The Missing Link in the Reading Debate

The debate du jour is about reading, begun in the Intelligence Squared debate, continued through Carol Burris's follow-up column, and followed up by literacy expert Russ Walsh. Okay, it's debate in the sense that disagreements between regular scientists and the Flat Earth Society are debates. Over at the Fordham, Robert Pondiscio offered his two cents which Mike Petrilli on twitter called a "debunking" of Burris, which is a generous reading of Pondiscio's work; apparently generosity is a Fordham trait, as Pondiscio says that AnnWhalen "intelligently critiqued" Burris's post, which is an extremely generous reading of a column that boils down to "neener neener, she's a big liar."

The Big Fat Question is this: should students be given reading materials that are at their actual reading level, at the reading level at which they're supposed to be, or at frustration level?

I'll cut to the chase, and give you Russ Walsh's answer which is, I believe, the correct one: all of the above, in a mixture best determined by the teacher who is working with the child.

But let me also explain why this debate is not going to go away, despite the fact that almost everyone involved is talking out of their butt, because there are some huge, Godzilla-sized gaps in our knowledge.

Reading is hard.

Our understanding of how the human brain does language is limited and inextricably bound up in questions such as how we truly connect wit other humans and know and perceive. Layer on top of that reading and writing, highly artificial and constructed versions of language living at the intersection of Knowledge Base Boulevard and Skill Set Street, and you get the most complex human activity, bar none.

So when someone says, "Chris can read," it's such an unparalleled oversimplification that even I, whose stock in trade is illustrative analogies, don't have a really good comparison. "I can play trombone" or "I can play basketball" come close-ish. But only close-ish.

This vast complexity means that whenever we talk about reading we are always either A) trying to squeeze an 800-pound gorilla into a breadbox, or B) talking about one limb of the elephant as if it's the whole African sub-continent. I'm an English teacher, not a reading specialist (the fact that such a thing exists also tells us something), but my wife is working her way to an advanced degree via reading, and looking over her shoulder confirms my belief. Folks who fall into the A trap tend to talk about reading as if unpacking the various layers of meaning in a piece of writing is as simple as unpacking a second grader's lunch box. But it's Group B that really makes a mess.

The elephant's toe

To make the unbelievably complex manageable, many folks simply stare at one tiny part. For instance, as a high school teacher, I will never get over DIBELS, a diagnostic test for which small children are told to read nonsense words, clumps of words that have no actual meaning.

But that's typical of much that goes on in the field-- we try to isolate one part of reading from the vast complex of reading behavior. So let's have students decode sounds that aren't words, or let's have students read short excerpts without any context-- better yet, let's make them boring and unrelatable so we'll know that students aren't tapping into prior knowledge or actual interest.

It's not that we can't learn useful things from the elephant's toe. But if we get so focused in the toe that we chop it off and take it back to the lab where we subsequently discover that it is bloodless and rotting-- well, we've lost the point entirely.

But we're stuck studying the elephant's toe, because that's what we're prepared to deal with. However, all this so far leads us to the hugest, most gigantic hole in all the reading discussions--

WE HAVE NO RELIABLE METHOD OF MEASURING TOTAL READING ABILITY

Look, we don't even know what it means to say, "Pat read Huckleberry Finn really well." Does it mean she could read every word out loud with correct pronunciation? Does it mean that she can recall character names and plot points? Does it mean that she can recognize the use of figurative language? Can she understand Twain's sarcasm and irony? Does she get the jokes and laugh at them? Does she recognize symbolic elements? Can she effectively discuss the final chapters and argue for their effectiveness or lack thereof? Can she understand how social, economic and racial issues in both the past and present context of the book?

All of them? Sure. Now design an assessment that measures all those. And make it something that's scaleable on a national level. And remember-- reading Huck Finn is just one type of reading.

When I started this piece, I was going to wade into all the research and fake research, but it all comes down to a line about "shows improved achievement in reading" which really means "got better scores on standardized test which measured a very narrow slice from the broad spectrum of skills." If you say to me, "We have proof that this approach leads to students who can read better," I am going to ask you what you mean by "better," because I don't think you know in any specific and quantifiable way.

Put another way, reading is a real world activity. The further you get from the real world, the less meaningful your study is going to be. If you lock an elephant up in a tiny cage, you can still learn some things about the elephant, but nothing remotely comparable to studying it in the wild.

If we are concerned about real reading by real humans in the real world as a real tool for real life, most of our research and testing data is junk.


That stuff that students have to do on standardized reading tests? It bears a superficial relationship to actual reading in the same way cybersex resembles actual sex. Some of the terminology and tools are the same, but they're used in ways that in some ways run directly counter to the real life applications.

Common Core mirages

We keep insisting that CCSS requires students to be taught in complex tests at or above grade level. I'll be damned if I can find anything in the standards that actually says that. 

I believe that fans of complex frustration-level reading as an instructional technique see the Core as a great opportunity to beat their favorite drum. But I think, once again, we're seeing the phenomenon of people seeing in Common Core just what they want to see.

About those levels

When we're discussing what level of reading a student should be doing, we need to acknowledge our methods of determining levels range from Good Enough To Get By all the way down to Unspeakably Stupid. Lexiles, for example, have been deservingly ridiculed for their stupid rankings. Ernest Hemmingway is always a go-to writer for these discussions because his language is spare, sparse and simple. A Farewell to Arms has a lower lexile ranking than The Hunger Games.

This is before we even get to issues of reading motivation. Give a student a book about a subject they love, and their passion for the topic will power them through tough reading. Give them a book that completely bores them at grade level, and they will have a terrible time.

Any discussion of what level a sixth grader should be reading assumes that we have an accurate master list of what books are sixth grade level books. We don't.


So what do we do?

We are on such a wrong path right now. What we know how to do-- what we can do very, very well-- is train students to do well on tests while simultaneously insuring that they will forever think of reading as an unpleasant, unrewarding activity that they will never, ever do, unless forced. The only way we could do more effective aversion training would be to give students a painful electric shock every time they touched a book.

I don't believe this is what anybody-- not traditionalists, not reformsters, not even thinky tank guys--wants. I do believe that some folks are so invested in the reformster agenda that they simply can't see what they're doing, but I don't think it's what they want.

Reading instruction will come best from trained and dedicated educators who have developed personal relationships with the students. Some narrow testing data will be useful as a diagnostic tool, but passing a test or proving that the student can and has read-- that can never be the point of the instruction. In fact-- and this is a topic for another day-- I truly and deeply believe that meaningful reading assessment is not scaleable at all.

Meaningful assessment might look like "Find some way to tell or show me what this book means to you" or just "Talk to me about what you read." And because some reading can produce a myriad of legitimate interpretations, any reading assessment with a set answer key will always be looking at the elephant's toe.

Reading instruction is also personal. Only someone who knows the student to know what his interests are, how deeply he is capable of reading and understanding, what prior knowledge he brings to the text, what interests him, how much frustration he can stand before cracking, how much of a "reader" he already is, how much help he needs to decode the text-- only someone who can know and process all that personally can make the right assignment.

Every good teacher knows that you have to meet the students where she is. Every good teacher knows what combination of hand-holding and butt-kicking is needed to move the student forward.

Those who insist that every student must read [only or mainly] frustrating material are not simply wrong-- they're deeply and completely committed to staring at the elephant's toe. They need to take a step back and look at the whole sub-continent. A good place to start would be looking at what experts like Russ Walsh have to say.