Having had the benefit of a nice road trip during which I was able to contemplate the issue I’ve been exploring in the last few posts—the virtues or lack thereof of letting learners figure things out for themselves—I’ve arrived at an unexpected conclusion. The answer to the question of whether this is the ideal pedagogical method, for teaching writing or many other things, is—drum roll—42!
No, seriously, the definitive answer is yes and no. Or, put differently, it depends. Or: on one hand, on the other. Or possibly: sometimes.
A quick recap: I’ve always wanted to learn programming. Told that Python was useful and accessible, I bought a $35 book. Within hours, I was just barely resisting the urge to hurl the book at the stupidly blinking computer screen. The author adopted the “throw them in and they’ll teach themselves to swim (or not)” school at its most extreme. He provided readers with code they were to dutifully copy, producing a simple game called “Find the Wumpus.” I copied, I played, I found the Wumpus. But throughout, I had to puzzle out for myself what different commands meant—for that matter, even how to write and run a command, which was one of the numerous things this author assumed I already knew how to do!
I showed this book to a mathematician friend adept at programming. He told me to go to Louisville and throw it off the Big 4 Bridge. “This is completely wrong. The way to teach programming is to provide short bits of code that illustrate specific commands and functions. Get another book.”
I already had, being a Very Smart Girl. I bought two on my Kindle. I perused the first one. Within just a few screens, I knew what operators were, and what some major ones did. I knew what functions were. (I already pretty much knew what variables were.) I knew the difference between a number and a string! (It’s just a matter of punctuation. If it’s inside quote marks, it’s “text” and it’s a “string,” Ain’t that cool?)
And yet.
I learned how to tell the computer to add 2 and 3 and get 5. I learned how to convert the price of an Apple computer into euros using functions. I learned how many spaces I could insert before a decimal.
No doubt there are people out there who need to do these things. Who want to do them. It was unclear to me why I would want to do them.
Here’s the upshot. The Find-the-Wumpus game, maddening though it was, gave me a sense of what I could eventually do with the various commands and functions—if only he’d told me what a few of them were. The other book told me what they were, but gave me only the most attenuated sense of what they would eventually allow me to do.
For all its value, this second approach, in isolation, felt flat and thin.
Here’s the connection to writing. Everybody in composition studies KNOWS (forgive the shout) that you can make students fill in the blanks in workbooks all day (The dogs has/have two bones) but you can’t answer their sighed question: what for? What’s all this technical vocabulary (“Verbs have to agree with their subjects in number”)* for? What’s it going to help me do?
That’s why letting people who “can’t” write write is so valuable, as Mike Rose told us years ago in Lives on the Boundary. When they’ve got a whole text, however in need of detail and polish, they have something. They’ve produced a product. As a whole thing, it takes on a life. Like Find-the-Wumpus, it gets them from point A of the instructions to point B of how the instructions can generate an actual thing. Having that thing in their hands—wow! I did this—provides a motivation that getting the right answer in the blank never will. It’s like getting through a first draft. “Now I’ve got something.” The detail work to come suddenly has a point.
The issue is that there are certain discrete things they do need to know (what a function is, why the correct answer is “have,” not “has”), because when the “thing” gets more complicated, they need tools for making decisions. Somewhere there’s a perfect (or at least workable) marriage between the detail and the gestalt.
I’m not convinced I ever quite found that marriage. The best approximation for me was the workshop process, similar to what happens in a good writing group. People produce a text. Then they share that text with others who respond as readers, and the technical vocabulary can sneak in as we jointly explore options for solving problems (“Here’s a chance to use parallel construction to emphasize your point”). I did find it hard in a regular classroom to find time for the more technical instruction among the many global points that spurred discussion, but in my writing group, some people specialize in the detail work. In the Python book, an analogue would have been taking apart the game line-by-line to show what each command did at a much less abstract level. Not just what you did, but how the various operators and functions helped you do it.
So, in my view, you have to let people actually produce something at the same time you’re trying to tell them how to produce it. Then they have to take the tools you’ve given them and try them out. That’s the self-teaching part.
*Yep, this ranks as technical. Even for adept native speakers, technical (unfamiliar) vocabulary includes such terms as “subjects” and “verbs.”