Intro

As I’ve said before, I’ve been homeschooling my kids for a long time now. Some subjects are fun to teach to them, and some are not. For the short time my older son was in public school (first grade), he was being taught writing in the way that seems fairly typical for public schools: he would be assigned some topic to write about, and given a length requirement (whether minimum or maximum or both) and told to write. In my experience, this way of teaching writing is frustrating for a lot of kids and a lot of them don’t ever “get it”.¹ My son found this task nigh impossible, and I found it baffling how he found it impossible when he could talk for hours and hours, very fluently, about the proposed topic. He clearly had thoughts about it, so what was preventing him from being able to write them down?

I’m going to use a recent viral tweet thread as a jumping off point to summarize here what I’ve learned about teaching kids to write.

Learning to Draw

The central thesis of that thread and the subsequent blog post is that copying other people’s artwork is a useful and reliable way to learn how to draw. The author notes that children in Japan are very proficient in drawing, without the drop off in the progression of drawing ability that is often seen elsewhere.

I taught in Japan, in public elementary schools, for a couple of years, and I noticed at the time that it is taken for granted that everyone can draw to some level of proficiency and that the style of the drawings done by most Japanese schoolchildren (and, often, their teachers) is very heavily influenced by manga and anime styles. They took it for granted that I would be able to draw and thereby create many of my own teaching materials such as flashcards and the like. So, I had to learn to do that, and I did it by doing the same things I saw my students doing: copying anime and manga. I wouldn’t say that I ever got really good at it, but I became proficient enough over the course of my teaching there that when I drew a “frog” flashcard to teach the 3rd graders the word ‘frog’, they could easily recognize it.

I have in the past summarized this as Americans tend to think of drawing as a talent while Japanese people tend to think of it as a skill. If something is a talent, some inborn you-either-have-it-or-you-don’t quality of you as a person, then teaching or copying or what have you isn’t going to help very much, and in my view, Americans treat too many subjects like this: drawing, writing, math. Viewing these as skills instead of talents means you believe that you can teach them, or at least, that they can be learned to some degree of proficiency. But, as Neil Cohn mentions in the thread and accompanying blog post, Japan has been much less subject to the pernicious influences of Rousseau.²

Handwriting by imitation

When we talk about learning to write, we typically mean two different things. The first is learning handwriting, whether printing or cursive or both, a subject that we teach exclusively (as far as I know) through having students copy letter shapes and word shapes in notebooks designed for the purpose. We give them paper that has squares and lines designed to help them get the proportions of letters correct, to teach them to make some letters rise above the midpoint of the line and some to stretch below the line. And to a first approximation, every student who is taught to do this learns to do it, although it doesn’t always come immediately or easily to every child. Many students make certain letters backwards, or confuse ‘p’ and ‘q’, and the like, sometimes for many years.

Perhaps more surprisingly, these kinds of mistakes are not very susceptible to correction. In fact, it isn’t clear that any errors children make in language acquisition are susceptible to correction. That is, you can correct a child making his ‘J’ backwards or saying “I sleeped” until you’re blue in the face, and it will have little impact on how long it takes the child to stop making the mistake.

And this isn’t only true of children. I have also taught English as a second language to adults and the mistakes adult language learners make when acquiring a second language, spoken or written, are also not very susceptible to correction! It is probably the case that very little of our linguistic skill is what you would call conscious and hence it is nearly impervious to conscious attempts to correct it. But imitation of the correct forms does, over time, correct even our worst habits.

Learning to speak your native tongue, learning to draw, learning handwriting, learning a second language all seem to have in common that the best way to learn to do it right, to develop basic proficiency if not creative and expressive genius, is imitation. I think in America that’s somewhat controversial for learning to draw but not very controversial for the others.

The copybook method

The other thing, the more usual thing, we mean when we talk about learning to write is learning how to make sentences, paragraphs, whole essays, on paper, and this, because we view it as a creative or expressive activity, is where we depart from what we know about learning handwriting. Here we think we need to give students a prompt and let them express themselves, and many teachers hesitate to offer too much guidance — except after the fact when they return papers marked up to “correct” everything that the student did wrong.

This is not the right way. This is what was happening in my son’s one year of public schooling. This was what we had to correct for once we started homeschooling in earnest.

In something like desperation, I looked around for another way. I didn’t want my son to hate writing — or, to be honest, I’m not sure I cared about that exactly, but I did have old-fashioned notions that it was one of the handful of skills every person should learn to do reasonably well. It’s fine if he’s not going to become a novelist; it’s not fine if he can’t put together a coherent paragraph.

So, we started using a copybook method. The basic idea is simple: students should literally copy good writing until the ways that punctuation, capitalization, and written grammatical structures appear on the page become completely habitual for the student, as we do with handwriting.

A typical lesson would go like this:

1. Read a short passage or poem to the child.
2. Ask them a series of questions aimed at getting them to summarize the main point of that passage. For very young children, you want short passages with a theme that can be summarized in a single sentence. Fables work particularly well for this, but descriptive passages from favorite books can also work. For older children, you want summaries to be longer, but you should still help guide their summaries by asking leading questions.
3. Write their summary down for them. Write it correctly, saying the correct version aloud as you write.
4. Have them copy what you wrote. Watch them while they do it and if you sense that anything is a struggle for them help them do it correctly rather than letting them make a mistake and then correcting them. This is a very important part of this process.

Over the course of the next few days, review their summary until they can write it correctly as it’s being dictated to them, where they now have to get it right without being able to see it. Again, help them where they need help to get something right rather than letting them write down something that is wrong.

For adults, this process is often tedious, and we adults typically think that since we find it tedious, the children will, too, and therefore it is bad and will teach children to hate writing. I’m not going to tell you that children do not find this tedious; I think some do and some don’t particularly. There is still a lot of novelty in it for young children, because this is a new experience for them, and many children are enchanted by the whole idea of reading and writing because they often understand what we have forgotten, that it is a remarkable, almost magic, thing that we invented. At least until they get disenchanted from it by teachers’ red pens.

What I am going to tell you is that you have to stop worrying about what children do and do not find tedious. These lessons are short; they will rarely take more than 10 minutes a day, at least until they are teenagers, and children will usually tolerate a tedious thing that takes a short time and only ever results in positive feedback and growth, particularly when the lessons are so carefully structured that you can guarantee that they will be short and neat.

There are books structured to help you do this³, with teacher scripts and pages for student writing included. Or you can do more of the planning yourself and use an ordinary notebook, although if you’re working with young kids who are also still learning to make letters correctly, I recommend a notebook that has guide lines for relative letter heights.

In part this is successful because of the literal physical copying, and the muscle memory and fluency in how good writing looks on the page, but the fact that the content of their first writing is also “copied” (being summaries of what others have already said) matters, too. It keeps the originality and novelty of the lesson tightly constrained.

Writing in this sense, as opposed to handwriting, is really several different tasks, including (although perhaps not limited to):

• The physical process of doing it;
• Getting the grammar and visual structure of written language correct;
• Expressing ideas.

The copybook method focuses on the first two of those and doing them until they are mechanical, and neglects the third. It’s the third one that people who already know how to write find useful and interesting, which is what makes this process seem tedious to us. But that constraint, limiting the amount of new stuff we expect children to do and learn at a given time, gives them the confidence-building freedom to master skills before moving on to increasingly more difficult and interesting tasks.⁴

Some advice for adults

As I said above, I used to teach adults how to read and write college-level English as a second language; now I teach adults how to read and write the Haskell programming language. And one of the most difficult aspects of my job is that adults don’t want to do things they have decided in advance are tedious. Programmers in particular think they are much too clever to need professional teachers and rote exercises. Much of the success of my first book was simply that it was full of exercises intended not to challenge and thrill experienced programmers but to force them (if they would do the exercises) to type things into their text editors until the grammar of the Haskell language sticks with them, until the way Haskell looks on the screen becomes a habit rather than something they have to google every five seconds.⁵

Doing the wrong thing and “beating your head against the wall” doesn’t work very well for learning. Correction just doesn’t work very well as a teaching tool. Maybe that’s how you learned; many of us did, just like many of us were spanked as children and yet we can still be open to the possibility that corporal punishment is neither necessary nor ideal as a child-rearing strategy. As Chris said, we’re humans and we can come up with better ways.

What works instead is doing the right thing, and doing the right thing over and over until it’s a habit, experiencing that positive feedback every time you do the right thing. Whatever you want to learn, copy it until it sticks.