Chapter 12 Endnotes

283  A look at the history: The history in this section is mainly from Clifford & Guthrie (1988), including the quotations.

284  unbelievable distrust and opposition: Clifford & Guthrie, 136.

284  The organization of the university: See Goldstein (2014). Teaching was a lower-status profession than the other fields, undertaken by lower-status individuals (women) who were paid lower-status wages. The status of a college or university or a specific program is mainly determined by its “inputs,” characteristics of the individuals who enroll, rather than “outputs,” how the graduates subsequently fare (thus the race to qualify as “highly selective” on the measures that determine the specious U.S. News & World Report rankings). Ed schools have lower-quality “inputs” in this country, though not in Finland.

284  “Teachers should demonstrate”: Bennett quoted in Clifford & Guthrie (1988), 16.

284  “kitten that ought to be drowned”: As quoted by Clifford & Guthrie (1988, 137). This ancient history does not reflect the current standing of the Harvard Graduate School of Education.

284  Degree programs in teaching: Adams, Bell, & Griffin (2007); Tyson & Park (2006).

285  Although that interpretation: Coleman (1966). The report documented large differences in educational opportunity (quality of teachers, school resources, and so forth) associated with race and socioeconomic status. It noted, “It is known that socioeconomic factors bear a strong relation to academic achievement.” The first finding reported under “relation of achievement to school characteristics” was that “when socioeconomic factors are statistically controlled…it appears that differences between schools account for only a small fraction of differences in pupil achievement” (21). This result does not indicate that “schools don’t matter,” but many interpreted it that way. See Clifford & Guthrie (1988) for discussion. The opposite analysis would be equally informative:  examine how much SES accounts for student performance  after differences in school quality have been “statistically controlled.”

286  Teachers complain: The complaint arises from the fact that the educational theory is taught by professors whose research focuses on alternative conceptions of learning and learning environments that might revolutionize education. The approaches are intended to be radically different from current practices and thus are of secondary interest to the prospective teacher.

286  Their students bear: A charismatic, seemingly authoritative figure whose ideas are bunk can do serious damage, but the effects are greatly magnified when combined with assurances that personal experience is as valid a source of evidence as systematic investigation. Ken Goodman conveyed that message in the guise of teacher self-empowerment. It encourages a populist epistemology that greatly undermines the utility of science. In Denialism, his book about the growth of antiscience belief systems, Michael Specter analyzed the impact of Dr. Andrew Weil, the famous guru of health and holistic medicine, in similar terms. Weil’s controversial ideas are also coupled with an emphasis on the primacy of personal experience. Spector notes, “It is much easier to dismiss a complete kook—there are thousands to choose from—than a respected physician who, interspersed with disquisitions about life forces and energy fields, occasionally has something useful to say.” As quoted by Janet Maslin, “Firing Bullets of Data at Cozy Anti-science,” New York Times, November 6, 2009,

286  They originate: My favorite “everybody’s an expert” example is from Richard Feynman (2014, 116), one of the most celebrated and revered scientists of modern times ( All the time you hear the question, “Why can’t Johnny read?” And the answer is, because of the spelling. The Phoenicians, 2,000; more, 3,000, 4,000 years ago, somewhere around there, were able to figure out from their language a scheme of describing the sounds with symbols. It was very simple. Each sound had a corresponding symbol, and each symbol, a corresponding sound. So that when you could see what the symbols’ sounds were, you could see what the words were supposed to sound like. It’s a marvelous invention. And in the period of time things have happened, and things have gotten out of whack in the English language. Why can’t we change the spelling? Who should do it if not the professors of English? If the professors of English will complain to me that the students who come to the universities, after all those years of study, still cannot spell “friend,” I say to them that something’s the matter with the way you spell “friend.” The comments convey strong beliefs but little understanding about spelling, reading, or English professors. If English were written in the Phoenician system, it would be unreadable because the Phoenician alphabet had no vowels. Friend is not notably difficult, because it is a lexical hermit, a high-frequency word with only one, lower-frequency neighbor, fiend. Such words are easy to discriminate from other words and thus easy to recognize. And English professors are the culprits? Surely Mr. Feynman was joking.

286  lack of tools: In-service teachers eventually hear about methods from speakers who are brought in for lectures and workshops. “Experts” travel state and national Chautauqua circuits pitching their systems and sharing special insights. No quality controls to speak of. Some examples: novel theories: http://www.the2sisters.com; vision therapists:; Wisconsin high school teacher Doug Buehl:

288  Flexner Report: Flexner, Pritchet, & Henry (1910). The plan for medical education synthesized and codified practices that had begun to be implemented at major medical schools. The report is discussed by Clifford & Guthrie (1988).

288  courses would emphasize: The readings could include Stanovich & Stanovich (2003), a highly readable guide to “using research and reason” in education.

289  a 1916 report: Flexner & Bachman (1916). Education was said to consist of “simple practical problems, which would quickly yield to experience, reading, common sense, and a good general education” (Flexner, 1930, 118).

289  So is demonizing: “A certain casual demonization of teachers has become sufficiently culturally prevalent that it passes for uncontroversial.” Rebecca Mead, “Chicago’s Teacher Problem, and Ours,” New Yorker, September 11, 2012,

290  New hybrid fields: For the British Royal Society’s view of neuroscience and education, see

290  Motivation is an old topic: See Carol Dweck, “The Power of Believing You Can Improve,” TED talk, November 2014,; Hulleman & Harackiewicz (2009).

291  One successful program: A new type of teacher education program: Goldie Blumenstyk, “After Years Lambasting Teacher-Ed Programs, Art Levine Creates One,” Chronicle of Higher Education, June 16, 2015,

291  Add a course: There are plenty of excellent resources for teachers to learn what they missed in ed school, if their circumstances allow. American Educator, a publication of the American Federation of Teachers, is top-notch. It has readable articles on current issues of importance to teachers written by educators and researchers. For several years it has included a column, Ask the Cognitive Scientist, by Daniel Willingham, whose explanations of basic science are models of clarity and accuracy, and his assessments of what is known are reliably sensible. He also writes books, such as Willingham (2015). Cunningham & Zibulsky (2014) is equally good, research based but very oriented to helping children read. Stanovich & Stanovich (2003) is an accessible introduction to how to think about scientific findings and make use of them in planning curricula. The National Reading Panel issued a pamphlet that explained the basic findings and recommended practices in accessible terms; the pdf is downloadable here: My book was preceded by Wolf (2007) and Dehaene (2010), both excellent.

292  best example: For the MTEL “Foundations of Reading” questions, see “Field 90: Foundations of Reading Sample Multiple-Choice Questions,” MTEL Test Information Guide,; for the “Reading Specialist” questions, see “Field 08: Reading Specialist Sample Multiple-Choice Questions,” MTEL Test Information Guide,

293  alternate routes: See Clifford & Guthrie (1988).

294  Linguistics 101: A course like this one, for example:

294  In reading, for example: Connor et al. (2007).

294  Some theorists think: Gee (2003).

294  heavily hyped “brain training”: Owen et al. (2010); Melby-Lervåg & Hulme (2013).

295  HMH, for example, markets: For HMH’s Read 180, see “About Read 180 Universal,” HMH,

296  Current proposals for reforming: Darling-Hammond & Richardson (2009); for the New Teacher Project, see

297  KIPP “no excuses”: Rachel Monahan, “Charter Schools Try to Retain Teachers with Mom-Friendly Policies,” Atlantic, November 11, 2014,

297  The excellent model: See “About Minnesota Reading Corps,” Minnesota Reading Corps,

297  “[The program] provides”: For information about AmeriCorps, see “AmeriCorps,” Corporation for National and Community Service,

298  Applications to TfA: On challenges facing TfA see Stephen Sawchuk, “At 25, Teach for America Enters Period of Change,” Education Week, January 15, 2016,; Olivia Blanchard, “I Quit Teach for America,” Atlantic, September 23, 2013,; Michael Zuckerman, “Is Teach for America Good for America?,” Harvard Magazine, December 18, 2013,; Kerry Kretchmar and Beth Sondel, “Organizing Resistance to Teach for America,” Rethinking Schools 28 (spring 2014), Teach for America is a major source of teachers for KIPP schools.

298  Evidence that contradicts: On polarization, see Sunstein (1999). For a short overview, see Hastie & Sunstein (2015).

299  “The empirical evidence”: James Surowiecki, “The Campaign of Magical Thinking,” New Yorker, March 21, 2016,

299  Brendan Nyhan and Jason Reifler: Nyhan & Reifler (2010).

299  we had the precursors: Teachers Advocating Whole Language was an active BBS, remnants of which are here:

300  origins of the theory: Adams (1998).

301  Figure 12.2: Many versions of the figure can be found here:

302  If the child has trouble: The approach is explained and demonstrated by a top practitioner, Catherine Compton-Lilly, here:

302  children’s comprehension of a story: Glenberg et al. (2004). The story in this study is so unusually written that it can’t be easily comprehended without obtaining additional information from somewhere.

303  Neurobiological studies: Hahn, Foxe, & Molholm (2014).

303  most worrisome aspect: The 3-cueing method is reminiscent of the fruit-yellow-round combination of probabilistic constraints described back in Chapter 5. The difference is that a skilled reader combines a very large number of very small, unlabeled statistical constraints, automatically and unconsciously. I used games to illustrate the combination of probabilistic constraints precisely because the process occurs with explicit awareness and the constraints can be described linguistically. There is a connection between these explicit (games) and implicit (lexical statistics) processes nonetheless: a researcher named David Rumelhart. The 3-cueing approach can be traced to a 1977 article of Rumelhart’s, highly influential in its time, “Toward an Interactive Model of Reading,” which described reading comprehension as “the process of applying simultaneous constraints at all levels.” Having described the general idea, Rumelhart asked, What are these constraints, and how are they learned and combined? He then led the revival of the neural network approach, conducting pioneering work on statistical learning procedures for multilayer networks and developing (with McClelland and Hinton) a general framework for applying such networks to human cognition. That framework was the basis for the reading models I’ve described. Who was David Rumelhart? A fascinating, pivotal figure in psychology, artificial intelligence, education, and neuroscience. Educators absorbed his idea that comprehension involves combining multiple types of information; the 3-cueing method applies this basic idea to words instead of texts, Rumelhart’s original interest. They didn’t take up the subsequent work on learning, sticking with intuitions about cues and explicit instruction in strategies for using them. The 3-cueing method is like doing constraint satisfaction without a net—that is, without a theory that specifies what “cues” are, how they are learned, how they are combined, and how they function in reading words and comprehending texts.