Since publication of the first edition in 1978, this book has become a standard work for all those interested in reading. This book is about the process of reading, the perceptual and language skills in involves, and the nature of the task facing anyone learning to read. Frank Smith’s aim is to help teachers towards the understanding they must have if they are to be successful, whatever approaches they adopt or materials they use. In his introduction to this second edition the author reinstates his opposition to ‘the programmatic approach to reading instruction’ which ‘ has clearly failed’. In the body of the book he focuses on the nature of reading without asserting how reading should be taught or improved at any level, although he argues that the solution to ‘reading problems’ cannot lie in any particular methodology or set of reading materials. This book therefore has crucial practical implications. No teacher is short of advice on what to do. But choices have to be made between differing and sometimes conflicting practical suggestions, methodologies and materials. An understanding of the nature of reading will help the teacher to make the right choice. In particular, the author introduces the reader of this new edition to the notion of ‘the literacy club’ which children need to join if they are to become successful readers. At the heart of Frank Smith’s analysis is the seemingly simple truth that it is only through reading that children learn to read. Much of the book is taken up with showing why this is so (confirming the intuition shared by many teachers). The argument follows that the teacher’s fundamental task is to make reading easy for the learner. An understanding of the nature of the process also helps the teacher decide what will make learning easy and what will make it difficult. In addition to the idea of ‘the literacy club’ the author has included newly researched material on early literacy and a new chapter on ‘the promise and threat’ of computers in literacy education. Other revisions take account of recent research, and the book remains primarily concerned with what teachers of reading must know to be successful, and what they must not do. It ranges over a number of theoretical disciplines, but the language is always non technical, the approach direct and the emphasis practical. READING FRANK SMITH Preface to the second edition In the first edition of this book, I described the ease with which children become literate when they are personally involved with people actually making use of the signs, labels, lists, newspapers, magazines and books in the world around them. In contrast, I examined the difficulty many children experience with formal reading instruction based on exercises, material and drills that are to a large extent nonsensical. The philosophy of this kind of instruction, which I now call 'programmatic', is that reading is a set of skills which can be taught and mastered in a pre determined sequence, provided there is a closely managed 'systems' approach with properly specified objectives and frequent tests. Programmatic instruction is the antithesis of meaningful language experience for teachers and children. It is primarily a method of control. Since the first edition was published in 1978 the issues have become more clearly demarcated and the conflict more acute. On the one hand, the programmatic approach to reading instruction has clearly failed. No one claims that children are reading better today than they were ten years ago, or even than they were twenty-five years ago, when the development of rigorously controlled instructional programmes for literacy education began to proliferate. That was the time when many influential educational leaders thought that the technology that would put man on the moon would also be a certain cure for illiteracy. One might think that most politicians and administrators by now would have recognized that the 'remedy' for illiteracy might instead be a contributing factor. But with the failure has come a clamor for even more programmes, for even tighter control of schools and teachers in the name of 'accountability'. All of this may sound reasonable but it constrains teachers to teach in the manner decreed by outside authorities who know nothing of the particular children in their classroom, of their unique individual interests, concerns and difficulties. Teachers are expected to conform to the programme, no matter how trivial, misconceived, and ultimately damaging to literacy the instruction might be. If the bleeding treatment does not seem to be restoring the patient to life, bleed some more. On the other hand, extensive research in many cultures has confirmed what many experienced teachers have known intuitively, that children become readers when they are engaged in situations where written language is being meaningfully used, much the way they learn spoken language from their association with people around them who use speech in meaningful ways. This is the opposite of programmatic instruction. The implications of this research have been slow to break through at decision making levels of education, primarily because they would replace stringent outside control of classroom activity with trust that teachers can teach and that children will learn if both are given reasonable autonomy. As the theoretical issues and conceptual differences have become more clearly defined, the conflict itself has become more acute. A new technological factor has speeded up the time scale and exacerbated the threat. I am referring to the computer, the ultimate programmatic instructional device. People who like programmatic instruction love computers, because they promise to teach all the exercises and drills more efficiently than teachers and at much less cost. Used in this way, computers could destroy both literacy and the teaching profession. Most of what I said in the first edition about children, learning and language remains unchanged. The nature of all of these remains the same, despite technological developments. We may endeavor to teach children differently, but their brains learn as they always did. Language is changing only in its most superficial aspects, and it is always developing in that way. Where recent research has led me to revise what I earlier said about children, learning or language in this book, it has generally been to provide new evidence or observations in support of particular points. A significant addition to this second edition is recent material on early literacy, enabling me to expand the topic of learning to read into a new chapter in its own right. I have also added a closing chapter on the promise and threat of computers in literacy education. It is not difficult to make reading impossible. I am not referring to such obvious disruptions as distracting the reader's attention, tearing out pages, defacing the print, switching off the lights, or even making the reading material a book written in a dull and incomprehensible manner, although a book that does not make sense will certainly be difficult to read. I mean that it is easy to make a book unreadable for a person who otherwise would be able to pick up that same book and read it fluently. One very effective way to produce incomprehensibility is to ensure that the person trying to read the book is apprehensive about making a mistake, for example while reading aloud. Reading is not easily accomplished if you are nervous about your performance. Equally handicapping can be the endeavor to memorize every trivial detail in order to avoid being caught out in a subsequent cross examination or written exercise, especially if the exercise is to be graded and the evaluation to become part of a permanent record that perhaps could make a difference to a career. Anxiety and the indiscriminate effort to achieve total recall both help to explain the widely experienced phenomenon of text-books that are nonsense before an examination yet transparently comprehensible after. If my catalogue of obstacles to reading comprehension sounds suspiciously like a description of the circumstances in which many students find themselves in school, the coincidence is entirely intentional, just as it is not difficult to make a book unreadable, so it is easy to make learning to read impossible. Even when there is a sincere intention to help children to read, the instruction can beset them with handicaps guaranteed to interfere with learning. Bad habits can be taught so effectively that whatever an individual tries to read in later life will be found incomprehensible. This book is primarily concerned with the process of reading, with the perceptual and language skills involved in reading and with the nature of the task confronting children attempting to learn to read. But the implications of the book are instructional. I shall try to show why it is that it is only through reading that children learn to read, and that a teacher's role must therefore be to make reading easy for every child, in particular I shall argue that children can learn to read only through materials and activities that make sense to them, that they can relate to what they already know or want to know. Anything that children cannot relate to what they already know will be nonsense to them, whether or not it is nonsense to the teacher. Expecting children to learn to read through nonsense is the easiest method of making learning to read impossible and unfortunately the most widespread. This book is addressed to anyone interested or involved in the topic of reading -- primarily to teachers and to students and faculty in colleges of education, although it should also be relevant to parents and to others who perhaps may be concerned about their own reading ability. How can one book be directed to such a varied audience? By focusing on the nature of reading, rather than asserting a particular point of view about reading instruction. This book contains no easy message about how reading might be taught or improved at any age level: in fact I argue that the solution to 'reading problems' cannot lie in any particular reading methodology or set of reading materials. The only practical educational conclusion that can be drawn from an analysis of the relatively few fundamental ideas that have come, gone and continually returned throughout over twenty centuries of reading instruction always with the result that some children have learned to read but others have failed is that the universal concern should change from what teachers should do to what teachers should know. Teacher training institutions seem often to expect teachers of reading to do their jobs without a basic understanding of the topic they are concerned with and teachers frequently achieve remarkable success under such conditions. Education would be in a far sorrier state today were it not for teachers' untutored intuitions and insights. But as far as their own formal education is concerned, most teachers are left with a patchwork of conflicting practical suggestions and a belief that the nature of reading is either too obvious to require analysis or too mysterious to be discussed. Yet the nature of reading cannot be ignored since the success of reading instruction must ultimately depend on whether the instruction makes reading easy or difficult. And in order to understand whether a particular method or set of materials will make reading easy or difficult, whether they will make sense or nonsense to a child, there is no alternative to acquiring an understanding of reading itself. The purpose of this book is to contribute to such an understanding of reading. The orientation is scientific; the assertions that I make about language and the brain are based on scientific evidence. But I have not overburdened the text with detailed references, footnotes or theoretical debate. My aim is readability: the technical discussions and supporting arguments can be found in my more specialized book Understanding Reading (third edition: Holt, Rinehart and Winston, 1981). My emphasis, as I have indicated, is on the importance of sense. Nothing can be taught unless it has the potential of making sense to the learner, and learning itself is nothing but the endeavor to make sense. The effort to teach or to inform, therefore, can be nothing but an endeavor to make sense, to be comprehensible. I hope indeed that this book will make sense to everyone concerned with literacy. F.S. Victoria 1985 1 Making sense of reading A preliminary glance through these pages might indicate that despite its title this book contains very little that is specifically on the topic of reading. But the first point to be stressed is that there is nothing about reading that is unique. In order to make sense of what takes place when we read, a number of more general topics must be explored. Reading and the brain There is nothing about reading that is unique, whether one considers the structure or the functions of the brain. Despite the diagrams that sometimes appear in textbooks, allocating specific responsibilities to different locations on the surface of the brain. Reading is not the exclusive concern of any one particular part. Anatomists and physiologists have not isolated a specialized 'reading center' in the brain. Many areas of the brain are active when we read, but none is involved with reading to the exclusion of anything else. Illness or injury may occasionally affect the working of the brain so that ability to read is disturbed, but almost certainly some more general activity involving either language or vision will also be impaired. There is also nothing about reading that is unique as far as intellectual processes are concerned. From the point of view of language, reading makes no demands that the brain does not meet in the comprehension of speech. And visually there is nothing in reading that the eyes and brain do not accomplish when we look around a room to locate an object or to distinguish one face from another. Researchers are discovering that in order to understand reading they must consider not just the eyes but also the mechanisms of memory and attention, anxiety, risk taking, the nature and uses of language, the comprehension of speech, interpersonal relations, socio cultural differences, learning in general and the learning of young children in particular. All these are topics that we shall be concerned with in this book, at least to the extent that they are relevant to reading. Fortunately there is a good deal in all these topics that is interesting in its own right. Fortunately also, when all these topics have been examined for the light they shed on reading, there is very little about reading itself that is left to be said. It is rather like climbing some hills; not until you reach the top do you get the best view of where you have been. When you have covered all the ground, not only have you gained insight into reading, but you also know a good deal about many other aspects of the human intellect, especially with respect to learning. Teachers in other areas find that such an analysis of reading also gives them a deeper understanding of the problems of comprehending subjects like science, music or mathematics. Reading and instruction Examination of a wide range of topics relevant to reading not only leaves little to be said about reading itself, it leaves little to be added about how reading should be taught. Instructional implications become self-evident. The more closely one delves into the nature of reading, the less one tends or needs to be dogmatic about what teachers ought to do in classrooms. Such an open minded attitude towards instruction runs counter to the majority of books about reading, and indeed to much of the research that has been carried out on reading. Partly because reading is a complex topic, but mainly I think because it is so widely regarded as a compelling educational problem, priority in most studies of reading is usually given to what might be done to improve instruction rather than to understand the process in the first place. As I suggested in the preface, the training of teachers often seems to assume that the problem is instruction that teachers should be told what they should do rather than what they should know. The training of teachers does not invariably encourage them to make their own decisions. Even books that purport to be concerned with the psychology or process of reading tend to be heavily weighted in favor of some particular theory about how reading should be taught. The authors are perhaps not entirely to blame. Often they are or have been practicing teachers who rarely have had the opportunity in the classroom to separate what they do from why they do it. But even 'experts' from outside fields (like psychology or linguistics) who write about what they think is relevant to reading are likely to be held responsible for translating their conclusions directly into practice. The belief that it is not enough to inform teachers -- that they should be instructed -- exists even among teachers themselves. To argue that questions like 'What is the best way to teach Reading?' are too simplistic and likely to provoke the ultimate challenge: Don't beat around the bush. Tell me what you would do if you had to face thirty five kids in a reading class on Monday morning.' This is rather like asking for a straight answer to the simple question of how far nails should be driven into wood. The only reasonable response to such a blanket question is rather impolite: 'If I were really responsible for teaching reading to thirty five children on Monday morning, I would make sure I knew enough about reading in general and about those children in particular that I would never have to ask an outsider such a question.’ Besides the teacher's problem is never the absence of advice. The world is full of experts willing to promote a favorite remedy for reading problems. But if one authority says Method A is best and a second prefers Method B, how can a third opinion possibly helps make up the teacher's mind? The teacher does not need advice, but understanding. The teacher still has to make the decision. And there is no point in expecting research to resolve the dilemma. Hundreds of methods of teaching reading exist, mostly minor variations on a few traditional themes, and literally tens of thousands of research studies have been done comparing one method or procedure or set of materials with another. And out of these tens of thousands of studies only one basic and incontrovertible conclusion can be drawn: all methods of teaching reading can achieve some success with some children some of the time. In the two thousand year recorded history of reading instruction, as far as I have been able to discover, no one has devised a method of teaching reading that has proved totally fruitless. No matter how bizarre an approach might seem to be, research studies can always be found claiming not only that it succeeded in teaching something, but also that it did so relatively better than some other method. All this research seems to prove just one thing that children are incredibly flexible and resilient. Children seem able to learn to read despite the method of instruction that is employed. At this point, credit must be given to teachers (who too often are held responsible only for the failures). Many teachers seem able to teach children to read, no matter what method they employ, although others, with some children at least, may follow exactly the same procedures in vain. In any case, the better teachers are always intelligently eclectic: they use what works when they see that it can work, even though they may not be sure why. Unhappily, although every method of reading instruction seems to achieve some success with some children, no method succeeds with all children. As we shall see, all methods of teaching reading must exact a price from the child attempting to learn, and in some circumstances therefore all methods can interfere with reading. So once again, what the teacher needs is an understanding of the particular possibilities and costs (to the child) of different methods and materials, an understanding of the particular children and what they are finding easy and difficult, and an understanding of reading and how children must learn to read. Without such understanding, teachers cannot make up their own minds about methods and materials, and are forced to fall back on the exhortations of experts or the importuning of publishers. Such teachers must work without knowing why they succeed or fail. Without understanding, instruction is founded on superstition. The teacher's dilemma in the selection of instructional techniques is also not likely to be resolved by the discovery of a new and ideal method for teaching all children. From the huge expenditures of commercial enterprises and government agencies in pursuit of a technology of reading instruction that will prove infallible, and preferably ‘teacher proof’, the only conclusion that can safely be drawn is that nothing could possibly be invented that is significantly better than or even different from the methods and materials we have always had available, even if they are dressed up for use with computers. Children have been learning to read for centuries without the benefit of technology. We need to discover only what we have been doing wrong and doing right in the past. Many people reading this book developed their skill in stuffy, badly lit and overcrowded classrooms, with desks nailed in rows to the floor, using ill printed and sanctimonious materials in an authoritarian atmosphere that would rarely be tolerated today. Today's children are not that much different from the children we were, and we learned to read without the advantages of systems research, behaviorally defined objectives or teacher account ability. The 'systems approach' involves breaking down reading into 'component skills and I propose to argue that 'breaking down reading' makes learning to read more difficult because it makes nonsense out of what should be sense. To learn to read children need to be helped to read. The issue is as simple and as difficult as that. Two basic necessities for learning to read are the availability of interesting material that makes sense to the learner and an understanding and more experienced reader as a guide. As far as materials are concerned, the problem is one of surfeit rather than scarcity. We are inundated with books, magazines, newspapers, comics, television commercials, computer display terminals, street signs, brand names, notices, bill boards, handouts, instruction manuals, menus, timetables, labels and wrappers, not to mention blank paper and a host of other writing surfaces ranging from walls to T shirts and a variety of marking instruments. The teacher's dilemma is again not so much to find as to select. I am not so sure about the supply of understanding adults. Certainly there are many well motivated people one in every classroom, I would trust. But whether they all have sufficient understanding, either about how children learn or what is involved in reading, is another matter. There is one final reason why blind faith in methods or materials will not solve problems of instruction in reading, a reason that will require the remainder of this book to explain. The final reason is this: children cannot be taught to read. A teacher's responsibility is not to teach children to read but to make it possible for them to learn to read. There is a difference. I want to show in this book that the real skills of reading, that have made readers out of you and me, are not skills that are formally taught at school or ever could be. We have acquired these skills only through the practice of reading. Most of the time we are not even aware of what these skills are, and they are certainly not explained in the majority of teacher training institutions. Learning to read I have said that there is nothing unique about reading physiologically, visually or linguistically: these are assertions that will be justified in the course of this book. I shall also try to show that there is nothing unique about learning to read. No special, exotic or particularly difficult learning skills are required, learning to read involves no learning ability that children