Friday, January 10, 2020

The Foundation of Real Writing

As I've mentioned before, we have the poor fortune to live in a golden age of bad writing instruction. There are a variety of reasons for this, ranging from the rise of high-stakes testing to some less-than-wonderful traditions to the widespread discomfort with writing instruction of many classroom teachers.

The last is probably the worst issue facing writing instruction. It's a curious thing; you don't find many band directors who don't play an instrument or many phys ed teachers who aren't involved in some sort of physical activity, but schools are loaded with folks who teach writing, but who never write themselves.

You can spot the different kinds of bad writing instruction by the foundations on which they are built, by the things that are treated as basic building blocks of writing. Here are the false foundations you're likely to find.

Text and a Question

Colemanism in action, this approach to writing starts with the idea that the only writing worth writing is in response to a text in order to answer a particular question that a teacher or test manufacturer has posed.

Look at the Big Standardized Tests of the past two decades and you find a remarkable phenomenon-- the writing test using multiple choice items. That's possible because one of the premises of this bad writing is that there is really only one correct answer. Every excellent answer to the prompt should use the same text evidence organized in the same way to support the same points; all of the best essay answers to the prompt should be essentially indistinguishable.

The other premise here is that the content of writing must be a reaction to someone else's writing. As with most types of bad writing, there is a germ of truth here-- great critical writing or strong explication do indeed start with someone else's writing. But then the author brings his or her own thoughts, ideas, reality and reactions to the piece. In Colemanism, nothing outside the fabled four corners of the text is allowed. Also in Colemanism, there is only one "true" interpretation of what lies within the four corners.

So this type of bad writing is an exercise in mind reading, a task that involves figuring out what the test manufacturer wants you to say and how they want you to say it, and then performing that response. It has words arranged in sentences and paragraphs, and that creates the impression that it is writing, but the fact that it can be translated into multiple choice questions (What is the main idea? Which of these statements can best be used to support the main idea? etc) is a dead giveaway-- this is not actually a writing task at all.

Structural Outline

An approach most commonly enshrined in the five-paragraph essay. You start by knowing how many paragraphs there will be in the essay; in more extreme examples, you also know before you start how many sentences each paragraph will contain. Which is patently nuts, but here we are.

It's not that the five-paragraph essay is useless; I've taught it myself, particularly early in my career when few of my colleagues were teaching writing at all. If your ideas are a formless soup, a bucket to hold them in can be helpful at first. But the five-paragraph essay is like training wheels; they might be-- might be--useful at first, but they go very quickly from being an aid to being a hindrance.

What you get is a more-involved fill-in-the-blank puzzle. The writer tries to answer the question "What can I use to fill in these paragraph-shaped blanks?' No real thinking about the topic is required.

In the very worst of these scenarios, we find state writing assessments that are "scored" by computer software. As has been shown repeatedly, computer software does not care about your ideas. It does not even care if your details are correct. All it can do is break down structure and vocabulary. Companies trying to sell this baloney will sometimes trot out "research" showing that the software gets the same results as human scorers, but the correct statement is that the software will get the same results as humans, if the humans are trained to score the essays just like software would.

But that's the appeal of a structure-based foundation; it gives the evaluator some simple, clear items to look for and "assess."

Sentences and Paragraphs

The previous two approaches may have been given a special boost by the high-stakes testing baloney, but this school is old, and still much-beloved by edu-experts. Folks like Judith Hochman, who declares in interviews and her many writing instruction guides that the foundation of writing is a sentence.

I don't mean to pick on Hochman, who is just a high-profile example (and enabler) of teachers across the country whose idea of writing instruction is first you do a unit on sentences, then move on to paragraphs, then on to full essays (and it should come as no surprise that the essays are often five-paragraph ones).

This is writing for school, not an authentic task, but a performance of writing-like tasks that is comfortable for all parties. But like the structural approach, it pushes students to start with the wrong question-- what can I write to fulfill this assignment. In its most traditional form it focuses on writing that is error-free, and as such is not so focused on what it does do, but on what it avoids. But mistake-avoidance is not a virtue in writing. First, it encourages timid, safe writing (don't attempt anything that might risk including an error). Second, the absence of mistakes is not the same as the presence of quality. A sports team can make zero mistakes and still lose. A musician can perform a piece with zero mistakes and still be excrutiatingly boring.

This can also lead to a performance tug-of-war, in which students try to figure out how small a technical performance they can get away with and teachers set artificial performance limits to thwart them ("This paper must be two pages long, size 12 font, margins exactly one inch, etc")

Does a writer need control of her technical tools? Sure-- but those tools need to be employed in something other than an empty display of tool use.

The Actual Foundation of Real Writing

An idea. Real writing starts with a deceptively simple question-- what do I want to say? In a school setting, it might be "what do I want to say about this topic" ore even "about this text." After that, we move on to "how can I best say that," a question for which there are hundreds of answers. The writer should pick the answer that best suits her.

Not all students will greet this approach warmly. My usual answer to "How long does this have to be" was "Long enough to make your point effectively, and no longer," which students do not find helpful when they're really asking, "What's the least I can get away with here?" Students will deliver versions of "Just tell me what hoop to jump through and how to jump so I can get my grade and move on." Because part of what's appealing and comfortable about the kinds of bad writing that students get used to in school is that thought is not required; in fact, real thinking about the content can get in the way of the performance you're supposed to give.

But think you must. Figure out what you have to say and how you want to explain it (and, maybe, to whom you want to say it, though focus on audience is overrated) and also what may be most important of all, saying it in your own voice. But the thinking is critical. Most of my students' writing problems were really thinking problems-- a failure to figure out what they wanted to say or how the supports fit into the larger picture (plus those parts that didn't belong, or seemed not to belong because the writer didn't show the connections clearly enough).

Does it make life harder for the teacher? Of course it does, because writing and writing instruction are squishy and messy and it is not possible to impose any standardization without having an effect on the process and product. Compromises are necessary for teacher sanity and survival, but make every one mindfully and conscious of the cost.

If you are serious about teaching real writing and not just the test-based or scholastic tradition versions of a performative writing-like activity, then you have to embrace the mess, open your classroom to more student control, and make your peace with whatever corners you have to cut. I've written elsewhere about some of the basic rules for teaching real writing-- hell, I have part of a book about this, but then John Warner wrote a book that says much of what I have to say and I felt redundant. So until I get wind back in my sails and some publisher offers me a juicy deal, read his book.

What's the point of teaching writing for real? Because of the various types of writing being discussed, it's the one that is most useful in the actual world. Writing for tests is good for tests. Writing for school is good for school, and for places that still pattern themselves on what they remember as "good" writing from their school days. And having a good technical command of structure and other tools is useful if you have a career that involves concealing baloney under a verbal smokescreen.

But for everything else, including, especially, bridging the gap between human beings with actual communication, writing for real is what's useful. It's not a performance or a show or a trick to be performed under artificial conditions; it's authentic, actual, real. Writing instruction is yet another area where teachers have to ask themselves whether they want to do what serves the institution or what serves the students.


  1. I'd argue a lot of what you say applies also to reading, and that we need to raise readers if we want to also raise writers. We don't do that my making them read four articles about The History of Cement. If we can make them love to read, we can get them thinking enough to write. It's a big ask, given the imprint Coleman has bought himself, largely with our money.

  2. For a comparable argument about teaching math as it is actually undertaken, see: