The time-consuming test prep with endless exercises on “finding the main idea” and “questioning the author,” exercises that are supposed to help improve verbal abilities, have become the chief cause of today’s curriculum narrowing.  Paradoxically then, emphasis on reading and reading tests have helped to cause low reading scores among school leavers.

 E D Hirsch recently in the Washington Post

The focus on the “skill” of reading has produced students who cannot read. Teachers cannot cultivate reading comprehension by forcing children to practice soul-deadening exercises like “finding the main idea” and “questioning the author.” Students would be better off gaining knowledge by studying real subject matters in a sensible, cumulative sequence. Instead, elementary schools are dominated by content-indifferent exercises that use random fictional texts on the erroneous assumption that reading comprehension is a formal skill akin to typing.

E D Hirsch recently in the Wall Street Journal

Over the years I have grown increasingly tired of Hirsch’s one-note samba about reading.  But I’ve kept my peace because the reading wars are endless, polemical, and easily bog one down in foolish debates. But these latest posts are just too over the top for me to remain mum.

For those unfamiliar with Hirsch’s critique, the only thing he thinks that truly matters is content knowledge. All of our ills – in reading, in learning generally, and in civic life – come down, in his view, to a failure of schools to teach a core and standard set of content.

But Hirsch now goes further, as the recent quotes reveal: he criticizes as unjustified any attempt to teach reading comprehension skills (and even critical thinking). Why? Because there are no such thing as “general” reading and thinking skills. He proposes, rather, that all understanding of text is almost exclusively content-dependent.

Yet, even his UVA colleague Daniel Willingham – whose research Hirsch often cites –  flatly disputes Hirsch’s claim:

The National Reading Panel conducted a comprehensive review of all of the 481 studies on reading strategies published between 1980 and 1998…. The Panel concluded that eight of the 16 strategies “appear to have a firm scientific basis for concluding that they improve comprehension in normal readers…. But make no mistake, when standardized reading tests are used, there is still a positive effect of teaching students reading strategies, and the effect is not trivial.

In fact, Hirsch’s position leads to an absurd conclusion (an echo of the famous paradox in Plato’s Meno, that you can never learn anything new): no merely “skilled” reader can be expected to make adequate meaning of a book on a topic with which they are unfamiliar. It’s as if no transfer of learning is possible.  (Hirsch is not strong on the topic of transfer; see below).

Main Idea. His constant bashing of teaching the skill of finding Main Idea is truly odd. (He uses this sentence or a variant of it in every piece of writing I can find by him in the last 10 years: “endless exercises on ‘finding the main idea’ and ‘questioning the author’, exercises that are supposed to help improve verbal abilities”). Yet, ironically, as I have long argued, a close look at all released test items on state, national, and international tests is that students are consistently terrible at identifying main idea and author purpose. On average, students only get such questions correct 50% of the time, in my review of released standardized tests. The results reveal over and over again that students cannot identify the key assumptions and conclusions – the main ideas that shape the text. They have great difficulty distinguishing key facts in the text from the (inferred) idea; they are too literal in their reading. Indeed, that’s a key reason why the Common Core stresses as it does the importance of Argument. And that’s why I support an increased emphasis on non-fiction in the new Standards.

Here are some examples:

Main idea FCAT 9th gr 2010-01-16 at 5.25.09 PM Cherry Blossom Main Idea FCAT





Consider Hirsch’s criticism: it’s as if he ignores the fact that argument has two components – fact-based evidence and logic. What the reading comprehension skills all provide, ultimately, are analytical approaches for piecing together the logic of a text. No one denies the role of factual knowledge in comprehension. (Would Hirsch please quote someone who does deny it, instead of setting up his straw man?) What the reading literature says and the reading Panel concluded is that vocabulary, background knowledge and reading strategies all work together to frame meaning. Yet, Hirsch insists on setting up his straw man:

We are told by reading specialists that the road to improved comprehension is through mastery of comprehension skills such as classifying, questioning the author, and finding the main idea. Specific content is secondary. Any appropriate, “authentic” content, it is said, will build vocabulary and develop comprehension abilities. This how-to conception of reading dominates current thinking; but we cannot make significant progress in reading until this conception loses its power over us. Children who lag behind are being subjected to endless practicing of strategy skills such as “finding the main idea.” Their slow progress induces our schools to add still more time to the literacy block—up to three hours a day in many places—during which time students practice empty exercises on trivial fictions that subtract from time that could be devoted to the substantive knowledge actually needed to gain reading comprehension.

I have highlighted the ‘straw’ to underscore how surprisingly (and ironically) un-fact-based his critiques often are. Who are these “specialists”? He never names them. What evidence suggests that this view “dominates” current thinking? He doesn’t ever say.

What the National Reading Panel said. More to the point, by his argument, it seems as if we must conclude that all failures of misunderstanding a text relate only to factual knowledge. But over and over in academic and civic life we see failures of logic and analysis that have little to do with how much factual knowledge is possessed by the reader or arguer.

Let’s look more closely at what the Reading Panel actually said (and which Willingham quotes approvingly):

In its review, the Panel identified 16 categories of text comprehension instruction of which 7 appear to have a solid scientific basis for concluding that these types of instruction improve comprehension in non-impaired readers. Some of these types of instruction are helpful when used alone, but many are more effective when used as part of a multiple strategy method. The types of instruction are:

  • Comprehension monitoring, where readers learn how to be aware of their understanding of the material;
  • Cooperative learning, where students learn reading strategies together;
  • Use of graphic and semantic organizers (including story maps), where readers make graphic representations of the material to assist comprehension;
  • Question answering, where readers answer questions posed by the teacher and receive immediate feedback;
  • Question generation, where readers ask themselves questions about various aspects of the story;
  • Story structure, where students are taught to use the structure of the story as a means of helping them recall story content in order to answer questions about what they have read; and
  • Summarization, where readers are taught to integrate ideas and generalize from the text information.

In general, the evidence suggests that teaching a combination of reading comprehension techniques is the most effective. When students use them appropriately, they assist in recall, question answering, question generation, and summarization of texts. When used in combination, these techniques can improve results in standardized comprehension tests.

This is a perfectly reasonable set of conclusions (and so were the proposals for follow-up study). But look at how Hirsch spins it:

The truth is that the recommendations from the NRP report about metacognitive strategies are misleading. The NRP report is highly incomplete in the very area, comprehension, on which so many sterile hours are being spent by the schools on so many fragmented and educationally trivial stories. The research citations in the NRP report ignore or deemphasize important studies that have established a central finding about reading comprehension—that the possession of relevant prior knowledge is the single most potent contributor to the comprehension of a text. The lack of relevant prior knowledge will hinder comprehension, no matter how many long hours a child has spent learning to monitor, question, or summarize.

Who said otherwise? What is misleading? Who said that background knowledge is not a factor? Where is the research that shows that most hours are spent on “sterile exercises” using “trivial stories.”? What specific research shows that prior content knowledge is “the single most potent contributor” to text understanding? In fact Willingham summarized the matter far more cautiously than Hirsch lets on:

Some researchers have suggested that prior knowledge is so important to memory that it can actually make up for or replace what we normally think of as aptitude. Some studies have administered the same memory task to high-aptitude and low-aptitude children, some of whom have prior knowledge of the subject matter and some of whom do not; the studies found that only prior knowledge is important (Britton, Stimson, Stennett, and Gülgöz, 1998; Recht and Leslie, 1988; Schneider, Korkle, and Weinert, 1989; Walker, 1988). But some researchers disagree. They report that, although prior knowledge always helps memory, it cannot eliminate the aptitude differences among people. Since everyone’s memory gets better with prior knowledge, assuming equal exposure to new knowledge (as in a classroom without extra support for slower students), the student with overall lower aptitude will still be behind the student with higher aptitude (Hall and Edmondson, 1992; Hambrick and Engle, 2002; Hambrick and Oswald, 2005; Schneider, Bjorklund, and Maier-Brückner, 1996). In the end, the issue is not settled, but as a practical matter of schooling, it doesn’t matter much. What matters is the central, undisputed finding: All students will learn more if they have greater background knowledge.

This whole either-or thinking about skill vs knowledge in reading leads Hirsch into silly territory about testing:

The solution to the test-prep conundrum is this: First, institute in every participating state the specific and coherent curriculum that the Common Core Standards explicitly call for. Then base the reading-test passages on those knowledge domains covered in the curriculum. That would not only be fairer to teachers and students, it would encourage interesting, substantive teaching and would over time induce a big uptick in students’ knowledge — and hence in their reading comprehension skills. That kind of test would be well worth prepping for. [emphasis added]

Now there’s a straw man if I ever saw one! Exercises on main idea are “soul-deadening” and “sterile” but teaching content knowledge will “encourage interesting, substantive teaching”! Has Hirsch not set foot in classrooms? Everywhere one turns there is endless and disengaging marching through content knowledge.  Cf Ferris Bueller’s Day Off. There is poor transfer of learning of both content knowledge and skill because the pre-dominant pedagogy is didactic teaching and passive inflexible learning instead of opportunities to investigate questions, apply prior learning to new situations, and reflect on what worked and what didn’t.

On transfer. In fact, Hirsch’s views on transfer of learning are downright puzzling, as I suggested above:

Reading, like riding a bike, is an ability we acquire as children and generally never lose. Some of us are more confident on two wheels than others, and some of us, we are told, are better readers than others…. When you think about your ability to read — if you think about it at all — the chances are good that you perceive it as not just a skill but a readily transferable skill…

The message has not yet reached American classrooms. A stubborn belief in reading comprehension as a transferable skill combined with the immense pressures of testing and accountability results in ever more time being wasted on scattered, trivial, and incoherent reading.

Think of reading as a two-lock box, requiring two keys to open. The first key is decoding skills. The second key is oral language, vocabulary, and domain — specific or background knowledge sufficient to understand what is being decoded. Even this simple understanding of reading enables us to see that the very idea of an abstract skill called “reading comprehension” is ill-informed. Yet most U.S. schools teach reading as if both decoding and comprehension are transferable skills. Worse, we test our children’s reading ability without regard to whether we have given them the requisite background knowledge they need to be successful.

Why aren’t meta-cognitive and other comprehension-aiding abilities mentioned as a third key – which was clearly the view of the National Reading Panel? Because it doesn’t fit Hirsch’s narrative.

Transfer never magically happens. Not even Hirsch’s precious background knowledge transfers unless there is practice, feedback, and tasks that demand such transfer. Here’s what the National Reading Panel had to say on transfer:

With respect to the scientific basis of the instruction of text comprehension, the Panel concludes that comprehension instruction can effectively motivate and teach readers to learn and to use comprehension strategies that benefit the reader.

These comprehension strategies yield increases in measures of near transfer such as recall, question answering and generation, and summarization of texts. These comprehension strategies, when used in combination, show general gains on standardized comprehension tests. Teachers can learn to teach students to use comprehension strategies in natural learning situations. Furthermore, when teachers teach these strategies, their students learn them and improve their reading comprehension.

A common aspect of individual and multiple-strategy instruction is the active involvement of motivated readers who read more text as a result of the instruction. These motivational and reading practice effects may be important to the success of multiple- strategy instruction.

The empirical evidence reviewed favors the conclusion that teaching of a variety of reading comprehension strategies leads to increased learning of the strategies, to specific transfer of learning, to increased retention and understanding of new passages, and, in some cases, to general improvements in comprehension.

The pressing need is for improved teacher training and skill in these areas:

The preparation of teachers to deliver comprehension strategy instruction is important to the success of teaching reading comprehension. As indicated by the Panel’s review of text comprehension, reading comprehension can be improved by teaching students to use specific cognitive strategies or to reason strategically when they encounter barriers to comprehension when reading. The goal of such training is the achievement of competent and self-regulated reading.

…[I]mplementation of the direct instruction approach to cognitive strategy instruction in the context of the actual classroom has proved problematic. For one thing, it is often difficult to communicate what is meant by “teaching strategies and not skills.” Several papers have been written whose purpose is to explicate exactly how teachers are taught to become teachers of comprehension strategies, and it appears that no small part of the challenge of training teachers comes from the difficulty of describing what is required of them. In addition, acquiring and practicing individual strategies in isolation and then attempting to provide transfer opportunities during the reading of connected text makes for rigid and awkward instruction.

Proficient reading involves much more than utilizing individual strategies; it involves a constant, ongoing adaptation of many cognitive processes. To help develop these processes in their students, teachers must be skillful in their instruction. Indeed, successful teachers of reading comprehension must respond flexibly and opportunistically to students’ needs for instructive feedback as they read. To be able to do this, teachers themselves must have a firm grasp not only of the strategies that they are teaching the children but also of instructional strategies that they can employ to achieve their goal. Many teachers find this type of teaching a challenge, most likely because they have not been prepared to do such teaching.

[note that these last few paragraphs support my critique of so-called reading strategies being taught as discrete skills in this prior blog entry. I agree with Hirsch: too often the strategies are taught as discrete skills, but that doesn’t invalidate the approach.]

This is of course a far different spin on Hirsch’s criticism of teaching reading skills. In effect, the Panel is saying: look, the research is clear on the benefits of teaching comprehension skills, but teachers do not adequately implement them nor have adequate training in how to teach readers to be cognitively strategic. So, transfer doesn’t happen. In other words, Hirsch conveniently overlooks the fact that the panels’ criticism was not of the value of comprehension strategies per se but the actual teaching of them.

They make an even stronger claim about instruction with respect to another of Hirsch’s favorite topic, mentioned above, vocabulary:

The need in vocabulary instruction research is great. Existing knowledge of vocabulary acquisition exceeds current knowledge of pedagogy. That is, a great deal is known about the ways in which vocabulary increases under highly controlled conditions, but much less is known about the ways in which such growth can be fostered in instructional contexts. There is a great need for the conduct of research on these topics in authentic school contexts, with real teachers, under real conditions.

[emphasis added]

As I mentioned, much of his analysis reflects Hirsch’s skewed view of educational history and the decline of reading ability. Here are three excerpts from different writings:

The root cause of this decline, starting in the 1960s, was a by-then-decades-old complacency on the part of school leaders and in the nation at large. By the early twentieth century worries about the stability of the Republic had subsided, and by the 1930s, under the enduring influence of European Romanticism, educational leaders had begun to convert the community-centered school of the nineteenth century to the child-centered school of the twentieth—a process that was complete by 1950.

Yet in the 1930s, American schools transformed themselves according to the principles of “progressive education,” which assume that students need to learn not a body of knowledge but “how-to” skills that (supposedly) enable them to pick up specific knowledge later on.

This contradictory and self-defeating situation has arisen because of a quirk in child-centered educational theory. Though it is opposed to imparting facts in a definite curriculum, it is not against inculcating all-purpose general skills—such as reading strategies and critical thinking. “Rote learning” and a set curriculum are to be regarded with scorn, but students may be subjected to drills in how-to skills… Many of the weekly hours that are assigned to language arts in the early grades are now being devoted to practicing reading strategies such as “questioning the author” and “finding the main idea.”

Never mind that since 1880 researchers were decrying the lack of fundamental knowledge of American students. Never mind that Progressive education died out during World War II and had limited impact on typical schools. Never mind that no reputable educator has ever scorned a “definite curriculum” in my life-time except a handful of radicals. But even if we grant a decline in American education, how can Hirsch justify such a statement as the following?

Teachers in a typical American classroom cannot rely on their students having acquired any specific item of knowledge. But effective classroom teaching depends on key prior knowledge being shared by all the members of the class. Without such shared knowledge, truly effective whole-class teaching cannot occur—no matter how potentially effective the teacher is.

Cannot rely on their students having acquired any specific item of knowledge? Really? And successful whole-class instruction is impossible without every student coming in knowing the same things?? How can such a claim be made without evidence, since it is so clearly not credible on its face in a country as diverse, mobile, and specialized as ours?

Not being taught? Or not being learned? So, finally, let’s zero in on Hirsch’s constant refrain: core content isn’t taught in our schools.  Why does Hirsch always write that key knowledge is not taught? What evidence does he have of this failure to teach content? In fact, a far stronger argument can be made, on the basis of sources as varied as the National Study of Student Engagement, Hattie’s meta-analysis, How People Learn from the National Academy of Science, and state standards and standardized test results based on them (not to mention a visit to any classroom in America) that content is routinely taught but not learned.

This is even true in math, the subject that Hirsch ironically tries to set up as a more effective case of content-based teaching leading to better results than reading. As I have often noted, math tests reveal repeatedly that students do not sufficiently understand such core content as the Pythagorean Theorem; they cannot “see” this idea to transfer it in novel problems. How does Hirsch explain such failures? The content was indeed taught, multiple times; however, it was clearly not understood and transferred. Surely the possibility of and the problem of deficits in transfer of learning are a more likely culprit than an absence of teaching of content. At the very least, Hirsch ought to justify this incessant claim of his that core knowledge is not taught in American schools.

I have spent 35 years in classrooms. Everywhere I go I see core content being taught. But it isn’t being well taught or well learned in many cases, echoing the National Reading Report, Hattie’s analysis in Visible Learning, and NAEP results. So, it’s time we retire Hirsch’s broken record about content not being taught. It’s well past time to focus on learning, not teaching, because regardless of one’s ideology the one undeniable fact that Hirsch and I can probably agree on is that students leave school with far less than they were taught, whether it is knowledge of the Algonquins (a piece of content in Hirsch’s core curriculum) or main idea.

PS: It’s already started – the either-or thinking that plagues this profession. A critic of my piece claims that I don’t value content knowledge. Nowhere do I say it or imply it. The aim of learning is not content knowledge; the aim is transfer of learning – including content knowledge – to new situations. Hirsch claims, without much evidence, that content knowledge is the determining feature in reading comprehension. I am responding to his narrow claim, period. As I noted and as the Reading Panel noted, reading well is a function of numerous factors and nowhere do they support the claim that background knowledge is the most powerful determinant of comprehension which is Hirsch’ claim.