Can the Reading Pendulum Be Swinging?

Two articles and some twitter chatter do not a trend make, but a guy can hope.

One of the worst trends to come out of modern reform has been the repackaging of reading as a content- and context-free series of skills, a series of tasks to be completed in a vacuum. This has always struck me as twelve kinds of wrong. Like preparing an athlete by having them practice movements in a padded room without a team, ball or field. Like having a band musician practice alone with just a mouthpiece. Like trying to tune the audio characteristics of a set of speakers inside an airless chamber. I could go on all day with the different kinds of crazy this is.

And I must be clear-- not all reformsters have been on this bus, ever. Robert Pondiscio hollered for years about "57 most important words" in the Common Core. Here they are:

By reading texts in history/social studies, science, and other disciplines, students build a foundation of knowledge in these fields that will also give them the background to be better readers in all content areas. Students can only gain this foundation when the curriculum is intentionally and coherently structured to develop rich content knowledge within and across grades. Students also acquire the habits of reading independently and closely, which are essential to their future success.

These words appear in a "note on range and content of student reading." They don't carry a great deal of weight compared to, say, the 43-page appendix that spends so much time on "text complexity,"  a term tied to the mechanical content-free conception of reading, suggesting that reading levels are just a matter of how tricky the decoding puzzle is (but unrelated to the actual content itself).

Podiscio and I long disagreed about the importance of those 57 words. The argument of many rich-content supporters of CCSS has always seemed to me to boil down to, "This must be crucial, because without this, the Core is bunk." My argument is that the 57 words are a minor cosmetic adjustment, a feeble attempt to patch a gaping hole, and the Core is, in fact, bunk. The most telling detail-- "rich content knowledge" appears nowhere on the CCSS-related Big Standardized Tests. As I've pointed out a million times, I can prepare my students to get good scores on the BS Test by spending the whole year reading excerpts from the newspaper and practicing BS Test-style multiple choice questions (and never learning anything about writing, but that's a gripe for another day).

Not only do the BS Tests not respect rich content knowledge-- they bend over backwards to negate it, deliberately choosing test items whose content  is highly unlikely to intersect with student prior knowledge (elementary test excerpt about Turkish village economics, anyone). BS Tests are designed to be the higher-complexity version of the dreaded DIBELS test, a test for young readers that tests their reading skills by asking them to read words that aren't actually words (because that way they can't "cheat" by already knowing the word and recognizing it).

But while Pondiscio and I and others have long disagreed about the Core (and charters and a bunch of other stuff), we agree about the importance of content when it comes to reading. Knowledge matters. Content matters. To state what seems obvious to me, it's hard to read or learn to read if you don't know much. It's easier to read or learn to read if you have a wealth of background knowledge. New learning is most easily acquired when it can be connected to old knowledge. So if you want to teach children to read, building up their storehouse of prior knowledge is a critical-- maybe the most critical-- thing you can do to build a foundation.

This point of view has never gone away, but it seems to be gaining traction lately.

Note, for instance, this piece from Louisiana's education chief and committed reformster John White. White's point is that the NAEP is a faulty test because it doesn't measure "what students know."

The trouble is that by not requiring knowledge of any specific book or facts, reading tests have contributed to the false impression that reading is mainly about having skills such as being able to summarize, and not about background knowledge. Walk into many English classrooms today and you will see students capably identifying an article’s main idea. But you’re less likely to find students learning the historical context for a novel or discussing the novel’s broader meaning. By not requiring knowledge, tests create no incentive for particular knowledge to be taught.

This is not fair to adolescents, who need knowledge to become effective adult readers. It’s particularly not fair to students from historically disadvantaged backgrounds, whose summer breaks rarely involve trips abroad or afternoons at museums, and who are thus at a disadvantage on any test that, whether it’s acknowledged or not, measures background knowledge. And it’s not good policy for a nation struggling with the influence of falsified news reports over its citizenry.

Meanwhile, Natalie Wexler took to the pages of the Atlantic to watch a panel ask (and answer) the question, "Why American students haven't gotten better at reading in twenty years." The answer again addresses the very basis of reading instruction:

On a daily basis, teachers have their students practice skills and strategies like “finding the main idea” or “making inferences.” And teachers select books that match the given skill rather than because of the text’s content. Rarely do the topics connect: Students might read a book about bridges one day, zebras the next, and clouds the day after that.

Cognitive scientists have known for decades that simply mastering comprehension skills doesn’t ensure a young student will be able to apply them to whatever texts they’re confronted with on standardized tests and in their studies later in life.

Wexler is only partly right-- students are more likely to be reading short excerpts of those books, rarely reading any entire works at all. But yes-- we've known for a while that this approach is backward. If you want to check out a golden oldie, take a look at the Recht and Leslie baseball study, in which it turns out that low reading skills plus high knowledge beats high reading skills and low knowledge. Here's the chart:

The panel included Dan Willingham, another writer I don't always agree with, but who gets it when it comes to the importance of content.

Current education practices show that reading comprehension is misunderstood. It’s treated like a general skill that can be applied with equal success to all texts. Rather, comprehension is intimately intertwined with knowledge.

As Wexler puts it:

The best way to boost students’ reading comprehension is to expand their knowledge and vocabulary by teaching them history, science, literature, and the arts, using curricula that that guide kids through a logical sequence from one year to the next: for example, Native Americans and Columbus in kindergarten; the colonial era and the American Revolution in first grade; the War of 1812 and the Civil War in second grade, and so on. That approach enables children to make sense of what they’re learning, and the repetition of concepts and vocabulary in different contexts makes it more likely they’ll retain information. Not to mention that learning content like this can be a lot more engaging for both students and teachers than the endless practice of illusory skills.

Yes, that, please.

Why might the pendulum finally swing back from reading "skills"? The answer is hinted at in the two articles [update: and you can add this one to the list]-- White wrote to get out in front of bad test results, and Wexler wrote to explain them. We've been doing reading "skills" for almost two decades, and it simply hasn't worked. Not only that, but it's been shown to not work by tests designed by the principles valued by reading skills fans. In other words, the set up the game according to their preferred rules, and they are still losing.

And finally some folks are asking, "What else could we be trying?" And lots of folks on all sides of the education debates already know the answer.

Content-based reading instruction will never be clean or easily implemented. For one thing, as soon as we start trying to come up with lists of content that "should be" taught, huge fiery debates will break out, and they will never, ever be settled. But we would at least be moving in the right direction (if we really want to get crazy, we could let students choose the content-based direction, thereby harnessing both the power of prior knowledge and the power of interest). Core knowledge cannot be tapped without unleashing and addressing political issues.

But we should not let disagreements about content scare us away from asserting clearly and effectively that content matters, and that it is properly the foundation of reading. Maybe if we keep insisting, the pendulum will finally swing.