Suppose that our education system were a corporation. You're the CEO, and you have a board meeting coming up. You're going to have to report your results. How are you doing? Well, on international comparative tests of reading, mathematics, and science among students in the developed countries, your students are LAST or nearly so. According to the Programme for International Student Assessment (PISA), American fifteen-year-olds rank 25^th in mathematics among 30 developed nations. NAEP results show that only 39 percent of our students are at or above the proficient level in math at Grade 8 and 23 percent at Grade 12. Of the kids that enter school at age 5 or 6, only 71 percent graduate from high school. Of those kids who graduate, 67 percent enroll in colleges or universities. Of those who enroll in colleges or universities, 25 percent graduate. So, if we multiply the percentages (.71 x .67 x .25.), we get the percentage of kids entering our school systems who graduate from college: 11.9. And those are the averages. The numbers for minority and economically disadvantaged students are much, much worse. How you going to justify those results to the board?

According to a 2007 study by the National Endowment of the Arts, 65 percent of college freshmen in the United States report reading "little or nothing" for pleasure. Only 56.6 percent of adults said they had read at least one book in the preceding year. And 70 percent of American adults have not visited a bookstore in 5 years. What can English teachers do to change these numbers?

According to the National Advisory Board for Mathematics, 78 percent of adult Americans cannot explain how to compute the interest to be paid on a loan. 71 percent cannot calculate the miles per gallon they got on a cross-country automobile trip. 58 percent cannot calculate a 10 percent tip. The percent of doctorates in science, technology, engineering, and mathematics granted by American universities to American citizens and permanent resident aliens has declined from 83.5 in 1966 to 59.8 in 2004. What consequences do these facts have for our long-term security here in the United States? Aren't these statistics at least as frightening as those for mortgage defaults?

Contemporary neuroscience tells us that the brain can be thought of as a society of thousands and thousands of interacting mechanisms for performing quite particular tasks, such as detecting edges of objects or recognizing distinguishing features of phonemes (e.g., the voicing that distinguishes Pat the bunny from Bat the bunny). Is it reasonable for us to posit, in place of a general intelligence factor (g), not seven or eight multiple intelligences but thousands of them that are hierarchically or otherwise ordered?

Think of a time when you learned a lot in a very short time and retained what you learned. Chances are that you were in a new environment--on a new job or sailing or diving  for the first time or in a new city or country. What does this tell you about learning? What's wrong with textbook writers' and editors' obsession with ensuring that the materials they create deal with what kids already know/are familiar with? Why would one of the greatest but least known of American educational theorists, the nineteenth-century U. S. Commissioner of Education W. T. Harris, have written of the importance to children of schooling that creates "a period of estrangement from the common and familiar"?

Studies in cognitive science show conclusively that most of our thinking—our cognitive processing—is not generally accessible to conscious introspection and so cannot be consciously directed by us. This is as true of so-called “higher-order thinking,” such as problem solving or decision making, as it is of, say, retrieval of a specific fact such as a person's name. Does the subconscious nature of most of our mental activity have consequences for how we design instruction? If so, what consequences?

Why are textbook publishers still, for the most part, selling goods when the rest of the economy has moved past commodities, past goods, past services, and on to selling experiences?

In the past few years, we’ve wired most U.S. public schools for the Internet, but with regard to hardware, software, and the technical training of teachers, our schools are still indistinguishable from the third world. Why is this so in one of the wealthiest countries in the world, especially given Negroponte’s $150 one laptop per child initiative being piloted in places like Ghana and Uruguay?

If you’ve ever tutored kids who are having trouble in math, you know that a given kid usually has a few highly specific problems that the teacher hasn’t noticed—she’s not lining up her decimal places, or she’s confusing the middle (the median) with the average (the mean), or she’s writing her 3s too close to other numbers and sometimes misreading them as 8s. So why has no one ever lined up all the mathematical algorithms that kids are supposed to learn from grades K-8 and done a systematic, comprehensive scientific study to create a specification, by frequency, of the errors kids make with regard to each of these algorithms? Why has such a thorough study not been made, given its obvious utility?

Is it possible that the so-called “New Math” of the 1960s failed not because it was a terrible curriculum but because there were no new math teachers? Do we have to continue fighting this traditionalist/constructivist battle? Isn't it possible that it's valuable both for kids to automatize number facts and common algorithms AND for them to understand a handful of central concepts in set theory and Boolean logic that are central to understanding what the mathematical enterprise itself is about? And wouldn't it be useful for kids to know these concepts, anyway, given their powerful applications in other disciplines such as computer science and electronics?

We now know that the parts of our brains that do planning and impulse control don't even start developing until we are about sixteen years old and aren't fully developed until we're in our mid twenties. Is it possible that there are other advanced activities (such as some kinds of conceptualization/understanding in mathematics), that aren't possible for most younger children because the mental machinery hasn't yet developed? If this were so, what consequences would it have for instructional design?

Given that writing is a recent cultural innovation (about 5,500 years old) and that there hasn’t been time for evolutionary development of dedicated structures in our brains for intuiting phoneme/grapheme correspondences, why would anyone have ever thought it was a good idea to throw out altogether direct instruction in phonics? Is it possible that there is a systemic failure to apply to practical pedagogy what scientists have discovered about learning?

The late twentieth century saw the development of the first truly scientific grammars. Before Chomsky, our scientific understanding of language was like our understanding of physics in the time of Aristotle. However, despite astonishing advances in linguistics during the past fifty years, almost NONE of that work is reflected in what we are teaching in those textbooks that still have grammar components. If the situation were the same in middle-school science texts, then those texts would still be teaching humors, the ether, phlogiston, and the vital force that an object uses up as it moves! Given that publishers "update" their texts yearly and given that schools pay lots and lots of money to buy these updates, why do the grammars that are taught (when they are taught), still embody the prescientific "folk" model of language? Again, is it possible that there is a systemic failure to apply to practical pedagogy what our scientists and researchers are discovering?

Given that we are born with machines in our heads for intuiting syntactic, morphological, phonemic, and semantic structures, and given that the language acquisition devices in our heads stop operating efficiently around the age of fourteen, why do we wait until then (late middle school or high school) to begin teaching foreign languages in this country? Don't we have it exactly backward?

Given that most of the linguistic structures in our heads are learned entirely unconsciously (e.g., no one ever taught you the rules for order of precedence of adjectives in English), what's wrong with our usual practice of limiting or eliminating entirely the novel (e.g., complex) syntactic structures and vocabulary from the language that we put before our young students? What are the consequences of raising kids in the INTENTIONALLY impoverished linguistic environments we create in our instructional systems, environments that rarely contain structures with which kids are as yet unfamiliar? Might our obsession with "leveling" and "readability" actually be causing kids irrevocable harm, especially because there is a window of opportunity for the operation of the students' innate language acquisition device that, once closed, cannot be reopened? How might we harness oral language to build passive fluency, given that most of our syntactic fluency is, in fact, passive? How should one balance the need to present to the student's internal language-learning machinery, which is almost entirely unconscious, the lexically and syntactically rich environment on which it must operate with the need to avoid creating cognitive overload that prevents learning? This may be the most important question in English language education, but it is one that an educator can't even begin to understand without familiarizing himself or herself with the relevant current scientific literature on psycholinguistics and language acquisition..

Obviously, we are theory-making creatures. Obviously, we construct models of the world. Obviously, it is primarily by such means that we learn. But why would anyone ever think that discovery education is incompatible with giving kids the basic tools with which to make discoveries? Sure, kids can learn a lot by going into the woods with the other Webalos and developing, on their own, a search procedure to find lost objects. But do you really want to require that they invent agriculture so they’ll have snacks to eat while they are out there? Education suffers greatly from the polarization between traditionalists on the one hand and constructivists on the other. Clearly, expertise involves a large degree of automaticity AND the ability to access and update large and complex models of the current situation in long-term working memory (LTWM). Something for the traditionalists: Lack of automaticity of key processes such as simple decoding or calculation increases cognitive load and decreases the ability of the learner to access models in long-term memory and compare them to the present situation. Something for the constructivists: Discovering structures and temporal and strategic relations on one's own increases the chance that they will become part of long-term memory and thus usable as LTWM, the model plus its updating. So, don't we need both in our teaching? Don't we need to give students practice to develop automaticity AND experiences that challenge them to recognize large-scale structures and relations on their own? We might not be able to turn every kid into Fermat or Gauss or one of the Bernoullis or von Neumann. Perhaps there are limitations on how far we can take a given kid based on his or her inherent aptitude. But we can certainly take most kids further than we are now taking them, especially if we educators stop shouting at one another from our respective camps and start attending to what studies of learning are actually telling us. The NCTM has belatedly recognized the importance of early instruction to ensure automaticity of recall of number facts and basic computations AND the necessity of focusing on a few topics in depth, but will real research, of the kind called for by the National Advisory Board, be conducted to answer, definitively, the questions posed by the opposing camps in the math wars?

Yes, I know that just about every kid has his or her genius-level aptitudes. These are to be found among those thousands and thousands of intelligences I spoke about earlier.

Why did the Education Department, when it set up the What Works Clearinghouse, call for across-the-board “gold standard” experimental research ( with controlled variables and random assignment to control and experimental groups) for all interventions when it is often impossible to conduct such research with real students and real teachers in the real world? Don’t these people remember the so-called “monster study,” and haven’t they learned that lesson? Obviously, there are some interventions that can be studied in this way and some that cannot. Often, ethical or practical considerations make doing randomized assignment to experimental and control groups impossible. In such situations, one must conduct observational studies. Why do we have this tendency in educational matters to propose blanket solutions to problems, and why are so few major independent, rigorous observational studies of particular interventions conducted? All the major textbook publishers now publish "research" on the efficacy of their programs. I've put quotation marks around that word for a reason.

Why would the state of Georgia call, as it did a number of years ago, for publishers to provide all the books for their next literature call on disk when getting permissions to reproduce these on disk was impossible? Why is it so difficult for publishers and states to talk to one another?

Is it possible that reasonably good ideas about instructional approaches, ones that might have value for some specific purpose, get  terrible across-the-board implementations that undermine their value? Is the devil sometimes in the details—in, for example, the time-consuming and confusing nature of many discovery lessons as presented in contemporary textbooks? Is it possible that sometimes we overdo it—taking what would make for a great extension activity and making it the primary mode of instruction?

All textbook adoption states require that publishers submit “readability scores” with their instructional materials. If readability formulae are so good, why does Jean-Paul Sartre’s Being and Nothingness, arguably one of the most difficult books ever written, have a score on the Flesch-Kincaid Scale corresponding to grade 11 or 12?

In the first half of the twentieth century, psychologists thought that they should be studying mental processes in the abstract, outside any particular knowledge domain, so they typically designed experiments that involved manipulation of content-free materials, such as nonsense syllables.We now know that for just about any task--recognition, retention, problem solving, decision making, inductive inference, analysis, synthesis, prediction, etc.--subjects perform better and gain in skill much more rapidly and thoroughly if they are working in a specific knowledge domain. So why are educators in the United States spending so much time trying to teach higher-order thinking skills, including metacognitive skills, IN THE ABSTRACT?

Metacognitive strategies are sometimes useful, of course. People are so terrible, for example, at probabilities that it really pays to think about one's own thinking when reasoning about them. But do metacognitive strategies also sometimes get in the way? Is it necessarily a good thing for the person catching a baseball to be thinking about the angle at which he or she is looking at the ball?

Why do most textbook lessons on writing research papers typically begin by having kids choose a topic and write a thesis statement? Doesn't one have to do an enormous amount of learning before having any idea what topic, much less what thesis, might be worth researching? Does our mania for "skills-based" approaches in the language arts have anything to do with people drawing the wrong conclusion from Bloom's having placed knowledge at the bottom of his hierarchy of cognitive skills? Isn't knowledge (of facts and procedures) at the bottom not because it is less important but because it is FOUNDATIONAL? And aren't some facts really interesting? (Darwin, when he was writing his book on earthworms, played the saxophone for a bunch of them to see if they could hear. The rocks at the top of Everest are marine sediments. The original draft of the Declaration of Independence attacked King George for promoting slavery. Most money isn't printed; it's created by lending and exists only in the form of electronic accounts. One could go on.)

Given that the most common difficulty that students have in reading comprehension is lack of requisite background knowledge (which is much more than just vocabulary), why are reading comprehension texts almost entirely devoted to instruction in finding the main idea, identifying cause-and-effect relationships, identifying sequences, making predictions, and drawing conclusions? How did reciprocal learning strategies for use in reading circles, a delightful bit of pedagogy with some real value but some real limitations, get elevated to the status of the whole of reading comprehension instruction? And why aren't reading comprehension texts addressing the significant issue of cognitive load, which is the second most important factor preventing kids from understanding given texts? (Even if the student has the requisite background knowledge to comprehend a passage, he or she may still give up in frustration simply because the passage delivers too much to process readily. Four kinds of cognitive load are relevant—1. that created by the amount of new knowledge presented by the passage, including new vocabulary defined in context; 2. that created by the complexity of the arguments and relations presented in the passage; 3. that created by the passage’s syntactic complexity; and 4. that created by unfamiliarity of the discoursive structures employed.)

Why don’t the people in the linguistics, AI, and cognitive psychology departments talk more with the people in the education departments? Don’t they have kids?

Given how much fun learning is, why do we have to MAKE kids do whatever it is that we are calling learning in our schools? (Hint: Kids don't do what we tell them. They do, eventually, what they've seen us do. The best thing we can do for students is to model for them what it is to be passionately engaged in our own life-long learning. That passion is contagious.)

Why do most of our college graduates find that they have to go back and get some other training in order to be at all employable? (Emerson said, and I’m paraphrasing here, “Give me a man who can write a Horatian Ode AND build a barn.”)

Why can’t the word poem be defined in a way that delineates all things, and only those things, that are called poems? Why is coming up with such a definition impossible, and what does this tell us about how the mind treats categories? And what consequences does this significant observation about how our minds work have for instructional design?

A famous study found that most of the kids in a recent Harvard graduating class couldn’t tell you why we have seasons (because of the tilt of the earth). How could this be so?

More Americans can name all the members of The Simpsons nuclear family than can name three of the rights guaranteed by the First Amendment. How can people with that level of knowledge of their government make informed decisions at the polls?

President Obama wouldn’t be allowed to take a position teaching seventh-grade civics. An emeritus professor of American Studies at, say, Northwestern, wouldn’t be allowed to take a position teaching eleventh-grade American literature. A career researcher in physics at Bell Labs wouldn’t be allowed to teach a high-school physics class. None would have the requisite “credentials.” In a time when we are all supposed to be life-long, self-directed learners who rise, on our own, far beyond our early credentialing, doesn’t this strike you as INSANE?

A recent 3.8 GPA graduate of a writing program at an elite northern university sat in an interview and told me that she was never taught the industry standard symbols for editing copy and that she had never used any of the standard stylebooks (Chicago, the AP, the MLA, etc.). Did her professors expect that she was going to write without editing? That she was going to work as a writer or editor without being able to mark copy? Is it possible that her professors did not know how to mark copy--did not know, for example, the differences between the marking procedures used for substantive editing and those used for proofreading? Does the scholarly disdain for “vocational” education go this deep?

Publishers of elementary and secondary school textbooks compete with one another based on giveaways of ancillary materials that support the basal texts that they sell. They sell the basals and give the other stuff away “for free”—great mountains of it—teacher’s guides, transparency sets, multimedia packages, workbooks, classroom management materials, diagnostic materials, and so on. Do the states buying these $120 basals actually think that they are getting these other materials, most of which are never used, for free?

Open a magazine, book, or newspaper at random, or choose a random paragraph from an article on the Net. Does it have a topic sentence? Does it have a concluding sentence? Usually, no. Why, then, do almost all writing textbooks define the paragraph as a group of sentences with a topic sentence and a concluding sentence and several sentences related to the topic sentence? How often, in our instruction, do we do just this sort of thing because we haven't stopped to think about whether that chestnut in our instruction is, in fact, true?

Middle-school and high-school textbooks are ALMOST NEVER written by the people whose names are on the spines and on the title pages. What conclusion are you meant to draw from these names appearing in those places? Does this misdirection bother you? Should it? Why?

Denis Diderot, H. G. Wells, and Vannever Bush all dreamed, long ago, of a universally accessible repository of all human knowledge. To what extent does the World Wide Web fall short of that dream, and why? If the Web ever becomes such a repository, will it replace textbooks and teachers entirely? Will textbooks become open source, freely available for the downloading and continually updated by means of crowd-sourcing?

Most educators agree that students should be placed in instruction at their “zone of proximal development,” that an ideal instructional system establishes a “learning community,” integrates self assessment and feedback with the instruction, allows for individualization and self pacing, and is universally accessible. It’s easy to do all those things online. So why aren’t we there yet?

Why does one of the best-selling middle-school and high-school literature programs in the country define incorrectly the basic terms used by literary scholars to describe the parts of a plot? Why haven’t a hundred thousand high-school English teachers sent horrified emails to the publisher of this program, and why haven’t state adoption committees caught the errors? Ed book houses employ a lot of smart, dedicated people, so what is there about the circumstances under which textbooks are produced and vetted that make for such poor quality control?

Why do the skills strands in most middle- and high-school literature texts not follow any developmental sequence? Why are skills taught, following literature selections, almost at random? What instructional idea gone awry in its application could possibly account for such an obviously odd phenomenon?

Why should each of the states be independently spending millions of dollars to create tests to figure out the levels at which their students are reading and doing math? Why do most of the so-called reading tests created by the states for reporting under No Child Left Behind consist mostly of writing assignments? Why don’t most of the math and science and social studies state tests cover in any kind of balanced, complete, representative way the curriculum in those subjects? Why aren’t most of these tests reliable or valid in the statistical sense? Why is there rarely any independent validation of the reliability and validity of these tests? Why do so many of these tests contain errors that are whoppers? What conclusion should we draw from the wide variances between state and NEAP scores? Doesn't that wide variance prove that the states' tests are incommensurable?

The Bill and Melinda Gates Foundation has been looking at education statistics, trying to figure out what improves teacher quality. We currently reward teachers based upon 1) number of years of service and 2) advanced degrees received. Guess what factors have NO STATISTICALLY SIGNIFICANT relationship to student outcomes? You guessed it: 1) number of years of service and 2) advanced degrees received. But, obviously, the factors need to be disaggregated, looked at more finely. I suspect that service in a variety of teaching settings with a variety of students and advanced degrees in the subject area of the instruction are strongly positively correlated with outcomes. (Good subjects for further research!)

Why do textbook publishers produce a Texas algebra and a Florida algebra and so on? Is algebra really different in Texas and in Florida? Isn’t mathematics a universal language? If you haven’t worked in the textbook industry, the answer is likely to astonish you.

Why are textbooks so expensive? We all know, of course, that trees are rarer and more expensive these days and that textbooks are still, for the most part, made of trees, but most people don’t think that paper, printing, and binding costs are the whole of the answer to this question. State administrators and the parents of college students often blame the rapaciousness of  publishers, but several of the major textbook publishers are suffocating under mountains of debt, which one wouldn’t expect if they were gouging their customers. What’s really going on here? Hint: Compare the number of independent textbook publishers in 1980 to the number today. You will be astonished at the degree of consolidation in the industry. Of course, the same sort of thing happened in banking. Look where THAT got us. And think of what it costs textbook publishers to ensure that their programs conform to the standards of fifty different states. And think of what it costs to make all those give-away ancillaries. Back in the 1960s, a textbook program often consisted of a basal text and a short, paperback teacher's guide. Now, it's a four-color, 1,400-page student edition with a 1,500-page teacher's edition with wrap-around annotations in a box with forty-five ancillary products, many printed in four-color, and the program has an accompanying video and an instructional website and diagnostic software and and and and and. Not good for students. Not good for teachers. Not good for LEAs and SEAs. Not good for publishers. Not good for anyone involved! The demands on publishers have created great barriers to entry for would-be new publishers, so innovation is stifled on two fronts--because there are fewer competitors and fewer significant new entrants. So, how do we change all this? How do we scale it all back, make instructional materials cheaper, ensure the financial health of the textbook houses, and reintroduce competition in the instructional materials industry?

One of the most widely used approaches to writing instruction in the U.S. today involves instruction in the "Six Traits of Good Writing": Ideas, Organization, Voice, Sentence Fluency, Word Choice, and Conventions. But isn't word choice THE primary determinant of voice and the major variable that one modifies to achieve a different voice (to speak in an informal or intimate register, for example, as opposed to a public and formal one)? Doesn't this bother any of the hundreds of thousands of people doing "Six Traits" training in our schools? If they were thinking operationally about instruction, wouldn't they have noticed this problem? (The reason why the Six Traits folks don't see the problem seems to be that they use the term voice in a narrow sense to mean those characteristics that set speech or writing by one person off from speech and writing by others, but if this is what they mean, why don't they use they traditional term for this trait, style, especially given that there is a long and venerable tradition of instruction in how to achieve one's own style that could be readily drawn upon?)

Most state language arts standards and many standards in other curricular areas call for instruction in making inferences. Why do these standards and the hundreds of thousands of lessons built on them and appearing in textbooks almost never recognize that there are very distinct types of inference--deductive and inductive--with very different rules and applications? Why do almost none of the lessons on making inferences in language arts texts today give students a procedure for making a deductive or inductive inference of any kind? Why do none of them teach kids anything about how to judge the truth or validity of inferences? Again, if textbook authors and editors and teachers of "inference skills" were thinking operationally about instruction, wouldn't they have have remedied this problem long ago? Why haven't they?
 

Have a look at a lesson in that supplemental language arts text that your school has adopted. Ask yourself, "Is there anything that the student can do after having read or worked through this lesson that he or she couldn't do before?" You will be astonished how often the answer to this question is "No." A lesson begins by telling the student that an inference is a conclusion. Then it gives a couple sample readings and some conclusions that can be drawn from those readings. Then it gives an activity that presents some sample readings and asks the student to make some inferences. But nowhere in the lesson is the student taught any procedure, any set of steps that can be carried out, in order to make an inference, and nowhere is the student taught any means for testing an inference once it is made. Or, to give another example, the student is told that good writing has the trait of exhibiting a unique voice and given a dull piece of writing and an exciting one, but nowhere in the lesson is the student taught anything that he or she can do to make a piece of writing less dull. Shouldn't a student emerge from a lesson able to do something that he or she couldn't do before?

Years ago, when I was a young teacher, I went to my first NCTE conference. I expected to find at that conference lots of workshops related to the content of this vast field of English studies. I expected workshops on syntactic structures in African-American dialects, on prosody in verse, on the types of dead metaphors used to describe mental processes, on common structures in journalistic writing, on motifs of the folktale, on the phonemic system of English, and so on. Imagine my surprise when I found that almost all the workshops were on methods! One could attend several days of these workshops and not add anything to one's store of knowledge about English per se. Why should there be this vast gulf between English studies on the one hand and English education studies on the other? Do people assume that English teachers already know everything there is to know about the content of their subject? To what extent has the preparation of English teachers become content free?

Every teacher makes up tests and calculates grades. How many of these teachers know the difference between a norm-referenced and a criterion-referenced test? How many can tell you what a p score is? How many can explain to you what reliability and validity mean? How many can calculate standard deviation? How many understand whether they are calculating their grades according to a norm-referenced or criterion-referenced paradigm? Given that these are essential skills for any teacher, why don't most teachers have these and other like skills? Aren't the education schools failing us in this respect?

The school reform movement seems predicated on the notion that creating standards and testing to ensure attainment of those standards will be sufficient to improve student outcomes. But isn't teacher quality even more important? If classrooms were filled with superb teachers, wouldn't high standards and accountability follow? Isn't the real problem that too many French teachers aren't fluent in French, that too many science teachers could not survive graduate-level science courses, that too many English teachers are not well read and knowledgeable about language? How can one teach what one does not know? Surely, the job of a teacher is not solely that of a generalist facilitator.

Let me put it this way: What would it take to make K-12 teaching one of the high-status professions in our culture? What would it take to cause a murmur to run through a cocktail party when a teacher enters the room (akin to what happens at the entrance of a rock star or actor or politician)? Can we pay teachers like grocery store clerks and still expect that we are going to have the best and the brightest in our classrooms? Isn't it the case that teacher quality is THE problem and that politicians don't want to face that fact because doing so would be expensive? How much more money can a qualified chemist make working for a pharmaceutical firm than he or she can make teaching high-school chemistry? But doesn't that mean that the average high-school chemistry teacher will not be a superbly qualified chemist?