues, and attempting to speak it only after they have listened long enough to have an idea of the basic framework? Now that’s just crazy.

I am afraid I don’t understand how anyone can say that children cannot be creative and original until they have a firmly established framework, unless these words are used as metaphors for “correct.”

I couldn’t understand how anyone could say that, either. I certainly didn’t.

Last, I want to consider Hammond’s reasons for “no Chinese” rules. I cannot agree with the implication that teachers need to speak some Chinese to be effective.

I couldn’t agree with such an implication, either. I certainly didn’t imply that. In fact, I pretty explicitly argued just the opposite, that an ability in Chinese is by no means necessary for ESL teachers to do their jobs effectively, that “this obstacle is probably not as difficult to overcome as it may seem”. How was that in any way unclear?

If this logic were applied to the wider context of language teaching, the only people employed in the English-as-a-second-language industry would be those with a knowledge of three or four relevant languages. This is clearly not the case.


I find his last point particularly telling: We may find that we don’t actually need the assistant to translate for us very often, he says. Excellent!

This looks like an opportunity to use language for what it is intended for: as a tool for communicating and negotiating meaning — if we could only endure the “bad habits.”

Why should we “endure” bad habits when we could just very simply prevent their formation in the first place? At least we agree that language is intended for communicating.

One last thing — there is at least one more reason for enforcing an “English only” policy: Publicize it and it becomes a pretty powerful marketing tool.

I find his last point particularly telling. This is indeed the bottom line. It’s a policy that attracts business because parents think if they put their kids in a school where they are not allowed to speak Chinese, it will help them to learn English quicker, when, in fact, what it does is to foster bad English. Hence my writing that article to the Taipei Times to warn them against that fallacious thinking.


Another person, Nathan Lindberg, also wrote a letter in reply to my article, which was published on September 9, 2010:

‘English-only’ rule helps

I cannot agree with Jeremy Hammond’s article (“Cram schools are bad for English,” Sept. 5, page 8). Hammond claims that the “no Chinese” rule enforced at some cram schools is essentially bad. Like many good ideas, the rule can be used improperly, causing students to resent English, but by itself I cannot see it leading only to bad grammar. In fact, I think it can be an important step to fluency.

Nathan cannot see how forcing children to use English words to speak Chinese (since they don’t actually know English) could cause them to reinforce bad grammar habits? Is this so hard to understand?

Hammond claims that forcing students to communicate in English too early will cause them to resort to their native language’s grammar.

Hammond “claims” this? Is Nathan denying this rather common sense observation?

I agree with John Coomber (Letters, Sept. 7, page 8); “forcing” students into speaking seems a bit strong, but discouraging them from using a language because they might use it improperly would discourage anyone from ever speaking a foreign language.

Where did I say anything about discouraging kids from speaking in English? I didn’t, of course. In fact, I explicitly stated: “This is not to say kids should be discouraged from using English”.

Hammond seems to be under the impression that language is learned in neat chunks. Once one feature is perfected, a student can move on to the next. However, no one learns this way. We all make mistakes when learning a language, mixing in our mother tongue’s grammar and butchering the language we are learning. It’s called “interlanguage” and it’s very natural. As students progress, they will notice their mistakes and eventually correct them.

I don’t know what Nathan’s first three sentences in this paragraph even mean or where he is getting that from my article, but his claim that students eventually correct their mistakes is dead wrong. As I noted, “Once such bad habits are developed — which happens very quickly and easily — it is extremely difficult to get students to ‘unlearn’ what is firmly implanted in their brains.” Ergo my following statement that “The way to avoid this problem is to provide students with a proper framework well before they are ever expected to invent their own conversational English.” Nathan apparently thinks this is a bad idea, that students should just go ahead and develop bad habits, and just correct them later. How that is supposed to be better is beyond me.

Since mistakes are natural when learning a second language, speaking can be intimidating. No one wants to speak incorrectly and be laughed at. As a student of Chinese, I know very well the experience of trying to say something serious and succeeding only in making everyone laugh. That is why using an English-only rule should be done carefully. Starting in small increments with controlled language can build students’ confidence. Teachers might teach a simple question such as “When is your birthday?” Then have students ask each other and record their answers. However, different situations demand different tactics.

I have never heard of a “No Chinese” rule in a buxiban (cram school) being implemented “in small increments”. “No Chinese” means “No Chinese”, period. As for the rest, once again, Nathan seems to be deliberately misunderstanding my point. What did I write? I explicitly stated: “However, the distinction must be made between the expectation that students use specific sentence patterns they are taught in the classroom and the expectation that students create their own sentences in conversation. The former expectation is reasonable and can reinforce good speaking habits; the latter is illogical and only reinforces bad English.” Was that not clear enough for Nathan to understand?

I taught English immersion kindergarten for six years. By the second half of the year, I required all my students to speak only English during the three daily classes I taught them. None of them had any problem, and today some are fluent, attending university or even studying overseas. Yet I must admit that I have met some of my former students who speak in “Chinglish,” with fossilized bad grammar habits. I do not blame this on too much English-speaking, but on a lack of it.

I don’t care how excellent a teacher Nathan might be, but to say that none of his kindergartners had “any problem” speaking English is just not a serious statement. Maybe he means that they had no problem answering “My name is (insert name)” in response to his “What is your name?” question. Wonderful. Now, what was that about his students speaking “in ‘Chinglish,’ with fossilized bad grammar habits”? I guess, then, some of them did have some problems, after all. He blames this on not having spoken enough English. This illustrates the fallacy of Nathan’s thinking perfectly. It is not a problem of too much vs. too little English speaking, it is a problem of reinforcing correct vs. reinforcing incorrect language usage. This is the central point that Nathan seems unable to comprehend.

I saw my students enter the public school system and have their speaking and listening skills deteriorate. It is easy to see why. I monitored one junior high school class in which the teacher spent almost the entire 50-minute class explaining the finer points of the past perfect tense in Chinese. Surely this is what necessitates English-only time in buxibans. Nobody has ever improved their spoken second language skills by listening to their first language.

Nathan has an excellent point. The public school system is also a big part of the problem. But to argue that this “necessitates English-only” in private cram schools is a non sequitur. Yes, public schools contribute to the problem, but this does not exonerate cram schools, and the case I am arguing is that the “No Chinese” rules of the buxibans is not a solution, but also part of the problem. I agree with Nathan that nobody has ever improved their second language skills by listening to their first language, but don’t see how that statement is relevant at all to any point I made in my article.

It is true that harsh punishment and requiring students to speak far beyond their level will only discourage students and make them resent English. However, done with encouragement and perhaps a token punishment, an English-only rule can be the firm push students need to use English. Often, for a deterrent, I have students make their own contract stating they will only speak English all class or during specific times. They also write their own punishments, which sometimes gets a little silly, but keeps the class good-natured, as it should be.

How is punishing kids for speaking their own language, instead of just not demanding them to try to use English words when the communicate with their Chinese grammar, a good thing, regardless of whether the punishment is “token” or “harsh”? Notice also that his “deterrent” logically implies that his students may at other times speak Chinese rather than English. If that was permissible at whatever school Nathan worked at, then it was by definition not an “English only” school, and hence his comments in that regard are completely irrelevant to the point I was making in my article.

Pin It on Pinterest

Share This