The New Reading Debates: Evidence from Science, Art, and Ideology
by Jeanne S. Chall - 1992
Examines recent debates on the teaching of reading from the standpoint of science, art, and ideology, focusing on the reading theories, research, and practice of the past two decades. The analysis emphasizes the issue of whole language versus phonics, noting the effects of whole language and phonics on reading achievement. (Source: ERIC)
During the past decade, teachers and parents have witnessed a series of heated debates on the most basic issues in the psychology and teaching of readingwhat to teach, how to teach, and even whether to teach. The debates have included all who study and teach readingthe researchers who conduct scientific studies of the psychology and pedagogy of reading, the teachers who teach children in the classroom, the college faculty who teach the teachers, the special educators who are responsible for the reading of students with various reading difficulties, and the policymakers, particularly those who make policy for students at risk for reading failurestudents from low-income families, children whose native language is not English, and children with learning disabilities.
Controversies in the teaching of reading are not new. A review of the history of American reading instruction finds, for example, heated debates in the 1920s regarding the value of silent over oral reading, whole words over phonics,1 the hazards of teaching the alphabet before teaching words, and the value of stories dictated by the children, rather than primers for the childs first reading text.
The new debates, as will be seen later, are quite similar to those of the 1920s. Although they are also concerned with reading beyond the beginning, their major focus, as in the 1920s, is on beginning reading. There seems to be an important difference, however. The professional literature of the 1920s and 1930s, compared with that of the 1980s, appears to be more reasonedan ironic twist since there was infinitely less research and theory on which to base the reasoning. In contrast, the reading literature of the 1980s and early 1990s uses stronger rhetoric and seems to base its positions more on ideology than on the available scientific and theoretical literature. The strong rhetoric and the ideological base for the suggested reforms of the 1980s have led more than one newspaper reporter to refer to these reading debates as the reading wars.
This article treats the current debates from the standpoint of the science (the available scientific basis for the proposed reforms, as compared with traditional approaches); the art (the practice that has worked); and the ideology (the values and attitudes that seem to lie behind preferences).
I will focus further on the reading theories, research, and practice of the past two decades because I think they are critical if we are to understand the present debates. I further limit my analysis to perhaps the most heated of the various issueswhole language versus phonics, with the understanding that some students of reading view whole language differently. For example, David Pearson views whole language in opposition to reading textbooks, that is, basal readers, not phonics.2 Others place whole language on the side of reading whole books, particularly literature, not short selections in the basal reader texts; still others stress that in whole language skills are not taught, but inferred from natural reading. For others, it means empowering teachers to teach reading as they wish. For still others, it means integrating the teaching of reading with writing, speaking, and listening. For a growing number, it means a philosophy of education and of life, not merely a method of teaching reading.3 It is difficult to discuss whole language since its meaning differs widely from person to person, and even includes, in some schools, teaching phonics and using basal readers as components of a whole language program.
There is a further problem in discussing whole languagethe tendency of its proponents to claim novelty for the use of many good practices when history tells us they have been in wide use for a long, long time. For example, many whole language proponents claim that their use of authentic literature is a unique feature of their program. Yet literature has been a part of reading instruction since Noah Websters spelling book. Although whole language proponents tend to blame phonics instruction on a paucity of literature in reading textbooks, one should note that during the past two decades, the amount and quality of the literature in the reading textbooks increased as the teaching of phonics increased. The faster pace of learning to recognize words when phonics is taught makes possible earlier use of more advanced, quality literature. Further, the combined use of reading, writing, language, and speaking claimed by many whole language enthusiasts as their discovery has been the basis of remedial instruction since the early 1920s, when it was called a multisensory approach.
For my analysis, therefore, I will consider whole language mainly with reference to its stand on phonics. As will be seen, I think it is perhaps the essential distinction between whole language and traditional approaches to teaching reading.4
I will discuss, first, the whole language/phonics issue by considering the reading research and theory of the past twenty years, the resulting applications, and the accompanying rhetoric. Second, I will consider the effects of whole language and phonics on reading achievement -whether they have brought improved or declining scores. Can these changing conceptions of reading, for example, help explain the trends in reading scores over the past twenty yearsthe gains and the declinesreported in national assessments?
LITERACY NEEDS OF THE PRESENT AND THE FUTURE
Before I take up these questions; it is important to acknowledge that the national concern for literacy is responsible, at least in part, for the present concerns with appropriate methods. We are reminded daily by many economists that we may slip from our status as a world-class nation if our work force does not become more literate. They remind us that, when we were a manufacturing nation, fewer people needed to be highly literate, but a high-tech economyone that produces and disseminates knowledge and symbols as well as things and goodsneeds more people who are highly literate. Indeed, there seems to be a growing mismatch between workers and jobs, with jobs that cannot be filled because workers are not literate enough. But it is not in work alone that there is a mismatch. Responsible citizenship also requires more literacy than it did two decades ago, and personal literacy needs seem also to grow with time. For example, the labels on food and medicine packages require considerable reading ability.5
Most reading researchers and school people are motivated to bring students to higher levels of literacy that will permit them to live in a more complex future world. At the present time this means the ability to read and write complex materials critically and creatively in order to solve problemsa standard of literacy that the National Assessment of Educational Progress (NAEP) h as labeled adept level, equivalent to the reading achievement of those in the junior and senior years of high school.6 In my developmental scheme of reading, I call it Stage 4, the ability to read difficult materials such as the New York Times or Time magazineor about a twelfth-grade level on standardized reading tests, a level that permits reading with understanding and critical reaction of difficult texts for information, inspiration, and problem solving.7
NEW READING CONCEPTIONS OF THE SEVENTIES AND EIGHTIES
The new reading conceptions of the past twenty years can be divided into two groups. One is concerned mainly with reading comprehension. It is strongly based in research. Most of the research on reading comprehension has been concerned with the reading of those in about grade 4 and higherthose who can already recognize and decode words. This research has further focused on the role of language (particularly vocabulary), background knowledge, and reasoning. One of the main findings coming out of these studies is the importance of background knowledge in reading comprehension. Thus, comprehension of texts depends not only on what is in the text, but on the prior knowledge the reader brings to the text.
These studies found also that vocabulary (word meanings) affects reading comprehension significantly in ways similar to that of background knowledge. In order to comprehend more difficult texts, readers need an extensive vocabulary. Readers familiarity with different text structures has also been found to be related to reading comprehension. Thus, text organization and cohesion are important factors of the text in enhancing reading comprehension.
Another factor associated with reading comprehension is metacognitionthe readers consciousness-of his or her comprehension process. Research on metacognition found that good and mature readers are more aware of their comprehension processes during reading and are more adept than poor readers at self-questioning and monitoring their own comprehension. At its most rudimentary level, this monitoring involves asking oneself whether one understands what is read.8
Comprehension research of this kind came-on the scene about the middle 1970s following two series of conferences supported by the U.S. Office of Education (USOE) that called for more focus on reading comprehensiona USOE conference series in Washington and the Conferences on Early Reading series at the University of Pittsburgh.9 These were followed by the establishment of the Center for the Study of Reading at the University of Illinois, whose express mission was to study reading comprehension, not early reading.
These conferences and the establishment of the center had a significant influence on the research, practice, and rhetoric of reading instruction in the late 1970s and 1980s. Most of the reading research of those years turned to essentials of reading comprehension and to students beyond the primary grades. Few of these studies-were concerned with word recognition or decoding as such or as a factor in reading comprehension. Thus, according to most of these studies, reading is mainly a problem of cognition and language. Since the studies tended to include only those students who could decode, difficulties with decoding were not found to be a factor in reading comprehension.
Most of the significant reading research of the 1970s and 1980s did not study beginning reading or students with reading problemsanother reason why these studies did not find word recognition and decoding important factors in reading. It should be noted that this came about largely from the assigned mission of the largest research centerto study reading comprehension. That others followed is not surprising, since individual scholars often follow the leaders in their field.
At about the same time, or perhaps a little earlier, another group also held that reading is composed mainly of language and cognition.10 This group viewed reading as a psycholinguistic guessing gamethat meaning and language were the primary components of reading and that, at all levels, reading was a matter of prediction and individual reaction to text. According to this viewpoint, accuracy of word recognition came from correct predictions. This conception of reading later became the foundation for the whole language movement. Those holding to this view based their conceptions not on empirical research, as did the reading comprehension group, but mainly on linguistic theories.
In distinction from the comprehension group, which focused on later reading, the main concern of the whole language group was for beginning reading. However, the bottom line as to what reading is and is not was essentially the same as that of the comprehension group. It is mainly, they hold, a language-reasoning process, one that is based on readers predictions as they read for meaning. The recognition and decoding of words is a minor process, used only by poor readers who do not read well for meaning. Therefore, at all levels, reading, they say, involves mainly language and reasoning. It is through the use of reading for meaning and communication, they hold, that the beginner acquires the ability to recognize and decode words, not by being taught the association between letters and sounds. Reading is learned best, whole language proponents claim, when learned as language is learnednaturally, always in context and not through deliberate teaching.
It is on the issue of teaching that the two groups tend to differ. The whole language group deemphasizes teaching, and particularly direct teaching. The comprehension group tends to favor some systematic, sequenced instruction, but on the reading comprehension aspects of reading. Both groups have in common a strong preference for viewing reading as meaning-gathering right from the start and at all stages of reading development, and both tend to overlook and even show a not too subtle contempt for instruction in the phonological aspects of readingeven for beginners and for those with learning disabilities and dyslexia.
Thus, during the past twenty years, the two major new conceptions of reading, which stemmed either from cognitive psychology or from psycholinguistics, focused on the primary importance of language and thought. The psycholinguistic-based reading theories of the whole language group first appeared in the early 1970s and seemed to have a slow start, with a greater thrust in the 1980s. The cognitive-based theories of the reading comprehension group first appeared around the middle 1970s and seemed to have a strong start. The comprehension group was strongly grounded in experimental research, while the whole language group has based its arguments more on philosophical and intuitive arguments, taking, at times, an anti-experimental position. The bottom lines of both groups are quite similar. For both, reading development depends essentially on language and reasoning development. To prevent problems in reading, both claim that one must focus on the readers language and thought, from the very beginning to the highest levels of proficiency.
What about recognizing and decoding words? Do they have a place in defining reading? Indeed, does not literacy itself depend on acquiring knowledge about print and skill in its use, since even preliterate children and preliterate peoples have language and thought, but cannot read? As for those who have the most severe problems in learning to readthose with learning disabilities or dyslexiawhy do they seem to do well with language and thought when tested orally, but not when reading?
In answering such questions, the whole language response has been that learning how to recognize print comes naturally from being read to and from practice in reading connected texts. It does not need to be taught. Indeed, if taught, they claim-with no evidence that it may interfere with reading comprehension.11
The cognitive psychologists also have few answers to these questions since their research until quite recently has been conducted mainly with students who had already learned to recognize and decode words. A recent shift is found in the synthesis of the research on beginning reading by Adams.12
HISTORICAL ROOTS OF NEW READING CONCEPTIONS
When one goes further back into reading research and practice, to the early 1900s, for example, one finds that, indeed, these so-called new approaches are not essentially new. The labels, the rhetoric, the experiments may be new, but the broad conception of reading as language and reasoning can be found as early as 1917 in a study by E. L. Thorndike on the errors of fifth-hand sixth-graders when reading expository texts silently.13 He found that reading comprehension requires a knowledge of the meanings of words used in the selections, as well as reasoning. His conclusion was that reading is reasoningfor readers who have already learned to identify the words. Thorndike recognized at this early date the importance for silent reading comprehension of background knowledge, word meanings, and metacognition. Other research in the ensuing eighty years has consistently shown that once decoding is mastered, reading comprehension depends strongly on knowledge, word meanings, and cognition.14
Thus, while the comprehension researchers have contributed much to our knowledge of reading, there seems to be a long tradition with regard to the main factors in reading comprehension. Indeed, although we trace reading comprehension research from 1917, it is well to realize that the authors of American reading textbooks of the eighteenth and nineteenth centuries were also seriously concerned with reading for meaning and communication.15
Whole language also has deep roots in the past. It very much resembles practices that began in the 1920s and that favored, for the early grades, experience charts (i.e., the childs own language production) for reading in place of reading textbooks. Then, too, the child was expected to infer sound-symbol relationsthe teaching of letter-sound relations directly was not favored by reading educators of the time. Instead, emphasis was placed on learning to recognize words. Similar to whole language proponents today, they claimed that the best route to word recognition and decoding was through reading for meaning right from the startwords, sentences, and stories.
From about 1920 to the late 1960s, the central conception of beginning reading was that language and cognition were the central components. From the late 1960s through the early 1980s, there was a return, for a brief time, to an earlier conception- that teaching letter-sound associations and reading stories was effective in learning to read and comprehend, Indeed, the research from which this concept came has been confirmed and reconfirmed over an eighty-year period.16
RESEARCH EVIDENCE ON PHONOLOGICAL FACTORS
During the past twenty years, there has been also a growing body of data on the importance of the phonological component in reading. Basing their work on theories of psycholinguistics, cognition, child development, and learning disabilities, many researchers focus on the centrality of phonology (phonics, decoding, word analysis) in reading development. If not learned early, phonological factors may interfere with the use of language and reasoning in reading development.17 For older students with reading problems, meaningful language and cognitive difficulties may not be the central difficultyit may be mainly phonological, which further leads to their difficulty with fluency and comprehension. Research on very young children has confirmed findings of the 1930s that phonological awareness of wordsfor example, segmentation and blending of the separate sounds in wordstends to be a more potent predictor of beginning reading than word meaning and intelligence.18
Those concerned with phonological factors in reading tend to view reading as developmental, with beginning reading essentially different from more mature reading.19 By way of contrast, the cognitive psychologists and the whole language proponents view beginning and later reading as essentially the same process.20
To sum up, the conceptions of beginning reading that gained acceptance in the late 1960s through the 1970s included phonological as well as cognitive and linguistic factors. It was during the 1980s that beginning reading was defined primarily as being comprised of linguistic and cognitive factors, although the research evident on the important of phonological factors had grown.
CHANGES IN PRACTICE
How have these conceptions played themselves out in practice? Did reading instruction change? If so, did the changes in practice lead to improvement or to decline in reading achievement?
Let me state at the outset that it is extremely difficult to know how reading instruction is practiced during given periods of time and in given places. One can only approximate and infer, since we do not have national surveys of how reading is taught in classrooms, although we have surveys of how well children achieve. Therefore, we must rely on indices that reflect practice. For Learning to Read: The Great Debate, I made judgments about the use of methods in classrooms by analyzing the reading textbooks and teachers manuals of widely used readers.21 Others have done the same.22
Have the reading textbooks changed over the past twenty years? There is some evidence that for the primary grades the most widely used basal readers of the 1970s, as compared with those of the early 1960s contained earlier and heavier instruction in phonics and a more extensive vocabulary grade for grade. The reading textbooks of the 1980s, on the other hand, as compared with those of the 1970s, seem to have less phonics and a relatively heavier emphasis on reading comprehension and word meanings even in the first grade.23
EFFECTS OF CHANGED PRACTICES ON STUDENT READING ACHIEVEMENT
Indices of reading achievement are easier to come by. From 1971 to 1988 there have been six national assessments of reading by the National Assessment of Educational Progress. Let us start with the NAEP scores for trends in reading achievement over the past twenty years and try to infer the possible influences of the new reading conceptions on instructional practice and on student reading achievement.
From The Reading Report Card, 1971-88, we learn that, from 1970 to 1980, there was a steady improvement in the reading comprehension of nine-year-olds. However, during the 1980s (1980 to 1988) the scores did not improve and may even have declined: Nine-year-olds assessed in 1988 read significantly better than their counterparts assessed in 1971. However, this progress was made during the 1970s.24
If we assume that practice followed the changes in emphasis of the basal readers for the primary grades noted above, we have a hypothesis for the increases in the 1970s and the lack of improvement and possible declines of the 1980s. The NAEP scores of nine-year-olds (fourth graders) increased during the 1970s when the children were exposed as first and second graders to phonologically based programs, and they ceased to improve and possibly declined in the 1980s when the phonics component in the early grades was less strong and the comprehension and word meaning components were greater. Thus, although the expectations of those holding the new viewpoints (comprehension and whole language) would be for increased scores on reading comprehension of nine-year-olds in the 1980s, when these components were stressed in the reading program, it would appear that the increased emphasis on comprehension in the beginning reading textbooks and a decrease in phonics teaching may have led to decreases, not increases, in reading comprehension of nine-year-olds. In contrast, the scores of the nine-year-olds during the 1970s, when there was a stronger emphasis on learning letter-sound relations and less focus on language and cognition, increased.
Are early advantages in reading maintained in later years? In the 1988 NAEP, increases were found among seventeen-year-olds. This is explained by NAEP as due, at least in part, to an early advantage in their reading scores when they were nine in the 1970s.25 Here again we have evidence that there was an increase in reading comprehension even among the seventeen-year-olds when their beginning reading programs had a greater emphasis on phonological factors, that is, on phonics (as in the 1970s).
Still, these are mainly trends. The NAEP report is cautious about interpreting these trends. For viable explanations, they call for additional research. I certainly would endorse such research and hope that it does not shy away, as do most NAEP reports, from comparing what goes on in schools and classroomsfor example, the methods and materials used, the instructional emphases, and the time spent on the various activitieswith students reading scores. Indeed, there is considerable evidence that methods and materials and other school factors do make a difference in students reading achievement.26 When such studies are undertaken, we may indeed find that the beginning reading programs of the 1980sprograms that put a greater emphasis from the start on reading comprehension and whole languagemay be related to the declines reported by NAEP in the scores of the nine-year-olds in the 1980s. We may also find that the beginning reading programs of the 1970s, which paid more attention to the phonological, to the alphabetic principle, to decoding, to phonics, much maligned today, may have contributed to the rising scores of the nine-year-olds in the 1970s and to the higher scores of the seventeen-year-olds in the 1980s, as reported in NAEPs most recent Report Card.
These findings from NAEP confirm those of other studies and theories of readingthat the best fit for the data is that reading is cumulative and developmental, and needs different instructional emphases for different stages of development. Beginning reading may look the same as mature reading, but it is quite different. Beginning reading has much to do with phonology. As reading develops, it has more to do with language and reasoning.27
What does all of this mean? First, I think it means that, although we have a very serious literacy problem, the sources of which are broad and deep in socioeconomic, cultural, and neurological factors, there is much that is in the hands of schools and teachers. We can, on the basis of what has been found, make adjustments in the curriculum to ensure better development of reading. It points, too, to the importance of basing practice on sound theory and research tested over extended periods of time, -and to the limited advantages and even disadvantages of basing practice on enthusiasms alone. The accumulated knowledge on literacy suggests that different aspects of-reading be emphasized at different stages of reading development, and that success at the beginning is essential since it influences not only early reading but reading at subsequent levels, even in high school and college. It demonstrates that a beginning reading program that does not give children knowledge and skill in recognizing and decoding words will have poor results.
The accumulated research indicates also that, while beginning reading is the responsibility of reading teachers, later reading cannot be left to reading and English teachers alone. Teachers in the various disciplines need to help students acquire the knowledge, the concepts, the special and general vocabularies, and the higher-level thinking required to read with adequate comprehension in the middle grades, high school, and college. It is the various curricular areasincluding Englishthat contribute to the knowledge, language, and cognition needed for continued reading development and for academic success, for work in high technology, for responsible citizenship, and for making personal choices in an increasingly complex world.
THE ROLE OF VALUES AND IDEOLOGIES
All of these suggestions, although based on recent research-and surveys, were valid at earlier times, based on earlier research and practice, Why do we seem to ask the same questions and get the same answers over and over again? Why, one may ask, do we not base our methods on existing researchespecially when there is so much of it, as in beginning reading?
For this we must consider more powerful forces at workvalues, ideologies, philosophies, and appealing rhetoric. Since the early 1920s, there has been a preference for reading methods that start with whole words or sentenceswhat I called earlier a meaning emphasis.28 These methods were adopted during the great educational reforms of the late 1800s and early 1900sthe early childhood education movement, progressive education, and the child-centered curriculum. Although meaning-emphasis methods can be traced back even earlier, it is during the 1920s that they gained widespread use and took on, as they do today, all of the qualities and values of love, care, and concern for children. These reforms claimed that reading for understanding was to be done from the start. They abhorred rote learning. Concern with print and phonological aspects of reading was seen then, as it is now, as pulling the reader away from understanding and toward rote learning, and therefore was to be avoided. The view in the 1920s was that concentrating on reading interesting stories (with little or no teaching of the forms and sounds of words) will result in better reading comprehension; further, that this procedure will result in lifetime reading, while learning phonics was viewed as dull and dreary and as discouraging the development of lifelong reading. Although the research of the past eighty years has refuted these claims, they persist. If they are relinquished for a period, they return as new discoveries, under new labels.
Why do these concepts of reading return again and again? Why are they so persistent? I propose that they are deep in our American culture and therefore difficult to change. These conceptions promise a quick and easy solution to learning to readreading without tears, reading full of joy. They are the magic bullet that is offered as a solution to the serious reading problems of our times. Further, phonics requires knowledge, effort, and work. The whole word or whole language way has always promised more joy, more fun, and less work for the childand for the teacher.
A one-stage theory of reading seems more attractive to Americans than a sequential, developmental theory. Since the whole language and reading comprehension proponents claim that reading is the same at the beginning and at the end, teachers are required to know less than for a developmental view of reading. If teachers know a few basic concepts about reading, the whole language and comprehension proponents would claim that they can teach beginners and advanced students alike.
Whole language, in particular, seems to say that a good heart goes a long way, and the less teaching the better. It fears structure more than no learning. Its major concern is that the higher cognitive processes be used in reading, right from the start, and it flees from the idea that there may be basics to be learned first.
These views are being debated in other school subjects as well. In math, the current thrust is toward learning concepts and away from computation. In history, the concern is for teaching broad ideas, not facts. Although the National Assessments indicate that our children have grave deficiencies in the most basic learnings,29 the current focus is away from teaching these and toward teaching concepts and higher mental processes. Thus, as for reading, the preferred emphasis today in most areas of the curriculum is on the higher cognitive processes from the start.
I propose that it is these particular views of the child, of the teacher, and of learningviews that have been with us for about a centurythat make us accept or reject a particular conception of reading. These views attract many teachers to whole language and to a comprehension focus even in grade one. It is a romantic view of learning. It is imbued with love and hope. But, sadly, it has proven to be less effective than a developmental view, and least effective for those who tend to be at risk for learning to readlow-income, minority children and those at risk for learning disability.30
Since the 1920s the prevailing ideology of reading has been one that views the child as self-motivated and joyous. This view holds that a child learns to read as naturally as he learns to speak, if only we encourage him to use his language and his cognition when he reads interesting books. Teaching skills and tools, especially those related to print, are to be avoided since they distract the learner from the naturalness of the process and the acquisition of meaning.
Very little is said about the children who have difficulty learning to read with this emphasis on language and cognition. Faced with the high incidence of failure, proponents of this approach answer that poor reading stems from weakness in language and inappropriate instructioninstruction that focuses on teaching the skills and tools.
The values and ideology I have briefly depicted can be found to underlie most reading programs from the 1920s on. From time to time there is a greater acceptance of the need for teaching skills and tools when it is realized that many children are falling behindparticularly those from low-income families, from minority groups, and those at risk for learning disability. Historically, however, these periods seem to be short-lived. Such a period existed during the 1970s, but by the 1980s the thrust was again toward the more romantic, charismatic, and global methodsmethods seen as natural and joyful. These methods are valued so strongly because we want them to solve not only our curriculum problems but our economic, political, and social problems as well.
I found an apt illustration of this view point in the words of James David Barber, a political scientist at Duke University: What has happened in American politics is its drift to sentimentality and emotion, which substitute for looking at the facts that face you at the moment.31
Definition of Terms
Phonics refers to the relation between letters and sounds. In lay terms, it means giving the right sounds for the letters-m printed words. Other words have been used to refer to phonics over the-years: phonetic analysis, word analysis, code-emphasis, decoding, the alphabetic principle, and phonological or phonemic awareness.
Basal readers are reading programs designed to provide total reading programs for students and teachers containing a system for teaching reading (in the teachers manuals), a collection of stories and selections for pupils to read (the readers), exercises for additional practice (the workbooks), and tests and other correlated books, games, and materials.
Higher cognitive processes refers to the higher mental abilities involved in reading comprehension as distinguished from word recognition. It also refers to the increasing need for knowledge and for high-level thinking as one advances from beginning to mature reading.
This article is based on a paper presented at the Orton Dyslexia Society Symposium on whole Language and Phonics, held in Minneapolis, March 1990, and published in a different form in All Language and the Creation of Literacy (Baltimore: The Orton Dyslexia Society, 1991), pp. 20-26. I became interested in reading debates with the publication in 1955 of Rudolf Fleschs Why Johnny Cant Read (New York: Harper &Row). The reactions to it were so polarized that even in the early 1960s the debates continued. In 1962, following a conference on beginning reading, I requested and was awarded a grant by the Carnegie Corporation to study the evidence on best methods for teaching beginners to read. The study, conducted from 1962 to 1965, involved analysis of the relevant research on beginning reading conducted in classrooms, laboratories, and clinics; analysis of the widely used materials for teaching beginning reading; interviews with their authors and editors; observation in hundreds of classrooms; talks with teachers and administrators; and analysis of selected classic studies of students who had difficulty learning to read. The results, which appeared in the book Learning to Read: The Great Debate, published by McGraw-Hill in 1967, indicated that, overall, the evidence favored what I called a code-emphasis approach for beginnersa phonic or linguistic approachone that put an emphasis on learning the alphabetic principle. The research evidence did not favor the prevailing meaning-emphasis approachone that stressed reading for meaning at the start. In 1983, I published an updated version (Learning to Read: The Great Debate, updated ed., McGraw-Hill, 1983) summarizing the relevant research from 1967 to 1983. It concluded that the evidence for a code-emphasis approach, one that put early focus on the alphabetic code rather than on the content and meaning of what is read, was even stronger than it was in 1967particularly for children at risk, those from low socioeconomic level families, children with early signs of dyslexia, and children of low ability.