E.D. Hirsch, Mike Smith and Linda Katz offer insights on reading development

If journalism is history’s first rough draft, then perhaps blogs like this one are journalism’s notes and outline. For me, this blog continues to be a wonderful learning opportunity, largely because of thoughtful readers who question my assumptions and provide me with information I have either forgotten or never seen.

In the few days since I posted my thoughts about early reading, I have received several (welcome) wake-up calls from E.D. Hirsch, Jr. (of Core Knowledge fame), Marshall ‘Mike’ Smith (former Deputy U.S. Secretary of Education under Bill Clinton), and Linda Katz (Director of the Children’s Literacy Initiative in Philadelphia).

Katz suggested that I was misleading the public by repeating the “30,000,000 word gap” assertion.

“I personally think it’s mathematically impossible,” she wrote. “It’s the equivalent of reading the densest children’s picture books you can find, every hour for 14 hours. You would have no voice, your baby would have no sleep or quiet feeding, no playtime with other kids, and it would be relentless.”

So we pulled out the original research and discovered that Hart and Risley made monthly visits, each lasting one hour, to selected homes of two- and three-year-olds and counted the words, over a course of  36 months. From that, they apparently assumed that what they heard in that monthly hour would apply over a 14-hour waking day for four years. For professional families, the 2,153 words per hour they recorded translates into 35 words per minute, every minute, in order to get to Hart & Risley’s total of nearly 45 million words, as opposed to the 13 million in welfare homes over the same four-year period.

I found myself humming a song from “Pajama Game:”

“Seven and a half cents doesn’t mean a heck of a lot, doesn’t mean a thing … but give it to me every hour, every day, every week, and that’s enough for me to be living like a king.”

Just as those singing workers are delusional, so too are those who imagine 14 hours a day of non-stop conversations. Linda Katz is right.

Besides, what actually happens in most educated families is very different. Yes, there are vocabulary-rich conversations, but there’s lots of quiet time too. Picture yourself sitting in the living room reading, your three-year-old daughter on the floor beside you, playing with toys or coloring. Occasionally you will reach over and tousle her hair, or she may lean against your leg. Stuff like that matters as much as anything else in child-rearing.

Content knowledge has proven to be the cornerstone of reading development over time.

So it’s not 30 million words, but there still is a vocabulary gap between affluent and poor. That’s where Mike Smith chimed in. It’s not about vocabulary, he pointed out. It’s about language, conversations, speaking and listening. Focusing on vocabulary can lead educators to forget how kids in affluent homes acquire that wonderful vocabulary — through conversations, not vocabulary drill. An important warning flag from someone who knows how policy can get screwed up if we aren’t careful.

And then Don Hirsch weighed in. Hirsch, the founder of the Core Knowledge movement and the author of Cultural Literacy, welcomes the national attention to our nation’s reading problems — but warned against focusing on the abstract concept of ‘reading.’

What matters most is not the skill set associated with reading (phonemic awareness, etc.), but vocabulary and content knowledge. Don sent me a copy of the speech he gave to a gathering of Virginia state officials, a marvelously clear statement of his principles. I’ve pulled out a memorable line:

“The persistent achievement gap between haves and have-nots in our society is chiefly a verbal gap. There is no greater practical attainment in the modern world than acquiring a bellyful of words. A large vocabulary is the single most reliable predictor of practical, real-world competence…”

The Virginia House of Delegates had passed a resolution honoring Don’s work.

Don explained what is called “the Matthew Effect” to Virginia’s legislators, perhaps hoping that the biblical reference would make them listen more carefully. “For unto every one that hath shall be given, and he shall have abundance: but from him that hath not shall be taken away even that which he hath.” What that means, Don said, is that the more you have learned, the more you are capable of learning and likely to learn. The reverse is also true: the less you know, the harder it is for you to acquire knowledge.

What does all this have to do with learning to read, I asked? Simple, he said: it’s all about content knowledge. You don’t just learn to read in the abstract. You learn facts, content, concrete information. There’s no ‘learning how to learn’ or teaching ‘critical thinking skills’ or ‘comprehension strategies,’ he warned, because those are a dead end. That approach might yield a temporary up-tick in reading scores, but no genuine lasting learning.

So what have I learned?

First, it’s not about numbers, although the vocabulary gap is real. Reading is the foundation, but what matters most is content knowledge. You have to read about something, whether it’s baseball or Patrick Henry or space travel or a pet dog. And it’s important that all children have common reading experiences — shared content. Finally, closing the vocabulary gap is best done in situations that replicate how vocabulary-rich children in the study acquired their larger vocabulary — through conversation, not in cold classrooms where drill is the M.O.

And the more you know, the easier it is to learn more.

17 thoughts on “E.D. Hirsch, Mike Smith and Linda Katz offer insights on reading development

  1. I agree that there isn’t a “silver bullet” to all of this. It really involves having well-prepared and supported teachers, in a very warm and nurturing environment, interacting in a highly-intentional way to help children develop language, vocabulary, background knowledge, alphabet knowledge, print awareness and the social/emotional skills that are aligned with success in school like learning to attend to instruction, to persist, to take directions from teachers and to solve problems with words. As Louisa Moats wrote, “Teaching reading is rocket science.”

    While the number of words estimated may be inflated, the difference is still an important gap that needs to be addressed to ensure that young children from disadvantaged backgrounds can significantly raise their trajectory of learning.

    Providing, supporting and sustaining this rich level of teaching and learning in preschool and per-kindergarten at scale is the first, important challenge as we think about how to ensure all children read at grade level by 3rd grade.


  2. Hey John,

    Great discussion here on a great topic. But I’d like to bring up one point that I often find missing from these talks our nation has about reading: almost no one in them has ever taught kids to read.

    Think about it: reading is fairly complex. In fact, we can’t see it happen even with a brain scan so we can only “guess” about whether and how it is occurring. And yet, when the subject of teaching kids to read comes up, everyone seems to have something very weighty and important to say — though few, I find, have ever actually taught even one child to read, let alone several hundred at all ages, abilities, grade levels, etc.

    I suggest we start paying attention to the people who have taught the most kids to read. Not necessarily the theorists, the pundits, the psychometricians, etc., but the actual human beings who’ve done the real work of teaching real kids to really read.

    I have been very lucky in the kind of work I have been allowed to do in education. And getting literally thousands of chances to teach thousands of different kids to read is one of them. Here’s what I’ve learned:

    1. It is not very hard to teach kids to read. We should start at age 3 or 4 and all kids should be reading and writing independently by kindergarten. There is no reason this cannot happen. The most common reason given for why this cannot happen is that kids are not ready. I ask people who say this if they’ve ever tried teaching small children to read. They say “no”. And I say, “They how would you know if they were ready or not?” This seems utterly logical to me, yet it fails to impress anyone in the dialog. Just another indication, it seems to me, of how irrational we are as a nation when it comes to literacy. Read McGuiness’s latest books if you want the scoop on how quickly kids could learn the “basic code” of reading and how early we could begin teaching it to them.

    2. Even though there are literally hundreds of reading programs out there, there is only one way to teach kids the phonics or decoding information they need at the start — it’s the way our language works and the way our brains work (as such, no program is required; the language is itself the program; though a program that follows the language would work just fine). If we do not base early decoding instruction on The Alphabetic Principle, we waste time and confuse our children. Yet fewer than 5% of our kids by my estimate are taught this way. Most get a “Sesame Street” letter-of-the-week approach or the good old “A is for apple” approach. Both are disastrous, yet they remain maddeningly popular.

    3. Reading must be combined with writing because they are complementary processes. This should be patently obvious to everyone but it is not. Kids do almost no writing as they begin to learn to read in American schools. Again, McGuiness and others are very clear on the value of combining reading and writing instruction.

    4. In order to ensure success for all, we must read approximately 500% more than we ask kids to read now, and much more of this reading must be individual reading with kids reading their own books. When there’s too much whole class reading, kids fail to develop independence. And when kids never get to pick books to read that they like, they tend not to like reading after a while. This is my own opinion based on the amount and type of reading I have had to assign in order to pull kids up to grade level over time. You won’t find a study on it because no one has ever studied how much reading kids need to do — especially disadvantaged kids. And even though Diane Ravitch contends that “No kids will pick Moby Dick on their own.” I have had this happen several times (along with many other classics) so I know that at least that old chestnut of an argument is wrong. Given responsible instruction, and exposure to good literature, kids read a ton of classics. Why? Because they’re good. But not every kid need read the same set of classics. Why? Because this limits the number of classics any single kid can read.

    5. While Mr. Hirsch is correct that reading is “domain specific” and that “background knowledge” is the #1 predictor of comprehension, he and his well-meaning and intelligent crew have put forth no serious work in the area of techniques for increasing kids’ background knowledge — except buying into the Core Knowledge curriculum, of course. There are actually many interesting ways to help kids gobble up knowledge at very quick rates — and retain it much longer than usual. So, if Mr. Hirsch is right, and he is, we need to be focusing on this aspect of teaching as much as any other as we help our kids become more literate. Core Knowledge is not the only way to go. It is merely one extremely conservative example of how to teach information. And it provides little help in terms of the techniques that might be used to teach it well.

    6. Reading fluency is extremely important for reading growth, especially in the primary years. Unfortunately, because of the popularity of the DIBELS system, and the general lack of knowledge about reading in America, most people equate fluency solely with reading rate or reading speed. In actuality, fluency depends on rate, phrasing, and expression, and the latter two are extremely important for comprehension. To put it succinctly, we lose many young readers because we don’t teach fluency explicitly.

    7. While Hirsch is correct that there really is no such thing as a reading comprehension test, our country seems to delight in creating and giving them. In fact, reading scores on traditional comprehension tests (which Hirsch shows are terribly by degree of background knowledge, not reading ability) are used to make hugely important decisions about kids, teachers, and schools. As a country, even our most thoughtful psychometricians don’t take this into account.

    8. Even though many people have worked hard to discredit reading strategy instruction, I’ve read the literature carefully and it does that a small amount of a certain type of reading strategy instruction works very well. Personally, I teach maybe six strategies at the most, and I teach them very thoroughly. A small number of well-chosen strategies, well-taught, can make an extraordinary difference, especially for kids who are many years below grade level. I do agree, however, that the “Reading Strategy Movement” has gotten ridiculous and that we certainly don’t need any more classes called “Reading Strategies” or any more books called “101 Reading Strategies That Really Work.”

    9. For all the sound and fury around reading, we don’t actually do much in our country to teach kids how to read at all – we just tell them to read and hope that they eventually get around to figuring it out. Taken in the aggregate, reading instruction in the US is random. This is largely because we don’t teach adults how to teach reading. It’s not hard. Especially at the beginning. Virtually all parents could teach their kids to read before they arrived at school. Our government would simply have to provide a very small amount of information to make this happen. In fact, one might think that our schools could provide this information to new families in their neighborhoods.

    10. Check out the percentage of kids in the lowest bracket of the 4th grade NAEP reading test. You’ll notice, as I did recently, that it’s just about exactly the same percentage as the percentage of kids who don’t graduate from high school. To me, that says that our newfound mania for cleaning up failed high schools is sorely misplaced. The game is won and lost in the primary years, so that’s where we should be spending our “school turnaround” dollars if we want to eventually do away with “dropout factories.”

    Thanks for the chance to contribute here. I always enjoy your work.

    Steve Peha
    President, Teaching That Makes Sense, Inc.


    • This is very thoughtful, no surprise. It makes me want to do some reporting on how and where (and how well) prospective teachers are taught to teach reading. Is this a useful line of inquiry, in your view?
      Regarding Core Knowledge schools, the ones that I have visited remain in my mind as beacons of what’s doable, places of joyous learning (and seemingly very satisfied teachers). I haven’t seen anything as positive or as scalable. Have you? Has anyone?


      • As a former classroom teacher and current university professor of pre-service teachers, I believe that I prepare my students well for teaching in their own classrooms. I use demonstration lessons using professional text that they need to comprehend and, at other times, text that they might use to teach their students to read and write. Different strategies to address surfacing or development of background knowledge, vocabulary development, comprehension, writing, etc. are implemented in each lesson. This allows the pre-service teachers to learn through practice the strategies they and their students should be using to teach/learn reading, science, social studies, etc.

        The next part of my plan is to research whether my students, upon leaving the university and getting their own classroom, actually implement what I have taught them in my courses. They might be reverting to teaching the way they were taught which might useful in some lessons, but certainly not every day of the school year.

        To make a long story, short, yes, we need some research and some reporting on how preservice teachers are prepared for the classroom and whether it is effective or not.


      • “This is very thoughtful, no surprise.”

        Thank you very much. Your work is always very thoughtful so it’s easy to add to it in a similar vein.

        “It makes me want to do some reporting on how and where (and how well) prospective teachers are taught to teach reading. Is this a useful line of inquiry, in your view?”

        I think “yes” if (and only if) you want to take on the establishment in ed schools with something really serious. From your standpoint as a trusted journalist, I think it would be both useful and, frankly, a little shocking.


        Personally, this is not an interesting line of inquiry for me because I’ve worked with so many undergrads, young pre-service teachers, and veteran “reading specialists” as well. As such, I sort of know the score. We don’t teach people how to teach reading. In fact, most of what we teach them makes it harder for kids to learn to read.

        I would never make light of how difficult teaching reading is. But I would also point out that all the resources are available today to learn to teach reading very well. And it doesn’t take a college degree to do it.

        Twenty years ago when I started, reading was something of a mystery. It’s not anymore.

        One area, though, that I have found to be the most useful is a “write-to-read” approach. I see extraordinary gains here and have written about this here and there.

        But because of the inappropriate splitting of reading from writing, this has never been studied to any great degree. It is however, consistent with available research on phonemic awareness, phonics, handwriting, and spelling.

        Most important to me are the hundreds of classrooms where I’ve seen this approach create staggeringly powerful results, especially for low-income kids. Unfortunately, because this is not a mainstream approach, and because it can be packaged into a program someone can sell, it simply doesn’t get traction.



    • Of course we need 101 reading strategies. Teachers need all the strategies they can get. No, they won’t use them all in one year or with one child, but children vary in needs and abilities. Using the same strategies for more than two weeks will produce groans from pre-teens. Their brain, like any other human brain, needs novelty to aide in learning. An effective teacher chooses intentionally the teaching/learning strategies that meet the lesson objective which should be written with student needs and standards in mind.

      And yes, I’m a reading teacher of public school students. I have been for more than 20 years. I agree that unless someone has been a teacher, I can’t imagine how he or she can make informed decisions about teaching.


    • Steve Peha confirms all the things I had figured out in a more informal way. In short I recommend everything he says highly, except one thing: “we don’t actually do much in our country to teach kids how to read at all…”

      I think it’s worse than that. Our Education Establishment deliberately uses bad methods and encourages semi-literacy. There is no way to explain the dreadful results for the past 80 years unless you assume the people in charge want dreadful results.

      I was just thinking this morning about Dr. Samuel Orton. He got a Rockefeller grant to study early tests with look-say. Apparently he was supposed to find that it was a great success. Instead he found that not only weren’t kid learning to read, they were being psychologically damaged. This was published about 1928. Everybody in the profession knew that this thing was a disaster. Starting in 1931 they flooded the public schools with a method they knew was useless. And they are still doing that to the degree they can get away with it.


  3. John,

    Thanks for including my comment. It was a little random in that I do admire the work of Hart and Risley, and ferverently promote the work and research on vocabulary, particularly Dan Willingham and Keith Stanovich.

    What I especially like about the Hart and Risley research is the danger of a child’s vocabulary not exceeding his or her parents’ vocabulary, and their finding about the perponderance of parent affirmatives (encouraging words) as you move up the income scale. You could envision a scenio where a mother drops her keys and a toddler grabs them. Mother 1 says, “No, bad.,” and grabs the keys back (discouragement no vocabulary). Mother (or father) 2 says, “Give me back the keys.” (neutral, and uses the word “keys.” Our savvy (and wealthier) parent says, “Oh, you found my keys, you smart little girl/boy. Can you say ‘keys?’ Let’s count the keys.” Replay that kind of conversation a few hundred thousand times, and you have some idea of the vocabulary gap.

    Great work as always, John.

    Linda Katz


    • So the question becomes, “How do we enable parents to be savvier?” See Steve Peha’s comments above about teaching parents how to teach reading. Do you agree?


      • I don’t think ‘savvy’ is the right thing to aim for. When educated parents talk with their kids, share ideas, discuss science and history, etc., they’re not doing it with the intention to “make the kids smart.”

        They’re just enjoying sharing their interests and watching the kids develop their own interests. The encouragement isn’t deliberate–it’s just fun.

        When the same parents are sick, tired, hungry or stressed out, the quality of conversation declines. (For instance, it’s much harder to be interested in your 3 year old’s dissertation on sharks and dolphins at 3am. You’re much more likely to reply with “It’s night now. Go to sleep.”)

        So, I see two issues:

        1. You’re not going to share academically oriented background knowledge with your kids unless it’s something you enjoy. The children of people who hate school aren’t going to have these discussions at home, no matter how ‘savvy’ we make the parents, because these are ‘spinach’ discussions to them. In the more educated homes, the same discussions are ‘popcorn and candy.’

        2. Tired, malnourished, overworked, stressed out parents are not interactive, happy parents. So improving quality of life MIGHT improve parent/child interactions.


      • You are right. I was glib and stand corrected. Still, there is a place for helping parents understand the importance of conversations with their children.


      • I think the first step is stable homes that lend themselves to non-stressed out parents. Unfortunately, this means working on living wages and encouraging two parent homes. (Single parents are NECESSARILY more exhausted and stressed because the job is really too big for one person!)

        Once there is leisure time, I think parents, even those who hated school as children, might have time to have interesting conversations–especially given how widespread cable tv is in low income neighborhoods, and the popularity of networks like Discovery and the History channel.

        In the past, we had a ‘middlebrow culture’ – so even the man who hand’t finished high school might spend his leisure time learning about history, science, arts and music. We’ve lost that– and I think culture change, not quick interventions will need to be at the root of our solution.


  4. I somewhat hesitate to enter this discussion, with some of the notables already involved. After all, I teach social studies at a high school level, which at least from Steve Peha’s point of view would seem to indicate that I might not have much to say on the topic, since he thinks the battles should be at the elementary level.

    Still, at one point I was enrolled in a Masters in Reading at Curry (UVa), while I was teaching reading and english to 8th graders as part of a program where I also taught them social studies. Thus I have some training in the field, probably just enough to make me dangerous.

    And I have taught social studies to students ranging from 7th graders to 12th graders. The skill levels have been inclusion classes where half the kids have IEPs, significant numbers of English language learners, and in my current 10th grade Advanced Placement US Government and Politics classes some preternaturally brilliant kids who enter the class with more background knowledge than many college graduates.

    That said, let me offer a personal perspective. It is drawn from my 16 years as a public school teacher, to be sure. It is also drawn from working with and observing nieces and nephews and now starting with grandnieces. It also includes my personal experience of teaching myself to read words and music when I was around 3 – no, I am not exaggerating. It includes having my reading accelerated as part of a small group so that by the time I left 2nd grade I was reading at what they said was a high school level.

    Reading is a skill, a pleasure, a burden, depending upon how it is approached. But it is more than one skill. Different kinds of texts need to be read differently. From my experience, this is sometimes especially difficult for boys, who if you give them a play tend to skip over things like stage directions and head right for the dialog. I know I did, and I saw it when teaching middle schoolers language arts.

    While content is important, it works in conjunction with reading strategies. How does one obtain the content from a particular piece of text? Does one read a philosophical tome the same way one approaches a story on a forthcoming superbowl? Sometimes what can be effective is for the students to attempt to discern the meaning of unknown terms in context, only then checking to see if they are right. Other times it is impossible to make sense without understanding the basic terminology, which can include words that in a different context have a very different meaning.

    I am going to somewhat disagree with Don Hirsch, but without bashing his approach. I think some of the skills of critical thinking can be taught with very little content knowledge. One can start with what a student already knows, provide material on which they can practice skills which can later help them to discern meaning from a text full of material they do not know. In fact, such skills can assist them in distinguishing between what they know, what they kind of know but probably should check, and what they do not know. The balance of these may help determine what approaches one takes with the text at hand.

    I will also suggest that reading is best learned not in isolation from its parallel, which is writing. Even students who may have mechanical problems in writing can now with technology use voice to create a starting text. In examining how what they have written does not fully match what they intended to communicate, students begin to learn skills about interpreting texts created by others.

    I also offer a caution of generational difference. I am something an old fogy, since I will be 65 near the end of May. How I learned to absorb and process information is very different than how the adolescents in my classes have learned. I am not sure my way is necessarily superior. I do not think we can address issues of vocabulary and comprehension in isolation from the culture in which our students operate, much of which is independent from what adults – parents or teachers – may want to mandate. Their methods of communication are often far more effective for them. It seems to me that rather than trying to restrict what they are already learning, from one another and on their own, we should be considering how we can take advantage of what is for them an exciting way of interacting with the world, including with text, and thus enables them to increase their vocabularies.


  5. Wow! This could be one of the most helpful blogs we’ve ever come across on the subject. Actually fantastic article! I’m also a specialist in this topic
    so I can understand your hard work.


  6. I think many thought that Hirsch believes knowing a bunch of facts is sufficient to be an educated person. He never said or wrote that. But the publicity about his work could leave casual observers of this byway in the culture wars with that impression, and insofar as I know, he did not do much to counteract the misimpression. That said, I certainly agree with your judgment that he has done much to enhance learning in the U.S. Cheers. Tom Ehrlich


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s