gg \4 Cornell University f Library The original of tliis book is in tine Cornell University Library. There are no known copyright restrictions in the United States on the use of the text. http://www.archive.org/details/cu31924012985044 THE PSYCHOLOGY AND PEDAGOGY OF READING THE MACMILLAN COMPANY NEW YORK • BOSTON • CHICAGO DALLAS • SAN FRANCISCO MACMILLAN & CO., Limited LONDON ■ BOMBAY • CALCUTTA MELBOURNE THE MACMILLAN CO. OF CANADA, Ltix TORONTO THE PSYCHOLOGY AND PEDAGOGY OF READING WITH A REVIEW OF THE HISTORY OF READING AND WRITING AND OF METHODS, TEXTS, AND HYGIENE IN READING BY EDMUND BURKE HUEY, A.M., Ph.D. PROFESSOR OF PSYCHOLOGY AND EDUCATION IN THE WESTERN UNIVERSITY OF PENNSYLVANIA IStio gfltfe THE MACMILLAN COMPANY 1920 All right! reserved LB 1513 Copyright, igo8. By the MACMILLAN COMPANY. Set up and electrotyped. Published January, 1908. (^\(\c^c^X. J. B. Gushing Co. — Berwick ifc Smith Oo. Norwood, Mass., U.S.A. MY FELLOWS IN RESEARCH WHOSE INVESTIGATIONS OF READING AND LANGUAGE ARE HERE JOINED WITH MY OWN THIS VOLUME IS PRESENTED BY THE AUTHOR IN THE HOPE THAT IT MAY RENDER SERVICE, AND WITH RESPECTFUL APPRECIATION OF THEIR PART IN ITS PRODUCTION PREFACE The writer's studies upon reading began nearly ten years ago, being first suggested by a question concerning the possibility of reading without inner pronunciation, raised by my friend and fellow-worker in the laboratory, now Professor G. M. Whipple of the University of Mis- souri. The reading process had long seemed to me to mirror the processes of thinking, and thus came to seem an appropriate subject for psychological analysis. Be- sides, the peculiar fatigue occasioned by reading caused a curiosity to know its sources, and the great variations and limitations in speed of reading suggested possibilities of improvement here. Such considerations gave birth to my experimental re- search. The field seemed clear. Diligent search in the literature showed only the preliminary experiments of Javal and his pupils, and those by Romanes and by Quantz, upon reading properly so called. Erdmann and Dodge were then completing their research, but I did not hear of their work until a year later. Reading thus offered to the experimentalist a practically unoccupied field. Ten years has given a development here of which experi- mental psychology may be proud. Dodge, Zeitler, Mess- Vm PREFACE mer, Dearborn, and others have thoroughly investigated important phases of reading, and the collected studies now present a very tolerable account of the main processes in- volved. It has therefore seemed to me that a conspectus should be made of this work, not to close the story but to furnish a new point of departure for further study, and to give perspective for new researches. Then it is due to education that from time to time the psychological investigations that have pedagogical bear- ings be edited, for such applications as education can helpfully make of them. And while engaged in this latter task, for reading, and falling in with much of the peda- gogical literature of the subject, it became ever more evi- dent that there was great need of bringing together the data not merely from the psychology of reading, but from the history of reading and of reading methods, from the current practice and points of view in the subject, and from the side of hygiene, drawing finally such conclu- sions as these collected data might warrant for the guidance of present and future practice in reading and learning to read. So the present volume has taken form, typical of books which, it seems to me, should be written for each of the more important school subjects, however poorly this volume may exemplify the type. Consider the need of this in the various subjects. Not to mention writing, a branch in which there is perhaps the most of needless con- PREFACE IX fusion and in which perhaps the greatest benefit would be derived from such a concentration of data, and num- ber, in which certain phases have already been well pre- sented, consider, for example, the value of such a treat- ment of geography. The psychological section would be mainly an outline of researches to be, but it would be of the greatest value to have these suggested, with our graduate departments full cf men looking for problems. The school subjects, ordinarily, involve some characteristic modes of mental and physiological functioning which fur- nish to psychology problems fruitful for psychology's own purposes. But pedagogically, what sort of symbols, for instance, are most effective instruments for thinking the earth, its divisions and dependencies ? Are actual experi- ences, the very appearances of mountains and cities seen in reality or constructed in miniature, the best geographical furniture for life's uses ? Or do symbols utterly unlike the realities, marks and colors upon maps and charts and globes, give us the most compact and convenient scheme for mind's dealing with the earth's forms ? Or are words, though totally unlike their objects, the best manipulators of meanings here, as they certainly are in some divisions of thinking? And what is the order in the development of capacity and interest, in the child and the race, for the various modes of symbolic presentation here, and for the various phases of geographical knowledge ? Such problems, but the first of a host that will suggest X PREFACE themselves to any competent investigator, are capable of at least partial solution from the data even now ac- cessible, if these be gathered from the various sources. And then, as every one knows, the history of geography and of geography teaching is full of valuable sugges- tion; and when presented vnth a review of present-day theory and practice and with the psychological data upon the subject, the whole cannot fail to give us a far better orientation and the possibility of distinct advance in this much-abused branch of study. It is to be hoped that the various departments of educational psychology, now becoming popular in our universities, will recognize the great service which they can render, both to psychology and education, by such organization and concentration of data concerning the various school subjects. Of course no two authors would select the same mate- rial for such a work upon reading. I have endeavored to present the most meaningful facts, and those researches in which more or less definite results have been reached. Completeness of treatment and of reference is out of the question in a subject having such various and intricate ramifications. Some of the pedagogical principles suggested by the psychological work are still in the " recept " stage. In working over the material one comes to feel their truth and their appUcability, but to attempt their logical state- ment or derivation would in many cases be premature PREFACE xi ajid would tend to arouse useless polemics. Teachers will usually be better satisfied to find them well founded empirically, and psychological values have guided me, at least implicitly, in their selection and use. In the work of collating and editing the data presented in the present volume, I have been greatly aided by the large number of writers and publishers who have cour- teously permitted me to publish so many extracts and illustrations from their works. To them is justly due a considerable share of the credit for whatever success the book may have. Professor Reeder deserves the main credit for gathering material and suggesting sources for my section on the history of reading methods and texts. The excellent volumes by Isaac Taylor, Hoffman, and Clodd were indispensable in preparing my sketch of the history of reading and writing. My thanks are especially due to Messrs. D. Appleton and Co., and to Kegan Paul, Trench, Triibner, and Co., for permission to use so many of their valuable illustrations. The American Book Co., the Fimk and Wagnalls Co., and the Macmillan Co. have also been especially indulgent. It will be noted that one of the chapters on the Hygiene of Reading has already appeared in the Popular Science Monthly. I wish to thank Professors E. C. Sanford, W. H. Burn- ham, W. F. Dearborn, and Henry D. Sheldon for sug- gestions from the reading of parts of the Ms., and Mr. Louis N. Wilson for efficient and kindly assistance 3ni PREFACE in the library and otherwise. I am also indebted to the genius of President Hall for much more than text or bibliography can well indicate. Professor and Mrs. Will Grant Chambers have given valuable criticisms and suggestions from a reading of the proofs, and the book owes much to their unfailing en- couragement and assistance. Mr. E. H. McClelland and Mrs. H. H. Fisher have kindly assisted with the revision of the proofs, and Miss Grace Kerr deserves special men- tion for patient care in typewriting most of the illegible Ms. To these and the other friends who have lightened the labor of the book's production, I express my grateful appreciation. In conclusion, it is a pleasure to acknowl- edge the intelligence and cheerful courtesy of the Mac- millan Co. and of the J. S. Gushing Co. in carrying out the plans of the author. E. B. H. Pittsburg, Pa., Christmas, 1907. CONTENTS INTRODUCTORY CHAPTER I rAGS The Mysteries and Problems of Reading . . . ,i PART I THE PSYCHOLOGY OF READING CHAPTER II The Work of the Eye in Reading • • CHAPTER III The Extent of Reading Matter Perceived during a Reading Pause 51 CHAPTER IV The Experimental Studies upon Visual Perception in Reading 71 CHAPTER V The Nature of the Perceptual Process in Reading . loa CHAPTER VI The Inner Speech of Reading and the Mental and Physical Characteristics of Speech . . -117 XIV CONTENTS CHAPTER VII FAGS The Functioning of Inner Speech in the Perception OF WHAT IS Read 142 CHAPTER VIII The Interpretation of what is read, and the Nature OF Meaning 152 CHAPTER IX The Rate of Reading 170 PART II THE HISTORY OF READING AND OF READING METHODS CHAPTER X The Beginnings of Reading, in the Interpretation of Gestures and Pictures 187 CHAPTER XI The Evolution of an Alphabet and of Reading by Alphabetic Symbols 203 CHAPTER XII The Evolution of the Printed Page .... 226 CHAPTER XIII The History of Reading Methods and Texts . . 240 CONTENTS XV PART III THE PEDAGOGY OF READING CHAPTER XIV rACB Present-day Methods and Texts in Elementary Read- ing 265 CHAPTER XV The Views of Representative Educators concerning Early Reading 301 CHAPTER XVI Learning to Read at Home 313 CHAPTER XVII Learning to Read at School. The Early Period . 336 CHAPTER XVIII Reading as a Discipline, and as Training in the Effective Use of Books 359 CHAPTER XIX What to Read; the Reading of Adolescents . . 371 PART IV THE HYGIENE OF READING CHAPTER XX Reading Fatigue 387 XVI CONTENTS CHAPTER XXI PAGB Hygienic Requirements in the Printing of Books and Papers 406 CONCLUSION CHAPTER XXn The Future of Reading and Printing. The Elimina- tion OF Waste 421 Bibliography 433 Index 447 L the psychology and pedagogy 'of reading INTRODUCTION CHAPTER I THE MYSTERIES AND PROBLEMS OF READING Reading, for our Anglo-Saxon forefathers, meant, coun- seling or advising oneself or others (A.-S. rcedan, to advise). To read was to get or to give counsel from a book, origi- nally from a piece of bark on which characters were inscribed, at least if the reputed connection of book and heech can be sustained. The accessory notion of talking aloud seems to have been implied in the word, as it was also in the Roman word for reading. To the Roman, on the other hand, reading meant gathering or choosing {lectio, reading, from lego, to gather) from what was written, suggesting that constant feeling of values which goes on in all effective reading. But reading had a meaning and was a common prac- tice long before the times of Anglo-Saxon or Roman. Men read in North Babylonia as long before Abraham's time as the latter precedes our own. In that land reading and writing had passed the pictograph stage eight thousand years ago. In Egypt, alphabet signs were used at least seven thousand years ago, and we know that with races the attainment of an alphabet is just the opposite of the ABC 2 INTRODUCTION stage of reading and writing. Indeed it is certain that even in that early period, in Egypt, Babylonia, and Crete, reading and writing were already of hoary antiquity, and had for these peoples already lost their beginnings in the mist and myth of a still more distant past. To the early peoples, reading was one of the most mysterious of the arts, both in its performance and in its origin. We recall how, even in modem times, Living- stone excited the wonder and awe of an African tribe as he daily perused a book that had survived the vicissitudes of travel. So incomprehensible, to these savages, was his performance with the book, that they finally stole it and ate it, as the best way they knew of "reading" it, of getting the white man's satisfaction from it. Among early peoples the mystery of reading naturally led to reverence for the printed word and book and for reading and the reader. Reading became a holy office, performed by individuals who possessed divine powers, and the book became a fetich. The written word was always of mysterious significance with the savage. It carried meanings through distances in space and time in such an utterly incomprehensible and apparently lawless fashion that he could not but fear and venerate it. A man might even be destroyed by doing certain prescribed things to his written name. The "winged words" of spoken language traversed the air unseen and were indeed objects of mystery, but they needed their written THE MYSTERIES AND PROBLEMS OF READING 3 symbols, more tangible and thus better fetiches, to become the objects of primitive worship. And the man who could deal in these symbols and use them for his purposes, he was next to the gods and might . rule in their stead. And so reading was long mainly in the hands of the priesthood, and written language, bear- ing the records of civilization and becoming the tangible subject-matter of learning, ministered to forms of worship and remained in the hands of the Church. Written lan- guage became the currency of civilization, and so of learn- ing and education. It was thought of as value in itself, and most commonly the Church "kept the bag." And so through this mystery of the printed word and this reverence for reading and for the book it came about that learning and education have ever been more or less holy things, and that the core of education, the reverenced part of it, has not been the arts, nor even the sciences as first- hand studies of reahty, but language and books. People's reverence for reading and writing helped to bring this about at first, and the clergy, always conservative, pre- served this ideal in the face of a "better reason," even until now. How dominant and ingrained is this ideal of education is evident in our names for, and common expressions con- cerning, learning. The learned man is a "man of letters," while the ignoramus is "unlettered." To say that one "cannot read and write" is to outlaw him in the common- 4 INTRODUCTION wealth of learning, while the acme of scholarship, until quite recently at least, has been stated in such terms as "Why, he reads seven languages," "He reads Greek at ■ sight," "He is well read," etc. "Learning" is still largely the ability to read and the reputation of having read, and reading keeps the momentum of the ages in which it was identical with learning. The modem dominance of this ideal is not merely because reading, especially reading of the classics, was for centuries the only road to culture if not to station. But to this day reading carries with it the faint but instinct- starting aroma of its old religious significance. Book is still Bible, perhaps as much as Bible is Book, to the average reader. All of us believe a little more readily if it can be said of the doctrine, "It is written." Students still worship the "power" of reading the difficult lan- guages and the remote literatures, and theological educa- tion, true to its history, still clings to reading and language as its main food. Little wonder, then, that Professor Dewey calls reading/ the Fetich of the Primary Grades. It is a mystery and a fetich with us all, and has not only blinded us to comparative values in the primary school course, but as subtly warps the judgment in a dozen other important directions. The written or printed "word," and especially, as of old, if it be written in strange char- acters, still awes us and controls us by its appeal to the old folk-soul, which is the deepest soul in us all. THE MYSTERIES AND PROBLEMS OF READING 5 Perhaps there is nothing that has been greatly venerated for long periods that is not really very wonderful, if one considers the essence of what is wondered at. And after all it does not seem so absurd that reading and the book should have been the worshipful wonder of the ages, and that they should still be reverenced. Real reading is still the noblest of the arts, the medium by which there still come to us the loftiest inspirations, the highest ideals, the purest feelings that have been allowed mankind, — a God-gift indeed, this written word and the power to interpret it. And reading itself, as a psycho-physiological process, is almost as good as a miracle. To the average reader the process by which he gets his pages read is not understood very much better than was the performance of Living- stone by the savage. Indeed, until twenty-five years ago, science could not give a very much better specific accoimt. The psychological part of the present study has grown mainly out of my own simple wonder at the process of reading, and out of my curiosity to know its mechanism; and this wonder has simply expressed itself in a different procedure than that of the African, has used the better tools of science and of scientific cooperation with other wonderers, in the endeavor to solve this mystery of the ages. Problem enough, this, for a life's work, to learn how we read ! A wonderful process, by which our thoughts and thought-wanderings to the finest shades of detail, the play of our inmost feelings and desires and 6 INTRODQCTION will, the subtle imagfc of the very innermost that we are, are reflected from us to another soul who reads us through our book. And a wonderful, awe-inspiring instrument, this book, that keeps this subtle likeness of its author, unfolding it part by part, changing it in phase at every page or sentence, yet all faithful to the original impress. Not in a day, not in ten thousand years even, as we shall find, has man been able to contrive the making of such a book or the manner of reading its page^. And so to completely analyze what we do when we read would almost be the acme of a psychologist's achieve- ments, for it would be to describe very many of the most intricate workings of the human mind, as well as to unravel the tangled story of the most remarkable specific performance that civilization has learned in all its history. The beginnings of such an analysis , and description are attempted, with the help of many co- workers, in the psychological chapters which follow. The strange and fascinating story of how the book and page have grown to be is sketched in the chapters on the history of reading, using the records of many patient scholars. So much for the appeal which reading makes to our psychological and historical interests, to our naive curi- osity and the concern which we have for the penetration of mysteries. Perhaps most of my psychological and historical chapters owe their origm to this motive, the THE MYSTERIES AND PROBLEMS OF READING 7 basis of much of our science for science' sake. But reading and books are of even greater concern to us for the reader's sake. What a habit, this, that has fastened itself upon us in modem times ! While the art of read- ing is indeed so very old, and while the practice vi^as even so prevalent among certain ancient peoples that learning to read is recorded to have been compulsory at one time upon all free Chaldeans, yet the absence of printed matter, in all early times, made it impossible that reading should be anything hke the habitual practice of the present, except among very limited numbers and special classes of individuals. But since the inven- tion of printing, the reading habit has steadily grown upon the whole civilized world ; and, furthered by modem laws for compulsory education, this habit has become the most striking and important artificial activity to which the human race has ever been moulded. Printed matter has been so diffused, and all that we do is so concerned with it, that a very considerable proportion of most people's waking time is taken up with the contemplation of reading symbols. This applies, of course, not merely to the read- ing of books and papers ; but in the car or on the driveway, in the street or at the railway station, advertisements, signs, notices, — what-not of printed matter, — keep one reading. At the opera or concert there is the libretto or program, and when interest palls, these are read and reread, even to the last silly advertisement. One is sel- 8 INTRODUCTION ' dom out of sight of some sort of matter to be read, and having formed the reading habit it has become second nature to read all that appears. Printed matter has be- come a stimulus which sets off, reflexly, a sensori-motor activity which goes off on sight anywhere, and which con- tinues imtil there is no more to be read or imtil one is checked by disgust or by a counter-interest. And yet this habit to which we subject ourselves and our children for so considerable a part of the time is an unnatural one, intensely artificial in many respects. The human eye and the human mind, the most delicate products of evolution, were evolved in adaptation to conditions quite other than those of reading. Such functionings as reading requires not having been foreseen in the con- struction of these organs, we need not be surprised that our continued and careless exercise of these unusual functions causes fatigue and, in very many cases, certain dangerous forms of degeneration. The very evident in- heritance of some of the more disastrous effects, such as myopia and nerve exhaustion, warns us of the danger of race degeneration from this source. For the sake of millions of tired readers, then, it may well be asked, and until recently the question has scarcely been raised. What are the unusual conditions and func- tionings that are enforced upon the organism in reading? Just what, indeed, do we do, with eye and mind and brain and nerves, when we read ? And what may be done THE MYSTERIES AND PROBLEMS OF READING 9 to avoid or minimize the dangers that come with this most universal and most artificial of habits ? The answer can only be given by an analysis of the process of reading and an examination of the essential nature and history of the page and booli. And then, as a school subject, reading is an old curiosity shop of absurd practices. Until as late as thirty-five years ago, in America, the blind devotion to the unreasoned and unreasonable ABC method of learning to read was as universal and as fetichistic as the worship of reading itself had been. Within this thirty-five years the ABC fetich has been put away in most quarters, and the results of trial and error with new devices have given us somewhat better methods, and best of all have put us in an inquiring and more humble frame of mind. But after all we have thus far been content with trial and error, too often allowing the publishers to be our jury, and a real rationalization of the process of inducting the child into the practice of reading has not been made. We have surely come to the place where we need to know just what the child normally does when he reads, in order to plan a natural and eco- nomical method of learning to read. We have come to the place where we need to pass in review all the methods that have been tried in all the centuries of reading, and to learn any little that we can from each. We need, too. to take a more profound survey, and to learn from the tortu- lO INTRODUCTION ous yet ever progressive path which the race has followed, in the hundreds of centuries in which it has been develop- ing reading and writing, the significant suggestions which this ethnic experience may have for our own further devel- opment of methods in reading and writing and printing. Whatever side of reading we consider we are challenged to investigate. For instance, we have long known that some readers read four times as fast as others of equal intelligence, and yet obtain better results. Yet we have remained content to completely ignore the question of rate, in teaching to read ; the only times it is mentioned, usually, being when the pupil is cautioned "not to read too fast." We know that the reading of hfe is almost exclusively silent reading. Yet in preparing for life we are instructed almost exclusively in reading aloud, and have not troubled ourselves to ask whether habits learned in reading aloud may not be hurtful in reading silently. We have learned comparatively recently that nearly if not quite all readers say over again within themselves all that they read. Yet no one has thought to determine whether purely visual reading, omitting this complex functioning of speech, may not be learned and be most economical of time and energy. We have made a fetich of our doctrine of formal discipline, and formal reading has kept its artificial place in our curriculum supported in part by this now fast-decaying prop. But when have com- petent persons taken the trouble to really analyze what THE MYSTERIES AND PROBLEMS OF READING II in mind is exercised or disciplined when we read or when we learn to read? And above all, how we have most wastefully failed to use the real opportunities that reading offers for discipline, the opportunities for training pupils to effective use of books and Hbrary, to selective reading, and to the prompt feeling for and use of values in what is read ! Indeed, when we ask seriously why we do, as we do, almost anything that we commonly do in reading or in learning to read, the answer is that so it has always been done. Why, indeed, should we read from side to side along a narrow hne, as the printers have found it con- venient to print, and not down or up hke the Chinese and old Egyptians? Why have we not turned inventive genius to the entirely possible task of makiag a page that can be read with one-fourth of the eye-work required by the page of the present? So slowly does thought find its way to the rationalization of the common things that we do in life. So rich are the possibihties for research in many lines. So vital to economy in education and in hfe is the particular group of preliminary researches on read- ing, the results of which we attempt to bring together in this volume. Though but the beginnings of what is to be done, and but faintly suggesting, if at all, the answers to many of the questions raised, we shall at least break ground for the more complete work that is to follow. And first we shall attempt an analysis and description of the reading process itself. PART I THE PSYCHOLOGY OF READING CHAPTER II THE WORK OF THE EYE IN READING If we sit directly before one who is reading and watch his eyes closely; or, more conveniently, perhaps, if we have him hold a hand mirror flat on the adjoining page as he reads, while we look over his shoulder, we will notice that his eyes move pretty regularly from side to side along the printed lines. We may count these sweeps from left to right, as a page is read, and at the end will find that they correspond to the number of lines on the page. The reader may insist that he reads several lines or even a paragraph with one sweep of the eye. He has perhaps grasped the thought of the lines or paragraph in one imitary act; and being quite unconscious of the movements of his eyes, he may very naturally suppose that he has taken but one "look" at the Unes. However, I have been able without fail to count the lines of the page from the movements of the reader's eyes, whenever the whole page was actually read, not skimmed ; and in using apparatus which recorded every movement, the eyes have been found in every case to move from side to side, line by line, without missing any line. As you watch the reading, you notice, too, that the eyes do not move continuously from left to right along the line, IS 1 6 THE PSYCHOLOGY OF READING but proceed by a succession of quick, short movements to the end, then return in one quick, usually imbroken movement to the left. You find all this very evident. And yet most of those who have studied the eye have ciuriously failed to note that the movement was discon- tinuous; and up to about 1879, when 'Trofessor Javal called attention to it, I find no mention of the fact in the literature. Indeed, I have not myself seen mention of these reading movements until 1898, except in the writ- ings of Professor Javal and some other French authors who took up his discovery, and in a paper published in 1895 by Professor Alexander Brown, of Edinburgh. It is a curious instance of the failure of scientists to make first- hand observations except along certain lines that have become habitual. You will find that there are at least two pauses for every line, and almost always more than that for lines of this length, — from three to five pauses, usually, and even more when the reading proceeds very slowly. The move- ments are so very quick that you may wonder as you try to follow them whether the reader has time to see anything during the movement, and you may forecast, as Professor Javal didj^what later experiments have seemed to prove, that there is practically no reading, or rather no direct seeing of the words and letters, except during the pauses. You will find it impossible to determine just what word is being looked at, or fixated, as we say, at any moment ; THE PSYCHOLOGY OF READING l^ and the reader himself cannot give a much more accurate account of this than can the onlooker, although he often supposes that he can. An attempt to count the pauses for each line will give rise to some curious difficulties, in case you "lay it on your conscience" to get the number right. If the movements occurred at regular time intervals, so many per second, you might get on very well with them; but it has ^been found that they vary greatly in extent, and that some of the pauses are very much longer than others. This irregularity prevents the rhythmic grouping which helps so much in counting, and forces one to make a particular counting reaction, say to tap with a pencil or to inwardly say "one, two," etc., for each movement. One may do this for a reader who is not very speedy, but I doubt if it is possible, even with special training, to make accu- rate coimts, by this method, of the movements of the eyes of persons who read rather fast. Besides the diffi- culty of the coimting itself, there is always the possibility of losing movements that occur while your own eye is in motion, as you are practically blind to what occurs while you are changing your own fixation; just as in boxing, one may be startled by a blow that started while his eye was moving, attracted by a feint. And we know the rule, as Professor Dodge observes, that the fencer should look his opponent in the eye, trusting to indirect vision for information about his movements. 1 8 THE PSYCHOLOGY OF READING Professor Javal, of the University of Paris, seems, as has been said, to have been the first to note thp actual character of the eye's movements in reading, i He con- cluded that there was a pause about every ten letters, and thought that this was about the amount that could be seen clearly at one fixation. lUe found that after reading he had after-images of straight gray lines correspond- ing to the parallel lines of print, and concluded that the eye's fixation point did not leave the line as it moved forward in reading. Finding that the upper half of the line was most important for reading, as can be seen at once by dividing a line in halves horizontally and com- paring the legibility of the upper and lower halves, he concluded, from this and other observations, that the fixation point moves along between the middle and top of the small letters. He also stated that the move- ment was such as to prevent the seeing of what was read except during the reading pauses. While not all of Professor Javal's observations are conclusive, he deserves more than does any one else the credit for making the initial discoveries in this field, and for initiating a considerable number of later studies. His own further work was prevented by his losing his sight, although, upon calling to see him a few years ago, I found him busily engaged in experimenting upon the reading of the blind. M. Lamare, working with Professor Javal, found it THE PSYCHOLOGY OF READING I9 simpler to count a series of sounds produced, in a micro- phone, by the eye-movements. The eyelid is displaced a little at each movement of the eye, and this gave the necessary stimulus in an electric circuit. Some informa- tion about the movements was thus obtained, although the author is careful to acknowledge the inadequacy of even this method of counting, and guards himself against making more than general conclusions. M. Landolt, continuing the study at the University of Paris, and ob- serving the movements directly, concluded that on an ^verage 1.55 words were read per reading pause, at the ordinary reading distance. Reading of a foreign lan- guage required more pauses, as did also the reading of detached words, numbers, and lists of proper names. He found that the small movements were very fatiguing, and that, since the angular excursion increases as the reading matter is brought nearer to the eye, this may accoxmt for the tendency of children to bring their books too near the eye. Relief is thus obtained from the fatigue incident to small-angled movements, but the work of the muscles of accommodation and convergence is correspond- ingly increased, with the resulting tendency to myopia. Doubling the distance of the page from the eye increased the number of movements in the ratio of nine to seven, the number of eye-movements seeming to depend upon the number of words per line rather than upon the visual angle subtended. Landolt 's method obliged him, as he 20 THE PSYCHOLOGY OF READING states, to have his readers read slowly, and more move- ments are made by the slow readers. Dr. Ahrens, at the University of Rostock, Germany, attached a light ivory cup to the cornea of a reader's eye; and fastening a bristle pointer to the cup, he at- tempted to get from this a tracing of the eye's movements written on a smoked surface. He was unsuccessful, but he had given a valuable suggestion. Dr. Lough, at Harvard University, and Professor Delabarre, at Brown University, at a considerably later time, attached a plaster of Paris cup to the cornea and obtained some records of the movements of the eye, but apparently obtained no record of the movements in read- ing. Erdmann and Dodge, in an extended investigation of reading made at the University of Halle, Germany, and published in 1898, studied the movements of the eye in reading, using the mirror method of direct observation referred to on a previous page. They found that the number of pauses did not vary greatly from line to line, for the same reader and with easy familiar reading matter. There were fewer pauses with familiar matter. In reading lines from a familiar philosophical treatise, printed in English, with lines 83 millimeters in length, Dodge, an American, averaged from three to five pauses per line, according to the familiarity of the passages read. Erdmann, a German, averaged from five to seven pauses THE PSYCHOLOGY OF READING 21 per line, under the same conditions, in reading a familiar German scientific work, with lines 122 milUmeters in length. In the initial readings. Dodge averaged five pauses per line and Erdmann seven. The variations above and below these latter averages were small and infrequent. More pauses were made in reading a foreign language. Proof-reading required about three times as many pauses as regular reading, in the case of Erdmann. In writing there seemed to be a pause for about every two letters, but they could not be sure that they^noted all of the move- ments here. By watching a reader's eye through a telescope arranged to permit measurement of horizontal distances observed, \ they found that the first pause was almost always within the hne, and that the last was still farther from the end of the line. J The more familiar a text, the greater was the indentation at the left, and more especially still at the right. They consider that the greater indentation at the right is because the previous context makes it easier to fill out the end of the line apperceptively, and also because the last section of the line is seen longer in indirect vision than is the first section, the reader getting no data from the begiiming of the line until he arrives at it. These authors made some observations which seemed to indicate that the fixations of the eye in reading were almost exclusively upon words, upon the middle of the word usually. However, the experiments were not 22 THE PSYCHOLOGY OF READING conclusive, and did not permit of the marking of the actual places fixated. Erdmann and Dodge did not measure the speed of the eye's movements in reading. Lamansky had measured the speed of the eye's movements in ^neral, by counting the number of after-images produced during a given movement of a pencil of light flashed into the eye at regular intervals through the perforations in a rotating disk. Dodge repeated these experiments, and then car- ried over the results obtained for the speed of eye-move- ments as they occur in moving voluntarily from one fixa- tion point to another, to the movements that occur in reading. The results, as pubhshed in 1898, did not agree with those of Lamansky, and indicated that the time re- quired for an eye-movement in reading was about .015 second. These experimenters had no means of measuring the duration of the reading pauses, but seem to have supposed them to be of tolerably uniform length for a given reading. The pauses have later, however, been shown to vary greatly in length. Professor Dodge has later succeeded in photographing upon a moving plate a beam of light reflected from the eye at different angles during its movement from one fijcation point to another, thus permitting a computation of the rate of movement. The tests that he made of the movement in reading showed that the forward movements varied from two to seven degrees, and that the time occu- THE PSYCHOLOGY OF READING 23 pied by these was, on an average for three readers, nearly .023 second. The return movement through twelve to fourteen degrees required a little less than .041 second. As the lines subtended an angle of sixteen degrees, the eye evidently passed over but three-fourths to seven-eighths of each line. In experimenting upon the psychology of reading, in 1897-1898, it seemed to me impossible, from my own observations and from those of all earher experimenters who had tried direct observations of the reader's eye, to get trustworthy account, by direct observation, of the speed, nature, and even number of the eye's movements in reading, of the length and variation of the reading pauses, etc. Nor could the reader himself give even so good an account. For him the succession of movements and pauses is practically non-existent, except when the eyes are very tired or in some way abnormal. In certain cases of abnormal vision, it is true, some valuable observations may be made by the reader. For example, a patient who had no use of the left halves of his retinae and thus could not see any letters that lay to the right of the point fixated, was quite conscious of the jerky forward movement of his eyes as section after section of new matter came into view. Again, some readers become quite conscious of the pres- ence of the muscae volitantes, or flitting spots that appear more or less in the vision of most people; and these give some notion of the jerky character of the movements. 24 THE PSYCHOLOGY OF READING Most of US, however, find ourselves getting over the page rather smoothly and continuously, apparently see- ing distinctly a considerable portion of it at once, and without interruptions of any kind. Words, letters, and letter-groups flash into greater distinctness from moment to moment, and there is some thought of a mental travers- ing of the lines.'"^,^ If we watch closely, we are apt to find some sort of iimer utterance of what is being read, and we have a notion of the meaning of it all, although we cannot very well describe this consciousness of meaning. Thus reading appears to the casual introspection of the reader. We find, however, that underneath this apparent simplicity, there is an astounding complexity of processes. These have been built up slowly, and by an immense amount of practice, until they have organized and settled into the smoothly running machinery of our present-day reading. The psychologist's analysis discloses a con- dition which impresses one, to use Francis Galton's figure, as "when the basement of our house happens to be under thorough ^sanitary repairs, and we realize for the first time the complex system of drains and gas and water pipes, flues, bell-wires, and so forth, upon which our com- fort depends, but which are usually hidden out of sight, and with whose existence, so long as they acted well, we had never troubled ourselves." ' As a beginning of such analysis of reading, it seemed ' "Inquiry into the Human Faculty," p. i86. THE PSYCHOLOGY OF READING 25 important to obtain a definite description of the work of the eye. For this purpose I arranged apparatus as fol- lows : A little plaster of Paris cup was moulded to fit the cornea accurately and smoothly, sand-papered until it was very hght and thin, and placed upon the front sur- face of the eye, the cup adhering tightly to the moist cornea. No inconvenience was felt, as the corneal sur- face was made insensitive by the use of a little holocain, or sometimes cocaine. A round hole in the cup permitted the observer to read with this eye, and the other eye was left free. A hght tubular lever of celloidin and glass connected the cup to an aluminum pointer, flat and thin, which responded instantly to the slightest movement of the eye; and, suspended over the smoked-paper surface of a moving drum-cylinder, the aluminum point traced a record of the eye's movement as the observer read. The drawing (Fig. i) shows the arrangement in the earlier and simpler form. The observer's head rested in a frame which was arranged to prevent movements that would interfere with the record, and which held an attachment to prevent the eyelids from interfering with the cup. The weight and friction of the recording apparatus was re- duced to a minimum, the weight of all the parts moved by the eye being but a little more than half a gram, while the weight resting directly upon the eye was less than one-seventh of a gram. During the reading, the reader was usually quite unconscious of there being an attach- 26 THE PSYCHOLOGY OF READING ment to his eye, and the reading proceeded as ghbly and easily as could be desired. Records were made from a large number of readings, with various leiigths of line, sizes of type, distances of reading matter, etc. Sometimes the reading was at nor- mal speed, sometimes as fast as the reader could possibly Fig. I read. The apparatus seemed to work equally well undei all the various conditions, even when the speed reached an average of twelve words per second. In order to measure the speed an electric current from an induction coil was passed through the pointer to the drum. This current was interrupted at very regular short intervals THE PSYCHOLOGY OF READING 27 by the vibrations of an electrically driven tuning-fork, the snap of the spark from the pointer's tip thus displacing a dot of soot on the papVr record at each interruption. As the pointer flitted over the drum during the reading, a tracing was thus produced like that shown in Figure 2. The tracing as taken of course magnified the actual move- ment several times. The tracings showed that the eye always traversed the page line by line, and always in a succession of quick movements and pauses when moving from left to right. Movements in retracal occurred but seldom, averaging about once in seven lines. Apparently the eye did not wander far above and below the line that was being read, although the arrangement of the apparatus could not show this certainly. As we have seen, Professor Javal thought that the eye's fixation point moved along the line between the middle and top of the small letters, but I do not think that his experiment with after-images establishes this. It seems to me probable that the fixation point varies more widely than this, but there is nothing to indi- cate that it wanders perceptibly above or below the line. The return sweep when a line was finished was usually without interruption, although about once in six lines a halt would be made near the end of the movement, ap- parently for the eye to get its bearings in the new line. These halts in the return movement are more numerous in reading long lines. The eye is apparently guided in 28 THE PSYCHOLOGY OF BEADING FtG. 2. Specimen Curve 0/ ' Spark * Record, This reproduction, cut by. s careful engraver upon a block on wliicit the original traCiug had been photographed,.shows with great accuracy the sort of record from which the times- of the eye movements have been determined. Thechief difference between the original and the repro- duction is in the breadth 'of the horizontal lines which are finer in the original. The curve shows the movements of the eye. in reading six lines, pre- ' ceded and followed by twO' free movements of the eye eachway, in which it wsis swept from -one end of the line to the other, the begin- ning and end alone being fixated. The broad vertical lines and the round blurs in the reading indicate pauses in the eye's movements, the successive sparks knocking the soot away from a considerable space. The, small dots standing alone or like beads upon .the horizontal lines, show the passage of single sparks, separated from each other by 0.0068 sec The breaks in the horizontal lines indicate that tbeL writing point was not at all times Tii. contact with the surface of the paper thOujgh near enough for the spark to leap across, as shown by the solitary dotsi The tracing shows clearly the fixation pauses in the course- of the line, the general tendency to -make, the "indentation" greater at the right. than at the leit, and the unbroken sweep of the return from right to left. Note. The cut and description ^re reproduced from the American Jamrnal of Psychology, Vol. XI, THE PSYCHOLOGY OF READING 29 making the return by the consciousness of the next line's beginning, seen dimly at the left as it starts. Such guid- ance would of course be less accurate as the line is longer, and this may necessitate both the baitings and the distrac- tion that we notice introspectively when we lose the line. In reading lines of the length shown in Figure 3, tjie subject-matter being of only ordinary difficulty, the small- Cach page thas honey-combed, was fastened closely upon a ■white paper background. The page was then marked off into four divisions. Two readings were .taken, separated by sev Fig. 3 est number of movements for any line was two, the largest seven. Usually there were from four to six. One reader averaged four and a half pauses per line for fifty-one lines. Another averaged a very little more. These readings were at the ordinary reading distance. Doubling the distance did not appreciably lessen the number of pauses per line. ney very seldom paid any attention to it. If they wanted to do cer- ain things, they would do it whether the constitution allowed them O or not. The only reason they had a constitution was because they Fig. 4 Of course this means that the angle of each eye-move- ment grows smaller as the book recedes, possibly with the increased tendency to fatigue that comes with small eye-movements, as experienced when you look along the pickets of a fence or the letters of a line, although in these latter cases the conditions are somewhat different. Using the smaller type shown ip Figure 4 slightly in- 30 THE PSYCHOLOGY OF READING creased the number of pauses and movements. T Using shorter lines of course reduced the number. _ A magazine article, with lines of the length and type shown in Fig. 5, gave an average of 3.6 pauses per line, there being always either three or four pauses for a hne. A newspaper article, give the murderer the benefit of a doubt, he felt as a man that the doubt could not really exist, and that Tebaldo had in- tentionally put him under the seal of con- FlG. s of lines and type as in Fig. 6, gave 3.8 pauses per line for one reader and 3.4 for another. The records for lines of all lengths show that the eye seldom moves along the whole extent of the line, but makes the first pause somewhat within the line and the last still farther within. From 78 per cent to 82 per cent of the line, on an average, was actually traversed by the eye, in reading such lines as are shown in Figures 3 and 4. The indentation was usually in the vicinity of 18 pet American, citizens could not have a part. 3ui I aw satisfled that It is all fjglit. Every one of tne great races that are blended In our national life has Us owa glorious traditions which, it Fig. 6 cent of the total line length, varying considerably from hne to hne, and being usually considerably greater at the right, although occasionally it was greater at the left. In some readings the first pause must have been in the first half of the first word in most of the lines, while in other THE PSYCHOLOGY OF READING 3 1 passages it must usually have been in the second or even in the third word. For a given passage, the reader seemed to fall into a way of indenting a certain amount at the beginning or end of a line, and kept this tendency through the passage. The extent of the forward movements in reading varied greatly in all cases, and the conclusion of Javal that the eye moves over the extent of matter that can be read at one pause, about the space of ten letters in his opinion, was shown to be unfounded. The specimen tracing above shows the great difference found in the extent of the movements, and is typical. A slight movement may be followed by one four times as extensive, and the move- ments are in general very irregular. The movements averaged from three to four degrees of arc in the various readings, with lines of the length shown in Figures 3 and 4. With these lines the return sweep usually trav- ersed twelve to thirteen degrees. The forward movements of the eye in reading were found to occupy a relatively constant time, somewhat irrespective of their extent. In one reading, the move- ments of the eye varied in extent as from four to twenty- six, while the times for these movements ranged from four to seven. Excluding three exceptionally short move- ments, the others in this same reading ranged in extent from seven and a half to twenty-six, while their times ranged only from six to seven. 32 THE PSYCHOLOGY OF READING The spark record indicated that the absolute time re- quired for each movement forward varied usually between .04 second and less than .048 second. The return move- ments usually required a little longer, usually from .051 to .058 second. Later experiments indicate, as we shall see, that the times are actually considerably shorter than this. The reading pauses were of very variable duration. In one of the readings the pauses averaged about .19 second, but the average variation was nearly .05 second. Another reading by the same observer showed an average pause of less than .11 second, with one-third less variation. This latter record was taken under circumstances which may have sometimes permitted the current to snap twice at the same spot, and if so, the pauses were really a little longer than as above. The other times given here for the pauses were measured from the displacement of the drum as measured by an electric time-marker connected with the laboratory clock and writing its record beside that of the eye's movement. Two additional readings by different observers gave an average pause of a little more than .18 second for each. The pauses in retracal and at interruptions of the return sweep are usually shorter than the reading pauses proper and are not included in the averages above. For rather fast readers, then, the pauses seem to average in the vicinity of .185 second; but the variation is so very THE PSYCHOLOGY OF READING 33 great that any average is misleading, and the pauses may really be of almost any length. The averages, however, show the interesting fact that the most of the reading time is used in a fixed gaze at the page and but an incon- siderable portion of the time is used in the eye's move- ments. In all the experiments whose results are given above, the readers read at their usual rate, silently, and for the thought. In another series of tests they were asked to read as fast as possible. This decreased the number of pauses per line, and also the duration of the pauses. The extent of each eye-movement was correspondingly in- creased. The speed of the movements was not increased in the least, and there was nothing in any of the experi- ments to indicate that the rate of movement is subject to direct voluntary control. Fast readers thus seem to perform less eye-work, their movements are less fatiguing in so far as large-angled movements may be easier than small ones, and they take less time for visual perception of the printed matter. These results are fairly congruent with those of Erdmann and Dodge as to most points, except as to the rate of move- ment. Rather fewer pauses per Une were found in my experiments than in those of either Erdmann and Dodge or of Javal and Landolt, for the reading of new matter. But the fact that Landolt's readers read slowly would account for the more numerous pauses in his case, and the 34 THE PSYCHOLOGY OF READING methofl of cojunting may have led to reading that was slower than normal in all these cases. - On the other hand, my records were from rather fast readers, and they may have mads somewhat fewer pauses than occur in ordinary reading. As to rate of movement, we have seen that the early measurements by Lamansky gave times that were not in agreement with the later ones by Dodge, using the same method. Dodge's still later photographic measurements show a -still slower rate of movement, and with consider- able individual differences. His times are still consider- ably shorter than those found in iny own experiments, averaging nearly .023 second for forward movements of from two to seven degrees, and nearly .041 second for return movements of from twelve to fourteen degrees. While a part- of this difference may be due to individual variations, the larger part of it is doubtless due to error somewhere. As we shall presently see, the recent meas- urements by Dr. Dearborn, using the Dodge method of photographic registration, confirm the later results of Dodge as stated above, and I accept these, therefore, as trustworthy conclusions on the rate of movement. I am unable even yet to certainly locate the source or sources of error in my own measurements of rate. All of Professor Dodge's criticisms of my apparatus were in my own mind in its construction; and it seemed to me, as well as to several psychologists and physicists who were experienced THE PSYCHOLOGY OF READING 35 in the manipulation of delicate apparatus, that there was httle likeUhood of any considerable error. Error, how- ever, there evidently has been; and I am at present inclined to think that most of it arose through a slight jaelding of the corneal surface imder the movement of the recording attachment. Very accurate determinations of the rate of eye-move- ment are of considerable importance along several lines of psychological inquiry, which, however, do not concern us here. The essential problem as to the movements in reading is to know whether they are of such a speed as to prevent our perceiving letters or words during the movement. This question is already practically settled, provided we can carry over to reading the general require- ments for fusion of visual stimuh. The rate of reading movements is much too fast to permit our getting such data during the movement as can appreciably help us in our reading, granting that the laws of fusion apply. Various experimenters have shown that when retinal stimulations are given La rapid succession, as when a disk of white and black sectors is rotated before the resting eye, the impressions fuse into one continuous impression when a rate of thirty to sixty stimulations per second is reached. But in reading at even a very slow speed, and supposing the rate of the eye- movement to be the slowest found in any of the measurements, the succession of black letter-strokes and white inter-spaces occurs at a rate which 36 THE PSYCHOLOGY OF READING is far faster than even sixty per second, and which would inevitably produce nothing but a light-gray blur, if the eye were still and the line shifted horizontally at a cor- responding rate. The experiment may be tried by hold- ing a pencil tip near to a sheet of printed matter, as a fixation mark; and giving the paper a fairly rapid move- ment from side to side. If the movement is in the direc- tion of the lines of print, the latter will appear as homoge- neous gray bands, with no letters or words recognizable. But why do we not see these gray bands as we read? Scarcely a trace of them has been reported by any ex- perimenter, except that Professor Dodge believed that he could detect faint traces. I have found no reader who had any consciousness of them, and have none my- self. We seem to see the letters and words as distinctly during the movements as at the pauses, and the visual field is unbroken to consciousness. Professor Cattail advanced the hypothesis that the visual organs respond to retinal changes more rapidly when the eye moves than when the objects are in motion, and that we really do see objects distinctly during the eye's movement. He believed that when we look from one end to the other of a row of books, for instance, we can note titles, etc., that were outside the range of distinct vision before the movement be- gan. Indeed, in moving the eye over ,black and white surfaces,, it seemed to him that no fusion occurred "even THE PSYCHOLOGY OF READING 37 when one thousand stimuh per second fall upon each retinal element." ' However, careful observation reveals the fact that in looking along a row of books or any other objects which we attempt to perceive, the eye makes a succession of short pauses and quick movements, as in reading, and it is not at all necessary to suppose that there was anything perceived during the movement. Professor Cattell supposed that a very large number of pauses would be necessary to allow of distinct perception, if no data were obtained during the movement ; but his estimate of one degree as the limit of the "field of distinct vision" is much too small, as will be seen later, the pauses even in reading occurring only at intervals of from three to four degrees, on an average. Professor Cattell's hypothesis seemed to be rendered untenable, as he himself frankly acknowledged in a later note in the Psychological Review, by some experiments of Professor Dodge upon the possibihties of vision during eye-movement. Professors Erdmann and Dodge had found that when one watches his own eye in a mirror, it is impossible, when the eye moves, to detect any trace of the movement. This furnishes at least a partial demon- stration of the apparent fact that we are not usually aware of what goes on in the field of vision while the eye is in motion. It cannot, however, be said to be a final test. ' Psychological Review, 1900, p. 325. 38 THE PSYCHOLOGY OF READING Professor Dodge made some further experiments to determine whether the eye is quite insensitive to impres- sions received as it moves. At first it appeared to be so. But after preliminary practice, and with more intense illumination, it was shown conclusively that distinctions of light and shade, and indeed of color, can be made in objects that are exposed only during the movement. Furthermore, under the special conditions of the experi- ment, it was found that lines of print exposed only during the eye's movement "gave a perfect though shadowy series of gray bands in a lighter gray background, in which individual letters or words were absolutely irrecognizable.'" Professor E. B. Holt, of Harvard, made a series of care- ful experiments to determine whether "voluntary move- ments of the eyes condition a momentary, visual, central anaesthesia," and concluded that they did. However, his results, even if finally conclusive on this point, cannot be carried over to the involuntary movements of the eye in reading.^ Professor R. S. Woodworth, of Columbia, on the basis of other experiments, maintains that the eye is not anaes- thetic during movement. He finds that, an image thrown on the retina during the eye's movement is correctly localized in space, that muscse, etc., are seen during the eye-movements, and cites the fact that an object which • Psychological Review, September, 1900, p. 463. ^Ibid., Monograph Supplement, January, 1903. THE PSYCHOLOGY OF READING 39 moves with the eye, at the eye's rate, is seen clearly dur- ing the movement.' Finally, Professor Dodge, in his article in the Psycho- logical Bulletin of June 15, 1905, says that he has "yet to meet with any imambiguous evidence of anaesthesia during eye-movement, either central or peripheral," and thinks the lack of clear perception during eye-movements must rest largely if not wholly on other grounds. Among the more important of these latter he mentions the per- sistence of the positive after-image for some three hun- dredths of a second, at about full intensity ; the inhibition by the following stimulation from the new fixation point, and the fact that the stimulations during movement are not objects of interest and are, therefore, ignored as are the muscEe, the fencing mask, and other such irrelevant stimuli. Evidently, then, the retina is sensitive to impressions received during the eye's involuntary movements, such as occur in reading, although it is not shown that it is more sensitive then than at other times. Evidently, too, fusion of impressions may occur during the eye's movement as when it is at rest. There remains the question as to why we are so completely unconscious of any fusion of the letters as we read. A conclusive answer cannot be made until there has ' Psychological Bulletin, February 15, 1906, and Proceedings oj Ameri- can Psychological Association, 1905-1906. 40 THE PSYCHOLOGY OF READING been further experimentation ; and, perhaps, a better un- derstanding of the conditions under which the conscious- ness of fusion arises. However, it should be remembered that the total stimulation given during the movement is very brief compared with that of the preceding reading pause, certainly not more than one-fourth as long in most cases. We know that a strong memory image of the read- ing stimulus tends to persist after the pause, as after any stimulus of such duration; and this would tend to per- sist in consciousness to the exclusion of the much fainter and briefer stimulations that occur during the movement. The stimulations occurring during the movement would have their own effect cut short prematurely by the intense stimulation of the succeeding pause, and would tend to pass imnoticed in consequence. Mainly, however, as has long been known, we habitually ignore stimulations and sensations which have no mean- ing for us, in favor of those which carry nieaning. Raw sensations are continually ignored in favor of their mean- ings, in all the sense spheres ; and indeed there is a tendency of consciousness to remove in the direction of the more and more remote suggestions from any given stimulus, at least so far as such removal is helpful in practice. The gray-blur stimulus produced during the movement car- ries no meaning of its own, and leads no whither as a sign of remoter meanings. It is faint, of most transient exist- ence, and remains beneath the threshold of clear conscious- THE PSYCHOLOGY OF READING 4I ness. Our ignoring of the movement stimulations is not exceptional in the psychology of perception. Stimula- tions constantly occur in various parts of our bodies, from pressure of our clothing and of the body's parts, stimula- tions which are often of considerable intensity, but which regularly pass unnoticed. The failure of most people to notice the entoptic phenomena, — the dark spots and strange shapes that are usually to be found in the field of vision, is, as Professor Dodge observes, a case of our systematically ignoring stimuU which would disturb our clear vision if attended to. I am in general quite of his opinion that the absence of the consciousness of fusion is "centrally as well as peripherally conditioned." These are the conclusions which, I think, the facts would oniinarily be said to warrant. I must confess, however, that I am, as yet, not entirely satisfied with them. Several minor facts brought out in the various experiments har- monize well with the view that the stimulations occurring during the eye's movement may be effective both upon the reflex mechanism of movement and upon the conscious content, and that they are correctly localized without fusion. Professor Holt's discovery that a stimulus acting upon the retina during eye-movement caused a reaction of the reflex mechanism in the direction of fixating the stimulus; Professor Woodworth's determination that an image thrown upon the retina during the movement is correctly localized in space ; these, with minor indications 42 THE PSYCHOLOGY OF READING from various sources, are suggestive of the latter view. Indeed, do we have the consciousness of visual fusion in any case where it is possible to apperceive the stimuli as they occur, where our concern is with apperceiving them, and where the conditions are such as to permit our giving them a correct orientation in space? There is little to indicate that a new orientation with reference to the reading stimuli arises in toto when the eye makes a new pause in reading. There is much to indicate that the orientation persists from pause to pause, and indeed dur- ing temporary closing of the eyes. A compensatory ad- justment of the field of vision would seem to be made as the eye moves forward, and it is doubtful if at any instant of the movement the spatial relation of the body and self to the various parts of the visual field is not felt as truly, although fleetingly, as when the eye is at rest. If so, then there is much reason to suppose that the stimulations re- ceived from the page as the eye moves are properly placed in order as they come, with no possibility of a blur. It is true that impressions made by each letter or smaller form upon any given retinal area would be very brief and shght ; and yet, if the stimulations are sufficient to be felt as a blur, they may well be also sufl5cient to serve as cues ^or their proper projection in space, without blur. So it may be that Professor Cattell was right in his conclusion if not in his method, and that our thought of the printed line gets the benefit of what stimulations occur THE PSYCHOLOGY OF READING 43 during eye-movement, and has no tendency to fuse the impressions because these are taken care of in advance. Fusion may occur only when there is inabihty to " think " the data given, in which case the consciousness arises of the data themselves, the black-white of the sensations. The cases in which fusion occurs have always, perhaps, been cases in which conditions were such as to prevent apper- ception or compensatory adjustment of the field of vision; and indeed the attempt to make such adjustment is seldom made ia such cases as produce fusion. These suggestions, however, are offered as but tentative, and of no scientific value except that they 'show, to my mind, that the absence of fusion during eye-movement presents a problem which is by no means closed at the present stage of experimenta- tion. Since most of the above account was written. Dr. Dearborn has published the results of a thorough study of the eye's movements and pauses in reading, based upon experiments at Columbia University in which Pro- fessor Dodge's method of photographic registration was used. Dr. Dearborn's results agree, in the main, with those of earlier experimenters, but they give additional information that is of much importance. It was found that "the more pauses there are in a line the shorter their lengths, on the average, and, vice versa, the fewer the pauses the larger any one pause is apt to be." "Viewed simply from the standpoint of speed of reading, it is in 44 THE PSYCHOLOGY OF READING general an advantage to read a given line with the smallest possible number of pauses, because while the elimination of a pause increases somewhat the average duration of the remaining pauses, the total time for the line is decreased, or remains constant." The number of pauses per line varies greatly, but is greater for the slow readers and when reading slowly. The eye readily falls into a brief "motor habit" of making a certain fixed number of pauses per line, for a given passage, independently of the nature of the subject-matter. "The eas? of the formation of motor, habits seems to be one of the characteristics of rapid readers as contrasted with slower ones." In mak- ing the adjustment for the return sweep at the end of the first line, and sometimes at the end of the second, the dependence is "solely on the peripheral local signs. The loiter the line, the more inexact these will naturally be." But, if the lines are not too long, "after the first or possibly the second horizontal movement, the resident muscular sensations of angular displacement govern the extent of movement of the succeeding return sweep. This is the basis of the^motor habit." It is important, therefore, that the lines be of only moderate length and that this length be approximately uniform, although Dr. Dearborn agrees with Professor Cattell that "a small indentation of a few millimeters, for example, of every other line . . . would help to differentiate the lines, and prevent their confusion." He finds that the modern primers and first readers con- THE PSYCHOLOGY OF READING 4J stantly violate this principle of uniformity, breaking up the line with their illustrations and often making a para- graph, with its unequal lines, for every sentence. Be- ' sides, the lines of the beginners' books are usually too long. A more or less uniform motor habit of eye-movement is to be acquired in the beginning, and the shorter lines of tmiform length are necessary for this. Lines of varying length "must naturally lead to a more cautious mode of eye-movement, hard to overcome later, and may cause unnecessarily slow readers." The average duration of the pauses was found to be imiformly less in the shorter lines than in the long ones. The total time per passage is also decreased in moderately short lines of right arrangement. It was discovered that the first pause in each line is distinctly longer than the others, especially in rapid reading and with rapid readers. Toi(7ard the end of the line, also, there is apt to be a pause of greater length than the average, although shorter than the first. At the initial long fixation and at the secondary long one the "attention expands," and by the former "a more general perception is secured of the ideas and words that follow in the Une. The succeeding fixations serve to amplify and fill out this general perception. Finally, this expanding of the field of attention is made more frequently and with greater ease in the short Une." In general. Dr. Dearborn thinks that differences in the rate of reading in the same individual and between differ- 46 THE PSYCHOLOGY OF READING ent individuals depend "largely, when other conditions are constant, upon the ease with which a regular, rhyth- mical naovement can be established and sustained." The peculiarities of such a movement are "first, a suc- cession of the same number of pauses per line, and sec- ondly, a certain fairly uniform arrangement in the order of long and short pauses," in which arrangement the first pause should be longest, with a secondary pause of in- creased length near the end of the line. When "shorter lines" are mentioned as best meeting these requirements, the reference is to lines of a length common in our news- papers,' or a little longer. Dr. Dearborn thinks favorably of "a line of seventy-five to eighty-five millimeters or about a third longer than the ordinary newspaper line of the New York dailies," although he recognizes that his data are not sufficient to warrant any conclusion upon this point. Dr. Dearborn's measurements of the rate of the eye's movement in reading agree substantially, as we have seen, with those of Professor Dodge, and these may be taken as representative for most readers, although the rate was found to vary considerably. Dearborn, therefore, con- cludes that there is "no distinct visual impression" during the eye's movement in reading. Some readers, however, were found to make a considerable number of slow shift- ing movements, often ten to twenty times slower than the usual movement. These shifting movements are THE PSYCHOLOGY OF READING 47 < really to be classed as unsteady fixations, the eye accom- panying the movement of attention as the latter shifted, and gathering data for perception as it moved. "The eye tends to follow each shift of the attention in order to bring the object nearer the fovea," and Dearborn beUeves that "the unsteadiness of fixations is due to the acute unbalance and general alertness of attention to peripheral excitation." "The attention is ahead and pulling the eye along." It was found that the "exact point that is fixated may be in any part of the words, or in the spacing between them." "It does not fall predominantly in the first part of words, nor does it occur more frequently in the first part of the sentence than in the last, and apparently pays little attention to many of the laws of apperception or the rules of the rhetorician." The exact points of fixation are "significant only as representing the point about which are grouped the 'block' of letters that are simul- taneously perceived as one word or phrase complex. It more often falls in the first third than at the centre of a given perception area." "The short connective and non- substantive words, the prepositional phrases and relative clauses, make the greatest demands upon perception, and thus require most fixations." " They necessitate the eye's coming out quite to the edge of the lines." These "transi- tive" parts of speech are "not associated in one phrase more regularly than in another, they cannot be fused into 48 THE PSYCHOLOGY OF READING a larger apperceptive unit, as the syllables into a word form, or 'phrase whole'; but each stands by itself and must be so perceived." "It is not the short words, as such, but the words which cannot be easily grouped with others, which necessitate separate fixation." Prepositions, conjunctions, etc., "occur now with one word and now with another; they cannot without danger of error be fused into larger wholes, and, for that reason, they must, except where the context gives the connection, be sep- arately perceived." The same was found to be true of numerals, lines of nonsense-letters, etc., and the explana- tion is similar. On the other hand, nouns, adjectives, and verbs, and especially words and phrases which are familiar to the apperception of the reader, or for which he has a particular memory, "allow of an exceptionally large 'jump' between fixation pauses." Children of from nine to eleven years, the only ages tested, were found to make more frequent pauses than adults, and generally longer pauses, although quite short pauses occur. While there is "some unsteadiness and refixation" their "accuracy of fixation appears as exact as that of adults." "The more purely physiological difficulties have been fairly well mastered. The rate of movement in the return sweeps and in the inter-fixation movements is not different from that of the adult." In the reading of a single adult who was tested for the effects of fatigue, it was found that after a hard day's THE PSYCHOLOGY OF READING 49 work with the eyes the reader made more pauses and longel ones than when the eyes were fresh next morning. Eye- fatigue was thus found to diminish the speed of reading, for this reader. Some further experiments showed that the eye becomes fatigued very quickly in making move- ments through a large angle, covering say one hundred and fifty millimeters on the page. At least the rate of movement became from one-third to one-fourth slower in the course of seven movements. This indication of fatigue did not appear to any considerable extent in making move- ments of about half this length, and thus furnishes an additional argument in favor of using lines of moderate length. I cannot agree, however, with Dr. Dearborn in his belief that these results negative the conclusion of Landolt, that small-angled movements are very fatiguing, as Landolt's reference is to the very short excursions such as are made in the smaller inter-fixation movements.' Landolt himself instanced the fatigue which comes when we attempt to count the palings on a closely picketed fence. Of course we must always recognize that such fatigue may be of the attention rather than of the eye-muscles. We may note in conclusion Dr. Dearborn's agreement with the writer in concluding that Javal's theory that the ' In a recent letter to the writer Dr. Dearborn tells me that his refer- ence to Landolt rested upon a misunderstanding. On the other hand, I agree with him that Landolt's test is not conclusive for the reading movements. 50 THE PSYCHOLOGY OF READING eye moves along between the middle and fop of the small letters is "a physical impossibiHty," although Dearborn's apparatus did not permit, as mine did not, definite measure- ment of movements in the vertical plane. Concluding here our survey of the eye's movements and pauses as mechanical processes, we will next consider the psychic processes of perception which occur in dealing with the data obtained during the reading pauses. And first we shall try to determine the amoimt of printed mat- ter that can be seen clearly at any moment in which the eye is at rest, say during a reading pause. CHAPTER III THE EXTENT OF READING MATTER PERCEIVED DURING A READING PAUSE \ In looking casually over a printed page, one is apt to think that a very considerable portion of it is seen at any moment with distinctness enough for reading. The amount that can be seen thus distinctly is smaller than is generally supposed. The illusion may come partly from the fact that the retina's ability to discriminate brightnesses,* or differences in the light and dark of the page, does not decrease from the point of clearest vision outward as far as is reached by both ends of the ordinary line, when one looks at the middle. So most of the page appears as bright in one part as another. Again, since by long experience we know that we can at once see distinctly any part of the field at wiU, and since we are usually un- conscious of the eye-movements which make this pos- sible, we naturally mistake the "reading range" given by several quick eye-movements, for that which is possible for the unmoved eye. However, if you will look fixedly at a letter in the middle of the page and will attempt to name the letters or words about it, without moving the eye for a single instant, you SI 52 THE PSYCHOLOGY OF READING will discover that the reading range of the unmoved eye is distinctly limited. Erdmann and Dodge, in trying this experiment on a page of German printed in good type, found that neither of them could see letters or words clearly beyond the ends of the lines represented in the diagram below, when fixating a letter at the central dot. Not all of even this amount could be seen clearly. Fig. 7 Some of you will be able to see all the letters distinctly over a little larger area than this; a few will not be able to do so well, for some readers are found to have a curi- ously limited range of distinct vision. However, if you have tried the experiment, you have doubtless noticed that, beyond the circle within which you recognize clearly most or all of the letters and words, there is also recognized, now dimly, now rather distinctly, an occasional letter of characteristic form, most often a capital, — or even a whole word of striking appearance, and this at a con- siderable distance from the fixation point. Erdmann and Dodge found this to occur at the distance of nearly or quite an ordinary line-length. Sometimes a word or a letter or group of letters flashes up momentarily from THE PSYCHOLOGY OF READING S3 the obscurity of indirect vision. Helmholtz found, in lighting a page for an instant by the, electric spark, that particular groups of letters would appear here and there in somewhat the same fashion. Some of you will see distinctly much farther than I have suggested as possible, because, as you interest yourselves in this or that part of the visual field, unconsciously your eyes move a little in that direction. This not only brings new matter within range, but freshens the retina as well, by changing the position of all its letter-images. It is difficult for an unpracticed observer to prevent such wandering of the eye; and even the most thoroughly trained observer cannot prevent a slight fluctuation of the fixation, which seems to occur almost constantly. We see an accentuation of this normal healthy condition in per- sons who are troubled with nystagmus, an affection of the eye characterized by a slight but plainly visible and constant change of fixation point, the patient being quite imconscious of the movement and supposing that he keeps his eye steadily fixated as he looks. The range of clear seeing about the fixation point is really a little greater than as given above. The retina is very quickly fatigued in maintaining a fixed stare for even a moment. The glance of the first instant shows the largest area, the border letters immediately growing in- distinct before we have had time to make them out or repeat their names if recognized. In actual reading, as 54 THK PSYCHOLOGY OF READING we shall see, the meaning or context fixes them before they fade. ' It was found by Cattell, Goldscheider and Miiller, Quantz, and various other experimenters, that when printed matter was exposed to the eye for a very short time, about one one-hundredth of a second, more could be read, or the same amount could be read more easily, ^than when the exposure was longer. Of course in such an exposure the retinal image remains for some time after 'the exposure ceases, just as you see the incandescent light after you have turned off the current. I proceeded to measure, with as much care as possible, the amount of printed matter that can be read at a single pause of the eye; and in doing so, relying on the experience of the investigators mentioned, I made a clear exposure of the printed lines for the slightly longer period of one sixty-sixth of a second. In actual reading, we largely disregard the lines above and below the one that is at the moment being read ; and as far as the eye is concerned the procedure is a matter c^ taking a succession of from three to five quick peeps at as many places in the hne. How much of the line can be read at a single peep of this kind? I carried through a series of experiments in which just such peeps were given, along lines of printed matter similar to those shown in Figure 3, taken from the American Journal of Psychology. The printed hnes were pasted on strips of cardboard and THE PSYCHOLOGY OF READING 55 were carefully joined from end to end so as to form, on each card, a continuous line of print. The exposures were made with the Cattell Fall Apparatus, an instru- ment in which a thin steel plate containing a rectangular horizontal cleft is arranged to fall close before the printed line, the latter being invisible except while the rectangular opening in the plate permitted a peep at the line in passing. The length of the exposure thus depended upon the height from which the plate had fallen before the cleft reached the printed Une, and could be regulated accordingly. The reader looked at a point close before a marked place in the line, two seconds before the plate fell, and^ thus had his eye fixed on this known point in the line during the exposure. The lines were seen at the usual reading distance, were well lighted, and the reader sat comfortably in a quiet room of the laboratory. The first series of peeps was taken at intervals of 1.75 centimeters, eleven and one-half letter spaces, in order as one would read, the reader saying aloud as much as he could see after each exposure. Then the intervals were increased to 2 centimeters, 2.25 centimeters, and so on up to and including intervals of 4 centimeters. The reader always had the benefit of knowing the context up to the section about to be exposed, the preceding sections being read to him if he had not made them out. It was found that some readers could read continuously for a considerable distance when the peeps were taken at 56 THE PSYCHOLOGY OF READING intervals of 2.5 centimeters, sixteen letter spaces, and not repeated. Of four readers tested three were able to do as well as this, except for occasional failure through inattention or the special conditions of the experimenting. When the peeps were at intervals of 4 centimeters, or separated by twenty-six letter spaces, none of the read- ers were able to read continuously for any distance, although all but one at one titaie or another read phrases of greater length than this, and that one read phrases as long ag 3.7 centimeters. When the peeps were taken here and there in new matter without giving the preceding context, it was foimd, to my surprise, that a little more was read, on an average, than when the context was known. With preceding context known, considerably more was read to the left than to the right of the fixation point, on an average. Without the context a little more was read to the right, although in neither case did this hold for all the readers. The average extent of matter read per exposure was of course much less than the amounts stated above. One reader read an average of ten letter spaces, another fifteen, a third eleven, and a fourth, with a curiously hmited range of distinct vision, averaged but five, — the average read- ing being ten letter spaces for these four readers, from many hundreds of exposures. If we include in our average the cases in which the beginning and the end of a section were correctly read but some intermediate part THE PSVCHOLOGY OF READING $7 was lost, the average extent was eleven and one-fourth letter spaces. Comparing the average amounts thus read in these peeps at the line with the average amounts read at the reading pauses in actual reading, we find that they are nearly the same, two representative readers being found to cover about ten letter spaces per reading pause, on an average, when reading similar lines. However, a comparison of averages has but a partial value, for the actual amounts read at each peep at the line or in each pause in ordinary reading varies very much from the average. For instance, in actual reading it was found that in rare cases two pauses sufficed for the reading of a whole Journal line (Fig. 3). The exposure experiments cited above show that in similarly rare in- stances each reader, even the one with the much limited average range, was able to read correctly nearly or quite the half of such a line at a single peep. The readers stated, too, that they actually saw the whole extent thus read at one exposure without conscious guessing or estimating. Some of these longest readings are given on page 59. We may conclude, then, that the reader's eye is usually capable of taking care of as much as nearly half a line of the length and type used in our experiments, provided conditions are favorable. The fact that much less was read at most of the exposures was due to causes some of which can best be stated in our later discussion. It was. 58 THE PSYCHOLOGY OF READING of course, inevitable that sometimes the reader's attention would wander at the moment of the exposure, and then little or nothing would be read. Again, parts of the line which were seen clearly for an instant would be forgotten before they could fix themselves in mind sufficiently to be reproduced. Then again, the amount that could be read varied very greatly with the character of the matter exposed. In general, the more the word groups resem- bled isolated words, as when divided by punctuation marks, the less easily they were read. Prepositional phrases, sub- stantives with a series of modifying adjectives or with a closely linked phrase modifier, and series of any kind which had a rhythmic swing, were preferred. Certain words, usu- ally rather unfamiliar ones, presented peculiar difficulties. It seemed almost impossible to bring about a recognition of them by repeated exposures when the reader failed to rec- ognize them at first. " Titillation" was exposed ten times successively before recognition had proceeded so far as to call it "tililation." All the readers had difficulty with this word. The letters would be clearly seen, but apparently could not be remembered long enough to enable the reader to construct the word. Raison d' itre caused similar trouble. It might be supposed that in a momentary peep at a line of print approximately as much would be read on one side of the fixation point as on the other, but this is by no means the case. The averages, it is true, do not show great differences, but it is very different with the results of THE PSYCHOLOGY OF READING 59 particular exposures. The following table shows the ex- tent in millimeters read on each side of the fixation point by two readers in consecutive peeps at the line : — Reader A Reader B Left Right Left Right 20 25 S 7 19 12 6 17 10 13 18 24 9 IS 9 IS 2 13 14 13 17 4 18 4 10 8 10 8 Fig. 8 Some of the longer readings by various readers in these experiments are given below, with the fixation points marked : — " Condition of consciousness by a brightly colored These muscular contraction Condition of consciousness So difiicult is. the of the relevant ones the whole body but also the movements The whole body converges by a brightly colored object typical form, known as are not of equal value these muscular contractions Fie. 9 6o THE PSYCHOLOGY OF READING It is interesting to note that in almost every case in which a large amount is read, far more is read to the right of the fixation point than to the left. We can hardly explain this by attributing it to an involuntary wandering of the eye's fixation toward the right. No tendency to wander to the right rather than to the left has been noted in other experiments, and there is much reason to think that the fixation point did not vary more than a letter space or two at most, except perhaps very rarely. One of the readers who is thoroughly practiced in such matters constantly stated the letter or space which he supposed he had fixated; and in practically all cases his statement was correct. Two of the other readers had had a fair amount of laboratory practice and exercised all possible care in preserving a constant fixation. The fourth, a graduate student in mathematics, was at least as careful as the others. The conditions, such as close grammatical connection, etc., mentioned earlier as making for larger total readings, tend similarly to give larger readings to one or another side of the fixation point, as they occur there. Then, too, the attention is by no means always directed to the same point as the eye's fixation, and there may then occur a greater readiness for dealing with matter seen at the right or left. Inr-the case of the larger readings in which so much more is read at the right than at the left, this is probably largely due to the tendency of our words, as of all mental contents, THE PSYCHOLOGY OF READING 6l to call up or to facilitate the perception of those associates which have habitually succeeded rather than preceded them. The words far at the right, although dimly seen, are helped into consciousness and preserved in memory by associative links from those that are clearly seen. Then, too, the latter half of a long word seen far at the left is not apt to suggest the first half, and the whole word is lost. The first half of a long word far at the right, on the other hand, is apt to suggest the whole word, both be- cause the flow of association has been in that direction and because the first half of a word is much the more important for the word's recognition, as we shall see later. The maximum amount which can be read during a read- ing pause has been measured by various other experi- menters, with fairly congruent results. Erdmann and Dodge found that a German reader, in a single exposure lasting one-tenth of a second, read correctly sentences con- sisting of from four to six words of two to ten letters each, and occasionally recognized a simple word even at the end of a sentence of seven words, containing twenty-six letters. The middle of the sentence was fixated in all cases. The type was larger than that used in the writer's experiments, and the other conditions somewhat different. Messmer, experimenting at the University of Zurich, found that most of his readers could read nearly as much as this, in certain cases. But Hke the writer, he found certain readers with a curiously limited reading range. 62 THE PSYCHOLOGY OF READING Cattell found one reader who could read as much as seven words at a single exposure, when the words composed a sentence and were given in two lines. His other readers, however, were usually limited to four words. Of course all these results show only the maximal amounts which can be perceived at a glance, corresponding roughly to what is read in a reading pause when pauses are least frequent. We have seen that the eye usually makes pauses so fre- quently as to keep well within this maximum, the average amount covered per reading pause approximating the average amount read in the momentary exposure tests. My readers will doubtless wonder that so much of sense matter can be read at a single glance or pause, when the particular letters can be made out only, as we have seen, within a limited radius about the fixation point. It is very likely, as I have suggested, that one can really make out letters somewhat farther from the fixation point than the Erdmann and Dodge experiment indicates. But there is no doubt that words of sentences are read at a distance from the fixation point at which letters are no longer recognizable. Similarly, Erdmann and Dodge found that words could be read at a distance from the reader which made the constituent letters unrecognizable when presented singly. Yet in .many of these cases the reader states that he sees clearly all letters of the words or sentences read ; and in my own experience, as also for Erdmann and Dodge, the letters of words seen THE PSYCHOLOGY OF READING 63 far from the fixation point seem about as distinct as the others. Nevertheless it seems certain that in the longer readings the parts most distant from the fixation point are not clearly seen except with the mind's eye; they are filled in mentally by suggestion from what can actually be seen, somewhat as we recognize a friend from a glimpse of his hat and cane or of his bowed form. Not only are words thus recognized when letters can no longer be made out, but Erdmann and Dodge found that even very familiar short sentences were sometimes recognized as wholes under conditions which prevented recognition of their constituent words. Careful distinction will have to be made, therefore, between what is actually "seen" during a reading pause, and what is mentally supplied, and we must review the experiments made to determine the reading range for meaningless letters, words, etc., with the suggestions which these furnish for an explanation of the apperceptive filling in which occurs in reading. AVhen a series of letters in nonsense arrangement is momentarily exposed to a reader, the exposure suffices, according to Erdmarm and Dodge, for the naming of but four or five letters, in the majority of cases, although as many as six or seven letters may, in exceptional instances, be read. Usually, when the series consists of six or seven letters, the first and last letters are clearly perceived, showing that the ey^ can recognize single letters at least as far into the periphery as this. The intervening letters were 64 THE PSYCHOLOGY OF READING possibly recognized, but failed to persist in memory until their names could be stated. Erdmann and Dodge found that while but four or five nonsense letters cojild regularly be read at' a single ex- posure, words consisting in the aggregate of four or five times as many letters were read under similar conditions. Their readers recognized, at a glance which lasted but one- tenth of a second, words having as many as twenty-two letters, in one instance, and twenty, nineteen, eighteen, seventeen, etc., letters in other trials by the various readers. They believe, however, that for the words, as for the nonsense letters, only an extent of six to seven letters is clearly perceived, although their readers believed that the letters of the words were seen, as letters, over three times this extent of space. It was found, too, that while nonsense words gave readings that were three or four times as large as readings from nonsense letters, the readings were still uniformly less than when the words were combined to form sentences. Zeitler, experimenting at the University of Leipsic, found that the most difiicult reading was of consonants in nonsense arrangement, such a,s v c p } n g I w. Of these, four to seven were read at an exposure, while with vowels interspersed from five to eight could be read. A series of familiar syllables joined continuously, as lencurbilber, losverkungwei, was next easier to read, six to ten letters being read at a glance. Progressively larger readings THE PSYCHOLOGY OF READING 65 were obtained from series of unfamiliar words, familial words, sentences, and familiar expressions or proverbs. Sentences consisting of four or five short words, with a total of twenty to thirty letters, were readily read. Single words having as many as nineteen to twenty-five letters each, as Bewusstseinszustand and Aujmerksamkeitsschwan- kung, were read at a first glance of one one-hundredth second. Evidently, then, the amount that can be read during a reading pause varies greatly with the nature of the reading matter. It may seem strange to some that this should be so, that the amount we see at a glance should depend upon whether the line makes sense, whether it is composed of sentences, nonsense words, nonsense sylla- bles, vowels, consonants, or what not. Let us determine first what the limiting factors may be on the side of eye structure and function, and later examine the psychic factors. In looking at a line of print, as at any object, an image of the line is formed upon the retina somewhat as is the image which we can see upon the ground glass of our camera when the focusing has been done properly. The image is inverted, of course, in both cases, and the mind at the outset must interpret the picture as representing a printed line that is "right side up" and "right side to." But let us examine further the nature of this inverted picture. We know that the rods and coneS; which alone 66 THK PSYCHOLOGY OF READING are sensitive to the light impression, are less and less abun- dant from the central fovea outward toward the periphery of the retina, and th^t the distinctness of the retinal image falls off rapidly, accordingly, as we go from the center. The visual field corresponds, therefore, as Helmholtz says in his " Physiologische Optik" (p. 87), "to a drawing in which, indeed, the most important part of the whole is carefully executed but the surrounding parts only sketched, and sketched the more roughly the farther they are from the main point. " The little depression in the retina, called the fovea centralis, in which the cones are closely packed together and in which, accordingly, the retinal picture is complete even in its smaller details, is only about one-fifth of a millimeter in diameter. It thus includes not more than perhaps three-fourths of a degree of the retinal image, cor- responding of course only to this small extent of arc on the printed line, about three letter spaces at the ordinary reading distance and with this type. The macula lutea, or "yellow spot," in which the fovea lies, is itself not more than about three milUmeters in horizontal diameter, in- cluding thus about eleven and one-fourth degrees of the retinal image, corresponding to about 6.3 centimeters or a little more than six words on this page, at the ordinary reading distance. As we leave the fovea there is a marked decrease in the distinctness of the image, a gradual blurring and losing THE PSYCHOLOGY OF READING 67 of details, due to the fast-diminishing numbers of sensi- tive retinal elements. Only the larger and larger figures of the picture appear as we go farther, and finally only the most general outlines of even the forms that are of con- siderable size are evident in the extreme outer parts. It is somewhat as though our camera plate were comparatively perfect within a very small circle at the center, but were perforated from this outward to form a sensitive netting having larger and larger meshes toward the edges of the plate. A print from such a negative would give some suggestion of the character of the retinal picture, and would be called exceedingly defective as a photograph. It will be readily understood, then, that, while for a space of six or eight letters the single letters can be made out independently of each other, with perhaps the dots and small marks in most of them, the small marks must in- evitably disappear as we go farther from the fixation point ; the small letters must gradually disappear except as their presence is suggested by what clews remain in the rough outline; later even the large letters and all but the most general outlines of words must be, in part at least, inferred from such clews of context, etc., as are to be had. So it is clear that the larger the amount read cfuring a reading pause, the more inevitably must the reading be by suggestion and inference from clews of whatsoever kind, internal or external. In reading, the deficient picture is filled in, retouched, by the mind, and the page is thus 68 THE PSYCHOLOGY OF READING made to present the familiar appearance of completeness in its details which we suppose to exist in the actual page. The defective retinal picture, taken in connection with all the other clews available to consciousness at the moment, means such a page, and we project this meaning outward, 'just as we fill in mentally the gap in the "visual field left by the blind spot. Two facts that especially concern u^ are very evident from what has been said: First, reading may and must go on by other means than the recognition of letters; second, the amount that can be read at a reading pause, and consequently the number of necessary pauses and movements per line and page, will vary with the nature of the matter read, with the associative connections existing between the letters, words, etc., and with the reader's familiarity with what is read, the latter enabling any part that may be cl^ar to help into consciousness other parts that are indistinct. Turning now to the psychic side, we find certain further hmitations upon the amount that can be read during a reading pause. And first, the recognition of any partic- ular object as such necessitates a unitary focusing of con- sciousness, practically an act of the attention. But it is a well-known fact that but very few acts of the attention can take place simultaneously or within the narrow limits of a reading pause. Let my readers try taking a momen- tary glance at a number of distinct objects, all of which THE PSYCHOLOGY OF READING 69 are in plain view at once, as at a group of pictures, the faces in a passing car, or the jumble of things in a work-basket. You will realize after a few trials that the number of separate recognition acts per moment has rather narrow limits, and is not merely a matter of how many objects can be imaged simultaneously on the retiaa. The extent of this span or range of the attention or apperception has been variously stated by those who have attempted to measure it, but is usually found to include not more than four or five unrelated impressions. To attempt to distribute the attention over more than about this number is only successful when we can in some way unitize them, when we can somehow relate them in our thought so that we are conscious of them in groups or as a whole having a unitary meaning. However, when groups of objects, no matter how complex in their details, have these details firmly organized into a imitary whole and are thought of as wholes, then about as many of these wholes can be attended to simultaneously as if they were simple objects. Accordingly we find that readers recog- nize, apparently in one pulse of attention, four or five unrelated marks of different shapes, four or five letters in which such marks are unitarily combined, or four or five unrelated words which are still higher unitary com- plexes of these letters. Even two or three short sentences which had come to be thought as units might possibly be recognized in one pulse of attention. Of course it 70 THE PSYCHOLOGY OF READING may be said, and it is doubtless iti some measure true, that a wandering of the attention may occur during a reading pause, that the recognitions that occur are not all simultaneous but are in some measure successive. This will be considered presently. But all admit that not more than a very few acts of recognition can occur, whether simultaneously or successively, within the limits of a reading pause, and this is our main concern here. Again, when we consider that the attention must con- cern itseK partly with the meaning, with the images, feel- ings, and conscious states generally which are aroused by the reading symbols, and in many cases with the artic- ulation of words, we begin to wonder that the mind can deal with so much data, rather than with so little, in any given moment of our reading. To summarize then, we are limited, in the amount that can be read during a reading pause, by the inadequacy of the retinal structure, by our inability to attend to more than a few parts of the total picture presented, and by the necessity of our attention's concerning itself with in- terpretations. CHAPTER IV THE EXPERIMENTAL STUDIES UPON VISUAL PERCEPTION IN READING We must next consider the mental processes concerned in perceiving what is before us on the page, and the means by which the mind takes note of what is there at such a very rapid rate. This raises, of course, the time-honored question of whether we read by letters or by words ; but we shall find that much more is involved than the settlement of this somewhat scholastic query. The fact that during a reading pause one may read as much as even twenty to thirty letters when combined in sense matter, and that one averages usually as much as ten letters, suggests, as I have indicated, that the reading must go on by some other means than the recognition of letter after letter as was once supposed. This old and deeply rooted assumption was foimded partly on the general behef that the eye passed from letter to letter along the line, the recognitions following the fixation point suc- cessively. The letter-recognition theory was strengthened, too, by certain data furnished by aphasic patients, data which were interpreted by speciahsts in speech defects into a theory of reading by letters which it is impossible to 72 THE PSYCHOLOGY OF READING hold in the face of what we now know about the eye's actual movements in reading. The eye being still while most of the data is received from the page, it is perfectly certain that stimulations from letters in various parts of the section before the reader affect him simultaneously, and that there cannot be separate acts of recognition for each letter. Professor Cattell early concluded, as a result of his ex- periments at Leipsic upon the amount which could be read in single short exposures, that we read in word-wholes and even, sometimes, in phrase or sentence wholes, and not by letters. This was evidently before the nature of the eye's movement was known to him, although the discon- tinuous character of the movement had already been determined by Professor Javal and his pupils. Cattell foimd that when single words were momentarily exposed, they were recognized as quickly as single letters, and indeed that it took longer to name letters than to name whole words, the exposures being made under conditions in which the times could be accurately measured. It was found that when sentences or phrases were ex- posed, they were either grasped as wholes or else scarcely any of the words or letters were read. This observation was strikingly confirmed in the writer's experiments in which sentences were momentarily exposed. Rarely were single letters read, even as forming the beginning or ends of words that were but partially recognized. THE PSYCHOLOGY OF READING 73 The readings were of whole words, and almost always of words connected in some sense fashion. The words appear very distinct, even, as Cattell says, "when the ob- server constructs an imaginary sentence from the traces he has taken up." Professor Cattell also found that the shortest exposure which would permit the recognition of single small letters and capitals sufficed also for the rec- ognition of short words, and that long words needed but one one-thousandth of a second more. The time needed for naming a word was considerably less than for naming a letter, and the time needed simply for recognition without naming was "only slightly longer for a word than for a single letter. We, therefore," he adds, "perceive the word as a whole." Again, he found that when imrelated letters or words were read aloud as fast as possible, the reading was about twice as slow as when the letters or words were combined into words or sentences respectively; this indicating that in the latter cases the reading was in larger wholes than letters. Erdmann and Dodge argue strongly for the theory of perception in word-wholes, on the basis of numerous and varied experiments. The length of the word and its char- acteristic general form as a visual whole seem to them to be the main means by which it is recognized by the prac- ticed readers. They base their argument mainly upon the following facts : First, words are recognized when lying too far from the fixation point to permit recognition of 74 THE PSYCHOLOGY OF READING their component letters. Second, words are recognized when formed of letters so small that the letters could not be singly identified. Third, in about half the cases tried, words were recognized at distances at which the letters, when exposed singly, could not be recognized. Fourth, in the latter experiment the words were more readily recognized when they were long, or of optically char- acteristic form. Fifth, when twenty-six selected words were learned thoroughly in a fixed order, as the alphabet is known, and then exposed beyond the distance at which the letters could be recognized, the words could be dis- tinguished and recognized in almost every instance. Sixth, words of four letters are named somewhat more quickly than single letters, and words of eight, twelve, and sixteen letters need comparatively little more time, the longest words needing only about one-fifth moire time than the shortest. These authors point out that it is not the constituent parts of any given form that make it recognizable, but it is the famihar total arrangement. Thus - • d is not recognized as 5 nor < I as K, although the constituent g n parts are presented. The arrangement d has all the a > e r elements of a familiar word and, indeed, in their usual order. But it is by no means the visual form recognized THE PSYCHOLOGY OF READING 75 at once in the word reading. Why should not a familiar word-form be recognized and named on sight just as a house or wall is recognized and named ? We do not, in the latter cases, take account of the constituent stories and bricks ; nor of aU the sticks and Umbs and leaves in recog- nizing a particular thicket or oak tree. The arrangement, the total form, is the main thing, whether in the recognition of letters, numbers, words, or objects of whatsoever sort. One may always analyze the whole into its parts and recog- nize each part singly, as we have done in the figures above. But we do not do this in actual reading any more than in regarding a landscape. So the argument runs, and much more might be said for it. We shall next consider another view, advanced by Goldscheider and Miiller, on the basis, of experiments made at Berlin. These experimenters, working earlier than Erdmann and Dodge, found first that when a group of simple imrelated strokes in various arrangements, as was exposed for one one-hundredth of a second, only four or at most five strokes could be recognized or Note. The cuts and quotations from Goldscheider ajid Muller are from their article "Zur Phys. und Path, des Lesens," in Zeiischrift /. Klin. Med., Bd. XXIII, p. 131 ff. 76 THE PSYCHOLOGY OF READING described and reproduced. When simple unrelated strokes were grouped into a regular symmetrical form, as although the whole had no definite meaning and was totally new, seven strokes could be similarly "recognized" and the arrangement given. When the strokes were combined into squares and a group of the squares was exposed in varied arrangements, as the form, orientation, and relative arrangement of two or three squares could be told at a glance, although this, of course, involved locating and describing eight or twelve strokes. With a symmetrical arrangement and similar orientation of the squares, as D □ D THE PSYCHOLOGY OF READING ^^ the whole group of five squares could be reproduced at the first-- glance, although this involved twenty strokes. Similarly with semicircles, ellipses, etc., the number that could be recognized and correctly reproduced, at a glance, increased regularly with their arrangement into forms that could be grasped unitarily. An arrangement like ',:■ was easily reproduced so far as the general form was con- cerned, but to state how each particular semicircle faced was very difficult. From the constant recurrence of various geometrical forms, in the world of things, we come to have a stock of ideas of these forms all ready to use, ready to be touched off by even very slight cues that may appear in any visual complex. This aliveness or acute apperception for total forms makes us negligent of the details that appear. They are too numerous to be attended to, and can come to consciousness with less expenditure of energy as parts or aspects of the total upon which the thought is mainly focused. So recognition by general forms rather than by particular details may be expected and will occur prefer- ably wherever the total arrangement has often recurred, 78 THE PSYCHOLOGY OF READING and where attention to certain details is not absolutely necessary for the determination of the recognition. The general form repeats itself oftenest, and so we are most ready for it.' Goldscheider and Miiller found that such a group as t.x-,V had to be exposed seven times before it could be repro- duced, while the same forms arranged into i'y'L were reproduced at the first or second glance. So when o/'P^V was exposed, but four or five characters could be per- ceived at a glance, but when these same forms appeared as 51 B^ ' Note. The theoretical view presented in this paragraph is not necessarily that of Goldscheider and Miiller. THE PSYCHOLOGY OF READING 79 the whole was recognized at the first or second glance. The habitual association of the parts into a unity, which makes the perception facile and the memory after the exposure easy, and the familiarity of the total form as an unanalyzed whole, work together as factors in these as in all such recognitions. Goldscheider and Miiller went on to expose series of nonsense letters, syllables, words, phrases, etc. They found that "an optical memory image" of a word was readily called forth by an incomplete series of its letters. Certain letters would be disregarded when present in the exposed word, or might be omitted and the recognition would still occur readily. The letters which seemed to be especially used in determining the recognition of any given word were named "determining letters." The others were named "indifferent letters." The places of the absent or disregarded letters would be filled in sub- jectively when the exposure was made, sometimes filled with the wrong forms even though the right letters were actually there. Whether wrong or right, the letters thus supplied were apt to seem as distinct on the page as did the others, and these authors quote with approval Professor Miinsterberg's conclusion ' that "reproduced sensations under favorable conditions caimot be distinguished from sense impressions." Exposure of C ntr m constantly sufl&ced for recognition of the German word Centrum, ' "Beitrage zur Experimentellen Psychologie," H. 4, s. 17 ff. 8o THE PSYCHOLOGY OF READING but ent um did not. Klangbild was recognized readily from Kl ngb Id, but not from Ian hild. M k do gave Mi- kado, but Mik 0, of the same word, gave only Mikosch. Ch te gave Charite at once, and other such familiar words were recognized when a few characteristic letters were given. To the determining letter class belongs the first letter of a word, almost always. If it is wanting, the recognition is apt to fail, especially if its absence breaks up an initial diphthong. Autor was never recognized from utor. Here the wrong sound seemed to be suggested for the u, resulting in such completions as tutor; eweis did not give Beweis as intended, but edelweiss; weifel did not give Zweifel, but Weibel; ia n se was completed to Wann- see, and the actual word Diagnose could not be made out. If the determining letters are left out of a word, there is left an "indifferent word-form" which sometimes per- mits a great number of different completions. Goldscheider and Miiller do not find that the con- sonants are the determining letters as against the vowels as indifferent letters, as some suppose. For instance, Diagnose was recognized with greater difficulty when D gn se was presented than from D a nose. The greater importance of the vowels in such a case may be due to the fact that the vowels give the clew to the number of syllables, and with this they awaken in us the memory of the rhythm and the accent. Or it may happen that the THE PSYCHOLOGY OF READING 8l vowel sound is of "determining significance" for the given word. The former would be apt to occur if the reader's inner speech habitually went on in motor terms, the latter if he were of the auditory type. The consonants, however, from their frequently projecting above or below the line, are apt to contribute more than the vowels to the char- acteristic form of the word. The kind of words which were usually suggested by the word skeletons presented in these exposures lead Gold- scheider and Miiller to conclude that the first suggestion from the sight of the determining letters is the sound of these letters, and that these sounds call forth or suggest, immediately, the sound of the whole word. They admit that the visual perception of the determining letters may sometimes be filled out at once with the remaining visual forms, and the word-sound then be aroused from this total visual form. But this, they think, occurs but seldom, and is a roundabout process. In general, these experimenters conclude that the more unfamiliar a sequence of letters may be, the more the perception of it proceeds by letters. With increase of fa- miliarity, fewer and fewer clews suffice to touch off the rec- ognition of the word or phrase, the tendency being toward reading in word- wholes. So reading is now by letters, now by groups of letters or by syllables, now by word-wholeSj all in the same sentence sometimes, or even in the same word, as the reader may most quickly attain his purpose. 82 THE PSYCHOLOGY OF READING In the case of reading by word- wholes, they call attention to the fact that the characteristic form of the word is conditioned by certain characteristic letters, namely, the determining letters. The reading of the blind, in the opinion of these authors, seems to illustrate this combinafion of methods of perceiving words. A practiced reader of the raised- letter pages goes ahead with the fingers of the right hand to examine the general outline of the word, while a finger of the left hand follows, gliding successively over the letters. Ordinarily, however, only a part of the letters are examined, while the finger passes over the others without touching the points. Intelligent and attentive blind readers state that they thus read but a part of the letters and conjecture the rest. Zeitler, experimenting at the University of Leipsic, made about six thousand exposures of groups of letters, words, sentences, etc., usually for very brief intervals. By making the exposures very short he thought he could best determine what letters, lejtter-groups, etc., stand out most prominently and are perceived when others are not. His experiments are therefore important as helping to determine what parts of reading matter are "determining" parts, or "dominating" parts as Zeitler prefers to call them. Zeitler found that in his brief exposures certain letters or letter-groups of a word, and indeed certain words of THE PSYCHOLOGY OF READING 83 exposed sentences, drew the attention to themselves and were apperceived. The apperception of these domi- nating parts or complexes is, he believes, the basis for the recognition of the word or sentence. These apperceived parts are at once supplemented by, filled out with, an inner mental contribution, associates that belong with the parts apperceived. The result is the blending of the outwardly given apperceptions with the inwardly arising associates into a total "assimilation," which constitutes the recognition of the word or sentence. "The word-form is indeed apparently assimilated as a whole, secondarily; but primarily, it is apperceived only in its dominating constituent parts." Zeitler admits, however, that or- dinarily we cannot distinguish these two processes of apperception and assimilation. His very short exposures, varying to suit the reader, ruled out, as he believed, most of the associative contribution, and caused the reader to strain his attention to the utmost upon the objective fac- tors, the matter actually exposed. It was found that the letters projecting above and below the line were recognized preferably. The vowels and small consonants were misread most often, the long consonants least often. In general, "the more characteristically" a letter is shaped, the more clearly is it recognized. As in the visual field with objects generally there are dominating points and lines which /get the attention, which reflexly attract the eye, and over which the eye preferably moves 84 THE PSYCHOLOGY OF READING or rests, just so when we regard words and sentences, the corresponding dominant parts here are these character- istically formed letters, "over whose high relief the eye springs along," although this last is true only figura- tively, he says, as the eye does not move during the exposure. As examples of the dominating letters may be cited the following, found to be the same for all of his (five?) readers : — Gold Gld Haut H t Fliege F Ig Woche W ch (ck) Streit St t Minute M t Cattell had already found that the different letters re- quired different times for their recognition, and that they were of different degrees of legibility. His observations here have a certain significant relation with those of Zeit- ler. It will be noticed that in the above examples the large letters are the dominating ones throughout, except in the characteristic combination ch. When much is read in the exposure of a sense passage, Zeitler finds that certain dominating " syllable complexes," usually those which contain the sense of the words, are apperceived and the rest is associatively supplied. If quite familiar sentences are exposed, there are dominating THE PSYCHOLOGY OF READING Sj words, sometimes, the perception of which suffices for the recognition of the whole sentence. Alterations in the "indifferent" words of the sentences, or even their ab- sence, may go unnoticed. The sentence is "assimilated" just the same. This experimenter opposes Goldscheider and Muller's conclusion that the perception of the determining or dominating letters arouses first the sound of these letters, the word-sound being fiUed out associatively from these sounds. Still, some of Zeitler's own experiments show that this occurs sometimes. Regularly, however, he thinks that the dominating complexes, when apperceived, are filled out directly into the visual form of the word or sentence. The dominating parts may be silent letters, or letters having a sound that is very different when heard singly than when combined in the given word. The conclusions of Cattell, Erdmann and Dodge, and others as to perception in word-wholes are also thought to be incorrect. The reading stimulus, when one looks momentarily at the page, is ordinarily not the whole sen- tence or the whole words printed there, at all. Externally, it is true, they are there and are of such and such total form, word-length, etc. But the real stimulus is the series of dominating letters or complexes. It is these which first affect consciousness and get the attention. It is these that are directly perceived. In Zeitler's opinion, then, word-length and total form 86 THE PSYCHOLOGY OF READING are not very important factors in the recognition. Foi instance, when the words in the left-hand column below were briefly exposed, the reader was uncertain whether he saw the one or the other of the words on the right, although they are markedly different in length : — Phantasie ^ Phalanstfere or Phalanstfere Skorpion Skioplikon or Skioptikon Pygmae Pygmalion or Pygmalion Such readings as Leoparden for Lepidodendron, Retoranda for Ritardando, Epimenides for Epaminondas, Polarstem for Phalanstfere, Agraphie for Agoraphobic show discrepancies in word-length and, quite often, in total form. The determining letters, however, are re- tained, only the indifferent letters being changed, omitted, or inserted. That the total word-form is not very fixed and rigid to THE PSYCHOLOGY OF BEADING 87 the consciousness is indicated by the inversions and per- mutations of even the letters that are dominant. Thus, Farbe was read Fabrik Meludie (exposed for Melodie) was read Medulla. Gefiidl (Gefuhl) was read Gefilde. Kulge (Kulpe) was read Klage. Fniede (Friede) was read Feinde. Analomie (Anatomie) was read Anomalie. It would seem to Zeitler that in the first perception of the dominant parts of a word these parts are not seen in any very fixed spatial arrangement, but are later put in place in a totg,l word-form when the full recognition com- pletes itself with the coming of the associative elements. At first "each dominating letter has a certain elbow-room in a space within which it can be changed about with its neighbors." There they hover, oscillating with the play of processes, until they become "anchored" in the places to which they are assigned in the total word- complex when this is once formed. "The letters are throughout not linked so fast to one another as they seem. The sense first welds them together." "The mere optical word-form is continually inclined to fall apart into its ele- ments, is held together only by the framework formed by the dominating letters. In this word-form the small and unimportant letters can be changed about quite irreg- ularly." "The word-form remains uncertain (labile), 88 , THE PSYCHOLOGY OF READING if it does not immediately receive its signification. First through the sense is the letter-complex established." Neither does Zeitler find that the perception of what is read during a reading pause occurs simultaneously for the various parts of the section read. The time of such a pause, very considerable as compared with the short times used in his exposure experiments, is quite sufficient to permit a wandering of the attention over what is read, and he fimds that such a wandering of the atten- tion actually occurred in the readings of his observers. The attention is upon the dominating letters or complexes, and wanders from one to another until all are apperceived. A dominating complex may consist of two or three neigh- boring letters, or may even be an entire short word in famihar sentences. In any case it seems to be perceived simultaneously over its various parts. Indeed, Zeitler admits that, when two distinct dominating complexes occur in different parts of a word, the attention may in cer- tain cases divide and be given to both simultaneously. But generally, in reading, "we arrange the dominating complexes successively one after another, similarly as we do the letters in the earliest reading by letters. The progress of the reading is only very fast, but it is none the less successive. With ordinary letter-after-letter reading, how- ever, this has nothing to do; instead, we arrange in a series one after the other the dominating letters and important complexes. This goes on, possibly, in a kind THE PSYCHOLOGY OF READING 85 of rhythmic succession, with continued variation in the rhythm." There is thus, according to Zeitler, a "very quick succes- sion of consciousness processes in reading," which indeed gives one the "illusion" of reading simultaneously what is seen at a glance, or during a reading pause. This il- lusion comes from long practice and from famiharity with the words. But with less famihar words, even the prac- ticed reader may establish for himself that the "simultane- ous reading exists only for the extent of a dominating com- plex." His reference here is only to visual perception in reading, and he remarks that the inner saying of what is read goes on successively, sound after sound. The following examples are given by Zeitler to illustrate the wandering of the attention which occurred when these words were exposed for periods of from one tenth to one fifth of a second, approximating the time of a reading pause. The strokes under the words indicate the impor- tant parts of the word-form, the bent arrows show the subjectively noticed course of the attention. All his readers seemed to agree that when these longer times were used, the readings were regularly successive and not simultaneous. That Cattell did not note any wandering of the attention and considered the readings to be simultaneous is due, Zeitler supposes, to the fact that Cattell's exposures lasted but one one-hundredth of a second, making the observa- 90 THE PSYCHOLOGY OF READING tion of such a wandering very difficult. Erdmann and Dodge, with their long exposures of one tenth of a second, were misled, he believes, by the illusion of the final simultaneous "assimilation," not distinguishing this from the slightly preceding successive apperceptions. Pharmako dyna mik. Kilimamscharo. Rochefoucauld. Demonstrationsversuch. Tagesbeleuchtung. Fig. gh- — Movements of tlie attention in Zeitler's readings. They were misled all the more from being mainly concerned with other factors. Messmer has more recently made a long series of experiments in the psychology of reading, at the Univer- sity of Zurich. He finds confirmation of Zeitler's con- clusions that perception in reading is mediated, for a certain type of readers at least, by "dominating" letters and complexes, and that there is a wandering THE PSYCHOLOGY OF READING 9I of the attention over these. He used very short ex- posures for the most part, as short as two one-thou- sandths of a second, making the time for each reader such as just sufi&ced to permit normal recognition. After practice, two one-thousandths of a second sufficed for all his readers. Messmer finds that the long letters which project above the line are usually the dominating ones. The attention concerns itself most with the upper half of the word, and the letters projecting below are not so important. The latter and the short letters are the ones most often mistaken in the readings from his exposures. Letters projecting below the line would be mistaken for vowels, as g for a, p for 0, etc. "They possess optically the value of small letters." The dominating parts of words and sentences are most apt to strike the eye and to get the attention. But accidental circumstances may sometimes make other parts more prominent. In the relatively long pauses of actual reading, very many if not all parts of the word can affect consciousness somewhat and thus give clews which help in the recognition, preventing the possibility of errors in filling out from the dominating letters, errors which actually arise, however, in reading from the short exposures. The dominating letters play the main r61e in recognition, but the others thus play an important part as well. The experiments indicated that readers may be either 92 THE PSYCHOLOGY OF READING of a subjective or objective type. The subjective type is characterized by a wandering attention which travels far from the fixation point, by a large associative contri- bution in perceiving, and by slight fidelity to the outward object. Readers of this type apperceive words from the total character of the word-form rather than from the dominating parts, these latter not differentiating from the whole. The objective readers, with characteristics which are the opposite of the above, recognize the dominating parts first, and the effect of the total form is minor. They read a smaller amount at a glance than do the subjective readers, but are less liable to error. It was found that during a reading pause there is first an impression of the whole word, as "hvely," "stiff," etc., for example, a feeling reaction to the total word-appear- ance. This may alone sufl5ce to set off the recognition of the word ; usually, however, with objective readers at any rate,, there follows a successive coming to consciousness of first the high dominant letters, then the low and "in- different" ones. Small letters adjoining a dominant letter may, by their proximity, help in forming a total configuration and may thus come to consciousness as part of a dominant complex. The effect upon consciousness of the total word-form as such is a simultaneous one, but the dominant parts come to consciousness successively. He agrees with Zeitler that in actual reading these suc- cessive acts of recognition follow each other so rapidly THE PSYCHOLOGY OF READING 93 that they appear simultaneous. He finds that word- length plays little part in characterizing words for chil- dren, and that it is usually less important for children than are the dominant complexes. Messmer's analysis of the "total character of words" has a considerable value. The three main factors are, first, breadth of the letters horizontally; second, height of the letters vertically; third, geometrical form of the letters. As to breadth, the letters are composed of one, two, or three vertical strokes, as i, h, m, or of forms oc- cupying one or another of these three horizontal spaces. Of one thousand consecutive letters on a German page, seven hundred and thirty are found to be small letters. While these are usually of about the same height, they have a variety of widths, as appears when words are printed in vertical arrangement, as W i m m Wimmem. e r n It might seem, therefore, that letter-width is a very im- portant factor in characterizing the total word-form. However, the differences in letter-width largely disappear in the total impression of word-length given in our hori- 94 THE Psychology of reading zontal arrangement of the letters. Differentiation in letter-width, therefore, seems to be of comparatively little value. Indeed, Messmer found frequent errors in perception due to combining parts of adjacent letters into a wrongly conjectured letter, or to making still other mis- taken groupings of the individual strokes. The uniform height of the small letters is the measure of the word's heiglit in the main, since these letters are in such a majority. But the long letters relieve the monot- ony by their projections, and thus characterize the word in the vertical meridian. Viewing the total word-form as to height, these long letters vary it and give a characteristic outline. Considering the total word-length, these letters break it up into sections, — "rhythmize" it, to use Messmer's expression. Note this effect in Ver- schiedenheiien, as compared with Zusammenreisen. The word is characterized, thirdly, by the geometrical form of the particular letters composing it. Disregarding their variations in height and breadth already referred to, the letters may be grouped, first, into those composed es- sentially of vertical strokes, as i, n, m, t, 1, f, h, r, j ; second, those composed essentially of curved lines, as o, e, c, s, a, g ; third, those composed essentially of both perpendicular strokes and curved lines, as b, d, q, p ; fourth, those com- posed essentially of oblique strokes, as, w, v, y, x, z, k, the last letter having also a perpendicular stroke. The first group includes almost half the letters, 469 per thousand THE PSYCHOLOGY OF ILEADING 95 as they occur consecutively on the page. The second group has over one-third, 371 per thousand. Only 61 per thousand belong to the third group, and 46 per thousand to the fourth. The remaining 63 per thousand were capital letters, more frequent, of course, in German than in English. The predominance of one or another of these classes of letters, in any given word, gives it a characteristic total appearance, as in wimmem and ubereinstimmen, for the first group, and ausgeschlossen, psychologisch, for the second. The former words have a unitariness of total character, giving an unbroken total impression. Their total form, however, is too little differentiated, and such words are most often misread and " most uncertainly recognized or falsely interpreted." They are "stiff" as contrasted with the words of the second group, whose letters have more individuality, and whose words are thus better differentiated and recognized with greater certainty. These characteristics of the first two groups are combined in such words as characteristisch, wis- senschaftliche, each containing about equal numbers of these two groups of letters. The total impression here is at least agreeable, and Messmer calls it "the most favorable total form," since it gives "the greatest harmony and most agreeable contrast." Groups two, three, and four include fewer of the al- phabet letters than are found in group one. As they also 96 THE PSYCHOLOGY OF READING occur less frequently in any given line, they thus have more individuaHty and differentiate the line better than the letters of group one. This is particularly true of the second group, the others partaking somewhat of the nature of the first group. In words which contain no long letters, as zusammenreisen above, certain letters having the greatest individuahty seem to serve as dominating letters, and to some extent break up or "rhythmize" the word. Messmer finds that readers do not get any very distintt notion of the length of the words exposed, very often mistaking a word for some much shorter one, occasionally -for a longer one. This was especially the case with his child readers, and he thinks that for them, at least, word-length is but a minor factor in word-perception. In experiments made some years ago I found that the first half of a word is of considerably greater importance for perception than is the latter half. If the reader will turn to page loo and will read down the last column of words as fast as possible, endeavoring to avoid lateral movement of the eyes, he will probably find himself fixating the words to the left of the center. If he will then read down the column again, fixating toward the end of the words, say three or four letters from the end, and again fixating near the beginning of the words, he will find the' reading much easier in the latter case. Indeed, in ordinary reading, I find myself much more conscious of the begiiming of words than of their other THE PSYCHOLOGY OF READING 97 parts, although I am not certain that my readers will be able to verify this by their introspection. It will be recalled that the beginning of a word was regularly found to be a determining or dominating part, in some of the exposure experiments. Indeed, the termi- nal letters are considerably more legible than the others, perhaps from being partially isolated. Fixate the middle of one of the long words and you will probably find your- self much more conscious of the end letters than of many of the intervening ones. The writer made a quantitative test of the comparative importance of the first and last halves of words by having readers read passages from which the first half of each word was carefully removed in the one case, and the second half in the other. Specimen lines are shown in Figure lo, the unmutilated passage being printed at the end. It was found that more words were made out, and in less time, when the first halves were read than when the latter halves alone ly ures f ch i eme? f is >u, en II )u dly, om 1 re ate he ages ad srms st ally pose so bine :se jws, tically o nd a ated py o ch ibutor. he jms low ely gest iing ics. ect Dse 3U re St ested n id id ay ers. at feat o su s sch I th do nc app t yc th wi yc kin fr £ mo pract stand st£ tl char at refc mo gres nee Ot purp i t com the vi«. statist at othei at t se s prir co t ea contri Tb ite bd met sugj lead top Sel the yc ai mo inter i ar ad an oth H es )t eal tical point. ded? ir id wise, ad 98 THE PSYCHOLOGY OF READING any features of sacli a scheme? If this does not .appeal 'tp you. then will you kindly, from a more practical standpoint, state, the changes and reforms most greatly needed ? Our pur- pose is to combine these views, sta.tistically and otherwise^! and to send a printed copy to each contributor. The items below merely suggest leading topics. Select those you ar^ most interested in and add any others. Fig. 10.' remained. The four readers tested averaged .49 words per second when reading from the first halves, as against .33 words per second when reading from the last halves. Among factors which cooperate to produce this result may be mentioned, first, the tendency of English to piace the accent upon the first part of the word, the accented part then tending to represent the word, at least the spoken ' word ; second, the preponderance of the number of suffixes over prefixes, the main root of the word tending to appear in the first part, thus rendering the first part more im- portant. It seems probable also, as a third factor, that the time-order in ordinary inter-association of syllables has much to do with the difference shown. This time-order has almost always been from the first part toward the latter, and, as has been shown by various experiments, associations do not work nearly so well in reversed time-order. The upper half of a word or letter is obviously more important for perception than is the lower half. This may be tested by comparing the difficulty of reading ' Reproduced from American Journal of Psychology, July, 1898. THE PSYCHOLOGY OF READING 99 the two mutilated passages below, the unmutilated pas- sage being in the same type: — Everybody knows the story of Mary and her little lamb; but not every one knows that Mary E. Sawyer, who was bom in Worcester county, was the heroine of the poem. ^X7Vlpn ATqrv txtqq m lif*-l<» trirl cKa f^Minfl e* Ti*>Tv_KnTn lomK n*>Qrlv f]f^ai^ vnii\\ liiir)cr*^T- on^ rnlrl Clio +*inHArlv T»ni*c<»'1 it Ko^V tr> lif*> flnH Ko/'pmo Hoirr\*-*iHltr otturVyt^A fr» }ioi* (rpntiA pliorcr** T'Vki loruK tttiq nAi- rnnctont rr*mT^qnirtn nnri T^lo\Tma*-/a an/1 txtog tn Iiai* nrViof g VViiCii ^YJ-aLy a luxu t^ui^\^ i\n. uci ivi-itaiious Liic Ltxiiiu xou xnjnu i±A\^ aioi\, a,±\,i^± ixv^x Lu liixr iiitcxioc uciiguL v/i Luc auxxuiuxa duu llic sui~ pxiSC Ul LUC LCCLCUCl. X UC iUULU W OS ^UL L/ULOiUC, (UiU IL WAILCLI VU iuc uuL/iaLCp itjr xvxai^ OUU lUUUWCU UCl UUUIC* Fig. II. Professor Javal, from watching the course of an after- image along the lines as he read, and for other reasons, concluded, as we have seen, that the eye 's fixation point moved along between the middle and top of the small letters, thus giving an advantage in perception to the upper half of the line. As already mentioned, I do not consider his experiments final on this point ; and it seems to me that the greater importance of the upper part is due rather to the words being better differentiated there than below, as is shown by Messmer's count of two hundred and thirty-eight letters projecting above the line to thirty-two below. Besides, we habitually find most ■meanings in the upper parts of objects; we ourselves are so placed and so oriented as to bring this about. lOO THE PSYCHOLOGY OF READING In considering whether we read by letters or by word- wholes, my readers may be assisted in their judgment by reading down the column of letters below, as fast as possible, either simply recognizing each or pronounc- ing it aloud, and then doing the same for the columns y pool analysis anthropology w rugs habitual independence u mark occupied histological s send inherent astronomical q list, probable tautological o more summoned paleontology m pick devotion consummation k stab remarked concomitance i neck overcome epistemology g your resolute irritability e dice elements somnambulism c font conclude minimization a earl numbered malleability z whit struggle emblematical X ants division permeability V role research etymological t sink original quantitative r rust involved ascertaining P ware obstacle definiteness ti fuss relative sociological ^ 1 tick physical legitimately J rasp pastness scientifical h mold lacteals institutions f hive sameness governmental d four distract emphatically Fig. THE PSYCHOLOGY OF READING lOl of words containing four, eight, and twelve letters re- spectively, comparing the rate and difficulty by the aid of a stop-watch, if one is at hand. It will be evident that the multiplication of letters makes proportionately little difference in the ease or speed of recognition. In my own experiments in which such Usts were read aloud as fast as possible, my four readers read the lists in the following times: — 50 letters in an average of 15.7 seconds. 50 four-letter words in an average of 17.3 seconds. 50 eight-letter words in an average of 19.6 seconds. 50 twelve-letter words in an average of 28.5 seconds. 50 sixteen -letter words in an average of 54.1 seconds. Since part of the slightly lessened speed of reading eight- letter words as compared with those of four letters must probably be due to the utterance of the additional syl- lable or syllables in the former case, it seems certain that the recognition of familar and comparatively short words is little affected by doubling the number of letters ; and this seems confirmatory of the view that such words are recognized in one unitary act, as wholes. The greatly lessened speed of reading the words of sixteen letters as compared with those of twelve is due in part to their being considerably less familiar. It is probably due in much greater part to the need of making lateral movements of the eye, these words occupying a considerable horizontal space when typewritten for use in the experiments. CHAPTER V THE NATURE OF THE PERCEPTUAL PROCESS IN READING It is very difficult to draw final conclusions concern- ing visual perception in reading, in the present stage of the investigations. I have, therefore, given at some length the views of the various experimenters, and have referred to many of the particular experiments upon which they are based, that my readers may be helped in drawing their own conclusions. I am glad to know that further studies are being undertaken in this field, and by comparing the results of these as they are published with this review of the work thus far, the truth will no doubt gradually appear, as to most of the problems. We are all working toward daylight in the matter, and many of the discrepancies of facts and theories are more apparent than real. A very important section of general psychology must here be worked out, constructed in the new rather than taken and apphed. How do we perceive anything? The whole stupendous problem rises at every turn, but is far too large even to be ade- quately stated here. The following conclusions seem to me to be warranted by the data now at hand. Goldscheider and Miiller were profoundly right when they said that readers perceive in various ways as their pur- THE PSYCHOLOGY OF READING 103 pose can be best attained. We must allow for consider able variety, not merely of individuals but of occasions. The manner of perceiving words must depend, for the child, very largely on how he is taught to perceive them in learning to read, and here, as we know, the methods are most diverse. To take a simple example, the writer still finds himself hesitant in naming or recognizing several capi- tal letters of the Greek alphabet, perhaps even incapable of recognizing one or two of them when seen in isolation. Yet Greek was a favorite study with him through years of college and secondary school. The reason for his per- sistent inattention to the letters is evident enough. He began Greek with the sentence method, and his attention was seldom called to the particular letters in reading. The perceptual process in the practiced reader doubt- less uses many a short-cut for which there can be no science. Such a reader has grown up with these letter- forms and word-forms as intimate parts of his environ- ment. He has made friends, boon companions, of them, in his own way, and differently with each. Some early absurd way of thinking about A, or about the appearance of the word cat or and, may have grown to be the in- variable feeling reaction which greets this form when it appears, may really be the core of the consciousness in its recognition. Various experiences and associa- tions have been woven in with the appearance of the various word-forms, differing from reader to reader. I04 THE PSYCHOLOGY OF READING These determine in part what will stand out most promi- nently in any given word. They bring out now the general outline of the total word-form, now this or that dominant letter or complex, now the word-length, etc., as one or the other of these may have formed the basis of our individual experiences with the given word. However, there are general features of the perceptual process which appear as we survey the collected data from all the experiments on reading. In the first place, perceiving is an act, a thing that we do, always and every- where, never a mere passive sensing of a group of passing sensations or impressions. It probably always involves actual innervation of muscles, and indeed coordinated and organized, we may say unitized, innervation of muscles. Certainly on the psychic side there is an active and more or less unitized movement of mind, a sense of inner activity. Perceiving being an act, it is, like all other things that we do, performed more easily with each repetition of the act. To perceive an entirely new word or other combina- tion of strokes requires considerable time, close attention, and is Ukely to be imperfectly done, just as when we attempt some new combination of movements, some new trick in the gymnasium or new "serve" at tennis. In eitjier case, repetition progressively frees the mind from attention to details, makes facile the total act, shortens the time, and reduces the extent to which consciousness must concern itself with the process. One may say that the THE PSYCHOLOGY OF READING I05 "memory image" helps in the later perception of tht word ; but it may well be that, as Goldscheider and Miiller put it, the memory is but an exercise or habit of the ap- perceptive activity (Uebung der Apperceptionsthatigkeit), — that we perceive better at the later trial just as we shoot better or skate better with practice. Again, as in the performance of any act, a perception may involve more and more complex constituent acts as these are progressively welded together by practice, and especially as they become synthesized to a total per- formance which may be set off from a single consciousness cue. It comes about, therefore, that just as the com- plicated but associatively concatenated and organized movements of hitting a target with a ball may be touched off by the mere sight of the target, in one attention-act, so the various activities involved in apperceiving a phrase or other word-group may become one complex but unitary act, and this act may be set off very simply by this or that cue or set of cues given from the page, and may be done with a minimum of consciousness concerning details. Again, perception is always a projection or localization outward of a consciousness which is aroused or suggested by the stimulations that have come inward, but which is conditioned strongly, also, from within. We have seen how, when some dominant parts of a word or sentence were exposed without the other parts, the reader would project the absent letters upon the page and would "see" 106 THE PSYCHOLOGY OF READING them as distinctly as when they were actually before hina. We have seen, too, how in every moment of our reading we project letters and parts of letters to fill up the gaps that are always left in the peripheral parts of our retinal image. We know how we project a consciousness content to serve for what should be imaged on the "blind spot." Such projection is as certain and as common as is any mental phenomenon. The simple fact is that the words and all the other objects that we ever see are thus thrown outward, projected upon a page in the case of reading, somewhat as a lantern might throw them outward upon a screen. In the case of perception it might be said that the mind furnishes the screen as well. It must be remembered that consciousness does not dwell in the retina or in retinal images. Objects may be pictured very well without any retina or optic nerve. For our purposes here consciousness may best be thought of as in the brain, totally in the dark as to physical environ- ment, constructing even its light as well as its forms and meanings according to the excitations that come in to it and their relations with those that have previously come in. I raise here no question of idealism, and there need be no discussion of metaphysics.' An outer world ' Of course the whole matter could be stated equally well in terms of James' radical empiricism, without affecting the argument here. I have come to consider the doctrine of James to be nearer the truth. How- ever, my thought about perception in reading is doubtless more intel- ligible as stated in te'rms of my working hypothesis of plain dualism. THE PSYCHOLOGY OF READING I07 may be there and may be quite as I think it, doubtless is a very great deal more than I think it. But it is simply an empirical fact that I do project my thought of it, that there is constructed a consciousness world. When visual forms, then, affect my retina, there come to the brain certain signs of their presence, position, etc. The character and destination of the incoming excitations from the retinal image are sufficient signs of the presence of the particular form. Its distance, size, and orientation in space are suggested from certain other signs, such as excitations from the muscles of accommodation and con- vergence, along with others which indicate the bodily position at the moment. The totality of signs, out- wardly and inwardly initiated, suggest or awaken a con- sciousness corresponding to the particular sign-combina- tion, a consciousness which is projected or placed, we say apperceived sometimes, in its proper place and relations, or rather a consciousness which is our seen world of the moment, including the page, the letters, etc. The signs indicative of the presence and nature of an object of perception may be either states of consciousness or merely neural conditions. In the case of a word per- ceived upon a printed page, it seems likely that the dis- tance and direction of the word from the reader, and the word's arrangement as spread out upon the page, are suggested mainly by states of the eye-muscles and tendencies to innervation peculiar to dealing with such a Io8 THE PSYCHOLOGY OF READING form in such a place, states and tendencies which are in the main neural only, except under artificial con- ditions of introspection. On the other hand, certain of the signs suggestive of the perceptual consciousness are themselves consciousness states, having, of course, neural excitations also as correlates if not as causes. Among these consciousness signs are the various intensities of blacks, whites, and grays which occur in the printed form, and the context imagery and feeling from what has just been read and from the general subject of thought and feehng for that moment. Part of these signs are operative precedent to the beginning of ad- equate stimulation from the particular word perceived. Such are the signs indicative of the word's distance and position, with a certain context consciousness lead- ing the reader to expect a certain kind of word, etc. The word is to this extent preperceived ; there is a "set" or "predisposition" in its direction which may need but a few supplemental signs to set off the proper perception. When we consider that the arrangement on the page, of the words and of their parts, is a construction within from cues which are probably given in non-spatial order, we are prepared for the statement of Zeitler that in the first awareness of the dominating letters of a word they are not seen in any very fixed spatial arrangement, and are only put in place and "anchored" there as the recognition THE PSYCHOLOGY OF READING IO9 completes itself with the coming of the associative con- tribution from within. The wonder is that the cues to the arrangement of the letters are not more often fallacious, and that misreadings like Krone for Kome, aneotic fof anoetic, Larabee for Labaree, actual cases which I have noticed recently, are so infrequent with most readers. In this view of perception one is inclined to accept what the experiments of Zeitler, Messmer, and others seem to show, that the first factors of perception in reading are not usually the total form, word-length, etc., but certain striking "dominant" parts, the appreciation of total word- form and word-length coming a Httle later as the recogni- tion is completed at the suggestion of these dominant cues. However, while the experiments of these investigators indicate the special part which the dominant letters and letter-groups play in setting off the word-recognitions,, we need by no means suppose that the former are always or usually apperceived as distinct letters in performing this function of special signs. Through their being the most obvious parts optically, and through habit, they have come to be most quickly operative in unlocking the word-recognitions; but in ordinary reading they would seem to have but a minimum of attention, per- forming their function automatically and without any apperceptive act that is distinct from that for the larger whole in which their recognition is subsumed. When that total recognition completes itself, however, we are no THE PSYCHOLOGY OF READING apt to be conscious of these dominant forms as the mosl prominent parts of the word. With some readers, however, and perhaps with all of us for many words, the total form, word-length, etc., seem to characterize the word and are apparently the first factors in its recognition. In these cases the stimu- lations from all the parts and points which signalize this total form are operative simultaneously as cues which set off the projection of this form, and this general out- line rather than a few particular dominant letter-shapes is the aspect of which we are apt to be most conscious in the total recognition. In such cases the recognition could well be set off by a skeleton drawing of the word showing no particular letter forms, and might well occur at distances at which particular letters were no longer recognizable as such. There is no question but that such perception can occur, for certain words and for certain readers, and that it does occur. But Erd- mann and Dodge have here apparently mistaken what is possible and many times actual for a usual and almost universal method of recognition. Here the testimony of the majority of careful experimenters is against them. As a matter of fact, the outline form of a word is a rather inconstant quantity. For a considerable part of our reading we concern ourselves with written symbols, in which, the word's total form is different, often very different, from reading to reading. Not only do the THE PSYCHOLOGY OF READING III though height and slant of the letters vary, but the spacings between them and, of course, the total length. If the reader will place side by side various printed, written, and type-written forms of the same word, the variations in the outline form will be evident, as in the example below. Of course the letter-forms change as well, and it might be difi&cult to determine whether these or the total form have the greater variation. The constant practice of writing words letter after letter, and the use of the letters in abbreviations, etc., tends to increase the consciousness of single letters as they appear in words, and thus to break up the consciousness of total word-form. Of course, too, the school practice in spelling and the synthetic methods of learning to read contribute strongly to the domi- nance of letter-units in the perception of words. Even in the more pronounced cases of letter con- sciousness, however, it is perfectly certain that words are not perceived by a successive recognition of letter after letter, or even by any simultaneous recognition of all the letters as such. By whatever cues the recognition may be set off, it is certainly a recognition of word-wholes, except when even these recognition units are subsumed under the recognition though Fig. 13. 112 THE PSYCHOLOGY OF READING of a still larger unit. The only question is as to what parts are especially operative as cues in setting off this recognition. Doubtless, for readers who are familiar with the letters, the recognition of word-wholes and of phrase or sentence wholes involves the inhibition of incipient recognitions which start for the letters or other constituent units. There is a hierarchy of recognition habits, the exercise of the higher drafting away the consciousness that would otherwise serve for completing the recognition of the par- ticular letters. Let us examine the case of perceiving a single letter as such, and then that of perceiving the larger and progressively more complex reading units. When a single letter is exposed and recognized as a letter, the simultaneously given stimulations from its various parts mutually reenforce each other, having been associatively knitted together in past experience. Doubt- less even in the case of the letter certain of its parts are more characteristic than others, and thus, having had the attention oftener, become especially effective in touching off the letter's recognition, and may even do so when the more indifferent parts of the letter are absent or are ignored. Doubtless we have dominant parts of letters as of words. The full recognition of the letter doubtless has in it a slight feeling attitude toward it as a total form, and carries with it some notion of the letter's sound and, more distantly perhaps, the letter's name. But now consider the recognition of a familiar word in THE PSYCHOLOGY OF READING II3 which this letter is contained, the observer knowing in ad- vance that a word and not a single letter is to be exposed. In this case there occur as before the coexcitations from the parts of the letter, knitting to an associative complex from habit and quite automatically, and tending to set off a recognition act for this letter. However, there oc- cur simultaneously the other groups of coexcitations from the other letters, each tending to set off its own letter rec- ognition, but tending also, when occurring in this partic- ular combination of letter-groups, to function with them in setting off the recognition of the word. The context "set" for words tips the balance in favor of the unitary recognition of the word, which drafts to itself the energy and consciousness which would otherwise have been given to the letters as such. With very famihar words, the letter recognitions are checked in their incipiency. With new words, the recognition of certain letters may quite com- plete itself before the whole word is known. With the familiar word, as with the letter, certain parts may be dominant parts, being first factors or more effective factors in setting off the word -recognition. And here we may, in part, accept the findings of Goldscheider and Miiller, Zeitler, and Messmer as to what the dominant parts are, and that these are first factors in initiating the word-recognition, for most readers and for most words. That these dominant parts have apperception acts prece- dent to or distinct from the apperception of the words as 114 THE PSYCHOLOGY OF READING wholes must be denied, it seems to me, for actual reading. Such separate apperception acts seem to occur mainly as artifacts of the experiments with short exposures and strained attention. That we may be conscious of the dom- inant parts earlier than of the other parts, inside the total recogr-ition act, seems to be the real fact, and the greater prominence of the dominant parts, in consciousness, is verifiable by introspection. Again, as to perceiving by total form: for some readers, and for all of us in the case of many familiar words, co- excitations from the various parts that make up a word's total outline, independently of their being parts of letters, mutually assist each other in acting as cues to touch off the recognition of the word. When this recognition occurs, we are conscious of the total form, as in the other case we were conscious of the dominant letters, earlier and more prominently than of the constituent letters of the word- form. Here again it seems that total form is not apper- ceived separately, but that, in one act of projection, the total form and the parts to fill it are placed, although this unitary act is not necessarily simultaneous any more than is the act of hitting a target. The visual recognition of a familiar phrase, as a phrase, is but a repetition of the process described above, the rec- ognitions of constitueht words as well as of letters in this case being partially inhibited in favor of the total recog- nition of the larger unit. Total visual form seems to be a THE PSYCHOLOGY OF READING II 5 less important factor in mediating tiie recognitions as the unit grows larger. Unitary recognition of phrases is very common in reading, for mentally the words do not stand entirely apart. The exigencies of printing have brought' about the division on the page of much that belongs to- gether in speech, and again many of our words are logi- cally phrases and might be printed as separate words. The psychological process of apperceiving these words or phrases would not change very greatly if they were printed differently. Very many compounds are written sometimes as separate words, sometimes as two words with a hyphen, again as a single word. Indeed, as we shall see elsewhere, the usual separation even of words upon the page, in Latin and Greek, came very late. In partial disregard, therefore, of the printer's divisions, there is naturally a gradual prog- ress, with practice, toward recognition in larger units, for those who learn first the recognition of letters and words. Larger and larger unitary reactions are set off as familiarity malces this possible, the same excitations coming to serve as cues for the larger recognitions instead of for the smaller, while the earlier processes or recognition habits, even when they do not atrophy, are performed automatically, con- sciousness ever tending to leave them for higher levels. We must remember, however, that there are continual reversions to older habits, consciousness descending to even the level of letter-recognitions, on occasion, and very often taking account of particular words. Here there seem to be Il6 THE PSYCHOLOGY OF READING very great individual differences, and these depend partly, although never wholly, on the methods by which the reader has learned to read. We are brought back to the con- clusion of Goldscheider and Miiller that we read by phrases, words, or letters as we may serve our purpose best. But we see, too, that the reader's acquirement of ease and power in reading comes through increasing ability to read in larger units. We cannot complete our account of visual perception in reading until we have first taken account of the part played by inner speech and by the consciousness of mean- ing. These have an important function in conditioning recognitions in reading. Meaning, indeed, dominates and unitizes the perception of words and phrases, as indeed, according to such writers as Stout at least, it dominates all perceptions. Zeitler's remark will be remembered, that the word's form first gets anchored or established as the sense is filled into it. This appears in perceiving phrases in which words are "seen" which are not there, but which make sense. The excitations from the page act here as cues to a meaning which reacts in the pro- jection of an equivalent expression. As for the inner vocalization of what is read, we shall find this, too, a powerful factor in welding together what is seen, and in keeping it together before the mind's eye until the full meaning dawns. It will be our next task to take account of this important constituent of the reading process. CHAPTER VI THE INNER SPEECH OF READING AND THE MENTAL AND PHYSICAL CHARACTERISTICS OF SPEECH The fact of inner speech forming a part of silent reading has not been disputed, so far as I am aware, by any one who has experimentally investigated the process of reading. Its presence has been established, for most readers, when adequate tests have been made. Its characteristics and functions have been variously described by many writers on general psychology and philosophy and on the psy- chology of language. Purely visual reading is quite possible, theoretically; and Secor, in a study made at Cornell University, found that some readers could read visually while whistling or doing other motor tasks that would hinder inner speech. We might perhaps all have learned a sort of visual reading, and might yet require ourselves to read so in a measure. But although there is an occasional reader in whom the inner speech is not very noticeable, and although it is a foreshortened and incomplete speech in most of us, yet it is perfectly certain that the inner hearing or pronounc- ing, or both, of what is read, is a constituent part of the reading of by far the most of people, as they ordinarily 117 Il8 THE PSYCHOLOGY OF READING and actually read. The evidence is cumulative from many sources, and cannot all be given h(ere, but there is no doubt as to the fact. We shall here consider some of the experi- ments which throw light upon the presence and character of this inner speech. In the writer's own experiments in which single unre- lated words were exposed for four seconds each, the reader to state just what was suggested as he saw each, the words were usually "mentally pronouhced" immediately after or accompanying the recognition of their visual form. When other words or phrases were suggested, as often occurred,,, these were almost always mentally pronounced. The conjunctive and relational words, definitive adjectives, etc., aroused few associations other than verbal ones, the latter usually being phrases of which the words customarily form a part. The inner pronunciation of these words and of the suggested phrases constituted much the most prominent part of the reader's consciousness of them. When sense matter was exposed similarly, giving the reader four seconds for each consecutive word or phrase, the words and phrases were almost always mentally pro- nounced, and usually with a strong feeling that they belonged with a preceding pronunciation or were to be followed by another, or both. In another series of experiments, the readers read equiv- alent pages from an interesting novel, by various methods assigned them. Sometimes the instructions were to read THE PSYCHOLOGY OF READING Iig "the way you like to read"; sometimes they were to "say it all to themselves" ; again they "read aloud" ; then they thought of "how it would sound " as they read ; and some- times they were directed to use lip-movement. Some- times the readings were at the ordinary and most comfort- able speed, and again they would be as fast as possible. The time required for the reading of each page was care- fully taken and will be referred to in discussing the rate of reading. In many cases these experiments brought the readers to an awareness of their inner speech in reading when this had gone unnoticed before. When asked to say the words over to themselves, they found that really this was what they had been doing all along in their ordinary reading, and was the one thing that they could not escape doing when they tried. In such cases, the time re- quired for reading a page in the assigned way would be nearly the same as when the page was read "the way you like to read." In the same way the reader's habit of hear- ing the sound of what he read, or of using lip-movement, etc., was often revealed. Of nearly thirty ^adults who were thus tested, the large majority found inner speech in some form to be a part of their ordinary reading. Purely visual reading was not established for any of the readers, although the test did not show that it was not present for a few. Motorizing with lips closed at the "comfortable" speed gave nearly the same average rate as when the reading was by the reader's "own method," 5.29 words per second I20 THE PSYCHOLOGY OF READING ' for the former and 5.35 for the latter, for twenty readers tested. Of twenty post-graduate students who were tested, but two or three used lip-movement when reading "as they liked." Many of the others who "motorized" said that the pronunciation was "up in the head," and it usually seemed to be without any very noticeable movements of the articulatory apparatus. For the readers tested in these experiments it seemed that the inner speech was a combination of auditory and motor elements, with one or the other predominating ac- cording to the reader's habitual mode of imaging. Some- times when the inner speech was very prominent it was dif&cult for the reader to say whether it was auditory or motor, although it seemed to him to be of but one kind. The fact is that what we say is always heard as well, and there comes to be an indissoluble union of the auditory and motor elements. Our hearing, too, has an active aspect which may go so far as to include an inner saying, or imitation, of what is heard. And so the auditory and motor types of readers are really apt to be audito-motor types, with one or the other aspect leading in many cases. ;| That the speech of silent reading is simpler than in reading aloud is indicated by the fact that it was faster for each of the readers tested, both when the reading was at normal speed and when it was as fast as possible. Read^ ing aloud was 66 per cent slower than reading silently, at the normal rate, and 56 per cent slower at the maximal THE PSYCHOLOGY OF READING 121 rate, on an average for twenty readers. In reading aloud, as in talking, generally, the words are practically all pronounced as the breath is expired, and there are pauses at the inspirations. The inner speech of reading, on the other hand, goes on during inspiration as well, and thus time is saved. Then, as Professor Dodge shows in his "Die Motorische Wortvorstellungen," the inner articu- lations do not call into play the chest and larynx muscles that are used in speaking aloud, and there is a shortening of the pronunciation, a slurring of the words, and in- deed the omission of many for some readers. Professor Dodge states that in his own ordinary silent reading almost every word is pronounced, but that in his fastest silent reading only the beginnings of words are pronounced. In his fastest reading of very familiar matter only cer- tain words were pronounced. His speed of reading seems to be determined by the speed with which his mo- tor word-ideas can follow one another. He finds, as I have found, that auditory elements are present in the reading of those who motorize, 'and that those who audit- ize are apt to have more or less of the motor present. Zeitler and Messmer did not investigate the inner speech of reading, but Messmer asserts that "in visual reading the auditory and motor centres work along," and that purely visual reading is normally not to be found. Quantz found lip-movement, and consequently inner speech, to be universal in the early reading of children. 122 THE PSYCHOLOGY OF READING The lip-movement decreases with practice and usually, although not always, disappears in the rapid and more in- telligent readers. He found that "lip-movement in silent reading is not an acquired habit, but a reflex action, the physical tendency to which is inherited." "It is a specific manifestation of the general psycho-physical law of dyna- mogenesis by which every mental state tends to express it- self in muscular movement." Reading without lip-move- ment is "an acquired habit," the natural thing being to use the lips, as almost all of us do in practice when we come to a difficult place requiring close attention. . My own observations indicate that the disappearance of the lip-movement is no indication of the absence of inner speech in reading. In my own case, the lips are seldom moved, but I can never escape the inner pronun- ciation that forms a part of all my reading. It would be easy to quote authority almost endlessly for the presence of inner speech in reading and, indeed, in most thinking. The simple fact is that the inner saying or hearing of what is read seems to be the core of ordinary reading, the "thing in itself," so far as there is such a part of such a complex process. It is so in all use of language. The spoken language is the language par excellence, as Professor Whitney says in his "Life and Growth of Lan- guage," "gesture and writing being its subordinates and auxiliaries." The child comes to his first reader with his habits of spoken language fairly well formed, and these THE PSYCHOLOGY OF READING I23 habits grow more deeply set with every year. His mean- ings inhere in this spoken language and belong but sec- ondarily to the printed symbols; and always, for most readers, we can say with M. Egger that "to read is, in effect, to translate writing into speech." And while this inner speech is but an abbreviated and reduced form of the speech of everyday life, a shadow copy as it were, it nevertheless retains the essential characteristics of the original. In order, therefore, to understand the inner speech of reading and its relation to the interpretative pro- cesses, it will be necessary to examine briefly the nature of speech generally, and its relation to thought and meaning. Language begins with the sentence, and this is the unit of language everywhere. A sentence is a unitary expres- sion of a thought. A thought may be expressed in a word sometimes, and this is then a sentence- word, as when Preyer's child put his milk cup down quickly and said "Hot!" This single word was to signify, "This drink is too hot." It was "a whole proposition in a syllable,", as Preyer says. The meaning that might have suggested "This drink is too hot" and that would have bathed every part of it as spoken, suggested only " Hot " as its expression, with this child who yet knew but httle of language. But there was more than the articulated word. The pitch, accent, modulation of voice, which characterize a sentence- word's pronunciation, are important factors in expressing the particular unitary meaning that is felt. "Papa" may 124 THE PSYCHOLOGY OF READING mean "Come here, papa," "Look out, papa," "Please do, papa," according to variations in the tone, etc., and according to the situation context. Often the modulation, accent, or rhythm are more expressive of the speaker's meaning than are the words as such, and the former factors belong to the sentence as a whole. The child, the primitive man, and indeed any speaker, when he would form a sentence, begins with a meaning, a total idea, as Wundt calls it, which he would express. This total idea is at first little differentiated and may find expression in a gesture, a tone, or a word, as when the earnestly spoken "Hot" was all that came. With more experience this total idea or consciousness situation, of being burned with milk, falls apart somewhat into the sub- notions milk — this — hot — drink, and gets a correspond- ingly analyzed expression in these several words, these expressing still, however, one unitary meaning. The part of the total idea that is most prominent in consciousness is apt to be expressed first, and in child speech, as with many primitive tribes, the words may come in any order according as the various aspects of the total idea suc- cessively become prominent in the speaker's mind. The child's language, however, is not an invention, but is learned by imitation ; and he accepts, in EngHsh, certain fixed habits of breaking up the total ideas into parts that are expressed in parts of speech, such as nouns, verbs, adjectives, etc. Certain habitual sequences of these parts THE PSYCHOLOGY OF READING 125 are also learned by imitation and his words come to fall into this habitual order. Thus, beginning with a total meaning and a total intention of expressing this meaning, the development is toward a more and more particular division of it into aspects or parts, and toward the expres- sion of these parts in words that are arranged in gram- matical sequence, this arranging and indeed the whole development being largely automatic, the result of asso- ciative habits learned gradually by experience and by imitation. But meaning leads, and the idea of the whole dominates the parts. The sentence is not naturally com- posed of words which originally existed independently, just as we shall find that the word is not a mere collection of syllables and letters. In the ancient languages a single expression would often be word and sentence together. Wundt says in his "Volker-Psychologie " ': "The Latin amavi is both word and sentence. The Romance languages resolve this thought into three words, ego habeo amatum, j'ai aime. Accordingly, if we compare, on the one hand, languages of an evidently more primitive development with those that are more developed, and if on the other hand we compare the earlier with the later stages of one and the same lan- guage, the differentiation of the parts of speech every- where shows itself as the process of gradually resolving the word out of the whole to which it belongs ; namely, the 'Vol. I, p. 561. 126 THE PSYCHOLOGY OF READING sentence, — the process which lends the word a relatively greater independence and fixes its grammatical form at the same time with its independent significance." There exist to-day languages in which the sentences are spoken with- out differentiation of either words or parts of speech, in a continuum of syllable sounds, or it might be said that the sentence is one long word. Our English and the kindred languages have made the analysis into parts of speech, words, etc., and our fashion of printing has made us very conscious of the results of this analysis. But in the living speech of conversation and thought these parts still inhere organically in the original sentence-wholes, and the actual structure is very different from the written or printed ex- pression, as we shall presently see. \ Genetically, then, as we might go on to show, the growth of living speech both in the race and in the child has been from the protoplasm of total meanings expressed in sentence-wholes, through a progressive anal- ysis to parts of speech and words, then to syllables and to elementary sounds. We shall later trace the anal- ogous development of the written and printed characters from primitive picture- wholes to characters representative of word-meanings, word-sounds, and finally to symbols for syllables and for elementary sounds. Let us now look more nearly at the processes that go on as we speak our English sentences to-day. In the first place it is certam that in ordinary speech THE PSYCHOLOGY OF READING 1 27 some thought of the whole sentence pervades every part as the part is spoken, and the part is felt in a perspective of the whole. This is true of the sentence's begirming as well as of its other parts, and some consciousness of the whole usually precedes even the initial utterance. Says Wundt in the volume mentioned (p. 563): "At the mo- ment in which I begin a sentence the whole of it stands already in my consciousness as a total idea." Wundt adds, however, that the sentence is then felt only in its main outlines, its constituent parts being at first dark, but coming out as the speaking goes on. "The process," he says, "is something like the sudden lighting of a complex picture, where one at first has only a general impression of the whole, and then successively of the particular parts, always seen in their relations to the whole." Only thus, Wundt thinks, can we explain the fact of a speaker going correctly through with a complex sentence without having reflected on it before. The total idea of what is to be said thus exists in consciousness precedent to the utterance, and dominates the utterance throughout. This total idea is not a mere sum of associations, but is an apperceptive unity. This unity becomes differentiated in the manner and in the direction indicated in its sentence expression, and the sentence is, according to Wundt, "the analysis into its parts of a whole that is present in consciousness." Ac- cordingly, he says, sentence formation is analytical, as it 128 THE PSYCHOLOGY OF READING I is a separation of the parts of a whole, but it is also syn* thetic in that it is an appearance of part after part in the focus of consciousness. "Above all, however," he adds, "it is an analytical process." ' Again, Wundt considers the sentence to be a "voluntary act," willed as a whole. True, it is a complex act, but the constituent movements of articulation, etc., go off auto- matically like the constituent mbvements of any other unitary performance. We "give the direction to the thought" and "the requisite words stream to us of them- selves"; that is, "they are awakened associatively from the first-excited word-ideas under the influence of the total idea that is present." "Psychologically considered the sentence is therefore at the same time both a simultaneous and a successive whole — a simultaneous since in every moment of its for- mation it is in consciousness in its entire extent, although particular secondary elements may occasionally disappear from this ; a successive, since the whole changes from mo- ment to moment in its consciousness content while definite ideas one after the other appear in the focus and the others grow darker." ^ Professor James bases his psychology of the sentence on his view of consciousness as a continuous stream of pro- cesses in which "breaks are produced by sudden contrasts in the quality of the successive segments." Conscious- ' "Volker Psychologie," Vol. II, p. 236. = jj^^^ Vol. II, p. 236 ff. THE PSYCHOLOGY OF READING 1 29 ness makes us aware of things, and things, being discrete and discontinuous, "pass before us in a train or chain, making often explosive appearances and rending each other in twain." But these do not break the f,(yw oj the thought that thinks them, with its continuum of feelings, bodily sensations, etc. There are, however, the apparent breaks, which are really "transitive places," places of rapid flow, of flight from the perch of one substantive resting-place to that of another, — to a conclusion perhaps or to a place in which the thought may bask in sensorial imagery. The consciousness life is like a bird's life, made up of an alter- nation of flights and perchings. " The rhythm of language expresses this, where every thought is expressed in a sen- tence and every sentence closed by a period." The places of flight are filled with thoughts of relations, static or dynamic, obtaining between the matters con- templated in the periods of comparative rest. Our thoughts and sentences are largely made up of these fugi- tive transitional processes, and are thus extremely hard to introspect. To attempt cutting such a sentence in the middle to get a look at it is, in James' figure, like catching a snow-flake crystal in the warm hand. The flake is no longer a crystal, but a drop. "So, instead of catching the feeling of relation moving to its term, we find we have caught some substantive thing, usually the last word we were pronouncing, statically taken, and with its function, tendency, and particular meaning in the sentence quite 130 THE PSYCHOLOGY OF READING evaporated." It is like "seizing a spinning top to catch its motion, or trying to turn up the gas quickly enough to see how the darkness looks." To these feelings of rela- tion, expressed or named by such words as and, if, but, and the like, far oftener not named but felt as the sentence moves forward, we will return when discussing interpre- tative processes in reading. According to Professor James, the speaker has "an intention of saying a thing before he has said it" "an entirely definite intention distinct from all other interitions, an absolutely distinct state of consciousness therefore, but with little sensorial imagery, — that welcomes right words as they come and rejects wrong ones." " One must admit that a good third of our psychic life consists in these rapid premonitory perspective views of schemes of thought not yet articulate. How comes it about that a man reading something aloud for the first time is able im- mediately to emphasize all his words aright, unless from the very first he have a sense of at least the form of the sen- tence yet to come, which sense is fused with his conscious- ness of the present word, and modifies its emphasis in his mind so as to make him give it the proper accent as he utters it ? Emphasis of this kind is almost altogether a matter of grammatical construction. If we read no more, we expect presently to come upon a than; if we read how- ever at the outset of a sentence, it is a yet, a still, or a never- theless, that we expect. A noun in a certain position THE PSYCHOLOGY OF READING 131 demands a verb in a certain mood and number, in another position it expects a relative pronoun. Adjectives call for nouns, verbs for adverbs, etc. And this foreboding of the coming grammatical scheme combined with each successive uttered word is so practically accurate that a reader incapable of understanding four ideas of the book he is reading aloud can nevertheless read it with the most delicately modulated expression of intelligence." ' In regarding the sentence as a total unity felt throughout as each part is uttered, and indeed in a measure existing in consciousness precedent to any utterance, James is in agreement with Wundt. "Even before we have opened our mouths to speak," James says, "the entire thought is present to our mind in the form of an intention to utter that sentence." Again, "after the last word of the sen- tence is spoken, all will admit that we again think its entire content as we inwardly realize its completed deliverance." The Pack of Cards is on the Table. Fig. 14 James' diagram, reproduced here, represents the prog- ress of consciousness throughout the utterance of such a sentence as, "The pack of cards is on the table," '"Psychology," Vol. I, pp. 253-254. 132 THE PSYCHOLOGY OF READING the horizontal line representing time, the spaces above representing the consciousness content during this time. Not only is the total thought of the sentence present at its beginning and at its end, but "all vertical sections made through any other parts of the diagram will be respectively filled with other ways of feeling the sentence's meaning,. Through 2, for example, the cards will be the part of the object most emphatically present to the mind ; through 4, the table. The stream is made higher in the drawing at its end than at its beginning, because the final way of feeling the content is fuller and richer than the initial way. As Joubert says, ' we only know just what we meant to say, after we have said it.' And as M. V. Egger remarks, 'before speaking one barely knows what one intends to say, but afterward one is filled with admiration and surprise at having said and thought it so well." The same object (of thought or total idea is "known everywhere now from the point of view, if we may so call it, of this word, now from the point of view of that. And in our feeling of each word there chimes an echo or foretaste of every other." f The total idea is " the overtone, halo, or fringe of the word as spoken in that sentence. It is never absent ; no word in an understood sentence comes to consciousness as a mere noise. We feel its meaning as it passes ; and although our object differs from one moment to another as to its verbal kernel or nucleus, yet it is similar THE PSYCHOLOGY OF READING I33 tliroughout the entire segment of the stream." Cut a thought in cross-section at any moment of its utter- ance, and "you will find, not the bald word in process of utterance, but that word suffused with the whole idea." ' I have quoted at length the words and figures of Professor James, as giving the most nearly correct and most graphic view that is yet obtainable of the organic unity of the sentence with its subordinated word-parts, on the side of consciousness and meaning. On the side of physical utterance the articulatory processes have been carefully analyzed and described, and here, too, the organic unity of the sentence-utterance and the subordination of the partial processes to this whole have been estabhshed beyond a doubt. The painstaking analysis made at Yale Uni- versity, by Dr. Wallin and Professor Scripture, of sentences spoken into a graphophone arranged for the purpose, have given us an especially accurate account of what we do when we talk, and the studies of various eminent philolo- gists corroborate their results. We shall therefore find it profitable to review the facts concerning speech as a me- chanical process. The utterance of even a single word requires the co- ordinated action of three distinct and very different groups of muscles. The breath is forced outward and regulated by the action of the large muscles of respiration of the chest ' " Psychology," Vol. I, p. 243 ff. 134 THE PSYCHOLOGY OF READING and abdomen. The sound is produced by the action oi the muscles of the larynx, tensing and controlling the two vocal cords or bands between whose edges the air passes. The particular character which the sound has when uttered as vowel or consonant, the articulation as it is called, is produced by the action of the muscles of the tongue and jaw and palate. Indeed, most of the other muscles of the body are involved in the movements of expression, ges- ture, posture, etc., which always form a constituent part of speech as it is actually spoken. Professor Scripture, in his "Elements of Experimental Phonetics," finds that this complex machinery for vocal utterance is in continuous action throughout the utterance of any word or phrase, with no interruptions such as letters, syllables, or even words suggest. "The word manly represents continuous action of the breath organs, con- tinuous action of the vocal cords, with a smooth rise and fall of pitch, continuous movement of the lips, tongue, and velum, through various positions." " In fact, the word is a continuous sound change, with no limits or minima of any noticeable kind." The word is to be considered as a fusion of a series of continuous changes, certain stages of which may be characterized as "m-n-l-i, etc. As far as the vocal movements are concerned the word is just as continuous as manned." "I do not believe a division of the flow of speech into separate blocks termed syllables has the slightest justification or the slightest phonetic THE PSYCHOLOGY OF READING I35 meaning." "A word is a continuous series of an infinite number of sounds, and the letters indicate, in an incomplete fashion, nothing more than certain characteristic points of this series." Even single vowel sounds were found to vary constantly in pitch, in talking though not in singing. It is true that, for the listener, there are really brief in- terruptions or moments of silence, not noticed as such, however; but "the motor activity of the speech organs goes on just as vigorously during the occlusion (silence) as before and after." The lips, tongue, etc., do not assume fixed positions at any moment, and "it seems therefore somewhat artificial to divide the words who'll, " e.g., " into three or five sounds; we may preferably say that for the sake of discussion five stages in the changing sound may be picked out as typical of the whole process." Professor Scripture likens this to taking kinetoscope pictures of a runner, treating his whole movement as a series of positions in which the runner remains at rest. " This treatment has its advantages for certain cases, but we should never lose sight of the fact that the true movement occurs otherwise." Not only is ,the utterance of a single word found to be such a unitary and indivisible act, but the experimenters find that words are not separated as they are spoken in sen- tences. Indeed, Professor Scripture asserts that "speech is a flow of auditory and motor energy with no possibility of division into separate blocks such as letters, syllables, 136 THE PSYCHOLOGY OF READING words, feet, etc., except in a purely arbitrary manner that does not represent the actual case." * Professor Sweet, the English philologist, in his "Primer of Phonetics" says: "The only division actually made in language is that into 'breath-groups.' We are unable to utter more than a certain number of sounds in succession without renewing the stock of air in our lungs. These breath-groups correspond partially to the logical division into sentences ; every sentence is necessarily a breath-group,^ but every breath-group need not be a complete sentence. " Within each breath-group there is no pause whatever, notwithstanding the popular idea that we make a pause between every two words. Thus in such a sentence as put on your hat, we hear clearly the 'recoil' or final breath- glide which follows the final t of hat, but the t of put runs on to the following vowel without any recoil, exactly as in the single word putting. In put back there is no glide at all after the t." The inhalation pauses afford momentary opportunity for rest, to both speaker and listener. The average num- ber of syllables uttered between pauses was found by Wallin to be about six, for reading prose, with a "range" of 1.85 syllables. However, on an average about thirty-two times this average number of syllables could be uttered at one exhalation. Wallin found that in impassioned speech this maximum was most nearly reached. '"Experimental Phonetics," pp. 279-593. THE PSYCHOLOGY OF READING 1 37 But if the spoken sentence is, except for inhalation pauses, an unbroken continuum as uttered, it is by no means an even or monotonous continuum, in natural living speech. There is a continual rise and fall of pitch and energy and a variation in quantity, producing a rhythm and melody of speech that is as characteristic and con- stant for prose as for verse, but not so marked. Indeed, when Dr. Wallin had persons read prose printed as poetry and poetry printed as prose, in a large number of cases they were unable to distinguish the one from the other. The poems of Browning and Tennyson were called prose and the prose of Bacon was called poetry. Certain variations in force or stress, in the continuous flow of speech, give the effect, as Sweet says, of its being broken up into syllables, even when there is not the slight- est pause. The vowels are the more sonorous sounds and thus help to make the syllable divisions. A vowel is " voiced breath, modified by some definite configuration of the superglottal passages, but without audible friction which would make it a consonant." In other words, "a vowel is voice modified by a resonance chamber, to wit the mouth." "Consonants," Sweet continues, "are the result of audible friction or stopping of the breath in some part of the mouth or throat. The main distinc- tion between vowels and consonants is that while in vowels the mouth configuration merely modifies the vocalized breath — which is therefore an essential element of them 138 THE PSYCHOLOGY OF READING — in consonants the narrowing or stopping of the mouth passage is the foundation of the sound, and the state of the glottis is something secondary." "Consonants can therefore be breathed as well as voiced, the mouth con- figuration alone being enough to produce a distinction without the help of voice. All consonants can be whis- pered." Between the consonants and the vowels are glides, — "transitional sounds, produced during the transition from one sound to another," the written letters repre- senting, as we have seen, but a portion of the sounds actually uttered. So the ear "learns to divide a breath-group into groups of vowels (or vowel equivalents, as the sonorous / in cattle), each flanked by consonants (or consonant equiva- lents)," giving a division into syllables. Sometimes variations in stress may suggest word-division, but often, and regularly in French, there is no such thing as word-stress or word-division. In French, says Sweet, the "sentences are cut up into syllables without any re- gard to the structure of the words they are made up of." ' The sound of any letter, therefore, and the nlovements necessary to prpduce it, depend partly upon the context in which the letter stands. The sound that occurs in context cannot usually be given in isolation, and in any case the vocal movements used for its isolated production are different, often very different, from the corresponding ' Sweet's " Primer of Phonetics," selections from various parts. THE PSYCHOLOGY OF READING I39 movements in the continuous utterance of a word or sen- tence. Similarly, a word sounded alone is somewhat different in sound and in the vocal movements required than when spoken as part of a sentence ; and a sentence whose words are uttered each for itself has, as we all know, a sound that is very different from the sound of a sentence uttered as such. Indeed the action in uttering a sentence is of the same kind as in skating, dancing, throwing a ball, or any such unitary complex of muscular movements. In throwing a ball, for instance, there is the sub-movement of grasp- ing the ball, itself quite complex; there is the raising and poising of the arm, with all the coordination needed for this performance, and there is the complicated final act of throwing. Now the grasping and poising and throw- ing might each be performed separately, with full atten- tion directed to itself. But who thinks of these separate movements as he aims and throws? The meaning of the total act, viz. the hitting of the mark, guides and controls and unitizes all, and renders each movement different from what it would be alone. Each subordinate movement is made in a perspective of the whole, just as, in perceiving, the consciousness of the part is profoundly modified by the consciousness of the whole. We have already seen how meaning welds the parts of a sentence into unity. We see now how the necessities of physical utterance contribute to the same end. 140 THE PSYCHOLOGY OF READING A main force tending to unitize the sentence both physically and psychically is the alternation of stresses, the rise and fall of pitch, and the variations in quantity, already referred to as factors in producing the rhythm, melody, and emphasis of speech. The total sentence- meaning inheres vitally and comes most to expression •in these characteristic variations in the sentence-flow. They are the hfe of the sentence, and give it its character- stamp. And these variations belong to the sentence as a whole, those made in one part having reference to those made in another, and all having a unitary significance. The written or printed representations of speech make no record of these its vital parts, except for an occasional use of italics, underscoring, or mark of accent. They fail, as well, to record the transitional sounds between letters and words, and the variations in letter-sounds according to variations in context. But the expressional variations, with all these others, come to life again when the written or printed characters are rendered into living speech in read- ing. In actual reading the rise and fall of pitch and inflec-' tion, the hurrying here and slowing there, what we have called the melody of speech, appears in the inner speech even more prominently than does the articulation of the particular sounds, wherever there is appreciation of the meaning of what is read. The inner saying of many a word, in rapid reading, is but a slurred remnant of its full soimd, a motor tally as it were. But it is a tally that THE PSYCHOLOGY OF READING I4I has its place and its time in the inner rhythmic sequence. Certain words may be omitted entirely, but they are usually words of relative unimportance, or they are parts which would be unstressed in ordinary speech and whose omission will not affect the natural swing of the sentence. The expression shell or form is the most persistent and essential part of the inner speech, certainly in my own case at least. I have thought it best to discuss rather fully the psychic and physical characteristics of speech and the relation of speech to meanings, both because I find these processes of speech, in their main lines, essential to reading as well, and because the pedagogy of phonetics and of reading methods generally needs the perspective which such a survey of the psycho-physics of speech can give, and needs it in this accessible form. We shall now, I hope, be better able to understand the part played by inner speech in the apperception of the printed sentence, and shall take up that discussion where we left it on an earlier page. CHAPTER VII THE FUNCTIONING OF INNER SPEECH IN THE PERCEPTION OF WHAT IS READ In so far as the cues from the printed page set off the recognition of phrases and sentences as wholes, they do this the more readily for the habits of sequence and ex- pectation which exist among words according to gram- matical and logical usages. We recall Professor James' account of how, in our English, the verb tends to follow its subject and precede its object, how the preposition tends to be followed by its related substantive, etc. Our words are thoroughly organized according to these general associative habits of our language, and when any given series has occurred in our reading, the sort of words and the sentence forms that belong in sequence with these are subexcited in advance of their appearance on the page, and need but slight cues from the page to cause them to spring into the perceptual consciousness. Indeed, hundreds of phrases and sentences have occurred so often in our speech that they have a place in mind as specific memory-wholes ; and as slight a glimpse is needed to start the recognition of these as when the tap of a cane 143 THE PSYCHOLOGY OF READING 143 suffices to announce the approach of our grandfather. But though the phrase or sentence has never occurred in reading before, in exactly its present wording, the inner readiness for it is ahnost as complete ; and it will inwardly complete itself almost as readily from a few visual cues, if it is cast in a familiar form, if its words and parts of speech stand in famihar grammatical sequences, so that each associatively helps the other to rise. and remain in con- sciousness. It is unitarily perceived quite as truly as if it existed as a specific memory-whole. All this would be true if the mechanism of perception in reading were purely visual, without the help of the inner speech. But the habits of inter-association and expectancy, which bind the units of our language into wholes that are ready to realize themselves when but a few of the constituent or context parts are suggested, are far most deeply founded in the audito-motor mechan- ism of speech. People ordinarily talk far more than they read, and the motor conditions which obtain in talking are much more conducive to the formation of lasting habits than those which obtain in reading. And even while peo- ple read, the talking habits are doubtless getting more efiFective practice in the inner speech than are the visual habits in visual perception, for all but persons who are strongly visual-minded. Even in silent thinking, clothed in a sort of language as the process always is, the organ- ization of our speech habits goes on perfecting itself, and 144 THE PSYCHOLOGY OF READING here again the more stably knitted if not indeed the more prominent language terms are motor and auditory. While, then, the purely visual habits of inter-associa- tion and expectancy play an important part in enlarging our perceptive range and in unitizing our phrases and sentences, the visual range is itself enlarged and its con- tent supported by the more stably organized inner utter- ance into which the visual percepts are constantly being translated. The carrying range or span of the inner speech is considerably larger than that of vision. No satisfactory measurements have been made of the amount of new "sense" matter that can be held in control at once by the psycho-physical mechanism of speech; but at least a couple of ordinary lines can be so held, and so perfectly that the relative stresses and melody of the original utterance can be faithfully reproduced. Indeed the rhythm and melody, by their binding the sentence together, become important factors in extending the Sprach Umfang, and the span is greater as the matter read is more rhythmical and more melodious in its com- position. But of prime importance, of course, in making possible a large range of inner speech in reading is the existence of the inter-association habits which we have been discussing, and the range is larger as the reading matter follows more closely the associative habits of the language. We remember, of course, through the form- ation of associative links, and if these are ready formed THE PSYCHOLOGY OF READING I45 for US in what is read, we can naturally carry more of it in memory at a single reading. The inner utterance doubtless begins as the words are visually recognized, but the full utterance of the phrases and sentences as such follows at a considerable distance behind the eye, a variable distance that is greater as the reading is faster, but depends also on other factors than rate. The reader may observe this separation of eye and voice as he turns his eye to a new page, several words of the preceding page usually remaining to be uttered. Dr. Quantz measured the amount of this eye-voice separation by slipping a card over a reader's page, at certain places predetermined by the experimenter but not known to the reader, and recording the number of words spoken after the view was cut off. He found that much depended upon where in the line the view was intercepted. "When the reader is pronouncing a word at the beginning of a line, the eye is on an average 7.4 words in advance of the voice; in the middle, 5.1 words; and at the end, 3.8, giving an average of 5.4 words." The space between is thus "very elastic," as he says, "expanding and contracting with each line, but with a uni- form regularity — except indeed where special conditions are introduced ; an unfamiliar word, for instance, would decrease the distance to zero, or a familiar phrase might increase it to a dozen words. After the long pause which a period allows, the eye lengthens its lead of the voice." 'X46 THE PSYCHOLOGY OF READING Quantz found a close correlation between the increase of eye-voice separation and the increase of rate in reading, and states that the "rapid silent readers read farthest ahead of the voice in reading aloud," and thinks that "a certain considerable distance between eye and voice is a condition of intelligent and intelligible reading." He also made some indecisive experiments to determine whether the consciousness of meaning goes ahead with the eye, and states it as probable that it does not. There seems to be a phase of inner utterance which follows the eye much more closely than this, but it is not the meaningful utterance of the sentence. It is true that further investigation has not justified Goldscheider and Miiller's conclusion that the letter-sounds are im- mediately suggested by the visual letter-forms, in ordinary reading. The word-sound seems usually to be suggested as a whole. But this sound of the word seems to be dimly suggested immediately accompanying or following the word's visual appearance. This initial motorization, present in my own case at least, is of the words in isola- tion, is accompanied by a slight feehng of the word's meaning, and seems to help hold the word in conscious- ness until enough others are given to combine with it in touching off the unitary utterance of the sentence which they form, with the total sentence meaning which dom- inates the inner utterance of this sentence as a sentence. The full inner utterance seems, then, to be suggested THE PSYCHOLOGY OF READING I47 partly by the individual word-sounds serving as cues, but the visual forms doubtless also serve directly as such cues. I have no doubt that the stability of the visual forms and their persistence in memory depend consider- ably on both the initial and final inner utterance. The full inner utterance thus "hangs fire," as Quantz shows, behind the eye until there are present enough visual and motor data to suggest the total meaning and the corresponding sentence expression, and this utter- ance then occurs with an inter-dependence of meaning and word-expression such as Professors Wundt and James have described for speech in general. There can be little doubt that the main meaning comes to conscious- ness only with the beginning of the sentence-utterance, and the reader does not feel that he has the complete sense imtil he has spoken it. He is almost sure to de- liberately say the passage over to himself if it is difficult, and persons who do not read very much must usually use an actual whisper, even in easy reading, if the meaning is to be obtained. That the general meaning dawns upon the reader precedent to the full sentence-utterance is evidenced by the many cases in which variant words of equivalent meaning are read, and also by the comparative ease with which a reader may paraphrase the thought of what he reads. This is especially noticeable in the case of a person reading a foreign language which he does not 148 THE PSYCHOLOGY OF READING pronounce easily but which he comprehends rather rapidly. Here the visual word and phrase percepts touch off total meanings which clothe themselves, as the meanings become articulate, in English sentences, and we have as a result the mongrel reading which passes for French or German in so many modem language classes. It is of the greatest service to the reader or listener that at each moment a considerable amount of what is be- ing read should hang suspended in the primary memory of the inner speech. It is doubtless true that without something of this there could be no comprehension of speech at all. When a considerable amount is thus suspended, the attention may wander backward and forward to get a fuller meaning where this is needed, with no fear of losing the minor parts, which are taken care of physiologically and may be taken into the focus of consciousness at will. Any careful introspection of actual reading will show that the main focus of atten- tion is often far behind the eye, concerned perhaps with the soimd of some word or phrase that is giving difficulty; and we know that the entire process of visually perceiv- ing and inwardly pronouncing may go on, for even an entire paragraph sometimes, with but very little of even marginal consciousness, the attention being absorbed in some thought suggested earlier, or perhaps in some irrelevant imagery. Indeed, any part of the reading THE PSYCHOLOGY OF READING 1 49 process or the whole of it may proceed automatically with but a minimum of consciousness, just as in walking, dancing, or other complex motor activity requiring adap- tive reactions to stimuli. The attention, the concern of the self about what is going on, may be here or there as there is need of it, and again is often centered where there is no need of it. Some prominent letter or other form in a preceding or succeeding line may flash into the focus more prominently than the advancing stream of visual forms that the eye is just revealing. The move- ment of the attention may thus be backward or forward, but of course is usually forward. Doubtless its actual advance has httle reference to the sequence of eye-move- ments and pauses. These are ordinarily indistinguish- able to consciousness, and the attention has to do with an unbroken line. The crest of the advancing con- sciousness seems often to be double or even quadruple, composed of visual, motor, or any other content with which the reader is concerned. In so far as the attention is upon the visual forms, it is of course apt to note pref- erably the larger, the more characteristic, or the more meaningful of the letters, letter-groups, and word-shapes that appear. It is thus apt to be especially concerned with the so-called determining or dominant forms. But I consider that the better differentia of these latter forms is their greater effectiveness in neurally conditioning the perceptual reactions. ISO THE PSYCHOLOGY OF READING There has been much controversy as to whether there is a movement of the attention during a reading pause. ' Much of this controversy concerns the technique of making the brief exposures of reading matter in the various experiments, and would be dif&cult to make intelligible or profitable to the general reader, in any brief compass. The behavior of the attention in such experiments must at any rate be very different from what occurs in the pauses of reading. And as to the proper length of the exposure, it is fovmd in measuring the amounts that can be read during a reading-pause that exposure-times differing most widely and conditions of lighting, etc., that are most variant give much the same re- sults. As to the question itself of whether the dawning of the visual forms in . consciousness occurs simultaneously or successively over the given areas, careful introspection of actual reading, probably the sanest though still an imperfect test, indicates that there is a successive coming to consciousness of the more striking forms at least. Doubtless the "forward push" of associative expectancy, |) on its visual side and supported, perhaps, by its audito- motor phase, would tend to pick up the prospective and just-appearing forms in a constant succession. Doubt- less, too, these same conditions bring about the simul- taneous appearance of many forms that go to make up expected phrases, etc.; but the general" effect must be to bridge the chasm, if there is any to bridge, between the THE PSYCHOLOGY OF READING 151 reading pauses, and to give a constantly but not uni- formly forward movement of the visual consciousness. Neurally, as I have elsewhere urged, it would seem that the cues from the various parts of the reading field work simultaneously, though doubtless with very un- equal effectiveness, in conditioning the perceptual reaction. It may be, indeed, that here too we have to do with suc- cessive functionings as well as simultaneous, but the experiments have not proved that the stimulations are effective first at the left, or in succession. The auto- matic functioning of these neural factors and habits, silent but effective workers behind the "stage-effects" which they arouse in consciousness, is worthy of a greater share of the attention in the discussion concerning how we perceive. Dr. F. Schumann illustrated this last year in his criticism of Erdmann and Dodge, rightly urging that when these experimenters read words at a dis- tance 'at which the particular letters could not be made out, the letters might still be efficient factors, as his own experiments seemed to show, in neurally conditioning the appearance of the total form. Thus it will be seen that this experiment leaves unsettled the question as to whether total form or letter-shapes or still other factors are mainly operative in arousing the perceptive reaction. The experiment deals with an effect and does not discover the cause. CHAPTER Vm THE INTERPRETATION OF WHAT IS READ, AND THE NATURE OF MEANING We shall now proceed to examine the processes con- cerned in the interpretation of what is read. How do we get the meaning of our reading, and in what does meaning consist? Our study of the relationship of speech to meaning has prepared the way for an under- standing of this vital part of the reading process, but we have not yet considered the part that is played by imagery. A series of experiments made by the writer gave data which throw some light upon this problem. The experiments were arranged as follows: Two printed articles of considerable interest and but moderate diffi- culty were selected, one an account of how a spider spins its web, the other a description of the arrangements made for the entertainment of Queen Elizabeth at Killing- worth Castle. The words of both articles were first pasted singly each on a square of cardboard, were shuf- fled, and were then exposed to readers for four seconds each, the reader looking at the word exposed and allowing associations to play about it as they would, reporting what had occurred as soon as the four seconds had passed. IS* THE PSYCHOLOGY OF READING i 1 53 Again, the lines of both articles were pasted consecu- tively end to end, on narrow slips of cardboard, and the successive words and phrases were exposed to readers, an additional word or phrase being presented at each exposure of four seconds, the preceding context being always in view as it is in ordinary reading. The reader again reported his associations, etc., as for the isolated words. The two series of experiments were separated in time sufficiently to prevent the memories of one from interfering seriously with those of the other. With the isolated words there was first, usually, an indefinable recognition of the visual form of the word as familiar. Accompanying or very closely following this, probably the latter though the readers could not be explicit, the word was usually "mentally pronounced." After this there was apt to come a mental pronimciation of some phrase or other word that had been associatively connected, as when by gave Sweet by-and-by, vertical gave vertical writing, etc. Often there would be but a dim suggestion of some familiar line of poetry, leaving the reader with a vague and tantalizing feeling of some- thing which he could not get. Quite often, especially with one of the readers, the characteristic feeling that belonged to such a suggested word or phrase was all that came, but this was vividly aroused. The words exposed tended to call up word-groups with which they had been rhythmically connected. Often the word that was exposed 154 THE PSYCHOLOGY OF READING was pronounced "interesting," "agreeable," "full of meaning," efc, and occasionally these judgments seemed to refer to the soimd or visual appearance of the word it- self. More usually the feeling seemed traceable to some particular associations or uses of the word in past experi- ence, though these latter would not appear above the threshold. As to imagery, there was almost none suggested by . the connective and relational words, the definite ad- jectives, etc., "the little words," as my readers called them. Because of this absence of imagery the exposure of these words was regarded with much displeasure, their isolated appearance seeming to be regarded as anomalous. They seldom aroused any ideas directly, and suggested few associations of any kind except verbal ones, usually phrases of which they constantly form a part. Occasionally they gave evidence of setting the readers' thoughts in some characteristic direction of expectancy, and doubtless the prepositions, especially, always had some very general influence in determining how the whole psycho-physical organism should face a coming related object. The amount of imagery suggested by the other classes of words that were presented in isolation was several times as great with some readers as with others. In general the auditory and motor elements exceeded the visual even for these words, the two former being mainly THE PSYCHOLOGY OF READING 15$ verbal. Both the verbal associates and the other images suggested by the isolated words were of the most varied character, the word often being taken in several quite different' senses in the course of the four seconds. There was a good deal of dawdling association, the words not taking much hold upon the mind. The readers were often surprised at the very different appearance and sound of the isolated word, suggesting how little the words' had previously been thought of for themselves apart from the sentences in which they figure. In the other series in which the words and phrases were exposed consecutively in context, the readers took a more active attitude, the associations were less varied but more numerous, and there were other very character- istic differences. A reader who had looked blankly at the word A when exposed singly, and had gotten no associations, had a rich content of associations when A appeared as the first word of a new paragraph. Be- sides, his feelings of expectancy, curiosity, strain, the "forward push" that was marked in all readers of the context exposures, were even more prominent than the definite associations. The mere statement that the word to be exposed is part of a sense passage limits the trend of association at the start. The limitation extends farther when the reader has caught the general topic discussed in the passage, and still farther when the ex- posed word is presented upon a verbal etjid ideational 156 THE PSYCHOLOGY OF READING background formed by the complete preceding context. lA the spider story, for example, after the mention of web-weaving, the word top no longer suggested top 0} hill, flag-staff, spinning-tops, etc., as when it was exposed in isolation, but now suggested the top of a post or gate- way, with the spider-situation in mind. The newly exposed word was usually mentally pro- nounced as before, and was "fitted into the preceding," as one reader very often put it, the new word seeming to contribute toward a notion of sentence vinity to which each additional element added a needed part. Im- mediately following this there was usually a filling out of the sentence or phrase so as to make sense with what came before, and when this did not actually occur, there was usually the "forward push," "forward tendency," "tendency to fill out," as it has been variously described by the readers. All emphasized the strength and com- parative constancy of this feeling, and mentioned it as perhaps the most striking thing observed in the experi- ments. It was usually not so strong at the beginning or end of sentences and paragraphs. Beside this "forward" feeling, the "little words" gave few except verbal associations. They seemed, as one reader several times remarked, to be but "verbal count- ers" in the sentence. This reader showed comparatively little tendency to visualize, throughout the experiments. However, he did visualize §ome of the main objects and THE PSYCHOLOGY OF READING 157 scenes referred to in the passages read, enough to form a vague background for the story, but in the main the story itself seemed to be thought in verbal terms. The other readers, however, had more of the visual element. In the story of the spider's weaving, for in- stance, a visual picture of the spider was early formed and remained throughout, although it was more or less modified to suit the different references to it as the story progressed. The spider itself was seen in a visual back- ground that had various components fused in a kaleido- scopic fashion into it, each time that the story gave addi- tional data, but still without any violent transitions being made. While this fluctuating spider-scene would some- times pass temporarily out of the attention field as some particular substantive called up scenes pecuhar to itself, it constantly returned and remained as a factor control- ling the course of association and expectation. The visualization was almost always static. The spider jumping was visualized as the spider ready to jump or just alighted. The thought of motion, when mentioned at all, seemed to be a consciousness of tendency to move- ment in the reader's organism. The agreement or disagreement of the exposed word with the trend of expectation produced by the preceding context was a matter of frequent remark by the readers, and was often a cause of considerable feeling on their part. The sequence was felt as right or wrong, giving 158 THE PSYCHOLOGY OF READING fulfillment or disappointment of expectation. If partic- ular words were expected, the ones that came, although different from the expected ones, were often called "all right," "still better," etc., showing that the real expecta- tion was rather of the expression of a desired meaning than of any particular words. The meanings of particular words were commented on from time to time, as when vertical gave an "up and down feeling," distant gave "a kind of feeling of what it meant," etc. Starting gave a dim thought of starting machinery, although of no particular machinery. Dis- tant, for another reader^ produced "a raising of eyes and looking off." Swings gave a "motion from left to right," with perhaps a slight head-movement. Many words gave a sense of location. Tribal gave a "sort of western location," and dress, patronage, alien, among others, were located in particular directions. Through- out much of the exposure-reading of the Queen Elizabeth story, one reader maintained an orientation of the events as occurring to the left and front of him. On the whole, the meanings seemed usually to be felt as belonging to the larger wholes, to the sentences and other large units. The words were mainly "counters," felt as having a part in the total, but their function being mainly to help tide one over to a place where a new meaning would be suggested or completed. The reader seldom escaped feeling the particular words in a per- THE PSYCHOLOGY OF READING 159 spective of the before and after, and was often much puz- zled, even baffled, to know how to deal with them as they stood with the total sense imcompleted. The absence of images, at least with the individual words, was rather marked. These are my more important observations from a considerable amount of such experimentation. They agree in the main with the conclusions of Professor Ribot from an extended and somewhat similar series of experi- ments in which words (general terms) were pronounced to listeners. The listener was to state immediately whether the word called up anything or nothing to the mind, and if anything, to state what it was. Out of a large number of replies, the most frequent was "nothing," the only sensory image present in consciousness being the sound of the word. In other cases, some concrete example of the term was imaged, sometimes with a visual image of the printed or written word. Sometimes the printed or written word would appear without other im- age. These were results for isolated words. Ribot also presented whole sentences, "abstract statements," and these gave a similar scantiness of imagery. In the latter test he considered only the sentence as a whole and did not look also for imagery from particular component words, as in my own experiments.' ' Ribot's experiments are reviewed in Stout's "Analytical Psychology," Vol. I, p. 82. l6o THE PSYCHOLOGY OF READING Professor Stout, in the chapter on "Implicit Appre- hension" in his "Analytical Psychology," points out that many of the early English and Scotch philosophical writers, as Hobbes, Berkeley, Hume, Stewart, and Dr. Campbell, "bore unequivocal testimony to the fact that the flow of words is for the most part imattended by a parallel flow of mental imagery." Burke had held that "in the ordinary course of (Conversation we are sufficiently understood without raising any images of the things concerning which we speak." Professor Stout himself holds that the presence of sensorial imagery is neither usual nor essential in the apprehension of spoken or written language. He does not think that even fleeting and shadowy images necessarily accompany the use of words in ordinary discourse, or are even usually present. Even when the image does appear, according to Professor Stout, it is often imessential and almost irrele- vant, as when I have the image of two persons talking when I hear the word understanding. The real meaning is not in this image, nor is the meaning of the word wealth in the image of a bale of goods that may be called up at the word's appearance. "In reality, imagery of this sort is part and parcel of the word itself considered as a sign rather than of the meaning which it signifies." At any rate. Professor Stout is certain that the part that images play in consciousness has been overestimated. Consciousness is not a picture-gallery, or a magic lantern THE PSYCHOLOGY OF READING l6l exhibition with slide displacing slide in rapid succession. We may suppose presentational states of consciousness not composed of sense imagery, but functioning analogously, just as revived images "have a representative value in some degree comparable to that of sense-perceptions, in spite of very great differences in respect of distinct- ness, vividness, and quality." Even in actual sense- perception, the meaning of the object is something quite other than the deliverance of the senses. According to Stout, "an imageless representation of the whole is conjoined with the sensible appearance as its psychic fringe." In this fringe lies the significance of the object, as of the image or the word. Apparently, Stout holds that a feeling of the possibility of reviving the experiences that one has had or may have with the object forms the basis of this fringe, of this halo of meaning. He says that if we dwell on a word, we have a presentiment that images are coming, followed by them shortly. "It is as if the multiplicity were some- how wrapped up in the distinctionless unity and were struggling to unfold itself." So he suggests the name implicit apprehension for "that apprehension of a whole which takes place without the discernment of its parts." Nearly all words stand for concept-wholes which are representative of many particulars. Even singular names are somewhat so, standing for a whole that represents varying phases in the history of the individual. Stout'j l62 THE PSYCHOLOGY OF READING main point in the whole discussion is that it is possible and usual to think such a representative whole "in its imity and distinctness without discerning all or even any of its component details." However, when a detail, say a sense image, does form a part of our thought of a word that is read or heard, it may, although irrelevant or un- essential, help the word to' hold the attention on the real meaning, which, however, is not the image. The image, Stout thinks, is a helpful instrument for fixing the atten- tion, as the sense-percept itself is. The name, however, is often superior to the image, in that it calls up the generic core of the concept, while an image is apt to be particular and thus comparatively unessential. In general, language, aside from its function as a means of communication, is most important as a means of directing the attention. It is a mobile "movement of fixation." "A word," says" Stout,' "is an instrument for thinking about the meaning which it expresses." There is more facile and accurate control over these articulatory "movements of fixation" than over the various muscular adjustments of the sense organs ; hence their advantage in the manipulation of meanings in our thinking. Stout finds that one "has almost as great a control over the internal articulation as over the external. The chief restriction appears to lie in the inability to make the represented sound as loud as the actual sensa- ' "Analytical Psychology," Vol. II, p. 194. THE PSYCHOLOGY OF READING 1 63 tion, but, apart from this, one may do almost as one likes with it." So we have come to think mainly in words, as the most rapid and accurate means of handling our mean- ings, of using our experiences. With children and unedu- cated persons, as we have seen, actual articulation is very common as they think or read. Older and educated persons resort to the inner articulation. The thinking may indeed be done in part with images as signs of mean- ings, but is far more usually and advantageously done with words as the instruments. So much for the part played by imagery in reading, a part that is far larger in the reading of young children than in that of adults and that is far larger in some than in others, but a part that is always secondary or auxiliary to the suggestion and control of meanings themselves. The consciousness of meaning itself belongs in the main to that group of mental states, the feelings, which I regard with Wundt as unanalyzables, or at least as having a large imanalyzable core or body. Each mean- ing-feeling is very much itself and imlike every other. Of the meanings felt with particular words as we read, most are, as James suggests, those of relations felt as existing between the larger objects of thought, feelings of and, if, but, by, etc. These feelings, fleeting and slight as they often are, and named only by the words which call them forth, are yet perhaps as definite and as much themselves as anger, pity, and the like. The 164 THE PSYCHOLOGY OF BEADING poverty of names substantively used for the subjective states makes us unduly oblivious of the real distinctions existing among these most meaningful of all conscious- ness conditions. Indeed the relational states themselves have usually escaped notice, as when M. Egger states that the word-sounds in speech make ten or twenty times more noise in consciousness than the other com- ponents of the consciousness. Careful observation, how- ever, reveals the fact that words are seldom or never heard as mere sounds ; what seems to be their sound is really mainly the meaning which we read into them,. as indeed is the case when we hear a tolhng bell, a tramp of horses, or a familiar hymn. When we are occasionally able to attend mainly to the raw sounds of a word di- vorced from word-associates and from meaning, the word soimds as strange and unlike itself as the isolated words appeared in my exposure experiments, or as any letter or figure appears when unduly stared at. The fact is that meaning is part and parcel of word- sound and of word-utterance, as these ordinarily occur in reading and in thinking ; that is, what we take for word- sound and word-utterance is largely word-meaning. And as meaning inheres in or is fused with the word's sound or utterance, so to get the meaning we naturally utter the word, incipiently for the most part, actually when the meaning is obscure. And not merely for the "little words," the words showing relationship, etc., do the mean- THE PSYCHOLOG'« OF BEADING 1 65 mgs lie in the passing feelings which they rouse. These relational words lead to the substantive resting-places of the sentence, where some sensory imagery is apt to occur, and should occur usually, to harmonize with the imagery awakened at the earlier substantive parts. But as Stout rightly argues, the meaning even here is not mainly in the image, but is in the feeling which attaches to the image and the word together as the feeling's sign. Often, as has been noted in the account of my experiments, we get this meaning-feeling without the word or image, often it is all we possibly can get, in those tantalizing cases of remembering only what a name is like, how it feels to say it, what its deeper significance is. Here we approach the pure meaning-consciousness as detached from articulation. So for reading as for thinking, we would agree with James when he says that "The definite images of traditional psychology form but the smallest part of our minds, as they actually live;" and with Flournoy,' who warns against supposing that the words and ideas suggested by a given word express the thought really contained in it. This he believes lies deeper. "The true psychological centre of the concept seems then to be not in the images called up, but in those confused feelings which serve them for a background, and which James has so well described under the names of fringe, suffusion, psychic overtones, etc." Feelings, and the motor reactions or tendencies ' L'Annie Psychologique, 1895, pp. 45-53. 1 66 THE PSYCHOLOGY OF READING from which feelings cannot be disjoined, are far more fundamental and usual than images, and these constitute the consciousness of meaning. There is an underlying feeling of meaning-in-itself, as it were, in reading any sentence that makes grammatical sense, and this quite independently of anything that the sentence tells. As James puts it, " Certain kinds of verbal associate, certain grammatical expectations fulfilled, stand for a good part of our impression that a sentence has a meaning. . . . Nonsense in grammatical form sounds half rational; sense with grammatical sequence upset sounds nonsensical." Often in our reading we are content, for considerable stretches, with this sense-meaning feeling, like children who listen with rapt attention to half-under- stood things, asking the meaning of none. "Their think- ing is in form just what ours is when it is rapid. Both of us make flying leaps over large portions of the sentences uttered, and we give attention only to substantive starting- points, turning-points, and conclusions here and there." ' Of specific meanings beyond this general feeling of "making sense," everything in my own experiments indicates that they are usually total meanings belonging to sentences or to unitary parts of sentence?, but felt dif- ferently as this or that particular word is being dealt with ; or we can say that the particular word's meaning is felt in a perspective of the total meaning. A relation can hardly •James' "Psychology," Vol. I, pp. 255-265. THE PSYCHOLOGY OF READING 167 be felt apart from the terms or objects related, a particular manner or intensity of action or being can hardly be sug- gested by an adverb apart from the thought of the action or being itself. And likewise a substantive that is thought of naturally has, in this thought, something of the sub- stantive's relationships. In short, it is total situations and performances that we think of and read of, and these often complex, always with various aspects and various relationships of parts. No single word names or describes the whole. When a single word is presented, therefore, it suggests but a part or an aspect of this total meaning and is felt as inadequate and artificial unless given in its sentence context. With meanings, as with vocal utterance, the sentence-meaning is the natural unit, and smaller divisions considered apart from this are felt as disjecta membra. We may safely conclude, then, that meanings in reading are mainly feeling-reactions and motor attitudes attaching most intimately to or fused with the inner utterance of the words and especially of the sentences that are read. And with the utterance in which the meanings mainly inhere, we must include the movements of emphasis, of inflection, of gesture, and of expression generally. We have referred to the peculiarly important part that these latter play in the expression of meaning. The feeling of these bodily postures, attitudes, gestures, etc., may well furnish the very body of much that we call meaning, according to the 1 68 THE PSYCHOLOGY OF READING principle which Professor James holds to be true for feelings generally. At any rate, they have a most intimate relation- ship with meanings, and deserve more careful observa- tional study than has yet been given them by students who would know the intimate structure of language. But we must also remember that meanings in reading are not reached solely through the inner utterance. The sentence-utterance, as we have seen, comes at some dis- tance behind the eye. But we have seen that with the first incipient thought of the word's isolated utterance there is present a suggestion of its meaning. Indeed, so strong is this with some words that they are said to "look like their meaning." It is so in my own case with the words venomous, God, with many proper names, and with very many other words. There is doubtless with most words some feeling characteristic of their visual form as such. And more or less of the word's significance is apt to be felt with this or immediately after. Indeed there seems to be a flash of the relation of this meaning to the preceding context-meaning, and to this extent some sense of total meaning seems to keep pace with the eye. This is indicated in the action of any reader as he finishes a page or a paragraph that he is reading aloud. He seems to feel that his work is done as his eye sees the last word, although the utterance is half a line or more behind. The total meaning, although not fully reahzed as yet, has begun to reahze itself, and the reader is satisfied as having its THE PSYCHOLOGY 01 READING 1 69 realization initiated and under control. The visual forms may often act as cues to touch off this total meaning immediately, this total meaning then guiding the suc- ceeding utterance of ' the sentence, which is partly an expression of the total meaning thus previously sug- gested and felt, and is partly also a means of bringing this meaning to fuU consciousness in its various aspects. CHAPTER IX THE RATE OF READING It remains finally for us to consider an aspect of the psychology of reading which is of the greatest importance practically and pedagogically, the rate of reading and the factors which condition speed. Romanes early experimented upon the matter, having "practiced readers" read paragraphs in a book containing "simple statements of simple facts," and noting the time needed for the reading. The moment the reading ceased, the reader wrote down all that could be remembered of what was read. Romanes found "astonishing difEerences in the rate of reading." "The differences may amount to 4 to I ; or otherwise stated, in a given time one indi- vidual may be able to read four times as fast as another. Moreover, it appeared that there was no relationship between slowness of reading and power of assimilation." ' On the contrary, "when all the efforts are directed to as- similating as much as possible in a given time, the rapid readers (as shown by their written notes) usually give a better account of the portion of the paragraph which has been compassed by the slow readers than the latter are able to give ; and the most rapid reader whom I have found is also the best at assimilating." 170 THE PSYCHOLOGY OF READING I7I "I shall further say that there is no relationship between rapidity of perception as thus tested and intellectual activity as tested by the general results of intellectual work ; for I have tried the experiment with several highly distinguished men in science and literature, most of whom I found to be slow readers." ^ Miss Adelaide M. Abell, in the Educational Review for October, 1894, reported experiments on the reading rate of forty Wellesley College girls, the work being done under the direction of Professor Calkins. The girls read a short story at a definite time not long before the class was to meet, and timed the reading, not knowing the aim of the experiment. When the class met, the readers wrote the story from memory, as nearly verbatim as possible. The slowest reader was found to have used in her reading six times as much time as the fastest. The reproduction test followed but a few hours after the reading, and was therefore taken as a test of comprehension rather than of memory. The results indicate that most of the readers "gain by relative slowness," but "two subjects head the list both in rapidity and comprehension," and these readers agree that except where matter is obscure "they grasp the thought more readily by rapid reading." "Of all three classes of readers — fast, moderate, and slow — some comprehend well and others fairly or poorly," showing that "comprehension may be independent of the ' Romanes' " Mental Evolution in Animals,'' pp. 136-137. 172 THE PSYCHOLOGY OF READING absolute rate of reading." Miss Abell found, however, that on the whole the swift reading saves time without necessarily decreasing comprehension. She believes that the rate will be increased by "increasing the rapidity of association, by repeating and multiplying associations, and by intensifying interest and attention." The actual or imagined pronunciation of the words read was found to be " a characteristic correlate " of the slow reading, and Miss Abell thinks this a hindrance when habitual, and a tendency that should be discouraged in children. "Another pecu- liarity," she adds, "of the slow readers among our sub- jects, is the reading of a word at a time, while the rapid readers grasp phrases, clauses, sometimes even sentences, at a glance." She concludes that although every individ- ual probably has his maximum rate, "determined by his natural quickness of comprehension and association, it is yet possible and desirable to some extent to increase the ordinary rate." Dr. J. O. Quantz tested fifty university students, juniors and seniors of the University of Wisconsin, to determine their normal and maximal rates in reading, and experi- mented upon them further to determine the factors and conditions upon which rate of reading depended. He found that his readers varied from 3.5 words per second for the slowest to 8.8 words per second for the fastest, when reading at normal speed. At maximal speed the rate ranged from 3.5 words per second to 12.2 words. This THE PSYCHOLOGY OF READING 1 73 was for silent readmg. In reading aloud at normal speed the rates were within a -range of 2.6 to 3.9 words per second. Generally, those who read fast at normal speed excelled in the maximal tests, and the slow readers were generally slow in both. In testing the ability to reproduce what was read, he found that the rapid readers were on an average about 37 per cent superior to the slow readers in the quahty of their work. "The superiority of the rapid reader is also shown by the fact that his memory of the substance of his reading is more exact than that of the slow reader. He introduces only two-thirds as many thoughts not found in the original selections." The use of lip-movement in readmg was not found to help in comprehension or in concentration of the attention, although it often occurred as a result of concentrating the attention. In general, lip-movement was found to be a serious hindrance to speed of reading, and consequently to intelligence of reading. "The ten slowest readers show almost double the amount of lip-movement that the ten most rapid do," while "not one of those whose reading is widest is a lip-mover to any extent which can be observed." "Extent of reading works directly against movement of lips, and is practically the only thing which does so, except among the medium lip-movers." Dr. Quantz found that persons who are of the visual type ■'are slightly more rapid readers than the auditory type." 174 THE PSYCHOLOGY OF READING Quickness of visual perception, shown in rapid recognition of colors, words, geometrical forms such as circles, squares, diamonds, etc., was found to be "an important factor in deciding one's rate of reading." Considering all the factors which he found to contribute to rapid reading, they are, "in order of importance, visual perception, practice in reading from childhood on, power of concen- tration, mental alertness estimated by rapidity of original composition, scholarly ability as decided by college rec- ords." These results were found in general for the read- ers tested, and as a particular confirmation the person whose reading was by far the most rapid of all those tested excelled markedly in practically all of the above conditions of rapid reading. It is interesting to note that the judg- ments of readers about themselves, as to whether they are slow, medium, or fast readers, agreed very well with the results of Dr. Quantz' tests of speed as taken later. I have carefully tested the reading rate of twenty-eight persons, and have tabulated the results for twenty of these who were graduate university students and whose reading, accordingly, had been pretty extensive. The reading was from an interesting novel, presenting no special difficulty, and the tests were made under conditions which approxi- mated as nearly as possible those of comfortable reading in one's own quiet room. The readers were found to range in rate from an average of 2.5 words per second for the slowest reader to an average of 9.8 words for the THE PSYCHOLOGY OF READING 1 75 fastest, when reading silently at their ordinary rate. When the silent reading was at maximal speed, the rates ranged from 3.5 to 13.5 words per second. In reading aloud, the average of the slowest reader was 2.2 words per second and that of the fastest 4.7, at the ordinary rate, and at maximal speed the corresponding range was from 2.9 to 6.4 words per second. The average rate of the twenty students when reading silently was 5.63 words per second at their ordinary speed and 8.21 at their maximal, while in reading aloud they averaged 3.55 words per second at their ordinary speed and 4.58 at their max- imal. Several of the readers averaged near the fastest rates given above, and the other readers were distributed somewhat evenly, as to rate, from the fastest rate to one somewhat above the slowest. Lip-movement was usual with only two or three of these twenty readers, but one of the fastest readers tested was a lip-mover. However, when those who were unaccus- tomed to lip-movement were asked to move their lips while reading, their speed was evidently hindered. The readers showed a strong rhythmic tendency. Each would fall into a reading pace that seemed most natural to him, and would then read page after page in almost exactly the same time. Quite usually the differences from page to page would not be over three or four seconds. Some of the readers showed surprising regularity, reading several successive pages in almost exactly the same time, although 176 THE PSYCHOLOGY OF READING they were quite unable to estimate the same time interval, failing by as much as twenty seconds in the attempt to estimate the time of reading the page. Habits of eye- movement are doubtless important factors in setting this pace. When the same number of words was printed in fewer lines, of the same length and the same size of type, they were read faster in just the proportion that the lines were fewer, suggesting that the eye has a habit of tak- ing about so much time for a line of a given length. I foimd by experimenting upon lines marked here and there by crosses, etc., for fixation points, as below, umbrella^beu it rained ;*4i)»t Sir Isaac New KW s upposed he ha4|eaten vvhe:i he s^ the chicken bJKts on h^ plate; '^^ ,that(Pii'sou forgot liiwweddiiig day. Stj^the fact retnain^fcat no perils of life is iv^ from. ns(iReabIe ^stractioj^ The |^ vith b«^ ia hand fo^ts to go tCtiinner afterise has run^^ie belli^e. young woiQui goes to diffefiit parts of the-^cnse, she ,knows U0why; middle, a^ hunts for the tbii^e on its fin« 'ger, or the p«?rtu its, mouth; while o]5Jage is troubled ^pat it caon^g^nd the glasses c^ts uose. Fig. 15 that the eye readily falls into a very uniform rate of prog- ress, corresponding more or less closely to its usual rate of reading, in traversing the hnes without readmg. Dr. Dearborn, as we have seen, finds that the eye falls into a brief motor habit of making a certain fixed number of pauses per line, or, sometimes, a certain succession of THE PSYCHOLOGY OF READING 1 77 pauses for a given passage, independently of variations in the subject-matter from line to line. The fast readers formed these "short-lived motor habits" much oftener than the slow readers, and the moderately short lines of uniform length give the best conditions for forming these habits, which seem to increase the speed of reading. We have already noted the emphasis placed by Dearborn upon the establishment of "a regular rhythmical movement" as a factor in increasing the reading rate. Dearborn tested the reading rate of a considerable num- ber of readers, and found that for a given class of reading matter the fastest reader read more than three times as fast as the slowest. He also carefully tested the readmg rates of three graduate students, one a mathematician, one a teacher in a secondary school, and one a psy- chologist, for various classes of reading matter selected from literature and science. Each of these persons had quite different rates for the different classes of matter read. However, the fastest reader in his most interesting subject, in this case the mathematician, read much the fastest in all the classes of matter, the secondary teacher read much more slowly than the mathematician in all the classes, and the psychologist read much more slowly still in all but one class. Dearborn draws from this data the conclusion "that one who reads rapidly in a given style and class of subject-matter will read somewhat pro- portionally faster than a slow reader, whatever, within 1 78 THE PSYCHOLOGY OF READING certain recognized limits, the nature of the style and subject-matter." After the reading of a passage in the reader's usual way, Dearborn had him read it again "as rapidly as possible consistent with getting the sense." The reader believed that he usually read at his maximum rate, and that this was particularly true of his first reading of this passage, the matter being especially interesting. It was found, however, that in the second reading nearly one-third of the total time was saved, the absolute time of nearly every pause being diminished, fewer pauses being made, and "the average distance of the eye's first pause from the left edge of the page noticeably increased." The length of the initial fixations in the lines was, however, even a lit- tle longer than before, permitting the preliminary general survey of the line which Dearborn finds to be needful for the more rapid reading. He finds, indeed, that "rapidity of reading is not necessarily correlated with regularity of movement," some of the fastest and slowest readers being found equally regular in movement. But "a wider 'span- ning' of attention," shown in the frequency of long pauses at the line's beginning and in the fewer fixations per line, is found to be characteristic of the more rapid readers. "The slow readers have a narrower span or working ex- tent of attention, and a greater total arc of movement." Both my own experiments and those of Dr. Dearborn indicate that there can usually be much improvement THE PSYCHOLOGY OF READING I7g in the rate of reading. There seems to be a rhythm into which each reader ordinarily falls in ordinary reading, and it seems doubtful whether ordinary practice changes this rhythm after it is well established. Doubtless many of us dawdle along in our reading at a plodding pace which was set and hardened in days of listless poring over unin- teresting tasks, or in imitation of the slow reading aloud which was so usually going on either with ourselves or with others in the school. And indeed, for the later school period, I quite share Dr. Dearborn's opinion "that the careful dwelling upon each word and phrase, which is the daily method of the classical student throughout many years of study, helps not a little in fixing such a habit of slow assimilation." Bad form in reading is doubtless as distressingly common as bad form in swimming, skating, or tennis. And we know that the form once set is apt to re- main for life in any of these activities, with all the corre- lated hmitations in speed and in quality of performance. I have considerably increased my own speed in reading by waking up to the fact that my rate was uimecessarily slow, and then persistently reading as fast as possible with well- concentrated attention, taking care to stop short of fatigue until the new pace was somewhat established. I thus reached a speed of a page per minute for such books as Ellis' "The Criminal," of the Contemporary Science Se- ries, maintaining this rate for a half-hour or so at a time, and with very good comprehension of what was read, x8o THE PSYCHOLOGY OF READING although after such reading a very hasty review of the leading points was the most satisfactory procedure. My earlier speed was not more than half as great. It is especially desirable that the reading rate and the conditions affecting it should be carefully determined for the children of the various school grades, and for various classes of reading matter, taking into account the apper- ceptive relation of the reader to what is read. That fast reading may come quite early is indicated by a test which I made of an eleven-year-old schoolgirl who was said to be "a great reader," and whose rate was found to surpass that of almost all the university men both at normal and at maximal speed. On the other hand, it seems perfectly certain that there is, among children, a great deal of dead- level plodding, with little thought of varying the speed ac- cording to the importance of what is read ; and investigation here, if carefully made and then acted upon pedagogically, may have the greatest value in lessening waste and in increasing effectiveness both in reading and thinking. A university friend, a mathematician, informs me that he has read the whole of a standard novel of 320 pages in two and one-fourth hours. The occasional though very rare instances of such rapid reading suggest that we may be far within our possibilities in dealing with printed symbols. I am inclined to think that at any such speed the meanings suggested immediately by the visual forms suffice for all but the more important parts, and that these THE PSYCHOLOGY OF READING l8l meanings are felt sufficiently, without inner utterance, to permit selection of what is more important, the more im- portant places themselves having a fleeting inner utterance to vivify their meaning. We must indeed experiment further before we can conclude against the possibility of mainly visual reading at the very high speeds. The inner speech in such cases must at any rate suffer a foreshorten- ing and atrophy of articulatory details which reduce it to Uttle more than a slight motor tallying as the meanings are felt or dwelt upon. / In general, the question of individual differences in reading is a very important one, and merits careful study upon still other phases than rate. For instance, the natural characteristics of children's reading, at various stages, must be worked out carefully, and the normal range of variations must be determined. Differences in the extent of the perceptual span, in the size and nature of the units in terms of which reading matter is per- ceived; differences in the amount of attention given to total form, to letters, etc.; "legato" versus "staccato" reading, a very important and typical difference that is wisely taken account of by Principal Russell in his teach- ing of reading at the Worcester State Normal School ; dif- ferences in phrasing, also made much of by Principal Russell, and important psychologically as well as peda- gogically; I should consider all of these promising sub- jects for investigation, were I to continue the general 1 82 THE PSYCHOLOGY OF READING study. Abnormalities of reading, typical or otherwise, are numerous and offer another rich field of study. Many of these abnormalities, to be found almost everywhere, are probably remediable or preventable in the light of careful analysis of normal reading, and the study of the cases may render much service to psychology as well as to education. The clinical method, following closely the fortunes of par- ticular and significant cases, seems to have great advan- tages here, as indeed it seems to me to have for very many of the problems of general psychology. Indeed, for all functions that are performed in reading, the determination of types, and of the normal as well as abnormal ranges of variation, becomes one of the im- portant duties of a differential psychology which would render real service to education, and the returns to psy- chology itself will be well worth while. Messmer's sug- gestion of subjective and objective types of readers names only the first of important type-differences. For example, in what typical ways may we be conscious of meaning? Professor Titchener, in a letter of nearly three years ago from which he kindly permits me to quote, holds that while "the 'meaning' of a concept is," as indeed appeared in Dr. Bagley's thesis study with Professor Titchener, "always carried by the 'fringe' of consciousness," the constituents of this fringe have typical and important variations. These go back for their source to the division of men "into two great classes, according as they have THE PSYCHOLOGY OF READING 1 83 or do not have a large amount of organic sensation in their general conscious make-up." For example, people for whom the organic sensations have practically disap- peared "remember old events without any stir-up of feel- ing." " These are minds of cultivated and bookish people, like those that Galton found to be lacking in visual im- agery." "These people are cold-blooded, as one says; they are intellectual and not emotional; they have no affective memory, but only an untoned reproductive memory; they are detached, impersonal, cool, reflective. That minds of this type exist I cannot have any doubt, although I myself belong to the former group. To say that these people do their conceptual thinking by means of felt organic attitudes simply goes against their own introspections. They find, in the extreme cases, that the fringe is verbal, just precisely as the center of conscious- ness is. They do not feel any attitude. It may possibly be (I say this hesitatingly) that their apprehension of meaning is purely physiological, — done by a not-felt attitude; at least, we have found cases of recognition in which neither felt attitude nor verbal fringe could be dis- covered by introspection, so that for all we could tell the act of recognition was a purely physiological, reflex matter : the organism fell into the recognitive attitude without any introspectively discoverable change in consciousness. This is imcertain ground; but of the existence of the verbal- fringe type I can, as I said, have no doubt. Doubtless, 184 THE PSYCHOLOGY OF READING it is a development from the other; and doubtless there are all stages between the extremes." I am glad that Professor Titchener has so ably stated these typical differences in this phase of the reading con- sciousness. Variations of great importance will be found in various other phases ; but in view of the fact that there has been so little observational study of cases, any attempt to treat further the subject of individual differences in reading would scarcely be profitable at this time. Enough has been said to suggest the importance of such study. The conclusions stated in the present volume seem to hold for the great majority of readers ; and the important contributions yet to be made by the study of individual differences are likely to be in the nature of additions rather than of contradictions. We must rest here our survey of the psychology of read- ing, and attend to other aspects of the general topic. There yet remain to be written many most interesting chapters on the psycho-physiologicatl phases of reading, which will be made possible as investigation proceeds further. The work that has already been done by many hands and in many lands illustrates well how the federated science of the world is making soUd progress with specific problems, and bears promise of a day when education shall rest on foundations better grounded than were the individual and unverified opinions about "Reading," for instance, even twenty-five years ago. PART II THE HISTORY OF READING AND OF READING METHODS CHAPTER X THE BEGINNINGS OF READING, IN THE INTERPRETATION OF GESTURES AND PICTURES , Wherever there has been civihzation there has been reading and writing, in the remote past as in the present. In North Babylonia, for example, written records have been discovered that are no less than six thousand years old, and these prove that writing and civilization were then by no means in their infancy. Clodd, in his " Story of the Alphabet," concludes that in Babylon writing had long passed the pictograph stage eight thousand years ago, and thinks that "Babylon carries the palm" in the age of writing. At least seven thousand years ago Egypt was reading a page that was at least partially alphabetic, show- ing that reading was even then an art that had been prac- ticed for ages. In Crete, inscriptions are being unearthed that go back to the early part of the third millennium before the Christian Era. In all these cases, and especially in Egypt and Babylonia, there are abundant indications that reading and writing were already most ancient practices, with the story of their origin enshrouded, as we have seen, in mystery, and told only in myth and legend. But the written records that have been preserved from these remote ages give sure signs of the true origin of the 187 1 88 THE HISTORY OF READING systems of writing used by these nations. They prove that in these early times reading and writing had much the same course of development that has been observed among later peoples and that is going on to-day among savage races so far as they are still uninterfered with. Various peoples and tribes on every continent have devel- oped systems of writing, independently. Some of these systems have reached a high state of completeness, some have been arrested at one or another stage, some are still in their rude beginnings. Yet so far as each has gone it resembles almost every other in the general lines of its development. One finds most striking resemblances, even in details, in comparing such widely separated systems as the Maya of Yucatan with the Egyptian, or the Ojibwa of North America with the Babylonian. Keeping in mind, then, this comparative agreement in the development of the different systems, I shall illus- trate various phases of the evolution of reading and writing by citations from various systems, as each may best serve my purpose, emphasizing, perhaps, the Egyptian as being typical and best known. Mankind began his reading with picture-books and his writing with picture-making, just as the child likes to begin. This seems to have been the case literally, when we recall that book probably once meant a piece of bark and that li- brary {liber, bark) and letters {lino, to smear or paint) bring down with them the smell of the woods. The first pictures, THE HISTORY OF READING 1 89 fiowever, were drawn in the air and were read as fast as drawn. They constituted a gesture-language. The spoken language and the gesture-language arose together as they do in the child, the words and the gestures being a joint means by which prehistoric man communicated with his fellows. So inseparable were these means of commimicating, for some tribes, that they found it diflncult or impossible to com- municate in the dark when the gestures could not be read. Primitive man became very expert in the use of gestures, and savages of to-day use them most effectively. Tylor, in his "Early History of Mankind " (p. 82), says that the natives of North America were as proficient in the use of the gesture-language as in that of picture-writing, much the same conditions having given rise to both. Professor Wundt believes that the languages of picture and of ges- ture grew up together, naturally influencing each other. Even among certain modem civilized peoples, notably among the Neapolitans, the gesture-language still plays an important part in everyday communication. Indeed, Professor Ribot quotes with some approval Dugald Stewart's assertion that "If men had been deprived of the organs of voice or the sense of hearing, there is no doubt that they would have invented an alphabet of visible signs wherewith to express all their ideas and sentiments." It is to be hoped that the notable tendency of children to live over again the use that the race has made of gestures may soon be made the subject of careful observational study. igo THE HISTORY OF READING The gesture-language is, in considerable part at least, a picture-language, a sort of drawing in the air. W. von Humboldt called it "a species of writing." Speaking of such a comparison, Tylor goes on to say (p. 82) : "There is indeed a very close relation between these two ways of expressing and communicating thought. Gesture can set forth thought with far greater speed and fulness than picture-writing, but it is inferior to it in having to place the different elements of a sentence in succession, in single file, so to speak; while by a picture the whole of an event- may be set in view at one glance, and that permanently, so as to serve as a message to a distant place or a record to a future time. But the imitation of visible qualities as a means of expressing ideas is common to both methods, and both belong to similar conditions of the human mind." From drawing in the air to drawing in the sand, or" on bark or stone or wood, would seem to be an easy transi- tion. In Central Brazil the natives were found to fashion an explanatory design in the sand when their gestures proved insufficient for conveyirig an idea. Hirn, in his "Origins of Art" (p. 156), says of this that "these designs are only a projection on a different surface of the ' hand- movements with which in their pantomimic language they describe the outlines of the objects in the air. One is tempted, therefore, to find in these transferred gestures the origin of pictorial art." He adds that "in some tribes THE HISTORY OF READING 19I — particularly among the North American Indians — the picture-signs have evidently been derived from the cor- responding gesture-signs." However, Professor Him and other authorities are uncertain whether the step was taken in this or in some other way. We cannot be sure whether the first pictures were made for purposes of communication or for the fun of the making, as when the child first scribbles. Certain it is that from very early times primitive man made pictures in the greatest abundance, and that by their means he communicated with his fellows. He attained to this means of communication independently in the most diverse parts of the earth, though the pictures, like the gestures, are remarkably alike throughout the world. Of this Tylor says (p. 88): "As the gesture-language is substantially the same among savage tribes all over the world, and also among children who cannot speak, so the picture-writings of savages are not only similar to one another but are like what children make untaught even in civilized countries. Like the universal language of ges- tures, the art of picture-writing tends to prove that the mind of the uncultured man works in much the same way at all times and everywhere." That the picture-writing is almost inconceivably ancient is shown by the many drawings that have been found of animals now extinct. Clodd, in the "Story of the Alpha- bet" (p. 22), writes of this : " Op fragments of bone, horn, 192 THE HISTORY OF READING schist, and other materials, the savage hunter of the Reindeer Period, using a pointed flint-flake, depicted alike himself and the wild animals which he hunted. From cavern-floors of France, Belgium, and other parts of Western Europe, whose deposits date from the Old Stone Age, there have been unearthed rude etchings of naked hardy men brandishing spears at wild horses, or creeping along the ground to hurl their weapons at the urus, or wild ox, or at the woolly-haired elephant. A portrait of this last-named, showing the creature's shaggy ears, long hair, and upwardly curved tusks, its feet being hidden in the surrounding high grass, is one of the most famous examples' of paleolithic art." In these rude pictures of tens of thousands of years ago lay the germs of the alphabets which have made civiliza- tions possible, and which have indeed slowly developed pari passu with these civilizations. We shall now sketch the typical features of this development. In the first stages £- in-^^^ of pictography the Fig. 16. — Record of starving hunter. drawings are made (From Clodd.i) ^^^^ ^jj^^g^ ^^^^.y I, a canoe. 2, man with hands outstretched, indicating " nothing." 3, the uplifted right hand Conceivable mate- means " food," or " to eat," and the left points to 4, the hut. rial, and for the ' This and the other cuts and quotations from Clodd are reproduced, by permission, from Clodd's " The Story of the Alphabet," copyrighf 1900, by D. Appleton and Co. THE HISTORY OF READING 193 most varied purposes. Some- times a hunter, out of food, would scratch upon a stick the picture-story of his destitution, pjid stick it in the ground on the trail nearest his dwelling. Sometimes upon some conspicuous rock his pictures indicated the game that was to be found in that locality. On grave-stones, the T7 o TT-j . pictures told of Fig. 18.— Hidatsa ^ pictograpb on a the prOWeSS of buffalo shoul- hunter and war- FiG. 17. — Tomb-board of In- dian Chief. der blade. " The trail of the animals and nor. "His totem, the reindeer, is reversed, and his own name, which means the White Fisher, /-v is not recorded. The seven some strokes note the seven war par- pursuers is shown great bowlder he ties whom he led; the three up- in the dotted lines. "ght strokes as many wounds Of the three heads WOuld express his received in battle. The horned the lowest is that head tells of a desperate fight of the seeker, who thought of his with a moose." — Clodd's " Story is depicted shout- r^ ■, , °f *-^^ Alphabet," p. 49. ing after his miss- God, as, perhaps. ing friends; then ^^ ^^^ he IS shown ad- ' Indian God Rock" still to be vancing and still geen on the bank of the Allegheny River shouting, till his call is returned south of Franklin, Pennsylvania. Again, from the spot where the hunters the pictures Scratched on the shoulder — ciodd, p. 58. blade of a buffalo killed in the hunt tell 194 THE HISTORY OF READING of the efforts to track companions who had gone on in the chase. And so on bark and wood and stone, on skulls and skins and bones and teeth, on surfaces formed of various fibers, and, with some tribes, on the human body ia tattooing, the pictures were made according to the exigency of the case or the whim of the artist. The investigators of children to-day find here, too, a most interesting parallel in the pictures and symbols carved and scratched and chalked everywhere in and about the schoolhouses of our earlier days. The pictures of primitive man were at first sketches por- traying directly objects to be found in the environment, and were rough sketches such as a child makes. Such lines and parts were drawn as stood for the object in the artist's thought, and imitative fidelity to the objective thing was not very essential. Of course this sketchiness is charac- teristic of the actual perception of objects, for savage and child alike, as indeed for all of us. A very few lines, angles, and other significant features make up the bulk of what we really note in casting a glance at any object. And perceptions, too, are full of things not really to be seen in the object, but standing for it in our thought. So the Indian drew the sound issuing from the mouth of the lost hunter. So the child makes the legs show through the clothing, etc. There is found in the primitive drawings a happy THE HISTORY OF READING I95 "hitting off" of the core of the thing, of the general and essential, such as occurs in the race myths, in the child's imitations, etc. We shall later show how the pictures became conventionaUzed, gradually, in their continued use. But first let us note that they did not long remain simply representations of objects of sense, or pictographs proper. They came to rep- ^ | € ' resent ideas and feelings of most varied ^"'- 19— Combat. (From Hoffman.*) kinds, became ideographs, as this class of pictures has been called. Thus, by metonymy, combat was pictured as in Figure 19. Figure 20 shows a drawing placed for warning at the foot of a steep and rocky trail, intimating that a goat may be able to climb the cliff, though at an angle of 45°, but that a horse would fall. By substitution of a part for the whole, various animals are represented by a drawing ^ y^ of the head, especially when 1 ^ ■ " ^ homed. The wild turkey is ^O^^^ represented by its three-toed Fig. 20. -Warning. New Mexico, imprint; the bear, by the (From Hoffman.) outline of its paw, large claws indicating a grizzly bear, while the absence of claws, or small ones, denoted the black bear. By metaphor, the ' This and the other cuts and quotations from Hoffman are repro- duced, by permission, from Hoffman's " Beginnings of Writing," copy- right, 1896, by D. Appleton & Co. <^ 196 THE HISTORY OF READING Egyptians represented the idea of mother by a vulture, this bird being supposed to nourish its young with its own blood. A king was pictured by a bee, the latter having a monarchical government. Hoffman says, in his "Be- ginnings of Writing" (p. 50), that "ideographs repre- *f senting abstract ideas, pictorially expressed, ^^-^"^ are more frequent in the pictography of f I I some tribes than the mere portraiture of \y^^ objects pure and simple." stored in a pit. Figure 21 shows the Dakota sign for (FromHoffman.) ^^undance, the circle signifying the pit in which buffalo meat was stored, as indicated by the out- line of a btoalo head within, with a forked stick extend- ing upward as used to support the drying pole. Hunger was sometimes indicated by a man with a heavy bar across breast or abdomen, as the seat of suf- fering; or with prominent ribs, as from emaciation. Often the gesture-signs were drawn, representing the corresponding idea. The cross, representing trade or exchange fig.~2. — Eat- (see Fig. 35), seems to have been an imi- lands^^tprom tation of the gesture for the same, and Hoffman.) so with drawings of the gesture-signs for eating, food, hpnger, etc., among both Indians and Egyptians. The drawings tended to become mere conventionalized symbols or symbolic signs of the object or idea signi- fied. Thus, for the Indians, a red tomahawk meant THE HISTORY OF READING 197 Fig. 23. — Snow. (From Hofiman.) war; a pipe, or hand clasped, meant peace. The Ojibwa Indians represented spring by trees with faint signs of buds, and winter as in Figure 23, the curved line representing the sky, with snow descending in zigzags, the whole meaning the "season of snow." Sometimes au- tumn seems to have been represented by leaves flitting over the ground. A month was sometimes a crescent. A day was a sim, or a sleep, represented in the latter case by a man in a reclining position. Figure 24 shows the Ojibwa ing. (From sign for moming, the curved line indi- ° ™*'^' eating the course of the sun, the short line signifying moming when at the left, midday when at the middle, evening when at the right. These latter signs seem to have been in imi- tation of the corresponding ges- tures. Another of the signs for moming was a radiant sun ap- pearing above the horizon line. Figure 25 is a Mexican repre- sentation of traveling, the course in this case having led across a stream, as indicated by the pad- dles used in crossing it. Soimd and speech were represented in various ways. Figure 26 shows the Ojibwa character for singing, the lines Fig . 25. — Traveling on foot and by water (Mexican). (From Hoffman.) igS THE HISTORY OF READING representing vocal utterance being repeated about the heart to denote joyous emotion. The Da- kotas sometimes represent whooping-cough by a number of lines issuing from the mouth as above, but the lines were longer and ing (Ojibwa°^' more divergent. Conversation was indi- (FromHoffman.) ^^^^^ ^^ another tribe as in Figure 27, the double voice lines signifying "speech from both figures." Primitive peoples have been thus versatile in represent- ing not only visual forms, but soimds, ac- tions, feelings, and the most abstract concep- tions. The characters were more and more conventionahzed with continued use, espe- ^'°- =7-— Con- '^ versation. cially among the peoples who attained any (From Hoffman.) degree of civilization. In many cases the characters quite lost their original resemblance to the thing signified and were mere arbitrary signs, to the writers, of an idea or its word-name, or_both. The ancient Chinese symbol for sim f *J thus became r i , some of the changes be- ing due to the greater ease of drawing straight lines with the Chinese brush-pen. So y 1 J, the character for moon, ■B became I 7. So the Accadian character ^.i^f, foi THE HISTORY OF BEADING I99 ^ sky, is a simplified form of sSj^^" 1 ^ s^^'^- '^^^ ideo- gram for Nineveh was ^* yF READING syllabary, a set of symbols for syllable sounds, each syllable constituting a word. The Aztecs, as we have seen, had by the use of the rebus broken up their proper names into syllables, as in the case of Itzcoatl ; and, by the acrological use of a character to stand for the first syllable of its name, had analyzed into syllables words that were other than proper names. Further the Aztecs did not go. The Mayas of Yucatan, alone of all the peoples of the New World, went still further and analyzed syllables, as we shall see. Syllabism is best illustrated perhaps, in the develop- ment of the Japanese writing from the Chinese. Unhke the Chinese, the Japanese language is polysyllabic. About the third century a.d., when Japan came in contact with the civihzation aiid rehgion of China, she adopted the Chinese characters, or verbal phonograms, as terms in which to write her own language. A selection was made of the phonograms which conveniently approximated the sounds of the Japanese syllables, and the entire Japanese language was written in these syllable-characters, much as the Aztecs had written their proper names. According to Taylor, the Japanese have but five vowel sounds and fifteen consonantal sounds, or seventy-five possible syl- labic combinations of a consonant followed by a vowel. As many of these possible combinations do not actually occur, less than fifty distinct syllabic signs suffice for the writing of all Japanese words. THE HISTORY OF READING 211 The Japanese have two syllabaries, both derived in- dependently from the Chinese before the end of the ninth century a.d. One has about three hundred signs and is rather cumbrous. The other "comprises only a single sign, written more or less cursively, for each of the forty- seven syllabic sounds in the Japanese language." The Chinese characters were much simplified, and all determi- natives, homophones, etc., omitted; so that the Japanese is, accordmg to Taylor, "one of the best syllabaries which has ever been constructed." ' "Here, however," continues Taylor, "the development has stopped short. The fact that during more than a thousand years it should never have occurred to a people so ingenious and inventive as the Jap- anese to develop their syllabary into an alphabet may suffice to show that the discovery of the alphabetic principle of writing is not such an easy or obvious matter as might be supposed." The Japanese are at last just beginning, as it would seem, the adoption of the alphabet that we use. The cuneiform writing of ancient Chaldea, Babylonia, and Assyria, as we have seen, went through the usual stages of pictograph and ideograph, and in very early times arrived at phonograms and a syllabary. Their language was polysyllabic, and it seems that sometimes a character which had come to denote the name of an object rather than the object itself, that is, had become a phonogram, was used further, by aerology, to denote • "The Alphabet," Vol. I, p. 36. 212 THE HISTORY OF READING simply the initial syllable of the word. Generally, how- ever, the characters for certain dissyllables, which by phonetic decay had worn to monosyllables, came to be used as phonographs for these monosyllabic sounds ; and these characters were then used, as by the Japanese, to write the syllables of the polysyllabic words. Thus, to write their word for "soul," pronounced nap-sat, they combined the syllabic sign ^ — . , nap, which originally meant "light," with the sign for sat, originally "mountain," giving the total character ^;^y ^ for "soul." Their language had many homophones, and determinatives had to be employed as with the Chinese. In the Assyrian cuneiform the mixture of variants, homophones, ideo- grams, determinatives, etc., made the writing clumsy and difficult to read. In the eighth century B.C., accord- ing to Taylor, the Proto-Medic tribes borrowed the cunei- form characters and effected a simplification somewhat as when the Japanese syllabary was constructed from the Chinese ideograms. They thus reduced the Assyrian cuneiform to a "comparatively simple and certain syllabary of ninety-six characters," retaining only about half a dozen of the determinative ideograms. But here again there was arrest. And most if not all of the forms of the cuneiform writing, except the Persian, stopped short of the construc- tion of a true alphabet, and were content with the use of characters for syllables only. THE HISTORY OF READING 213 The Egyptian writing, it is certain from the evidence of the monuments, went through the usual stages of primi- tive pictographs and ideographs, had its homophones and determinatives hke the Chinese, and went through its stages of rebus, aerology, and syllabic signs. Up to this point, as Taylor says,^ it "offers a remarkable parallel to the development of other primitive methods of writing, such as the cuneiform or Chinese." But the Egyptians went further, and analyzed the syllable. Indeed, in the very oldest of all the Egyptian inscriptions, according to Taylor, the inscription of King Sent, which indeed he believes to be "the oldest written record in existence," three alphabetic characters are employed to spell the monarch's name, which reads AAAMA P "Two of our English letters," he thinks, " n and d, are derived, in strict historical filiation, from two of the alphabetic signs, AAAMA and c=^2=." These and some other origi- nals of our letters he thus finds to be "older than the pyramids — older probably than any other existing monument of human civilization with the possible excep- tion of the signs of the zodiac." ^ Thus early had the Egyptians reached an ultimate ■ I, p. 60. ' These and the other cuts and quotations from Taylor are repro- duced, by permission, from Taylor's " The Alphabet," copyright by Edward Arnold, London. ' I, p. 62. 214 THE HISTORY OF READING analysis of the word and syllable into letter-sdunds, into vowels and consonants. The difficulty of such analysis must indeed have been very great, as shown by the fact that astute peoples like the Babylonians, Assyrians, Medes, Chinese, and Japanese never succeeded in making it. "Symbols for vowel sounds are found in the syllabaries of some of these nations, but the more difficult conception of a consonant was not attained or even approached. Easy as it seems to ourselves who are familiar with it, the notion of a consonant, a sound that cannot be sounded except in conjunction with some other sound different from itself, is by no means so simple as it may appear. It involves the decomposition of the syllable into its ultimate phonetic elements — the mental isolation, for instance, of the unpronounceable sound t, which is common to the articulations tea, tie, toe, and two, and yet is not identical with any of them." ^ Taylor thinks the Egyptians were aided in making this analysis by the nature of their vowel sounds. These seem to have been of a rather indeterminate character, like the italicized vowels in about, assert, bird, hut, dowble. Their words were very often written without the vowel-signs, the vowel being perhaps regarded as inherent in the pre- ceding consonant. So only initial and final vowels were necessarily written down, except for emphasis. For exam- ple, their character - \ | . originally represented ses, a bolt, 'Taylor, "The Alphabet," Vol. I, p. 62. THE HISTORY OF READING 215 and came to stand for the syllable se. With \\, a character used for the vowel i, the combination I\ is read si, " the vowel sound of e being elided, so that the symbol ■■ \ t has here the power of a pure consonant. It may be regarded as probable that it was in some such manner that the difficult conception of a consonant grew up, slowly and almost unconsciously." ^ The principle of aerology helped constantly in the analysis. Almost any one of the four hundred Egyptian phonograms could be employed, it seems, to denote the initial sound of the corresponding word. Gradually, however, for any given alphabetic sound, one or two or three of the more easily written characters representing words beginning with this sound came to commonly stand for the sound. It is as though we should take seriously the rhyme, "A is an Archer who shot at a frog, B is a Butcher who has a big dog," and reversing the terms, make the picture of an archer stand for the letter A and that of a butcher for the letter B, ignoring all the other words and things that A and B might stand for. So the Egyptians came to represent all their sounds by an alphabet of forty-five characters, having sometimes two or three characters for the same sound, as with c and k, c and s, etc., in our alphabet. There was a further simpUfication in practice until the Egyptian alphabet, as ordinarily used, consisted of but twenty- ' Vol. I, p. 65. 2l6 THE HISTORY OF READING five letters. Figure 30 shows the Egyptian hieroglyphic alphabet, with the letter-names and approximate equiv- EQVPTIAN HIEROGLYPHIC ALPHABET. Vdan. ■Kam.. Nonnrt OunKtflrs. TirllnH. 1 a eagle ^ >. a leed 1 < a an)) B. t i {Parallels W f i flouble leed /]/ 7 * chick bowl ^ 'c;* u • * tlirons a » 7 angle /] 10 X sieve I «- 11 i maander ra f- i knotted cord I 1 a t gemicircle Ji ] » f hand ^— , K t K snake tongs' ^ P 1 " « cbaiiback , .. J, inandntdl eanlui shutter Mi 9 ^ 10 i a / leg cenistM i ^ a r mouth <=> » t lioness -2a Fig.- 30. (From Taylor's " The Alphabet.") alents in our alphabet. The pictorial origin of the Egyptian letters is evident enough. THE HISTORY OF READING 217 Thus very early the Egyptians came to have this set of simple signs, sufficient in itself to express all that they thought and wrote. But they never realized its sufficiency, and continued to use their ideographs, syllabic signs, etc., side by side with their letters. As Taylor says, we "find a word spelt out alphabetically, a needless syllabic sign is then added, and this is followed by an unnecessary ideogram. So many crutches were thought necessary, that walking became an art of the utmost difficuhy." ^ All that re- mained to be done was to reject the superfluous mass of ideograms, homophones, syllables, and what not, and use the nearly perfect alphabet which had at last evolved itself. But the scribes clung to their ancient characters with a greater tenacity even than we do to our silent letters, and the writing of Egypt remained a confusion, their magnificent discovery going begging for a nation that could make use of it. During the latter half of the nineteenth century it was generally supposed that the modified Egyptian alphabet had been borrowed by the Semites, and had been put into convenient form by that businesslike Semitic people, the Phoenicians. Thence transmitted to the Greeks in com- mercial intercourse, it had been further modified and handed on to the Romans ; and thence, as we know, came our Latin script. Recent investigations, however, particu- larly the excavations in Crete reported by Sir Arthur • I, p. 68. 2l8 THE HISTORY OF READING Evans, render the theory of Egypto-Phoenician origin ex- tremely doubtful, if not impossible. Greece is far older than has been thought. A flourishing civilization has been shown to have existed in the ^Egean at least nearly 3000 B.C., with centers in Crete and probably later in Mycense. There was intimate intercourse between this civilization and that of Egypt about 2500 B.C. Works of art found at Mycenas show that Greece and Assyria were in contact fifteen hundred years before Homer's time, though Greek art had even then its own characteristic features.' Mr. Hogarth, in his "Authority and Archaeology" (p. 230), says: "Man in Hellas was more highly civilized before history than when history begins to record his state ; and there existed human society in the Hellenic area, organized and productive, to a period so remote that its origins were more distant from the age of Pericles than that age is from our own. We have probably to deal with a total period of civilization in the ^gean not much shorter than in the Nile Valley." And these people possessed an indigenous system of picture-writing and a system of signs which were at least syllabic, perhaps in some degree alphabetic. The ^gean script seems to have been in use long before Phoenicia existed. The ^Egean civilization only fell when Mycenae, its later center, though Crete was probably its place of origin, was overrun by the Dorians in the twelfth ' See Clodd, "Stoiy of the Alphabet," p. 187. tHE HISTORY OF READING 2ig century B.C. Phoenician history, on the other hand, hardly goes back of 1600 B.C., and Phoenicia's chance for commercial importance seems only to have come with the fall of the Mycenaean civilization. Between this time and the rise of the later Greece that we know, Phoenicia was dominant in the Mediterranean, and seems to have taken the alphabetic material that was to be found and given it a more practical form. However, she used ma- terials that were very much older than herself, and derived perhaps as much from' Greece as from Egypt and from other sources. The Cretan signs have similarities with the Eg)^tian and with the Cypriote or Cyprus syllabary, and with the little-known Hittite. But while there are certain indications of a common origin, it cannot be said that one is derived from the other. The busy Phoenicians adapted and unified the existing systems, changing and perhaps borrowing characters as needed, and the alphabet of later Greece was the result. Of some of the characters of this alphabet it seems possi- ble to trace the actual pictorial origin. For example, the Egyptian hieroglyphic owl, mulak, becoming the sign for its initial letter m, was conventionalized successively by the Egyptian scribes to the second, third, and fourth forms shown in Figure 31. The first character on the lower line is from the Semitic ; and, Taylor thinks, is a modification of the Egyptian form. Then follow thr^e successive Greek forms of the same, reaching our capital M. The 220 THE HISTORY OF READING ears of the owl still show in the upper peaks of the M, and the beak shows in the angle between. So the character AAAAAA, originally a picture of a "water-line," became the Egyptian n; and it is said that our n has come from it. And so for D and others of our letters. Whether or not these particular lines of Fig. 31.' — Successive Forms of the Letter M. descent are the true ones, certain it is that our letters and all letters seem to have been the result of just such evolution from primitive pictures ; and whether or not we are ever able to determine the particular ancestry of the whole Greek alphabet, we know that the development of writing, cul- minating in the production' of alphabets, has proceeded practically everywhere through the stages already sketched • This and the other cuts and quotations from Judd are reproduced, by permission, from Judd's " Genetic Psychology for Teachers," copy- right, 1903, by D. Appleton & Co. THE HISTORY OF READING 221 as typical. The final stage, the analysis of the word to its elementary sounds and their representation by an alphabet, seems to have been reached by only a very few indeed of the world's peoples, these few dwelling in and about the Mediterranean ; with the exception, perhaps, of the Mayas of Yucatan, who reached high-water mark in the New World by attaining, apparently, to the use of a few real alphabetic characters. By borrowing and by derivation from these few sources, some two hundred and fifty alphabets have come into being, from first to last. Some fifty of these survive, says Clodd, about half being found in India, and the rest being mainly variations of the Roman, Arabic, and Chinese scripts, the Roman constantly taking a further lead. The analysis into elementary sounds was made differ- ently, and in different degrees of completeness, by the dif- ferent peoples. A language may possess not more than a dozen consonants, as in the case of the Finnic or Oeguese, "or it may have a dehcate gradation of sounds like the Sanskrit, which requires no less than thirty-three conso- nants and fourteen vowels for its adequate expression. Some languages are especially rich in sibilants, others in gutturals, or nasals, or dentals, or hquids, or vowels. Hence either more or fewer symbols of a particular class are required." ' The Semitic alphabet was mainly consonantal, and so are ' Taylor, Vol. II, p. 368. 222 THE HISTORY OF READING usually the scripts of Asia which are derived from it, the vowels being only partially indicated. The Greeks made a further and better analysis, putting superfluous charac- ters to new uses and using separate lettefs to represent the vowels, "so that there might be a visible sign for every audible sound of the human voice." But great as is our debt to the Greeks for the improved alphabet which they JLLEVOLMSlMyi HICVllXDllEIM SV DXB n S PAT lA. BELGICiVytlMOL Fig. 32. — Roman Capitals. (From Judd.) bequeathed to us, we know that their ideal was only roughly approximated; and our use to-day of a large number of diacritical marks attests the persistent deficiencies of our alphabet. From the Greek alphabet to our own the steps are few and well known. The Romans adopted a form of the Greek alphabet in use among Greek colonists in Southern Italy ; THE HISTORY OF READING 223 and after certain modifications with use and as influenced by the contact with the later Greek, this alphabet became the vehicle of culture throughout Western Europe. For inscriptions on monuments and for other writing demanding prominence, the Romans used the forms shown in Figure 32, practically our modern capital letters, which we use very similarly. These simple, sharp-angled letters were very legible, and could very readily be chiselled on hard materials, but could never be written rapidly. The Romans early developed another set of forms, a rapid, running hand used in business and in correspondence, a specimen of which is shown in Figure 33. Then, as now, greater speed seems to have meant decreased legibility. Fig. 33. — Roman Cursive Script. (From Judd.) A compromise form between the capitals and the cursive, or running hand, was developed later in the uncials, as they were called, specimens of which are shown in Figure 34. Through the many centuries in which letters were made 224 THE HISTORY OF READING only by hand, down to the invention of printing, the forms varied very greatly, according as legibility, beauty, or ease and speed of writing were desired. When printing came, the makers of types selected forms that pleased them from the handwritings of the time, and letters soon began to take stereotyped forms. The German printers made the unfortunate choice of a complicated and compara- tively illegible Gothic script, and German readers still quoscuooco-; MOssesxriehJ T)$esiTu(v>ue Fig. 34. — Roman Uncials. (From Judd.) suffer the consequences. The English printers borrowed a beautiful running script from the Italians. This was an imitation, by the fifteenth century Italian printers, of the beautiful minuscule letters, which were small, cursive forms of the large uncials, the "inch" letters or "crooked" letters shown above. By this happy chance of the early English printers, Anglo-Saxon readers have, in the present somewhat modified forms of this script, a set of symbols which are easier to read and more convenient to use than any other forms. THE HISTORY OF READING 225 Thus by the slow processes of evolution, through varia- tion and selection, the characters used in reading have developed through the ages, — from the rude pictures of the cave-dweller to the printed characters of the modem type-setting machine. It is remarkable how small a part conscious purpose has had in this development, how little rationalization there has been of the characters. They have been a growth, as language has been ; and they have been allowed to carry down with them, from remote an- tiquity, useless but interesting marks of their origin, and rudiments of their stages of growth. CHAPTER XII THE EVOLUTION OF THE PRINTED PAGE In the earliest stages of pictography the pictures are arranged in almost every conceivable' fashion. The read- er's eye may traverse a page of picture-story with no con- straint as to direction of movement or sequence of atten- tion to the various symbols and parts. This is illustrated in the picture-letter shown in Figure 35, in which a A Picture Letter. Mandan Indian offers to a fur trader the skins of a buffalo, fish-otter, and fisher, in exchange { + ) for a gun and thirty beaver skins. In Figure 36 I quote another exam- ple, with description, from Deniker's "Races of Man." We go back to this protoplasmic free arrangement in 'From Wundt's " Volker-Psychologie. Bie Sprache," published by W. Englemann, Leipsig. 236 THE HISTORY OF READING 227 our modem cartoons, and in certain advertisements. It is the all-at-once view that we take of objects and situa- tions as directly experienced by the eye, or as we recall 'M.^-. ^Wc'nl S^tn miUdi. Fig. 36. (After Schoolcraft. •) — Petition of Chippeway Indians to the President of the United States. Example of pictography. "The petition is painted in sym- bolic colours (blue for water, white for the road, etc.) on a piece of bark. Figure i represents the principal petitioning chief, the totem of whose clan is an em- blematic and ancestral animal (see Chapter VII.), the crane; the animals which follow are the totems of his co-petitioners. Their eyes are all connected with his to express unity of view (6), their hearts with his to express unity of feeling. The eye of the crane, sym- bol of the principal chief, is moreover the point of departure of two Unes : one directed towards the Presi- dent (claim) and the other towards the lakes (object of claim)." them in our imagery. This primitive lack of fixed order in picture-stories is paralleled by the young child's lack of fixed order in his speech. To him the order of words is nothing, at first. As Dr. Lukens says,^ "He wants to say it all at once, anyhow, just as he thinks it all at once.'" And so likewise does he draw. He is fond of making • Reproduced by permission of Walter Scott Pub. Co., Ltd. ' Pedagogical Seminary, Vol. Ill, p. 459. 228 THE HISTORY OF READING Fig. 37 The Story of Joseph. picture-stories, and they are of this go-as-you-please order, as in the "Story of Joseph and his Brethren," told in pic- tures to Professor Lukens by some children. , Even in the primitive pic- it '^ % *ft I ^ 4 n ture-writing, however, there is Fig. 38. — Record of Departure evident a tendency to present (Innuit). (From Clodd.) . the symbols serially. Figure 38, from Clodd, illustrates this serial arrangement. Figure 39 is a himting story engraved by an Esquimau of Alaska on an ivory whip, and shows the same ar- rangement. ^N.Y. Teachers Magazine, April, 1899. THE HISTORY OF READING 229 In making picture-narratives, this serial order would naturally suggest itself, expressing in a spatial sequence the temporal succession of ideas in the writer's mind. His story would often be more inteUigible, too, when Fio. 39. (After Mallery-Hoffman.') — Journal of the Voyage of an Esquimau of Alaska. Example of pictography. The first figure (i) represents the story- teller himself, his right hand making the gesture which indicates " I," and his left, turned in the direction in which he is going, means "go." Continuing ovu: translation, we read the subsequent figures as follows : — (2) "in a boat" (paddle raised); (3) "sleep" (hand on the head) " one night " (the left hand shows a finger) ; (4) "(on) an island with a hut in the middle" (the little point); (5) " I going (farther) ;" (6) "(arrive at) an (other) isle inhabited" (without a point); (7) "spend (there) two nights;" (8) "hunt with har- poon;" (9) "a seal;" (10) "hunt with bow;" (11) "return in canoe with another person" {fwo oars directed backward) ; (12) "(to) the hut of the encamp- ment." (Deniker's " Races of Man," p. 138.) the order in which it was to be read was thus indi- cated. Very often, too, the material upon which the writing was done would favor the serial presentation, as in writing upon long strips of bark, or upon teeth, bones, or sticks. As the picture vmting developed and became more definite, the characters came to be arranged almost ex- clusively in series, in more or less regular lines ; and these lines came to have habitual directions which tended to • Reproduced by permission of Walter Scott Pub. Co., Ltd. 230 THE HISTORY OF READING become fixed for any given system. The Egyptian hiero- glyphs were sometimes ^arranged in horizontal lines, some- times in vertical columns. There was no fixed rule as to the direction in which they were to be written, but they I © ! Jk 5 1^. Fig. 40.* Egyptian Hieroglypli3. were read in the opposite direction to that in which the animals' heads pointed. Thus in the first extract above, from Budge's "Egyptian Language" (p. 11), "we notice that the men, the chicken, the owl, the hawk, and the hares, all face to the left ; to read these we must read from ' Reproduced by permission of Kegan Paul, Trench, Trubner, & Co. London. THE HISTORY OF READING 23 1 left to right, i.e., towards them." In the second extract the arrangement is in vertical hnes, to be read similarly in each hne. The Hittites read from right to left and then returning, left to right, as the ox plows. According to Hoffman the groups of characters composing their words, on the other hand, were read from top to bottom. With the Easter Islanders, the reading began in the lower left-hand comer and proceeded to the right, to the end of the line of picture-characters. Then, according to Hoffman, to read the next line above, the tablet was turned upside down and thus read again from left to right, to the end, when the reading of the third line began as the first. The Semitic writing, in general, was from right to left hke the modem Hebrew. If we should collate the various ways of arranging the reading symbols, it would be seen that almost every conceivable arrangement has been used, but that the tendencies have been everywhere toward arrange- ment in vertical or horizontal lines. Turning to the history of our own arrangement of char- acters, we find that the early Greek reading (not, however, that of the Mycenaean civilization, in respect to which there is imcertainty) was from right to left in each line, as with the Semites. Later, the reading came to be from right to left in the first line, from left to right in the second, etc. The characters faced, too, in the direction in which the reading was done, as in the inscription below, which 232 THE HISTORY OF READING read as printed in the third line, and which Taylor sup- posed was "the oldest Greek sentence in existence." Still later, the reading came to be from left to right in the first line, returning from right to left. This arrangement seemed to be more convenient for the scribe and was gen- erally adopted in consequence. Finally, the still more con- venient habit prevailed of writing and reading from left to right for all the lines, and this has continued to the present. On ancient inscriptions the words were sometimes sepa- rated from each other by dots or points. The early Fig. 41. Right to Left Reading in Early Greek. practice in Greek and Latin literary texts, however, was usually to write continuously without spaces or other divisions between the words. This, says Thompson, in his "Greek and Latin Palaeography" (p. 67), "was cer- tainly by far the more ordinary method, and in the imcial vellum manuscripts of the earlier Middle Ages it may be said to have been the only method that' was followed. In the documents of ordinary life the distinction of words was, from early times, more frequently though still only partially observed." Even when separation of the words THE HISTORY OF READING 233 gradually appeared, the prepositions were still attached to their related word, and there was always a tendency to de- tach a final letter and to attach it to the following word. It was hardly before the eleventh century that a perfect system of separately written words was established in Latin manuscripts. As early as Aristotle's time, according to Thompson, paragraphs were separated by a horizontal stroke or other mark drawn between the lines at their beginnings. Later the first letter of the new paragraph was placed farther to the left and also came to be enlarged, and thus the separation stroke came to be unnecessary and disappeared. Division of words at the end of the line was often avoided by writing the last letters smaller, or by linking two or more letters in a monogrammatic form. When the word had to be divided, it was an ancient practice to break off with a complete syllable. This was con- tinued in the later Greek and Latin manuscripts, though with many exceptions. The hyphen connecting parts thus divided did not appear until the eleventh century, although a point was used somewhat earlier, for the same purpose. The further breaking up of the written sen- tence by punctuation marks, quotation marks, etc., occurred gradually, principally during the eariier cen- turies of our era.' • See Thompson's "Greek and Latin Palseography, " pp. 67—71. 234 THE HISTORY OF EEADING The form of the modem book had its begimiings in the wooden, wax-coated tablets, something Hke our school slates, which were used from the earhest times in Greece and Rome. They were used for "literary composition, school exercises, accounts or rough memoranda." Two or more would be fastened together by ring hinges at the side, the raised margins of the tablet protecting the writing from being erased. Some such folded tablet seems to have existed even in "Homer's day.* Little booklets of tablets, called codices, came into very general use by the Romans, for correspondence, legal documents, etc. The convenience of this form made it gradually supplant the roll form that had been generally used among them, and codices, or books composed of vellum sheets instead of waxed tablets, became common at Rome even in the earliest centuries of our era. As the book form became more general, papyrus was also used for the purpose, as well as vellum. The arrangement of lines into two or more columns on the page was early adopted in these codices. Ordinarily a page had two columas, but three or four were also allowed. Thompson states that the three- column arrangement seems to have been "generally abandoned after the sixth century." Paper proper, which of course is very different from papyrus, although known to the Chinese at a most remote ' "Iliad," VI, 169; referred to by Thompson, p. 20. THE HISTORY OF READING 233 period, was not introduced into Europe until the eighth century, and came from the Arabs. It does not seem to have been used to any great extent by Europeans until the twelfth and thirteenth centuries. In the fourteenth century, according to Thompson, it "began to rival vellum as a material for books and in the course of the fifteenth century it gradually superseded it." Manu- scripts came to be composed, sometimes, of paper with a sheet of vellum forming the outer leaves of the quire. And so, step by step, was evolved the modem book, with the present arrangement of pages and columns, and Unes divided by spaces and marks into sentences, words, etc. The invention of printing stereotyped the forms that had up to that time foimd most favor or that were the most convenient for the mechanics of printing. This caused, to some extent, an arrest of the free develop- ment of forms of writing. At any rate, there resulted an immense limitation of the possibilities of variation, since very many had written where few could print. The history of the printed book and its adornment has been written elsewhere and cannot profitably be even sketched in this volume. Some references to the ten- dencies as to line-length, spacing, form of type, etc., will be found in a later chapter on present-day requirements of the printer. It would be verv interesting and suggestive if we could have a quantitative statement of the comparative amounts 236 THE HISTORY OF READING of actual reading done by the general population at various times in the world's history, and at various stages in the development of writing. Too little is known of culture history to furnish a safe basis for such estimates. The simpler forms of picture-writing could of course be read by all intelligent persons without special train- ing; so that at this stage practically all were readers, just as all young children are picture-readers. With conventionalizing of the characters and transfer of mean- ings, special training became more and more necessary and reading tended to become limited to certain privi- leged classes, especially, as was noted in our introduc- tion, to the priesthood. We have already referred to the difficulty of learning the Chinese syllabary; and in spite of the great reverence which the Chinese have for education, it is said that not more than one out of ten Chinamen can read. The Babylonian and Egyptian were also very difficult to learn, with their large number of characters and their many complicated features. Yet the Babylonians seem to have made education, in- cluding tablet-writing, compulsory on all free Chaldeans. Libraries were founded in all the chief cities of Babylonia. They had librarians, kept the books or tablets method- ically arranged and numbered, took out books by handing the librarian a ticket inscribed with the requisite number, etc. The great number of their writings which have come down to us through so many thousands of years THE HISTORY OF READING 237 mdicates that reading and writmg must have been very general among them. Egypt had an immense Hterature and great libraries, and manuscripts without number have been preserved to our ovsrn time. Among the upper classes of Greece and Rome, the povirer to read and write must have been almost as common as with ourselves, but the lower classes were illiterate. During the Middle Ages readers were very few indeed, in proportion to the total population, and down to the invention of printing it can hardly be said that any very large proportion of the people were able to read, or did read habitually. Indeed it seems certain that there could hardly have been, in any of the older civilizations, any remote approach to the number of readers or to the amount of reading per capita found in these days of printed books and papers. The difficulty of learning the complicated systems of writing of the early times, which would necessarily prevent the mass of the people from ever learning more than the barest rudiments at most, did not of course apply to Greece and Rome. But the cost of the materials from which books were made, and the fact that every one must be hand-made, and by what one may call skilled labor, would necessarily preclude their possession or use by the great majority of the people, who were even far less able to have luxuries in that day than in our own. Even among the privi- 238 THE HISTORY OF READING leged classes, books and other manuscripts were, from their comparatively great expense, necessarily much less abundant than in our time. The absence of inexpensive writing material was very important. The papyrus, made from an Egyptian rush or reed of that name, could not be produced cheaply; and although a sort of paper was made in early times from cotton, this material was very perishable and un- satisfactory. Forsyth, in his " History of Ancient Manu- scripts" (pp. 25-27), says that the use of linen rags for the manufacture of paper "was wholly unknown to the ancients. Indeed, they did not understand the manu- facture of flax at all, even if they possessed the plant." He quotes De Quincey as asserting that the ancients had repeatedly discovered the art. of printing. "The art which multiplied the legends upon a coin or medal had, in effect, anticipated the art of printing. It was an art, this typographic mystery, which awoke and went to sleep many times over from mere defect of materials. Not the defect of typography as an art, but the defect of paper as a material for keeping this art in motion. There lies the reason, as Dr. Whately most truly observes, why printed books had no existence amongst the Greeks of Pericles or afterward amongst the Romans of Cicero. And why was there no paper? The common reason, applying to both countries, was the want of linen rags, and that want arose from the universal habit of wearing THE HISTORY OF READING 239 woolen garments. . . . How desperate," he continues, "must have been the bankruptcy at Athens in all mate- rials for receiving the records of thoughts when we find a polished people having no better tickets or cards for con- veying their sentiments to the public than shells." Hence, as we know, came our word ostracize, from the practice of marking upon shells (ostraca) the votes for civil banishment. A similar poverty of material was shown by the Romans, according to Forsyth, in their use of "tickets of admission to the gladiatorial shows just hke tickets of admission to our own theatres," except that they were made of httle oblong pieces of lead, some of which have come down to us and are now in the British Museum. We find, in fact, that printing came very soon after paper had come into general use among the European nations; and the cheapening and increase of reading- matter through the two discoveries have been very great indeed. In consequence, as we have seen, reading and the reading habit ~ have become practically universal, in all civilized countries. In later chapters we shall have to consider certain disquieting results that come from this tremendous modem development of reading. CHAPTER XIII THE HISTORY OF READING METHODS AND TEXTS With the development of syllabaries and alphabets came reading in the modem sense, and also methods of learning to read. Among the early peoples who used an alphabet each letter was used for a definite pur- pose to represent a definite sound, and this mdde the letters of much greater importance than at present, and tended to the practice of reading and learning to read by letters. The ABC method of learning to read became general among the Greeks and Romans, and persisted to recent times in the Western world, though here and there an ineffective protest was made by educational reformers. It was different in some parts, at least, of the Orient, where the method of teaching to read was to place a book in the hands of the child from which he repeated the words in concert with his comrades until he knew them by heart, learning by imitation, in word and sentence wholes. Renan, in his "Life of Jesus," thinks that Jesus was thus taught to read. The Greeks and Romans, in teaching the child his letters, taught the combination of letters into syllables and words, and then of words into sentences. Various 240 THE HISTORY OF READING 24I devices were used, at times, for getting the pupil over the difficult alphabet stage. In one case a Greek pur- chased twenty-four slaves as playmates for his stupid boy, giving to each the name of a letter in the Greek alphabet. Quintilian, a.d. 68, advised giving the young child blocks and tablets containing the letters, to play with, and that he should be allowed to trace with a pen the forms of the letters as engraved on ivory tablets. And so there were innumerable devices for teaching the alphabet. A popular method of a later century was the gingerbread method, described as follows by Mat- thew Prior : ' — "To Master John the English maid A horn book gives of gingerbread, And that the child may learn the better, As he can name he eats the letter. Proceeding thus with vast delight He spells and gnaws from left to right." Basedow (1723-1790), who taught that the child should learn to read by playing, strongly advocated this gingerbread method. The school should have a special school baker. "The children must have breakfast, and it is not necessary for any child to eat the alphabet more than three weeks. The cost of shaping the dough Into letters is less than one-half peimy daily for each '"Alma," Canto two, quoted from Reader's "Development of School-Readers. " 242 THE HISTORY OF READING child. This makes three pence a week or for four weeks a groschen. The acquisition is entirely worth so much and is possible even to the poor children." ' Various mechanical devices were contrived to facili- tate the manipulation of letters in script and print, in grouping them into syllables, words, and other com- binations. Other devices were primarily to interest the child in the letters. The development of methods proper will be traced further after we have given an account of the development of primers and reading texts. The early primers were all books of religious instruc- tion, and their content was determined and limited by the authority of the Church. In the Abecedarien of the ninth century the alphabet and ab, eh, ib colunans were followed by the Credo and Paternoster; later the Ave Maria and, soon after the thirteenth century, the Benedicite and Gratias were included. From Charles the Great until Luther, no other material than the above appeared in school readers. The early primers of the Reformation were not only school books but manuals of church service. The German word for primer, Fibel, appeared in 1419, and signifies a little Bible. Henry the Eighth forbade the printing of un- authorized primers while a Catholic, and issued his '■'Reform Primers" in the interest of the true doctrine ' Kehr's " Geschichte des Leseunterrichts, " p. 59. THE HISTORY OF READING 243 when he became a Protestant. "Alphabet and creed became united in one book which became the forerunner on the one hand of the book of Common Prayer, and on the other of the modem school primer." ' The first Protestant primer, however, by Philip Me- lancthon, had no inconsiderable quantity of secular material. In addition to the usual Catholic content and some extracts from the New Testament, there were fourteen pages of the sayings of the wise men of Greece. Luther's primer followed the fashion of the Catholic primers of the time. The ABC book by Schulte, pub- lished in 1532, made one of the earliest attempts to adapt to the child's interests. The letters were presented with pictures and in rhymes, introducing the jingle in which the child soul revels. A form, that was much followed in the early English primers, ran as follows: — "H h Hase H h Hammer. Gebratne Hasen sind nicht boes. Der Hammer gibt Gar harte stoess. K k Katze. K k Kamm. Die Schlaue Katze frisst die Maeus. Der Kamm herunter dringt die Laues." ' The Puritans brought with them to America an A B C Catechism which was succeeded by the famous New England Primer, about 1690. "For more than one * Reeder, " Development of School Readers," p. 10. ^ Ibid., p. 12. 244 THE HISTORY OF READING hundred years," says Reader, "the New England Primel had the field in America against all coiners, and for half a century longer it continued to be used in the schools." Its total sales are estimated to have been not less than three million copies. This primer was a Church book, but had enough of secular matter to make it "a step in the direction of a secularization of the course of study." It contained the alphabet, lists of the vowels and con- sonants, lists of syllables such as db, el, ib, etc., lis|:s of words for spelling arranged according to the number of syllables; rhymes with illustrative wood-cuts for the letters in order, as in the cut ; moral injunctions, prayers, catechisms, etc., for the children, including the "Now I lay me down to sleep," which was apparently written for this primer and which has come to be "the dearest prayer of American childhood." This little book, pres- ent with the Bible in every home, had a profound in- fluence on the moral and religious thought of the whole coimtry. It is said to have been "the daily companion of President John Adams throughout his long career." When it went down, after more than a century of un- disputed sway, it continued to exert "an abiding influence upon the quality of its numerous successors." Beginning as early as 1450, the Horn Book, as it was called, came to be more and more the means by which the English child learned his first use of letters and words. It was used extensively in England down to the beginning THE HISTORY OF READING 245 6, Hh- T'^ >jO ffiTJ S" o a o 3 c S'M a ti ^g p- » Er« E.M P>3 - 9 iw ^•s. S " 3 W » 7, 3 a P-a lire A John rmissi ■ ? S ' 5* P' §2^ (5 — TO 3 o o o' O !» » "3 GQ » - ^ 3 o cI hhs 246 THE HISTORY OF READING of the nineteenth century, and in our schools as well. The cut below, from Johnson's " Old-Time Schools ,-AabcdefghjjkIinnopqiJI Irfit u vw xy z& «e}oo| |ABC0i fghuk.la:?.»op( |RSTUVV/XYZ aeiou aeioa lib eb ci^ j ad ed ; J od uc? dad; dido da j Ilnihe Nwieofibe Fatht/ and of the J I Son, and of tJie Holy Qhoft. jiimn. kUR Faiher.vvhichsLrtJiL "Heaven.hallowedbethyl |f'-'3rtK , 'i'yKingdomcome.thyl 8 Wyiii be lone onEartb.as it is ial pie»ven. GivcusthisDayouj^ idaily Breai!; 1*^4 forgive usourf |Trerpsale9,8', wc. forgive them] %hst trefpafs agaiaft us : Andf lead! a» not intoTeiBpcatt(3n,bGt I 'e!iver uj fmia Evil. >fe^;j. i ,' Fig. 43. — a Typical Horn Book. and School Books, " shows all that there was of it. It was a paddle, with a card of printed matter tacked upon it under a protecting sheet of horn. THE HISTORY OF READING 247 In England the Battledore paddles came to be trans- formed into wooden primers in a similar manner, and were used both for play and for lessons. In America the little girls of colonial times very often wrought out their own primers with needle and thread, in samplers ■ ^' '■- ■^''- ' * " J!L ^^ ' ' '\k\mn oprfn rr \y xw ^^^<^ ' SM<^'^S;'0 ^ 37PS- VJWMAAMMAMMIHP . .1- m*- frccurc vvu^^ur mme can loii ; fct to fpend my Ume in riO>igfit ^^^^'S^^;^^;?^^^?'^^^!^^ P Q Fig. 44. — A Sampler. (From Johnson.) containing the alphabet with vowels and consonants, bible quotations, prayers, verses, and sometimes illustra- tions, in various designs and styles of type. As the New England Primer declined in America, the spelling-book took its place as the book for beginners. The spelling-book combined the alphabet, primer, speller, 248 THE HISTORY OF READING and reader in one book, and often included other subjects as well. Webster's Spelling Book, published in 1783, soon displaced the few spellers previously introduced, and came to be used almost universally throughout the country. In 1785 five hundred copies a week were being sold, in 1818 the total number had reached five million, and to 1847 the total sales had amounted to forty-seven million. In i88g Commissioner Harris stated that twelve hundred thousand copies were then being sold annually, and that it was "the most generally used of all school text-books." In 1900 it was still being sold at tne rate of hundreds of thousands annually. The book contained long lists of words arranged according to length, a large number of names of persons and places, illustrated fables for reading lessons, and short sentences for beginners in reading. Supplementary matter, such as numbers, abbreviations, moral instructions, a cate- chism, etc., appeared variously in various editions. Arti- ficial in its arrangement of words, thought, and vocabu- lary, most ill-adapted to the needs of its users and to the various ages of children, it yet served an important pur- pose in its earlier days, and through its universal use, in "reducing a dozen local dialects to one harmonious language," and bringing about "that remarkable uni- formity of pronunciation in our country which is so often spoken of with surprise by English travellers." ' ' Scudder, "Life of Noah Webster," pp. 38-39, quoted by Reader. THE HISTORY OF READLNG 249 The list of geographical names below, quoted by Reeder from Webster's Spelling Book, illustrates the choice of words, arrangement, etc. : — A bac' o Cat a ra' qua Schuy' ler Wa que fa no' ga A bit' i bis Cat te hunk' Scoo' due Win' ni pic A ca' di a Chab a quid' ic Shen' brun Win ni pis o' gy A quae' nac Chat a ho' chy Sho' dack Wy a lu' sing Fig. 44V2. — An Old Spelling Lesson. Webster seems to have published the first American school reader, about the same time as his speller. Be- sides selections intended directly to instruct the youth in morals and rehgion, it contained dialogues, narratives, and many selections from American statesmen and patriots of those revolutionary times. Webster's reader was not so successful as his speller. Several rival readers, made on somewhat the same plan, divided the field with him. The preference of their makers for the productions of American genius "resulted in the selection of much that was commonplace and the omission of most that was really great." Primers of various sorts seem to have abounded during the early part of the nineteenth century, but they took httle account of method in teaching beginners to read. If they contained anything beyond the illustrated alpha- bet, it was the catechism or other moral or religious con- tent. But by the end of the first quarter of the century, primer-makers began to attend to method and adapta- tion as well as to matter. For example, Keagy's Pes- 2 so THE HISTORY OF READING talozzian primer, of 1826, contained a series of "thinking lessons," a beginning of object-lessons. "The size, shape, color, number, origin, and use of common articles of the household, the street, and the field were to become rallying points for pleasing and useful thoughts. Ex- ercising the pupils in handling the objects was recom- mended wherever practicable. It was probably the first primer published in this country in which there was a distinct purpose to make use of the child's en- vironment in an educative manner."^ The plan and arrangement, however, were exceedingly crude. Early in the nineteenth century the readers began to be graded somewhat as to subject-matter, appearing in series of two and three books each. There would be an Introduction, a Middle Book, and a Sequel, a Primer and a Spelling Book completing the series. In 1828 Putnam's series introduced a custom that has been much imitated, that of doing the work of a dic- tionary in defining the difficult words and phrases. Wor- cester's series of readers, published in 1828, contained a primer which seems to have been "the first American primer to advocate the word-method." Of this more further on. Pierpont's series of readers, beginning about 1823, omitted the usual treatises on inflection, emphasis, accent, punctuation, etc., as being little used, insisting that "reading, hke conversation, is learned from example ' Reeder, p. 43. THE HISTORY OF READING 25 1 rather than by rule." The selections were taken mainly from the writings of Jefferson, Patrick Henry, Webster, Irving, etc., in the behef that our country had a "char- acter of its own," physically and morally, which should be learned by the children while at school. The series was long a popular one, and on its merits, with little pushing by publisher or author, and set an excellent example in its choice of "literature as the proper field for, subject-matter." The chief competitors of Webster's first American Reader had been Bingham's "Columbian Orator," Bing- ham's "American Preceptor," and the "English Reader" by Lindley Murray, an English author, with the Introduc- tion and Sequel to the "English Reader." The "English Reader" continued to be very largely used in American schools during the first half of the nineteenth century. In literary worth it much surpassed its early American rivals. It had many selections from the best English poets, but still more of moral and didactic matter, prov- erbs, Bible stories, dissertations on Virtue, Friendship, Comforts of Religion, etc., pathetic pieces, public speeches, etc., with little adaptation to the needs of the young. A series of readers by Cobb, begun about 1831, made some effort to interest the child by means of stories, in- formation about animals, etc., and the author made a strong appeal to American patriotism in support of his readers as against the "English Reader," then so generally 252 THE HISTORY OF READING used. By 1844 more than six million copies of Cobb's Readers had been sold. First Readers gradually took the place of spellers as introductory to reading, though the spellers remained in constant use. The readers came to give much instruction in correct articulation and in elocution generally. The vari- ous series that appeared each embodied some characteristic feature which publishers made the most of, as nowadays. McGuffey's six-book series, which appeared in 1850, has, according to Reeder, "probably attained the largest sale and widest distribution of any series yet produced in America. In range of subject-matter it swept almost the entire field of human interest, morals, economics, politics, literature, history, science, and philosophy. Many a profound and lasting impression was made upon the lives of children and youths by the well-chosen selections of this series, and valuable lessons of industry, thrift, econ- omy, kindness, generosity, honesty, courage, and duty found expressions in the after lives of millions of boys and girls who read and re-read these books, to the in- fluence of which such lessons were directly traceable." From i860 to 1880 the character of school readers seems to have undergone little change, but changes in method were taking place. The word-method had again been advocated in the Bumstead Readers of 1843, ^iid ap- peared again in 1860 as "new and original" in the "Word Builder," the first book of the National Series of readers. THE HISTORY OF READING 253 Reading books had been taken into the service of the school subjects as early as 1824, in the "Agricultural Reader" by Daniel Adams. In 1827 appeared a "His- torical Reader," by Rev. J. L. Blake, and reading was later taken into the service of the various sciences, nota- bly in the Willson seven-book series of i860. The latter seems to have marked the culmination of the tendency to utilitarian specialization in the choice of subject-matter for readers, as against the literary excellence shown in such readers as Pierpont's and in Murray's "English Reader." "In the new series and supplementary readers, which began to appear about 1880, hterature took the field and since then has held it against all comers." It was the beginning of a new epoch in the development of school readers.' Science and the other departments of knowl- edge, while rather losing their place in the reading-books, have been given their own place in the present enriched elementary curriculum. Since 1880 the subject-matter of readers has been taken mainly from the field of hterature, and the problem has been one of selection, arrangement, and adaptation within this field, the tendency being toward the use of literary wholes instead of the earher selection of scraps. The scrap compilations of the school readers were scathingly denounced by Horace Mann as early as 1849, but with httle effect. President Eliot forcibly re- newed the criticism in his article in the Educational Review ' Reeder, "Development of School Readers," p. 56. 254 THE HISTORY OF READING of July, 1 89 1, arguing for the use of real literature and literary wholes in the readers as against the literary scraps and trash of most of the books. With the appearance of the supplementary readers, about 1880, came a tendency more and more to present literary wholes, condensations of such classics as "Hiawatha," "Robinson Crusoe," "Ivan- hoe," etc. Most of the present-day series of readers are based on literature as the subject-matter, and the "Heart of Oak" series, a six-book series edited by Charles Eliot Norton, perhaps marks the extreme of this tendency to "read for literature's sake," as contrasted with the other extreme represented in the Willson books. After this review of the contents of school readers, let us now return to the history of methods of learning to read. The alphabet method, in spite of occasional protest, was almost universally used from the Greek and Roman times until some thirty years ago, and of course has not been entirely discarded even yet. In this method the child learned first the names of the large and small letters, and their order in the alphabet. This was task enough, uninteresting as it was to many, to keep them employed for some months, or even in some cases for a year or more Then the combinations like ab, eh, ib, were spelled out and pronounced, and then three-letter combinations hke glo, flo, pag, etc., in all of which the early pages of the old spellers abounded. Then monosyllables and gradually longer and longer words were used. SpeUing the word THE HISTORY OF READING 255 preceded its pronunciation, until it was known well. "It was assumed that there was a necessary connection between naming the letters of a word and pronouncing the word." "No other approach to the pronunciation of the printed symbol was imagined by the great majority of teachers." '■ The alphabet method had early modifications in Europe on the side of phonetics. As early as 1534 Ickelsamer had a device of "placing the picture of an animal, its printed name, and the letter whose sound was most like the animal's voice or cry in parallel columns. Against the picture of a dog, for example, was placed the growhng r, against a bird the twittering z," etc.^ Later, A was associated with Apple, B with Boy, etc., and in this cen- tury we have seen various imitative picturings of the sounds of the letters, as of w by a cow lowing, sch by children driving away hens, etc. The philanthropinists, in Germany, had their boys personate the letters by their dress and actions; for ex- ample, / by "dressing in helmet, big necktie, and stilts," w by twisting their bodies into its shape, etc. Such methods had, as one of their results, the lessening of at- tention to the letter's name, in favor of its sound or visual form. Germany much earlier than America began to realize that spelling was not the only or the best approach ' ' Reeder, "Development oi School Readers," p. 63. 2 HaU, "How to Teach Reading," p. 2. 256 THE HISTORY OF READING to reading, but the spelling, method held its ground there until well into the nineteenth century. ^ Outside of the illustrations of the alphabet which we have noticed, the first illustrated schoolbook seems to have been Comenius' "Visible World, or a Nomen- clature, and pictures of all the chief things that are in the world, and of men's employments therein; in above an 150 Copper Cuts." This book, the "Orbis Pictus" jTifans ejular, e^ £ The Infant cr'ieth. Ventui fiat, // The Wind bloaeik An/er gmgnt, gaga The Goofe gagletb. Ojhalat, h((hha'h The mouth ireathethm- £e Ff Hh Fig. 45. — Part of an Illustrated Alphabet in the "Visible World." (From Johnson.) as it is usually known, published in Nuremberg in 1657 or 1658, was for more than a hundred years the most popular text-book in Europe. Reeder calls it "the first attempt at object-lesson instruction, and the beginning of the word-method in teaching reading. It was translated into ten European and four Asiatic languages" (p. 67). In the "Orbis Pictus" each subject had its picture, with explanatory sentences below in Latin and in EngUsh THE HISTORY OF READING 257 The BarbersShop. LXXV. To;iJirha» lie Bather, i. in rAff Barbers-fliop, 5. cuttethofftheHsit and the Beard vithapairofSizzais, 3. er Jhaveth with a Razor, which he takfth out of hit Cafe, 4. /ind he waflieth me ffver a Bafon, $. with Suds ruimiiig cut of a Laver, 6. andalfo with Sppe, 7. tmdwipeth him with a Towel, 8. combeth him with a Comb, 9. and curleth him with a Crifping Iron, lo. Sometimes he cutteth a Vein w/Vfc <« Pen-knife, 11, where tbeBlndf$imtbm,il> TonfoT, I. inTonftritia, 2, tondet Cr/n« & Barbam Forcipe, 3. vel radic Novaciil3, quam e Tieca, 4. deptaaaV Et lavat fuper Fe/ww, 5, i.»*h;w defluentc hCuttitrnio, 6. ut & Sapone, 7. & tergit Linteo, 8. peftir PeHine, g. crifpac CaUmifiro, lo. Interdum Venam ftcae ^calpeBo, ir. ubi Sanguis proiSttliuIar, 12. Fkj. 46. — a Page showing the Method of Teaching in the "Visible World.' (From Johnson.) 258 THE HISTORY OF READING or other language. In his preface Comenius says: "The very looking upon the thing pictured suggesting the name of the thing will tell the child how the title of the picture is to be read. And thus the whole book being gone over by the bare titles of the pictures, reading cannot but be learned — and indeed, too, without using any ordinary tedious speUing — that most troublesome torture of wits." However, Comenius was far beyond his times, and his book was little used as such a method of learning to read. There were glimpses of better things in the phonetic system of the Jansenists, and in the primer of Gedike, in 1791, which advised teaching words before letters, as the natural order is from the whole to the parts ; but none of these had appreciable effect in changing current ABC practice until Jacotot (17 70-1840) advocated the word- method as a part of his system, and set forth clearly the arguments for it. In America, Worcester's Primer, in 1828, seems to have been the first beginners' book to recognize any other than the alphabet method. The author says in his preface : "It is not, perhaps, very important that a child should know the letters before it begins to read. It may learn first to read words by seeing them, hearing them pro- nounced, and having their meanings illustrated; and afterward it may learn to analyze them or name the letters of which they are composed." Bumstead, in the first THE HISTORY OF READING ^59 book of his series of readers published in 1840-1843, stood stoutly for the word-method, and urged that a scholar be never required to spell a word "before he has so far learned it as to be able to read it." Horace Mann had already advocated the word-method for years, and ridi- culed the en-o — no, pee-you-tee — put, tee-aitch-ee — the, way of begiiming reading, as it was taught in Webster's Spelling Book. As early as 1790 Dr. Thornton, head of the Patent Office in Washington, had issued a pamphlet proposing that letters be named as they sound; and, as there are more soimds than letters, he introduced new letters to supply deficiencies, making a phonetic system such as we have seen much of in recent years. But the ABC method and the reading by spelling went on with little disturbance from these protests. Reeder calls attention to the fact that even in advanced reading "analysis played the leading r61e" (p. 78). Pupils would "spell and define the words, tell their synonyms and op- posites, write and paraphrase the sentence or paragraph, analyze and reduce it to its simplest sentences," etc., sometimes spending twenty to thirty minutes on six or eight lines. J. Russell Webb, author of the Normal Readers, did much to bring about the adoption of the word-method, and by 1870 it began to be adopted by progressive teachers in various parts of the coimtry, and gradually grew in favor. 26o THE HISTORY OF READING The phonic method, so early used by the Jansenists, helped also to displace 'the alphabet method. In the phonic method, the words are spelled by producing the succession of sounds forming them. As there are some forty-four sounds, new characters must be added to the usual twenty-six if the system is to be complete. If the child is able to successively reproduce the sounds of the letters as they stand in a word, he can learn for him- self to pronounce new words as they appear, a great ad- vantage of the phonic method over the word method. As for the alphabet method, it was easy to show that knowing or saying the letters' names gave no clew, necessarily, to a new word's sound. The phonic method was tried in various parts of the country, and met with great success for a time, developing into what came to be known as the phonetic method, notably in the "Pronouncing Orthography" system of Dr. Edwin Leigh, pubhshed in 1864 and patented four years later. In this system the letters were given various special forms to represent their dififerent sounds, these forms being slight modifications of the ordinary form. Silent letters were printed, but in hair lines. The method is further described and illustrated on a later page.* This system was used in a series of readers by Leigh, and in several other series, including McGuffey's. It was introduced into the schools of St. Louis, New York, / Washington, Boston, and other large cities. It met with ■ See Fig. 48. THE HISTORY OF READING 261 great success, but only for a short time. The "pronounc- ing print" was hard on the eyes, requiring an unnaturally close inspection of each letter, in the beginning ; besides, it made trouble for the printer, distracted from attention to the thought in reading, and caused confusion in the at- tempt to use two alphabets. The sentence method was more or less used, here and there, as early as 1870, and indeed was advocated by occa- sional writers very much earlier, as we have noted. It was not very generally used until as late as 1885 or 1890. Since then there have appeared a very great variety of modifica- tions and mixtures of all these methods, devices for mak- ing them interesting to the child, arrangements for corre- latmg the begiimings of reading with writing, drawing, number work, etc. There has been development simul- taneously along so many and so conflicting lines that historical treatment seems impossible in any brief compass. I shall, however, endeavor to briefly describe the methods that are now in most general use or that have much of promise or suggestion, and shall not6 the present trend of practice among the better teachers of reading. PART III THE PEDAGOGY OF READING CHAPTER XIV PRESENT-DAY METHODS AND TEXTS IN ELEMENTARY READING The methods of learning to read that are in common use to-day may be classed as alphabetic, phonic, phonetic, word, sentence, and combination methods. The special systems of teaching to read, which now pass under the names of their authors, are usually but specially adapted means of using one or another of these standard methods. A brief account of these standards will therefore pave the way for an accoimt of the concrete systems now in vogue. The alphabet method, used almost imiversally in Greece and Rome, and in European countries generally imtil well into the nineteenth century, and which was nearly universal in America until about 1870, is now chiefly of historical interest. However, there are innumerable cor- ners of our country, a little removed from the centers and thoroughfares of civilization, in which the alphabet method is still "the good old way." In this method, as we have seen, the names of the printed or written: letters are first taught, and the order of the letters in the alphabet. Some- times the sounds of the letters are also taught. Then nonsense syllables hke ab, ib, ob are spelled and pro- 26s 266 THE PEDAGOGY OF READING nounced ; then combinations of three letters, monosyllabic words, dissyllables, etc., follow, the word usually being spelled before it is pronounced. Just how naming the letters was supposed to assist in pronouncing the word it is difficult to see. The value of the practice in. learning to spell doubtless had much to do with blinding centuries of teachers to its uselessness for the reading of words and sentences. However, in dealing thus constantly ^with the letters and their combinations, the pupil necessarily acquired a familiarity with the soimds represented by each letter, whether purposely taught these or not. And thus this method always combined something of phonics as well. The phonic method, used by the Jansenists in the Port Royal Schools, long neglected but advocated again by Thornton in 1790, began, as we have seen, to be exten- sively used as a special method in this country in the system of Leigh, about 1 870-1 873. It is a spelling method, but the word is spelled by its elementary sounds and not by the letter-names. The word is slowly pronounced until its constituent sounds come to consciousness, and these sounds are associated with the letters representing them. Drill in this sound analysis trains the articulation, trains the ear and the ability to sound the letters of any new word, and gives the power to pronounce it by blending the soimds suggested, — provided there are no silent let- ters and provided the sounded letters represent but one THE PEDAGOGY OF READING 267 sound. This seldom occurs, and the reader of new words must be helped out by context or conjecture. Both Thornton and Leigh met the difficulty by contriving addi- tional characters to represent the other sounds after one 2 And bofh Jesus was called, and his disci'- pies, te ih.T^ marriage. 3 And when iJiay wanted wine^ tk^d mntiier of Jbsus soiifh unte Lim, TBiay have no wine. 4 J"esus saifh unte her, Woman, what have I te de with ttiBe? mine honr is not yet come. 5 His mr>ti.er setith unte thB servants, What- soev'er h^ saifh unte you, de it. 6 And thare were set thare six waterpots of stone, after the manner of th"e purifi^igg of ■tiiB Jeujs, containing twe er fknee firkins apiBce*. 7 J^sus saith unte them, Fill the water- pots with water. And tiiey filled liiem up te 1h.B brim. 8 And h^ saith unte liiem, Di*aw out now, and bear unte th^ guvernor of thB feast. And I3iay bare it. Fig. 47. — A Specimen of Leigh's Print. sound each had been allowed to the twenty-six regular letters. The forty-four or more sounds used in Enghsh needed as many characters, and when these were furnished the method came to be known as the phonetic, to distin- guish it from the simpler phonic. Leigh made the addi- 268 THE PEDAGOGY OF READING tional characters by slightly modifying the existing letter- forms, and silent letters were printed in hair lines, as shown in these extracts (Figs. 47 and 48) from his article in the Report of the National Educational Association, 1873. A special form of a letter is used for each sound of it. The hair-line letters are silent. The pronunciation is according to our- staadard dic- tionaries, .Webster and Worcesteft < The 8 pairs of .vowels, the diphthongs, and the semi- vowels (w y) are- eel it,. ale ell, air at. art ask ;. urn up, or on, old foZks, fool foot, ice oil our sue, use. we-ye. e 1, a e, a a, a.a ; u u, e o, o o, e o. ioi cm u,u. wy! The aspirates, liquids, nasals, and the 8 pairs of consonants are— . " f&r, (sirg?4nt). " fist " iAri, Hiii^.- " fit, as, fe?t, pique. ( Vol. II, pp. 442-444- THE PEDAGOGY OF READING 377 chroniclers and romancers compare with the Homeridje; the quest of the Grail with the argon au tic expedition for the Golden Fleece; Vivian with Circe; Merlin with Nestor; Asgard with Oljmipus. The northern myths are more sublime and less beautiful; content predominates more over form; there is more of the best spirit of modem romance, and woman's position is higher. This rich field represents perhaps the brightest spot of the dark ages and the best expression of feudalism. It teaches the highest reverence for woman- hood, piety, valor, loyalty, courtesy, mtmificence, justice, and obe- dience. The very life blood of chivalry is heroism. Here we find the origin of most of the modem ideas of a gentleman, who is tender, generous, and helpful, as well as brave; the spirit which has given us Bayard and Sidney, as well as the pure, spotless, ideal knight, Sir Galahad. These stories are not mechanically manufactured, but they grew slowly and naturally in the soul of the race. They, too, shape and direct fear, love, pity, anger, essentially aright. The Anglo-Saxon writer never legislates more wisely for the feelings or for the imagination than when he is inspired by and uses this material well. It stirs those subtle perceptions, where deep truths sleep in the youthful soul before they come to full consciousness. Although they have no very definite geography or date, so that such events and persons existed nowhere, they might be realized anywhere. To the mind at this stage of growth nothing seems quite complete or quite actual. The air whispers secrets of something about to hap- pen, because to nascent faculties the whole world seems a little mys- tic, though very friendly. It is this kind of muihos that is the mother of poetry, religion, art, and, to some extent, of morals, philosophy, and science. It is not very examinable material, for it works too deeply and unconsciously, and the best and largest objects of the soul have not yet come to consciousness at this age, but the great lines of cleavage between right and wrong, beauty and ugliness, truth and falsehood, are being controlled, and the spiritual faculties developed. Morals and aesthetics, which are never so inseparable as at this period, are here found in normal union." . . , "If 378 THE PEDAGOGY OF READING we have anjrwhere the material for an ethnic Bible left at the moa' interesting and promising stages of incompleteness by the advent of the alien culture material brought to the Teutonic races by Chris- tianity, it is here. I have looked over eight of the best known popular digests of all or principal parts of this matter and many lesser para- phrases, but do not find quite the right treatment, and I believe that a great duty is laid upon high school teachers now ; namely, that of reediting this matter into form that shall "be no less than canonical for their pupils. Pedagogic art is often, as Walter Pater says of art in general, the removal of rubbish. Excrescences must be elimi nated, the gold recoined, its culture power brought out, till, if the ideal were fully realized, the teacher would almost become a bard of these heroic tales, with a mind saturated with all available literature, pictures, and even music bearing on it, requiring written and oral reproduction from the pupils to see what sinks deepest. Some would measure the progress of culture by the work of reinterpreting on ever higher planes the mythic tradition of a race, and how this is done for youth is a good criterion of pedagogic progress." Perhaps we shall have, by-and-by, such a collecting and editing of this material as has been done for German young people in "Das Deutsche Lesebuch," a ten- volume work of over thirty-five hundred pages. In the preparation of this great "Reader," "many men for years went over the history of German literature, from the Eddas and Nibelungenlied down, including a few living writers, care- fully selecting saga, legends, Mdrchen, fables, proverbs, hymns, a few prayers, Bible tales, conundrums, jests, and humorous tales, with many digests, epitomes, and condensations of great standards, quotations, epic, lyric, and dramatic poetry, adventure, exploration, biography, THE PEDAGOGY OF ILEADING 379 with sketches of the life of each writer quoted, with a large final volume on the history of German litera- ture." ' Until this is done for English literature, and indeed always, more or less, we must make our selections with the help of trained librarians, who more and more are becoming ready and efficient assistants and advisers ia directing the reading of youth. Extended discussion of the selection of reading-matter belongs rather in a trea- tise on the study of literature than in such a volume as this; but the few suggestions that have here been made seem to belong here properly in view of the widespread neglect of the real nature of youth in the] choice by the schools of what shall be read.' Finally, we may briefly summarize the practical peda- gogical conclusions which have seemed to be warranted in our study. I. The home is the natural place for learning to read, in connection with the child's introduction to literature through story-telling, picture-reading, etc. The child will make much use of reading and writing in his plays, using both pictures and words. The picture writing and read- ' "Adolescence," Vol. II, p. 480, note. ' " The Children's Hour," a ten-volume series edited by Eva March Tappan and just issued by Houghton, Mifflin & Co., comes to my notice as this goes to press. It seems to fiU many of the requirements of a comprehensive and careful selection of the best literature for children. 380 THE PEDAGOGY OJ' READING/ ing of primitive peoples has a wealth of suggestion for such practice. 2. The school should cease to make primary reading the fetich that it long has been, and should construct a primary course in which reading and writing will be learned secondarily, and only as they serve a purpose felt as such by the pupil, the reading being always for meanings. 3. The technique of reading should not appear in the early years, and the very little early work that should be tol- erated in phonics should be entirely distinct from reading. 4. The child should never be permitted to read for the sake of reading, as a formal process or end in itself. The reading should always be for the intrinsic interest or value of what is read, reading never being done or thought of as "an exercise." Word^pronouncing will therefore always be secondary to getting whole sentence-meanings, and this from the very first. 5. There should therefore be much more practice in silent reading than in reading aloud, the latter being practiced not as an exercise in reading, but in the effective use of oral language. 6. Until the speech habits are well formed, the school should have much more of oral work other than reading, than of work involving reading. Grammar and other analytical study of language should play little part in train- ing to the correct use of the mother-tongue, in all the lower grades. THE PEDAGOGY OF READING 381 7. The learning of real literature should begin in the home and in the very first days of school, and should con- tinue uninterruptedly, the literature being presented by the living voice and with the help of pictures and drama- tization, for a good while, the children reading for them- selves as fast as their interest demands. School readers, especially primers, should largely disappear, except as they may be competent editings of the real literature of the mother-tongue, presented in literary wholes, or as they may be records of the children's own experiences and thoughts, or as they may be books needed for information in the everyday life of the school. The children should learn to read books, papers, records, letters, etc., as need arises in their life, just as adults do, and they should be trained to do such reading eEFectively. 8. The children should from the first read as fast as the nature of the matter read and their purpose with it will per- mit, but without hurry. Speed drills in the effective gather- ing of meaning from what is read will be very beneficial. 9. The reading of the mother-tongue may be done so as to discipline the mind at least as effectively as in the reading of ancient languages. To this end the pupil should be practiced in grasping the essential meanings, in selecting and gathering from books and papers what they have for his purposes, in ignoring the irrelevant, and in feeling values always. 10. Most of the time usually given to "exercises" in 382 THE PEDAGOGY OF READING reading aloud, etc., will be far more productive if spent in learning the effective use of the library, of indexes, books of reference, periodicals, in learning to make notes, ab- stracts, reviews, and to make effective use of these for the reader's purposes. 11. Far more extensive reading should be done in the upper grades and in the high school, as compared with the usual intensive analytical study of a few texts and authors. Analysis generally should give place to synthesis until the college period at least; and especially at ado- lescence the individual tastes, even though capricious, should be given as wide a range of choice as is possible. 12. The reading and hearing of literature is to be de- pended upon to impregnate the soul with the race's highest ideals and tastes. To this end reading, as the study of literature, should be of what our race has voted best, or classic, in its successive stages of culture, the child and youth roughly recapitulating these stages in reading interests and needs. The literature of Teutonic feudalism . and chivalry and of mediaeval romanticism seems especially suited to the nature and interests of adolescents. 13. Reading of the mother-tongue, learned and always used as a means and not as an end, done effectively and as rapidly as is natural and possible, done so as to serve as an effective discipline, real reading, is to increase rather than to diminish in comparative importance among the studies of the school. It wiU absorb many of the values THE PEDAGOGY OF READING 383 hitherto set mainly or exclusively upon classical study, and largely displacing the classics will become our most effective means of growth in culture and ideals ; just as we pursue the sciences, on the other hand, for information, for control of nature, and for the peculiar discipline which they afford. PART IV THE HYGIENE OF READING CHAPTER XX READING FATIGUE Reading makes certain severe demands upon the psycho-physical organism, demands which were not fore- seen in the evolution of that organism. These demands fall most heavily upon the eye, upon the mechanism of iimer speech, upon mind and brain in the rapid functioning of attention, apperception, association, imagery, feeling, etc., and upon the general nervous mechanism. The causes of the pecuhar fatigue experienced after continued reading have not all been satisfactorily made out as yet, and the writer hopes to make thlis the subject of later treatment. Provisionally, we may here point out certain functionings more or less pecuhar to reading which condition part of the fatigue and degeneration that is thus induced. In the case of the retina, in the first place, as Javal long ago pointed out,* the stimulations in reading constantly fall on approximately the same regions of the retina, tend- ing to give, as he believes, the same fatiguing effect that is so noticeable in after-image observations. In the ordinary work for which the eye was evolved the stimula- ' Revue Scientifique, 1879. 387 388 THE HYGIENE OF READING tions of the different retinal regions are varied and redis- tributed from moment to moment. Javal finds, too, that the eyes have much trouble in the effort to make the asymmetrical accommodations needed in near work, especially in reading long lines, when the fixation point is often much nearer to one eye than to the other and the comparative distance is constantly changing. He considers this an important condition of the fatigue and strain of reading, and a strong argument for the use of shorter lines. Among the more imusual and probably fatiguing func- tionings of the eye in reading is, of course, the excessive number of eye-movements necessitated. A page that we read in a minute or two has required perhaps one hundred and fifty of these quick movements and stops, while the eye in ordinary looking at objects at a Uttle distance would make but a fraction of this number. And not only is the number of these rapid movements greatly in excess of the normal, but during each reading pause the muscles must maintain, with rifle-aim precision and steadiness, a "set" of the eye whi^h will prevent blurring of the letters. That there is in spite of this a certain wavering has been repeat- edly shown. But it is perfectly certain that the reading pause requires a much greater accuracy of fixation than would be necessary in ordinary seeing, and there is much reason to believe that these fixations are trying both to the eye-muscles and to the attention. THE HYGIENE OF READING 389 It is true that whenever we go about, whether walking or moving rapidly in a car, our eyes constantly move whenever we look at objects that are stationary or that have a different motion than our own, the movement being in the endeavor to keep points in the objects fixated. But this movement, at least in walking or driving, is much slower than the reaction movements of reading, does not ordinarily require the same accuracy of fixation, and is of a more free and varied character. When the move- ments must be rapid and frequent, as in watching objects from a railroad train, they are, as we know, very fatiguing, more so than the movements of reading. Indeed, as Pro- fessor Dodge urges in his article on "The Act of Vision," ' it is doubtless less fatiguing to read than to look out of the windows, in a swiftly running train. The fatigue pecuhar to reading on the cars seems, as he suggests, to be mainly due to the constant blur causing muscular strain in the " vain and persistent attempts to correct the blur by changes in the convexity of the eye lens," and, we may add, by adjusting the convergence. The amount of this eye-strain, of course, depends upon the amoimt of jolting and vibration. It varies in amoimt, also, with different ways of holding the reading-matter. To really rest the eyes while riding it will be better, as Professor Dodge sug- gests, to look at objects within the car, or at any objects having the same motion; and, if the scenery must be ' Harper's Magazine, May, 1902. 39° THE HYGIENE OF READING observed, look forward or back rather than to the side, and attend to the more distant objects. Reading on the cars is always at least a little more trying than reading at home, and is a neurally expensive process at best. Readers must not be misled by Professor Dodge's statement that "the eye muscles are at rest more ,than nine-tenths of the time as we read " ^ into minimizing the fatiguing effects of reading in general. The muscles are "at rest" only in the sense that they are trying to main- tain the eye in stable equilibrium, which in reading may possibly mean even more trying work than when they are in rapid movement. But doubtless the most dangerous artificial condition produced by reading is the great amount of near work that is thus forced upon an organ that was planned for dealing mainly with objects at some little distance. The tremendous development of myopia among the peoples who read and are educated and its comparative absence among the others, its usual appearance at about the time at which the reading and other near work of the school begins, its progressive increase up the school grades, and its greater prevalence and degree when the lighting and other conditions are particularly bad, all point to reading and the other near work of the schools as a prime factor in producing this dangerous form of degeneration. The prime cause of myopia is eye-strain, either in the children • " Act of Vision,'' note. THE HYGIENE OF READING 391 that are affected or in their ancestors. The strain maj occur either in the oculo-motor muscles or in the ciHary muscle, or in both. Javal opposes the common theory that myopia comes from strain of the oculo-motor muscles in convergence, arguing that this violates principles of physics, and holding that some blind persons become progressively myopic. He finds the key to a truer theory in the fact that some myopics have power to change the length of the eyebaU as a means of accommodating for distance, instead of by changing he curvature of the lens, and this change in the length of the eyeball becomes permanent. In any case we know that myopia always means a lengthened eyeball, and that muscular strain of some kind has produced it. We are certain, too, that near work is a main cause of this strain. Cohn, in his "Hygiene of the Eye," pp. 46-53, says: "All ocuHsts agree that protracted near work with a bad light is one of the circum- stances most favorable to the origin and development of short sight." "In looking at near objects the accommo- dation is strongly excited, the choroid strained, the con- vergence forced, and the head bent forward." The strain- ing of the ciliary muscle in accommodating for the near object stretches and pulls the choroid and so induces, near the optic nerve, choroidal attenuation and atrophy. The stooping of the head in near work "produces a congestion in the veins which carry off the blood from the eye. Hence arise irritating conditions and over distentions with blood 392 THE HYGIENE OF READING in the back part of the eyeball, and these may bring on a yielding of the choroid and the sclera." "Short sight is almost always accompanied by atrophy of the choroid, which increases, as has been proved by experience, with the increase of short sight." This atrophy of the choroid at the posterior pole "gradually approaches the yellow spot," and "when once that is attacked by the disease, the central sight is extinguished." "Not less dangerous is the detachment of the retina from which so many highly myopic people suffer," this last being "the last step to incurable blindness." Bonders finds that " in youth almost every kind of myopia is progressive." "This age is the critical period for the short-sighted ey^ ; if during youth the defect does not greatly increase, it may become stationary ; but if it once develops into a higher degree, it is difficult to put limits to its further advance. It is, then, in youth that injurious exciting influences must be most anxiously guarded against." ' As myopia is not only very prevalent in the schools but is absolutely incurable, the importance of avoiding the conditions which originate and develop the disease is obvious enough. It may be remarked that one of these conditions is the fact that many of the children have astigmatism, due to a difference in curvature in the two meridians. Myopia often arises from the strain in trying to accommodate, in reading, to prevent the blur ' " Anomalien der Refraction und Accommodation, " p. 289. THE HYGIENE OF READING 393 due to this condition. Of course proper fitting with cylin- drical lenses will largely remove this blurring and the consequent strain. Near work causes strain of the ciliary muscle not only by the excessive degree of contraction necessary for proper focusing, but by the constancy of that contraction, by what Javal calls the "permanent tension of accommodation," which he finds to be an important factor in producing reading fatigue and myopia. The near work other than reading is not usually so fatiguing or so productive of myo- pia because the tension is not kept so constant. Of course in the ordinary seeing for which the eye was evolved the tension varies constantly. Javal finds that this near work of reading is one of the most common causes not only of myopia but of strabismus. The n^ar work is especially harmful when it is continued for long periods at a time. The congestion then becomes very considerable and the muscles are strained to their Umit, both conditions increasing the intra-ocular pressure and tending, as Cohn shows (p. 109), to break the timics at the weakest part, the posterior. Shorter periods of work should be the rule, with complete rest from near work in the intervals. Cohn calls attention to the fact that dispensing with afternoon school sessions often means five consecutive hours in the morning, perhaps with inter- vals of only five minutes between classes. He urges in- stead an interval of fifteen minutes after every hour and 394 THE HYGIENE OF READING one of half an hour after three hours. "For the bodily health and eyesight it would generally be better to return to the old arrangement of three hours school in the morn- ing and two hours school in the afternoon." ' Next to prolonged near work, especially with small objects, Cohn finds that bad lighting is most conducive to eye-strain, and next to this is bad seats, causing im- proper Jjostures in reading. The arrangement in which there is a plus distance between the desk and seat, leading the pupil to stoop over, with the resulting congestion of the eye, is especially to-be avoided. The desk-top must be at a proper angle and the whole arrangement suited to the height of the pupil. If artificial illumination must be used, the light should be shaded and not too near, the heat from a gas or oil lamp, especially, tending to heat the eye, drying the cornea and causing general congestion with its tendency to myopia. Cohn prefers a welLshaded electric light, as being much cooler than oil or gas. It is very important that the Ught should be steady, and it should come from over the shoulder or from the side. It is important in writing that the ink should be black and that the paper be placed at sufficient distance. The school assignment of home tasks should be sparing, especially in the earlier years. The home conditions as to light, seats, paper, etc., are often very bad, and much harm may result from doing school work there. It is important that if ' "Hygiene of the Eye," p. 216. THE HYGIENE OF READING 395 the children are already myopic, they be sent to a good oculist instead of being left, as often happens, to the mercy of ignorant opticians or spectacle peddlers. Investigations have shown that a very large percentage of myopic school children are wearing improper glasses. In view of the facts advanced concerning myopia and the other defects of vision, and the evident part that the near work of the school plays in producing and aggravating these, we have additional reason for agreeing with Cohn that the reading and other near work of the lower grades, especially, should be strictly hmited, and that the language work here should be largely oral. Indeed we have in the needs of the eye itself quite sufiScient reasons for de- manding a radical change in the traditional primary course. It must be remembered, too, that eye-strain is in the closest relations with nerve-strain, and that we seldom or never have the former without the latter. The weaken- ing of eyes by the near work of the early grades means the weakening of the entire psycho-physical organism, and the fact that these conditions commonly become hereditary warns us of the danger of race degeneration coming from this abuse of the school. It is not to be supposed that the eye is incapable of adapt- ing itself, in time, to the artificial conditions incident to reading. Violent environmental changes have been fre- quent in the course of human evolution, and the organs have met these changes with suitable adaptations of func^ 396 THE HYGIENE OF READING tion and structure. It will be so with the new activities that are really required by our complex civilization. There is no need for pessimism on this score if we regard the final outcome. But such adaptations require time, and disaster awaits many in the transition period. When we find, there- fore, that the danger is chiefly in the early period of growth, a period that is in any case better suited to active employments, and when we find too, that the danger at all points can be greatly lessened by proper attention to plain hygienic requirements, it is simply the part of wis- dom to act upon the warning so sharply given by the my- opia and asthenopia which are so prevalent among our young people. A second class of disturbances which the organism suf- fers from early reading are those of speech, and these arise chiefly from reading aloud. As a result of two censuses of stutterers, taken by Dr. Hartwell, director of Physical Training in the schools of Boston, he reported that "out of every thousand children in the public schools of Boston at least seven stutter or stammer." This, of course, does not take account of the very many others who are back- ward, hesitative, bungling, or otherwise deficient in their speech, but who cannot be said to stutter. The causes are often the same for all. Hartwell finds that the period of the second dentition, at about seven years, is a period of disturbance in the nervous system and is a period that is most liable to language disturbance. He thinks , that THE HYGIENE OF READING 397 stuttering "is largely due to faulty or misguided methods of instruction in speaking and reading." Clouston, in his "Neuroses of Development," makes "stuttering and back- wardness of speech" one of the prominent neuroses liable to occur in this early period, and remarks that " it is dur- ing this very period that most children enter school, and are launched upon intellectual pursuits by being taught the rudiments of the art of reading aloud." This takes atten- tion and mental effort, and "it is a comparatively easy matter to induce stammering among Abecedarians." A. Melville Bell, the well-known "inventor of visible speech," called schools the "nurseries of stuttering," and wrote in 1866 that " with a proper initiatory training and school sur- veillance, stammering and its train of silent errors would be altogether imknown." Hartwell calls the elementary schools "the breeding ground" of the stuttering habit. He found a "marked increase of stuttering among pupils of the primary schools as compared with pupils in the kindergartens," and thinks it "highly significant that the amount of stuttering, both in boys and girls, is greatly augmented at the very time when instruction in reading aloud is begun." It may be remarked, by the way, that from three to four times as many boys as girls are found to stutter habitually. Hartwell finds that "any one or all of the organs con- cerned in producing speech may be affected in one who stutters," but that "the respiratory muscles are almost 398 THE HYGIENE OF READING certain to be at fault." Unless these are set right the work on the throat and mouth muscles seems to be "largely wasted." He finds that those most successful in curing stuttering have "instinctively" begun with gymnastic |ex- ercises of the breathing muscles, and later have developed phonation, then articulation, — from fundamental to accessory. The means of cure suggest the means of pre- vention, and Hartwell is doubtless right when he urges that free play and gymnastics will prevent much of it. He dwells on the importance of building up the funda- mental system of muscles before working hard with the peripheral muscles, with the eyes, the articulatory appa- ratus, the fingers, etc. "The highest level centers, in the cortex of the brain, represent the most special, precise, elaborate, and varied of our peripheral muscles," and these, as Flechsig shows, cannot function so early as the centers for the more fundamental movements. We are thus reminded again of the necessity of making the early work of the school largely motor, with little fine work of the fingers or eyes or speech organs. Of course in the case of the speech organs the danger is not from the exercise of the muscles themselves, but from the attempts to control them through the higher centers. Prevalent methods in phonics and in teaching to pronounce and to read aloud call the child's attention to the par- ticular movements and processes concerned in speaking, and this consciousness of the "how" of speaking arises THE HYGIENE OF READING 399 whenever, in reading or talking, the thought is directed to anything else than meanings. Any analytic work of this sort, done before the speech habits have well set, brings in its train the abnormal fimctionings that always attend the attempts of consciousness to tamper with processes which are meant to fimction automatically. Doubtless the most unusual functionings upon which the psycho-physical organism has fallen in reading are those of mind and brain. Reading fatigue is mainly fatigue of mind and not of eye, though the eye-movements and tensions and over-stimulations are conditioning factors in this fatigue, just as movements of one sort or another probably condition all mental activity. We know how frequently and how closely mental and neural exhaus- tion is correlated with painful vision and especially with difficulty in reading. Over-use of the eyes in reading and in other near work brings with it a reduction of the general stock of nerve energy such as results from over-use of the mind. Doubtless the nervous mechanism which functions in vision, and especially in the vision of reading, is involved as well in many of the more general functionings of mind, and especially in many of the mental processes that go on in reading. The fatigue, therefore, of mind and eye mutually condition each other, some, at least of the fatigue that seems to be purely of the mind being due to the abnormal or excessive functionings of the eye, — functionings which, however, may never come 400 THE HYGIENE OE READING to consciousness as such, even when abnormally per- formea. However, much of the mental fatigue that comes from reading is conditioned otherwise than by tfie work of the eye. One of the most important characteristics of the mind's work in reading is the unusual amount of attention that is required. There is a certain constant "set" of the mind upon the book or upon the general procedure in reading, having, as its bodily basis (and Ribot and the other aiialysts of the attention have shown that attention always has certain muscular adjustments and tensions as its basis), certain tensions of the muscles of the neck and head and eye, doubtless with others that are basal to maintaining the mental attitude characteristic of read- ing. We notice the presence and volume of this atten- tion set when we involimtarily relax and perhaps sigh, yawn, and feel general rehef after a period of continuous reading. Besides this general set of the attention, there is a continued succession of particular sets and quick changes of the attention as the mind fastens upon one after another of the "total ideas" expressed in the sentences read. The attention is mainly upon these total idea-meanings, but these are different with every sentence, and often have sharply varying phases with the sentence's subdivisions as well. Now to be conscious of things focally, to attend, is a normal functioning of mind, practiced in ordinary THE HYGIENE OF BEADING 40I activities without special fatigue. But in reading, these successive sets of the attention are so very numerous, are forced upon the mind at such a pace, that they must sooner or later become fatiguing. Besides the fatigue due to the rapid succession of apperceptive-attention acts concerned in apprehending the meanings, a certain variable amount of consciousness attends the rapidly succeeding acts of looking at the line itself. The extent to which conscious- ness concerns itself with the actual "looking," and con- sequently the extent to which reading fatigue is due to these visual attention-acts, varies with readers and with subject-matter. Normally this visual "looking" should be largely automatic, except when we are dealing with some special kinds of subject-matter, and of course sup- posing that the matter is properly printed. But it is a fact that very many readers are much concerned with the mere visual looking at the line, and the rapid succession of atten- tion-acts thus necessitated is an important condition of their fatigue in reading. There is no doubt that for all readers the rapid succession of eye-movements and pauses, though unconscious in themselves, condition somewhat a general state of mental tension or attention, fluctuating perhaps somewhat with the movements and pauses, and tending to fatigue when long continued. Besides, reading involves certain general forms of mental as well as physical activity and attitude, in addition to the general set of the attention which we have mentioned, which repeat 402 THE HYGIENE OF READING /themselves constantly and are doubtless wearing by theil very monotony. The pace at which mental content moves in reading, and the large amount of mental content which is aroused from moment to moment, furnish additional conditions of read- ing fatigue. There is a continual shifting and resetting of the kaleidoscope of imagery, feeling, and motor attitude, with a rapidity of the flow of associations, verbal in the main, that is unparalleled in the ordinary life of observa- tion and action for which mind was developed. Not only must a really vast amount of mental machmery fimction in the mental construction of the words upon any page, in their visual, motor, and auditory elements and in their meaning implications, . but a still larger number of words must be sub-aroused, almost to the point of actual construction, as associative expectancy points in their direction before the particular form appears. Large por- tions of the mind's total vocabulary may thus have to keep "fired up," in reading certain classes of matter, and the total quantity of psycho-neural fimctioning is thus much in excess in reading. The fact that the content is mainly verbal doubtless contributes all the more to the fatigue. Words, as Stout so well shows in his Analytical Psychology, are the most admirable instruments for thinking because they are susceptible of such nice and rapid control; but this very nicety of control makes it "near work" for the mind, and as fatiguing as all finely THE HYGIENE OF READING 403 discriminative work tends to be. Words, too, are con- ceptual, abstract, and their use in reading involves more or less of the fatigue that comes with continued use of highly generalized experience. The rapidity with which such thinking may be done is a source of illusion as to the amount of energy that is being expended, and thus becomes a source of danger from fatigue. In reading, thought may run with "seven league boots," and we are apt not to have the natural reminders when the mind has done enough. Of course such thinking is far less fatiguing for its being a game of "follow your leader," and much more of it can be done with safety than when the mind must blaze its own trail. But many sympathetic readers read almost as actively and constructively as though doing the thinking on their own initiative, and for them reading is quickly though often insidiously reductive of nervous energy. In cases of nerve exhaustion, with this type of reader, the feelings of head strain and of being wrought up with nervous tension come quickly on attempting to read, and are especially aggravated by doing hurried reading. The trouble is often referred mainly to the eyes, and much time and money are wasted in attempting to correct these. Eye fatigue and mind fatigue here have their common denominator in nerve exhaustion. We must remember that, even for all classes of readers, reading even at its best and pleasantest requires the expenditure of much 404 THE HYGIENE OF BEADING energy and is reductive of nerve reserve. Though often a helpfully recreative employment, it lacks the freedom and the rejuvenating effect of free play. Our organ- ism is always working at a considerable expense while taking such recreation. I have elsewhere urged the great advantages of rapid and selective reading. On the side of hygiene we must remember that such reading will be more fatiguing, when continued for long periods, and especially in the first attempts to hurry the pace and vary the method. It is to be remembered, however, that the fast rate becomes habitual as well as does the slow rate, and the hurry feel- ing then disappears. In the selective reading, too, while certain parts are thought more intensely, far more is quietly ignored, and the feeling of values and using them as they appear becomes a habit of mind which functions almost as automatically as other aspects of the reading process. The slow reader who with painstaking " thorough- ness" works as hard at one line as at the next expends so much energy in lifting dead weight and in handling useless debris that his work is doubtless more fatiguing, to obtain equivalent results, than that of the rapid and selective reader. Of course, all selective mental activity, in which the mind really acts, judges, and constructs for itself, whether in reading or otherwise, naturally causes far more fatigue per hour than when the performance is mainly passive. We cannot get something for nothing even in THE HYGIENE OF READING 405 psychic economics. Here, as everywhere, a hard pace means a short course and frequent change. It should be remembered, however, that a rapid pace is not always the hardest of paces, for organisms that are trained to it. And it is a safe and needed rule for all kinds of reading that it should never be continued uninterruptedly for long periods of time, seldom indeed for longer periods than two hours at the most. CHAPTER XXI HYGIENIC REQUIREMENTS IN THE PRINTING OF BOOKS AND PAPERS Probably the most important and most feasible means of lessening the fatigue and strain of reading is by bring- ing it about, so far as possible, that all books and papers shall be printed in such type and arrangement as shall fall within certain recognized limits of hygienic require- ment. As to some of the requirements which should be made of the printer we are still uncertain, and further experimental investigation rather than the present excess of opinion is in order and is cryingly needed. Of other requirements we can now be certain, and these should be enforced rigorously, in the printing of schoolbooks and government publications at least. If enforced here, they will tend to extend to all printing. The size of the type is perhaps the most important single factor. The experiments of Grififing and Franz showed that fatigue increases rapidly as the size of the type de- creases, even for sizes above eleven point, or above a height of 1.5 millimeters for the short letters like v, s, etc. The various investigators are generally agreed that this should be made a minimum for the height of the short 406 THE HYGIENE OF READING 407 letters. Matter printed in this size of type is read faster and individual words are recognized more quickly than where the type is smaller. Besides, Grifi5ng and Franz foimd that the effect of insufl&cient illumination is less marked with the larger type. Preferably the height of the small letter should be somewhat above the minimum stated, though when the height is much above 2 milli- meters Weber's experiments indicated that the speed of reading is decreased. The thickness of the vertical strokes of the letters should not be less than .25 millimeter, according to Cohn ; prefer- ably .3 miUimeter, according to Sack. This thickness of the letters has been foimd by Javal and others to be a very important factor in increasing legibihty, and thus in decreasing fatigue. Griffing and Franz found, however, that hair lines might form parts of the letter without decreasing the legibility provided the other parts were thick. They find it possible, however, that such hair lines may increase fatigue. The minimum of thickness stated above should be insisted on for the main lines. The space within the letters between the vertical strokes should not be less than .3 millimeter, according to most investigators. Sack finds .5 millimeter to be preferable. There is probably little to be gained by increasing the distance between the letters beyond that which is usual in the better printed books of the present time. Burgerstein and Netolitzky would require that this distance should be 4o8 THE HYGIENE OF READING greater than the distance between two "neighboring ground strokes" of a letter, and Sack, would make the minimum distance .5 to .75 millimeter. Burgerstein and Netolitzky would not allow more than six or seven letters per running centimeter and would require as much as 2 millimeters between words. With these requirements Sack is in agreement. It should be remembered that any very u n- usual separation of the letters of a word is distracting and should be avoided. These minimal norms, as stated by Burgerstein and Netolitzky, should be made require- ments, except that possibly the distance between letters is not so important as they urge. The minimum of six or seven letters per running centimeter is a convenient ap- proximate gauge which can be quickly applied and is not too stringent. Griffing and Franz found that legibility increased some- what, though not greatly, with increase in the distance between the lines, with the leading, as it is called. Cohn thinks it important that there should be a minimum leading of 2.5 millimeters, and Sack requires the same. Javal does not find that leading increases legibility ap- preciably, and thinks that the space used for this purpose would far better be given to an increased size of letter without leading. The leading is doubtless a mistake when the size of type is below the requirements made above. The size of type should by all means be increased instead, as this is by far the most important of the^factors THE HYGIENE OF READING 409 conditioning fatigue. However, a certain amount of lead- ing should be required in schoolbooks, at least, but hardly more than Cohn's minimum of 2.5 millimeters. As to length of lines there is a general consensus in favor of the shorter as against the longer lines, with a tendency to favor 90 millimeters as a maximum, some plac- ing the maximum at 100 millimeters. The latter is doubt- less too high. Javal, who has studied the matter very carefully, insists that the maximum should be consider- ably below even 90 millimeters. As already noted, he names as one of the principal causes of fatigue in reading, and a cause tending to produce and aggravate myopia, the considerable amount of asymmetrical accommodation required as the eye moves along a long line, the amount increasing always with the length of the line. Even with the page squarely before the reader, vmless he makes con- stant and fatiguing movements of the head while reading, the reading-matter is always farther from one eye than from the other, except at the middle point of the line, and the reader strains to accommodate for both distances, especially for objects held so near as is the page in reading. Against the long lines is also to be urged the difiSculty and distraction incident to finding the place at each turn to the next line, increasing always as the lines are longer. Besides, the longer lines require a greater extent of eye- movement for a given amount of reading. This comes from the fact, verified by various experimenters, that the 4IO THE HYGIENE OF READING eye does not traverse the whole line in reading, but begins within the line and usually makes its last pause still farther within, the reader reading the first and last parts of the line in indirect vision. The amount of this indentation tends to be a constant amoimt somewhat irrespective of the line's length, and is consequently a larger proportion of the line's length in the shorter lines. There is thus an important lessening of eye-work in using the shorter lines. Indeed I found that readers could read matter printed in lines of 25 millimeters in one downward sweep without any lateral movement of the eyes. With lines 30 millimeters long the lateral movement was sometimes almost nil, and seemed to be due mainly to habit. In reading such lines in this way the eye's extent of movement is hardly more than one-fourth or one-fifth the amoimt needed for the same matter when printed in long lines. With the shorter lines, generally, more words were read per fixation than with the longer ones. A magazine column having lines 60.5 millimeters long was in one case read at the rate of 3.63 words per fixation, while columns having lines 98 to 121 millimetres long required a fixation for every two words. Lines of a length approximating 60 millimeters are usual in newspapers, and in my experi- ments were read with a minimum of eye-movement. The makers of the modem newspaper have felt the reaction of readers more, perhaps, than have the makers of books. Out of this experience has evolved the present practice THE HYGIENE OF READING 4II of printing newspapers in narrow columns, the line-lengths of which are perhaps as near the optimum as can be determined at present, when we consider that much shorter lines give great inconvenience to the printer. For books, also, the newspaper line-length is near an optimum so far as ease and speed of reading are the con- ditions to be considered. In the case of large books, where the question becomes one of printing in one or in two columns per page, the latter alternative should imdoubt- edly be chosen. For books of ordinary sizes a somewhat longer line may be used where this will contribute to con- venience or beauty ; but a book should not be used whose lines are more than 90 millimetres in length, and somewhat shorter lines are generally to be preferred. One of the great advantages of the shorter lines is that they constantly permit the reader to see in indirect vision what his eye has just passed as well as what is just coming. Though the words of this related matter may not be clearly perceived, they furnish visual clews which keep the reading range further extended at each moment, a most desirable condition for all reading and especially for fast reading or for skimming. With such lines a hurried reader may glance straight down a page with only an occasional short stop and may yet be sure that he has gathered the gist of everything. Dr. Dearborn, in experiments made recently at Columbia University, foimd that the eye makes its longest pause near 412 THE HYGIENE OF EEADING the beginning of the line, thus permitting a preliminary general survey. A secondary pause of more than average duration is made near the end of the line, perhaps partially in review. He finds that lines of only moderate length facilitate these general surveys better than the longer lines, and finds also that they facilitate a rhythmical regularity of eye-movement, both being conditions which contribute to speed and ease of reading. His tests showed that such lines (a little longer than newspaper lines) were read at greater speed and with shorter pauses than lines of twice the length. Dearborn argues, and correctly I think, in favor of imiformity in the length of lines, particularly in books for children. The reader drops quickly into a habit of making a constant number of movements and pauses per line, for a given passage, and broken lines confuse and prevent the formation of such temporary habits. However, a slight indentation every other line may, he thuiks, be of distinct advantage. Dearborn thinks that a line of 75 to 85 milh- meters combines a good many advantages, and we are certainly safe in putting 90 millimeters as a maximum, with a preference for lines of 60 to 80 miUimeters. The smaller books, which can be easily held in the hand during the reading, are to be preferred, and on the whole have grown in popular favor. The larger books usually have to lie on a support which exposes the letters at an angle, greatly lessening their legibility and producing the equivalent of a material decrease in the size of type. THE HYGIENE OF READING 413 As to the forms of particular letters, many changes are cryingly needed. However, further investigation is needed before we are warranted in requiring changes of the printer. We know that such letters as t, z, o, s, e, c, i, are compara- tively illegible. C, e, and o are often confused with each other, and i with /, h with k, etc. This confusion can be avoided by making certain changes in these letters, and their legibility can be increased. Certain excellent recom- mendations of changes in particular letters have been made by Javal, Cohn, Sanford, and others. However, there are many things to be considered in making such changes, and further thorough and mature investigation is needed before any letter is permanently changed. The whole matter should be placed in the hands of a competent specialist or committee of specialists, to be worked over experimentally and advised upon in the light of the psychology of reading, the history of typography, aesthetic considerations, the convenience of printing, and the lessons of experience generally. Changes should not be made on the single basis of experiments upon the com- parative legibility of isolated letter-forms. A letter whose legibihty in isolation is bad may sometimes contribute most to the legibility of the total word-form. Studies now being made of the comparative legibility of letters as seen in context will doubtless throw light on this point. The subject is too complex to permit the adoption of recom- mendations that are based on study, however careful, oi 4l4 THE HYGIENE OF READING any single aspect, or on anything that does not include a careful study of all the factors. It is high time that there should be a rationalization of these printed letter-forms that have come down to us in such a happy-go-lucky fashion, and it is to be hoped that either the Carnegie Institut on or some department of research in a well- equipped imiversity may take hold of the matter and see that the work is thoroughly done. Among further printing requirements that are important and that should be insisted on, the letters should have sharp, clear-cut outlines, and should be deep black, The paper should be pure white, but without gloss, the latter being especially trying to the eyes. According to Cohn and Sack the paper should have a minimum thickness of .075 milUmeter. Paper of a sUghtly yellowish tinge is probably not in- jurious and is preferred by Javal. But in general the legibihty depends on the contrast between the black of the printed forms and the white of their background, and colored or gray papers lessen this difference and thus dimin- ish legibiUty. Pure white light gives the greatest legibility. The print of one side must not show through from the other, and the printing must be so done that it will not affect the evenness of surface of the other side. It is important that wall charts and maps should not contain more names than are absolutely necessary for purposes of instruction, and that these should be in large, THE HYGIENE OF READING 415 clear type ; or the most important names for reference at a distance and by classes may be in the large type, with the others in type fulfilling the requirements for school- books, and for use by individuals at the ordinary reading distance from the chart or map. Burgerstein and Ne- tohtzky advise that school maps should not present the physical and poUtical features on the same map, ia the interest of greater legibihty. Names printed on colored map surfaces need to be in larger rather than ia smaller type than that used in books if legibihty is to be main- tained, as any other background than white means dimin- ished legibility. The writing upon slates is considerably less legible than that upon good white paper. In the case of blackboards the surface is apt to be gray after erasing, and this, of course, lessens the legibility very considerably. It is important that the blackboard surface be deep black, without gloss from reflection so far as this is possible, and that it be kept clean, avoiding the gray effect. Teachers and pupils should acquire the habit of vsrriting on the blackboard in a large plain hand, as the greater distance at which the writing is read and the usually diminished legibihty makes this of importance, and especially in the primary school grades. In stating the requirements above I have had in mind the needs of adult readers and of the older school children. The younger children must have a type much larger than the minima there stated. The reading of young children 41 6 THE HYGIENE OF READING has not been sufficiently studied to warrant a final state- ment of what should be required in the printing of their books. As the most usable approximate statement of what may properly be insisted on, and for the sake of vmiformity, I quote here the requirements made by Shaw in his "School Hygiene," with his illustrative examples. These requirements are none too stringent, except that sometimes some of the leading may well be sacrificed in favor of a type that is a little larger, for the third and fourth grades especially : — "For the first year the size of the type should be at least 2.6 millimeters and the width of leading 4.5 millimeters, as shown in this example: — Then there is a turn in the road. The long train runs over the bridge and swings round behind a hill. The children cannot see it now. For the second and the third year, the letters should not be smaller than 2 mm., with a leading of 4 mm. Some of the more carefully made books for the second and the third years are printed in letters of this size, as shown in the following example: — She must climb the tree. She held on, first to one branch and then to another, and THE HYGIENE OF READING 417 tried to reach the golden plums. Her hands, her face, and her feet were scratched and torn by the thorns. Try as hard as she could, she For the fourth year, the letters should be at least 1.8 mm., with leading 3.6 mm., as follows: — On the way down, an Indian who was ill a canoe stole something from the ship. One of the crew saw the Indian commit the theft, and, picking up a gun, shot and killed him. This made the other Indians very angry and Hudson had several fights with them." For some grades succeeding this the type should be kept well above the minimal requirements for adult readers. Examinations of the schoolbooks in use in Germany, Russia, and other European coimtries, made at various times and places, have shown that usually from fifty to eighty-five per cent of the books came short of hygienic requirements American books are somewhat better, but include very many that are very bad. Even when the principal part of the book is in good type, there wiU often be large sections printed in a type so small as to be very injurious. The dictionaries and other books of reference have notoriously small print, and those with he smaller and poorer types should be mercilessly discriminated against. As Shaw rightly says, "Principals, teachers, 2e 41 8 THE HYGIENE OF READING and school superintendents should possess a millimeter measure and a magnifying glass and should subject every book presented for their examination to a test to determine whether the size of the letters and the width of the leading are of such dimensions as will not prove injurious to the eyes of children. If every book, no matter what its merits, were rejected if its type were too small, the makers of such books would very quickly bring out new editions with a proper size of type." CONCLUSION READING AND PRINTING OF THE FUTURE CHAPTER XXII THE FUTUKE OF READING AND PRINTING. THE ELIMINA- TION OF WASTE Reading is the means by which the world does a large part of its work. The printed page is a contrivance used for hours daily by tens of milhons of people. The slightest improvement either in the page or in the method of read- ing means the rendering of a great service to the human race. Human thought has been busy rationalizing. It has rationalized the traditional methods of transportation and locomotion imtil we have the steam and electric loco- motive and the economy and comfort of modem travel. Means of communication at a distance have had the keenest and most persistent efforts of inventive genius, and the modem marvels in telegraphy and telephony are the results. Even printing and the making of books has had attentive study and continuous improvement until wonders of the printer's art are within easy reach of all. Yet with it all the essential characteristics of the printed page itself, and of the reading process by which we gather its meaning for so many hours of the working day, have never been rationalized in the interest of the reader's time or energy or comfort. 4»i 422 CONCLUSION To take a trite example, note the spelling of our printed page. Like the ancient Egyptians, who spelled out their •words in letters and then laboriously added the useless picture hieroglyph, we compel "practical" modems, pressed for time as they are, to traverse one-fifth or one- sixth more of printed matter than is needful, on every page, in order that a few scholastics may enjoy a luxurious thrill from the sight of the silent-letter rehcs. Again, we have never seriously worked out to a conclusion what form of any letter would give the greatest legibility, and we have never used the results of the meager though very valuable inves- tigations already made in this field. Indeed, who knows but that Broca and Sulzer may be right in their contention' that the extreme of simplicity, in letter-forms, means the maximum ease of recognition, recommending therefore such forms as Q .-— ^/> J -^ P* U) < L- / for the capitals, and I j I \/ Lm \\ \J /I for the small letters. Certainly the letter-forms that have come down to us through the ages have never been pruned to meet the readers needs, though the writer and printer have made conservative changes for their own convenience. There is not the slightest doubt that forms can be devised which will be much more legible than these ancient tra- ditional symbols. From the point of view of Messmer's • La, Nature, Paris, February 13, 1904. READING AND PRINTING OF THE FUTURE 423 analysis of letter-forms, it may well be that the legibility of words will be increased by adding to the number of characteristically formed, or dominant, letters. We cer- tainly have many words whose letters are optically very similar, and these words would be much improved by such additions. And then we have never canvassed the possibilities of improving the total word-form, for particular words. We know how the German use of initial capitals, for instance, and the imitation of this practice by such writers as Carlyle, gives greater prominence to the capitaUzed words. If by using capitals or by changing the shape, size, or even color of constituent letters we bring into prominence the total word-form and characterize it better, total form will thus come to play a still larger part than at present in mediating the recognition of what is read. Such recog- nition in larger imits favors speed in reading and lessens the strain on eye and mind. The special temporary characterization of the important words or phrases in any given article, by changes in type, etc., may also aid much in speed and ease of reading whenever the reader's aim is selective, purposing to get quickly the kernels or gist of the matter read. Even the present somewhat crude use of such characterization, by our daily newspapers, shows that such a method meets a need of busy readers. Any ar- rangement which makes comprehensive skimming an easy matter will be of great benefit for large parts of our reading. 424 CONCLUSION We are likely, indeed, soon to consider the possibilities of a total rearrangement of our printed symbols, in the interest of economy of time, energy, and effectiveness in getting thought from the page. The history of writing has a wealth of suggestion here which may well be pon- dered. Consider, for instance, the Egyptian representa- tion of the name "King Sent," as compared with the narrow row of little black strokes by which modem printers present the words. In the figure, the 5 (vertical hook within the oblong, at the 1 right), N (upper left wave line), and T (D or T, the hand) represent "Sent," while the surrounding oblong represents "King." Such bunching of the letters of words into a characteristic total form, with even such substitutions of simple forms, like oblongs, squares, etc., for oft repeated words, would agreeably distribute the stimulations on t' e retina, would perrnit the eye to cover several times as much reading-matter with the present extent of movement, would encourage the more facile and more speedy reading in total forms, — would have, on the whole, advantages so vital that the possibilities of such rearrangement are at least well worth careful consideration. Consider, for instance, the traditional arrangement of our words into straight horizontal lines, already referred to in our introductory chapter. We read so because, a good many thousand years ago, the scribes foimd it con- venient to write their characters that way. The Egyptian, READING AND PRINTING OF THE FUTURE 425 on the other hand, found the arrangement in vertical lines to be very readable, and the Chinese and Japanese still prefer it. The writer made an extended series of experi- ments, a few years ago,' to determine the relative speed of reading-matter printed by our present method as against reading equivalent matter arranged in columns of words, the individual words in the latter case standing horizontally, one below the other, down the page. The results showed that for reading aloud at maximal speed nonsense matter could be read as quickly in vertical arrangement as in horizontal. With sense matter the vertical reading was only from seven to ten per cent slower. In silent reading the vertical method reduced the speed considerably, the distraction from the novel arrangement seeming to have more effect in the silent reading. But in view of the tre- mendous amount of practice which each reader had had with the horizontal lines and the consequent distraction produced by the imusual vertical arrangement, it seemed entirely likely that the vertical arrangement might ulti- mately give even- greater speed than the horizontal. The further great advantages would be that in reading down vertical columns in the Japanese fashion the eye may cross-section its words, reading the lateral parts in indirect vision and thus getting the visual data needed for reading four or five words with the movement now needed to traverse one lengthwise. Not only might we thus ' American Journal of Psychology, IX, pp. 375-386. 426 CONCLUSION save at least three-fourths of the fatiguing eye-movements and three-fourths of the pauses as well, but we would constantly be able to use the upper and lower retinal periphery in getting far more data than at present for the perception of the immediate context just past and just coming. The reading range of each moment would thus be materially increased, with the accruing advantages. It is probable, too, that with such an arrangement of printed symbols there would be much less distraction in "keeping the place," less distraction from the remotely related matter in the neighboring lines, and greater freedom in the choice of fixation places. Of course, on the other hand, such an arrangement would have certain disad- vantages both to the printer and reader. But enough has been said to show that this is an important direction for experimentation in the rationalization of the printed page. The economy that may come from a more general and effective use of illustrations is already being recognized, and the psychology and pedagogy of picture-printing of itself is worth a volume. ' What a development here already since the days, in the memory of men now living, when illustrated books were a rarity, and when boys were flogged for bringing them to school ! Johnson gives an interesting account of this in his "Old-Time Schools and School Books." How very much of the reader's time may be saved by judicious use of graphic methods of presentation, by charts, maps, globes, the stereopticon, etc. READING AND PRINTING OF THE FUTURE 427 What possibilities for beautifying and increasing the effectiveness of our printed page, and for easing the work of the eye, may come from using the wealth of suggestion in ancient and modem pictography, and indeed ia modem cartooning and advertising. The history of Egypt, of China, and of other ancient peoples shows that they endured untold wastes for hun- dreds and even thousands of years because they failed to rationalize their systems of reading and writing, regarding the traditional ways as "good enough." It behooves the modem world to appropriate the benefits that are sure to come from eliminating our own very evident wastes in these same arts. Our printed page, as we have seen, may be made a far more economical one. And with this improved page, with a simpler alphabet and a natural phonetic system of spelling, much of the present waste in learning to read will disappear. A thorough rationali- zation of the methods of learning to read, on the basis of the psychology and history of reading, will give additional economy. When our schools take up their proper work of teaching all readers to utiUze the hbrary and all printed matter effectively and rapidly, and to substitute real selective reading for mechanical plodding at the customary uniform, "aloud" pace, another great waste will disappear from the school and from life. In this training the oppor- tunities for mental discipline in reading, largely neglected hitherto, will be utilized, and certain subjects long studied 428 CONCLUSION mainly for the sake of discipline may consequently be advantageously omitted. Further, it may be said that the learning of foreign languages, ancient or modem, will in many quarters imdergo considerable revision in the direction of economy, when the facts are clearly grasped as to what constitutes the essence of natural reading. It will then be even more clearly and demonstrably evident than hitherto that much of our academic "reading" of languages has been but a gloss and hollow parody upon reading, lacking the free rhythm and melody play of inner speech, lacking the dominance of all parts by total imifyuig idea-meanings, without the habits of associative expectancy that are absolutely essential for the coAtrol of a language either in speaking or in reading, and want- ing still other earmarks of real reading. A certain waste of the reader's time has already been lessened by the improvement in the style of writing Eng- lish, since the Elizabethan times at least. Professor L. A. Sherman, in his "Analytics of Literature" (p. 256 ff.), calls attention to the fact that the early English prose had either crabbed or heavy sentences, which demanded "re-reading" or even "pondering" before the meaning would reveal itself. He finds that "ordinary modern prose, on the other hand, is clear, and almost as effective to the understanding as oral speech." Still he admits that few of us to-day really "write as idiomatically and naturally as we speak," and most men's written language is very different from READING AND PRINTING OF THE FUTURE 429 their spoken language. This difference is partly due to the slowness of handwriting, which makes an author finical, self-conscious, and unnatural. On the whole, however, Professor Sherman finds that "from the lyrical bards to the present the language of books and the lan- guage of men have been growing rapidly alike." If this development continues, and the modem habit of dictating at a natural rate to stenographers and even to grapho- phones may hasten it, reading, as the translation of what is written into natural inner speech, may have further facilitation by this change, and perhaps from still further changes in composing to be suggested from studies, already overdue, in the psychology of style. Indeed there are those who go further than any of these legitimately warranted prophecies of future economy in the time and effort of the reader, and predict the displace- ment of much of reading, in toto, by some more direct means of recording and communicating. Just as the telegrapher's message was at first universally read from the tape, by the eye, but has come to be read far more expeditiously by the ear; so, it is argued, -writing and reading may be short-circuited, and an author may talk his thought directly into some sort of graphophone-film book which will render it again to listeners, at will ; repro- ducing all the essential characteristics of the author's speech, which, as we have seen, are not recorded by written language and which the reader must construct for 430 CONCLUSION himself at a considerable expense of energy. This lattei proposition is, of course, as yet, the wildest of speculations. But the plainly possible changes in the direction of ehmi- nating present wastes in reading are so important that they demand the early institution of organized research upon the various problems, to determine, at least, the ideals toward which we should strive in the making of our page and in the practices of reading and learning to read. All these problems are complex and demand maturity of judgment, as well as n^astery of technique, in their solu- tion. And they are problems in which the points of view of the psychologist, the philologist, and the educator must receive a practical synthesis. Too often, as in the working out of systems of phonetic spelling by philologists, a system excellent from the philological or logical standpoint has lacked fitness to the psychic or hygiehic conditions in- volved in reading, or it has lacked the pedagogical adap- tations needed to permit its making a successful appeal to the masses. The need is, first, for more of particular researches such as we have had, on specific problems, to furnish much more of fact and of suggestion. Second, the problems of determining optima, along the more im- portant lines already suggested, should be placed in the hands of committees of competent specialists, to be worked out as a part of the duties of our institutions for higher research, if necessary with government supervision and provision. It is important that we should have before us, READING AND PRINTING OF THE FUTURE 43I as early as possible, correct and authoritatively promul- gated ideals in all these matters. Conformity to these will come but gradually, but will come the earlier for their being definitely stated. The increasingly effective means for the dissemination of information is making possible the hastening of reforms; and the possibiUties of controlling conditions as to reading and even printing, through the government supervision of the practice of the schools, gives promise of early improvement in conditions when once the specialists have reached final conclusions. When the world comes to put aside the false sentiment and traditions that have clouded the subject of reading, and sees it as the everyday means of doing a large part of our work; and when the proposed reforms are shown to mean definite savings of time, money, and health, and definite improvement in mental habits as well, practical sense will sooner or later see to it that they are duly installed. BIBLIOGRAPHY Bibliography of the works referred to in this book, and of a few others which may be of service. For most of the topics the bibliography is, of course, not even approximately complete. It introduces, however, to the literature that has been found most helpful in my study of the subject. Abell, Adelaide M. Rapid Reading. Educational Review, 8, 283. Ahrens. Die Bewegung der Augen beim Schreiben. Rostock, 1891. Bagley, W. C. Apperception of the Spoken Sentence. American Journal of Psychology, Vol. XII, pp. 80-130. Ballet. Le Langage Interieur. Bamvun, Edith C. Reading, in the series of articles on the curricu- lum of the Horace Mann School. Teachers College Record, January and September, rgo6. The work of the higher grades is to be presented in later numbers. Bawden, H. Heath. A Study of Lapses. Psych. Review Mono- graph Supplements, Vol. Ill, No. 4, April, 1900. Baxt, N. Ueber die Zeit welche nothig ist, damit ein Gesichts- eindruck zum Bewusstsein kommt. Pfluger's Archiv jiir Physiologic, Vol. IV, 1871, pp. 325-336. Becker. Experimentelle und kritische Beitrage zur Psychologic des Lesens bei kurzen Expositionzeiten. Zeitschrijt jilr Psych, und Phys. der Sinnesorgane. Bd. 36, H. lu. 2, pp. 19-73. Brown, Alex. C. The Relation between the Movements of the Eyes and the Movements of the Head. Oxford, 1895. pp. 28. See also Nature, Vol. LII, 1895, p. 184. Calkins, Mary W. Association. Psych. Review Monograph Sup- plements, No. 2, February, 1896. Cattail, J. McK. Brain, Vol. VIII, p. 305 ff. Philosoph. Studien, II u. III. Mind, 1886, p. 65 and p. 531 ff. Mind, i88g. 2F 433 434 BIBLIOGRAPHY Cattell, J. McK. Time and Space in Vision. Psych. Review, Vol. VII, 1900, pp. 325-343. Dearborn, Walter F. The Psychology of Rfeading. Columbia Univ. Contribs. to Philosophy and Psychology, Vol. XIV, No. I. New York, The Science Press. Delabarre, E. B. A Method of Recording Eye Movements. Amer- ican Journal of Psychology, July, 1898. Dodge, Ra3miond. Die Motorischen Wortvorstellungen. Halle, 1896. pp. 65. Visual Perception during Eye Movement. Psych. Review, Vol. VII, 1900. Eye Movements and the Perception of Motion. Psych. Review, January, 1904. Five T3rpes of Eye Movement. American Journal of Physiology, Vol. VIII, pp. 307-329. Act of Vision. Harpers, May, r902. The Illusion of Clear Vision during Eye Movement. Psycho- logical Bulletin, June 15, 1905. An Experunental Study of Visual Fixation. Psychological Review Monograph Supplements, Vol. VIII, No. 4, November, 1907. This article was received after Part One had been electrotyped. See Erdmann und Dodge. Egger, V. La Parole Interieure. Paris, 1904. Erdmann, B. Die Psychologischen Grundlagen der Beziehungen zwischen Sprechen und Denken. Archiv f. Systemat. Philos , III, 1897. Erdmann, B., und Dodge, R. Psychologische Untersuchungen uber das Lesen, auf ExperimenteUer Grundlage. Halle, 1898. Ezner. Entwurf zu einer Physiol. Erklarung der Psychischen Erscheinungen. Erster Theil. Leipsic und Wien, 1894. Versuche iiber die zu einer Gesichtswahmehmung nothige Zeit. Sitzungsbericht der Acad. d. Wissenchaft in Wien, Bd. LVIII, Abth. II, 1868. I BIBLIOGRAPHY 435 Floumoy. Temps de Lecture et Omission. V Annie Psycholo- gigue, II, 1896, pp. 45-53. Gale, Harlow. Psychology of Advertising. Psychological Studies, No. I, July, 1900. Published by the author, Minneapolis, Minn. Galton, Francis. Inquiries into Human Faculty. New York, 1883. Macmillan Co. Gildersleeve, B. L. Pindar (Preface). Olympian and Pythian Odes, 1885. Am. Book Co. Goldscheider, A., und MUUer, R. F. Zur Phys. und Path, des Lesens. Zeitschrift f. Klin. Med., Bd. XXHI, p. 131-. Gutzmann. Das Stottem, p. 178 ff. Frankfurt-a.-M., 1898. Die Praktische Anwendung der Sprachphysiologie beim ersten Leseunterricht. Berlin, 1897. Eelmboltz. Optique Physiologique. Physiologische Optik. Voss, Hamburg, 1896. Holt, E. B. Eye Movement and Central Anaesthesia. Psych. Review Monograph Supplements, Vol. IV, January, 1903. Huey, E. B. Preliminary Experiments in the Physiology and Psy- chology of Reading. American Journal oj Psychology, July, 1898. Psychology and Physiology of Reading. American Journal oj Psychology, Vol. XI, pp. 1-20, and Vol. XII. James, William. Psychology (Advanced Course), Vols. I-II. Henry Holt & Co. Javal, Emile. Notice sur les Travaux Scientifiques de Emile Javal. The author, Paris. (The Notice reviews the twenty-five articles which constitute the bibliography of Professor Javal's personal researches. See also the various numbers of Les Comptes rendus de I'Ecole des Hautes Etudes, Paris.) Conditions de la Lecture facile. Comptes rendus de la SocUtS de Biologie, 1879, p. 8. Sur la Physiologic de la Lecture. Annates d'Oculistique, 1878 et 18/9. (Several articles.) Physiology of Writing. Eng. trans. Harison, No. 59 Fifth Ave., New York. 43*5 BIBLIOGRAPHY Lamansky, S. PflUger's Archiv, 1869, p. 418 ff. Lamare. Des Mouvements des Yeux pendant la Lecture. Comptes rendus de la SociSte frangaise d'Ophtalmologie, 1893, p. 354 ff. Landolt. Nouvelles recherches sur la physiologic des mouvements des yeux. Archives d'Ophtalmologie, II, 1891, pp. 385-395. Meringer und Mayer. Versprechen und Verlesen. Stuttgart, 1895. Messmer, Oskar. Zur Psychologie des Lesens bei Kinder und Erwachsenen. Archiv fur die gesamte Psychologie, December, 1903, Bd. II, H. 2 u. 3, pp. 190-298. Meumann, E. Die Entstehung der ersten Wortbedeutungen beim Kinder. Wundt's Philosoph. Studien, XX. Festschrift 2. Miinsterberg, Hugo. Beitrage zu Exper. Psych., H. 4, p. 17 ff. Kllsbury, W. B. A Study in Apperception. American Journal of Psychology, Vol. VIII, No. 3, April, 1897. Preyer, W. Development of the Intellect. Appleton & Co. 1898. Quantz, J. O. Problems in the Psychology of Reading. Psych. Review Monograph Supplements, Vol. II, No. i, December, 1897. Ribot. Enqugte sur les Idfes G6n6Tales. Revue PhUosophique, October, 1891. The Evolution of General Ideas. Open Court Co., Chicago, 1899. Romanes, G. J. Mental Evolution in Animals. Appleton & Co., New York. Schumann, F. Psychologie des Lesens. Bericht iiber den II Kongress jUr Exper. Psych, in Wurzburg, 1906. pp. 31. Scripture, E. W. Elements of Experimental Phonetics. Chas. Scribner's Sons, New York. 1902. Secor. Visual Reading. American Journal of Psychology, Vol. XI, pp. 225-236. Stout, G. F. Analytical Psychology, Vols. I-II. London, 1896. The Macmillan Co. Manual of Psychology. Hinds & Noble, New York. Stratton, G. M. Eye Movements and the ^Esthetics of Visual Form. Wundt's Philosoph. Studien, XX. Festschrift 2. BIBLIOGRAPHY 437 Strieker. Studien iiber die Sprachvorstellungen. Sweet, Henry. Primer of Phonetics. Oxford Univ. Press, London. Henry Frowde, Agent, New York. ■ A Practical Study of Languages. Holt & Co., 1900. Talbot. American Journal of Psychology, Vol. VIII, p. 414. (De- scribes an attempt at training in visualization.) Taylor, Clifton O. Ueber das Verstehen von Worten und Satzen. Zeitschrijt j. Physiol, u. Psych, der Sinnes-Organe, Nov. 30, 1905. Wallin, J. E. Wallace. Researches on the Rhythm of Speech. Studies jrom Yale Psych. Lab., Vol. IX, 1901. pp. 143. Whitney, W. D. Life and Growth of Language. New York, 1880. Appleton & Co. Wirth, W. Zur Theorie des Bewusstseins Umfanges und seiner Messung. Wundl's PhUosoph. Studien, XX. Festschrift, 2. Woodworth, R. S. Vision and Localization during Eye Move- ments. Psychological Bulletin, Vol. Ill, No. 2, Feb. 15, 1906. in Proceedings Am. Psych. Assoc, r905-i9o6. Wimdt, Wilhelm. Volker-Psychologie, Die Sprache. W. Engel- mann, Leipsic. Zeitler, Julius. Tachistoskopische Versuche iiber das Lesen. Wundt's PhUosoph. Studien, Bd. XVI, H. 3, pp. 380-463. THE HISTORY OF READING AND WRITING Bouchot. The Printed Book; its history, illustration, and adorn- ment, from the days of Gutenberg to the present time. Trans- lated and enlarged by Bigemore. London, 1887. 312 pages. Budge. Egyptian Language. Kegan Paul, Trench, Triibner & Co., Ltd. London, 1899. Chambers. Encyclopaedia, article on Babylonia, p. 632. Clodd, Edward. Story of the Alphabet. Appleton & Co., New York. Copyright, 1900. Deniker. The Races of Man. Walter Scott Pub. Co., Felling-on Tyne, England. 1900. 438 BIBLIOGRAPHY Evans, Sir Arthur. Primitive Pictographs. Journal of HellenU Studies, XrV, ii. 1894. Cretan Pictographs and Pre-Phcenician Script. 1896. Further Discoveries of Cretan and ^gean Script. 1898. Fors3rth, W. History of Ancient Manuscripts. John Murray! London, 1872. ^ Him, Yrjo. Origins of Art. London, 1900. The Macmillan Co. Hoffman. Beginnings of Writing. Appleton & Co., New York. Copyright, 1896. Hogarth. Authority and Archeology. John Murray. London, 1899. Homer. Iliad, vi, 169. Judd. Genetic Psychology for Teachers. Appleton & Co. Copy- right, 1903. Plomer, H. R. Short History of English Printing. London, 1900. Ravrlings, Gertrude B. The Story of Books. Appleton & Co., New York. Taylor, Isaac. The Alphabet, Vols. I-II. Copjrright by Edward Arnold, publisher. London. Thompson. Greek and Latin Palaeography. Appleton & Co., New York. 1894. Tylor, E. B. Early History of Mankind. Holt & Co., New York. 1878. Vogt, C. L'Ecriture, etc. Rev. Scientifique, June 26, 1880, p. 1221 ff. Wimdt, W. Volker-Psychologie, Die Sprache. W. Engelmann. Leipsic. THE HISTORY OF READING METHODS Comenjus. The Great Didactic. Translated by Keatinge. London, 1896, pp. 468. Orbis Pictus. Edited by C. W. Bardeen. pp. 197. Bardeen, Syracuse, 1891. Fechner, H. Die Methoden des ersten Leseunterrichts. Wiegandt und Grieben, Berlin, 1882. BIBLIOGRAPHY 439 Fechner, H. Grundriss der Geschichte der Wichtigsten Leselehr- arten. Wiegandt und Grieben, Berlin, 1884. pp. 72. Johnson. Old-Time Schools and School Books. TheMacmillan Co. Leigh. Pronouncing Orthography. N. E. A. Report, 1873, pp. 207- 219. Mann, Horace. Seventh Report of the Secretary of the Massa- chusetts Board of Education. Published by the state. New England Primer. Boston, 1777. Reprinted by Ginn & Co. Prior, Matthew. Alma, Canto II. Reeder, R. R. Historical Development of School Readers and oi Method in Teaching Reading. Columbia Univ. Contribs. to Philos., Psych., and Ed., Vol. VIII, No. 2. The Macmillan Co., 1900, pp. 92. Scudder. Life of Noah Webster. Houghton, Mifflin & Co., Boston. Thornton. Cadmus, or a Treatise on Written Language. Common School Journal, Vol. II, 1840. METHODS AND GENERAL PEDAGOGY OF READING Adler, Felix. Moral Instruction of Children. Appleton &Co., 1895. Arnold, Sarah L. Waymarks for Teachers. Silver, Burdett Co.. 1894, pp. 152-186. Reading, How to Teach it. Silver, Burdett Co. Atlantic, July, rSgo, Vol. CXLIII, Pace in Reading. Broca and Sulzer. Reading and the Alphabet. La Nature, Feb. 13, 1904. Reviewed in Lit. Digest, March 12, 1904, p. 367. Bryan, E. B. Nascent Stages. Fed. Sem., Vol. VII, p. 357. Bullock, R. W. Some Observations on Children's Reading. N. E. A. Report, 1897, p. loig fE. Burt, Mary E. On Teaching Children to Read. New Eng. Mag., Dec, 1889, pp. 426-429. Experiments in the Teaching of Reading. The Dial, March 16, 1893. Chadwick, in New York Teachers' Monographs, June, 1902. 440 BIBLIOGRAPHY Chambers, Will Grant. How Words get Meaning. Fed. Sem., March, 1904. Clark, S. H. How to Read Aloud. The author, University of Chicago. Conradi. Children's Interests in Words, Slang, Stories, etc. Fed. Sem., Vol X, pp. 359-404- ; Cooke, Flora J. Articles in the El. School Teacher, October, 1900, p. Ill ff. and April, 1904, p. 544 ff. Dale, F. H. The Teaching of the Mother Tongue in Germany. Special Reports Ed. Dept. Great Britain. London, 1896-1897, PP- 536-578- Dewey, John. The Primary Education Fetich. Forum, Vol. XXV, pp. 315-328. in New York Teachers' Monographs, November, 1898. Eliot, Chas. W. Educational Review, July, 1891. Educational Reform. The Century Co., New York. Everett, Cora E., in New York Teachers' Monographs, June, 1902. Farnham. The Sentence Method of Teaching Reading. C. W. Bardeen, Syracuse, N.Y. 1887. v^ Ftink & Wagnalls. Standard Dictionary, pp. 2104-2107, and elsewhere.' Funk & Wagnalls Co., New York. Gordon, Emma K. The Comprehensive Method of Teaching Reading. Heath & Co. Copyright, 1902. Hancock. Children's Tendencies in the Use of Written Language Forms. N. W. Monthly, Vol. VIH, pp. 646-649. Iredell, Harriet. Eleanor learns to Read. Education, Vol. XIX, 1898, pp. 233-238. Judd, Chas. H. Genetic Psychology for Teachers, Chaps. VII-VIII. Copyright, 1903, by D. Appleton & Co. Kehr, K. Geschichte der Methodik des deutschen Volks-schul- unterrichts, Vol. I, on Lese-unterricht. Geschichte der Methodik. Lukens, H T. Children's Drawings in Early Years. Fed. Sem., Vol. Ill, pp. 79-110. BIBLIOGRAPHY 44 1 Lukens, H. T. The Joseph Stoiy. New York Teacher's Mag., April, 1899. March, Francis A. The Spelling Reform. U. S. Bureau of EA Report, 1893. McMurry, C. A. Special Method in Primary Reading and Oral Work with Stories, 1903. Special Method in the Reading of Complete English Classics, 1903. The Macmillan Company, New York. Merry, Robert. Book of Puzzles. New York Teachers' Monographs, June, 1902, and November, 1898. (Devoted to reading and language work.) Parker, Francis W. Talks on Pedagogics. A. S. Barnes & Co., New York. 1894. Patrick, G. T. W. Should Children under Ten learn to Read and Write. Pop. Set. Monthly, January, 1899, pp. 382-391. Peck, H. T. Psychology of the Printed Page. Cosmopolitan, Vol. XXXI, p. 161. Pollard, Rebecca. Synthetic Method of Reading and Spelling. Copyright by Am. Book Co., New York. Reinj W. Encyklopadisches Handbuch der Padagogik, Vol. IV. Das Lesen. Scott, W. D. The Theory of Advertising. Boston, 1903. Scripture, Mrs. E. W. In the Japanese Way. Outlook, 1897, Vol. LV, pp. 556-557- Scudder. Childhood in Literature and Art. Houghton, Mifflin & Co. Search. The Ideal School. Appleton & Co. Sherman, L. A. Analytics of Literature. Ginn & Co. Boston, 1893. Trettien, A. W. Psychology of the Language Interest of Children. Fed. Sent., June, 1904. University Quarterly, Vol. XX, p. 120. Science of Reading. Vaile, E.G. Reading as an Intellectual Process. Pop. Sci. Monthly, Vol. VIII, pp. 212-225. Ward, E. G. Rational Method in Reading. Silver, Burdett & Co., New York. in New York Teachers' Monographs, November, 1898. 442 BIBLIOGRAPHY Wheaton, Margaret, in New York Teachers' Monographs, Novembei; 1898.' ^ PRIMERS AND READERS Alger, Edith G. Primer of Work and Play. Boston, 1901. Amold-Forster, H. O. The Citizen Reader. London, 1885. Beard and Van Marter. Bible Symbols, or The Bible in Pictures. Hertel, Jenkins & Co., Chicago. Burt, Mary E. Poems that Every Child Should Know. Double- day, Page, & Co., New York. Deutsches Lesebuch fiir Hohere Anstalten, Vols. I-V. Puller, Sarah. An Illustrated Primer. Copyright, 1898, by D. C. Heath & Co., Boston. Funk & Wagnalls. Standard First and Second Reader. Funk & Wagnalls Co., New York. / Judson and Bender. Graded Literature Readers, First and Second Books. Maynard, Merrill & Co., 1899. McMuiry and Cook. Songs of the Treetop and Meadow. Bloom- ington. 111., 1899. ^ McMurry, Lida B. Classic Stories for the Little Ones. Blooming- ton, HI., 1904. Norton, Chas. Eliot. Heart of Oak Books. Heath & Co., Boston. . 1894. Shearer, J. W. The Combination Speller. Published by F. B. Johnson, Richmond, Va. Smith, Jessie R. Four True Stories of Life and Adventure. The Story of Washington. Copyright by E. H. Harison, publisher. New York. Smythe, E. Louise. Old Time Stories Retold for Children. Am. Book Co., New York. Spears and Augsbiu-g. Preparing to Read. New England Pub. Co., Boston. 1892. Summers, Maud. The Thought Reader, Bk. I. Ginn & Co., Boston. BIBLIOGRAPHY 443 Taylor, Frances Lilian. First Reader. Am. Book Co., igoo. The Werner Primer. Am. Book Co. Turpin. Classic Fables. Maynard, Merrill & Co. Williams, Sherman. Choice Literature, Bk. I. Am. Book Co., 1898. WHAT TO READ Burton, Richard. Literature for Children. N. Am. Revieiv, September, 1898, pp. 278-286. Chase, Susan F. Adolescence, Choice of Reading Matter. Journal of Adotescence, January, February, and March, 1901, and Ch. Study Monthly, Vol. VI, pp. 322-328. Dana, J. C. A Librarian's Experience on what the American People are Reading. Outlook, Dec. 5, 1903. Elmendorf, H. E. Some Things a Boy of Seventeen should have had an Opportunity to Read. Rev. of Reviews, December, 1903, PP- 713-717- Griffith, George. Ch. Study Monthly, February, 1899. Hall, G. S. How to Teach Reading and What to Read in Schools. Heath & Co., 1874, pp. 40. Adolescence, Vol. 11. D. Appleton & Co. Copyright, 1904. Haultain, Arnold. How to Read. Blackwood's Mag., Feb. i, 1896. Irving. Home Reading of School Children. Ped. Sent., Vol. VII, No. I. Tappan, Eva March. The Children's Hour; a comprehensive and careful selection of the best literature for children. Ten vols. Houghton, Mifflin, & Co., 1907. Vostrovsky. Reading Tastes. Ped. Sem., Vol. VI, No. 4. Williams, Sherman. Supplementary Reading. New York Teachers^ Monographs, November, 1898. Wissler, Clark. Children's Interests in Readmg. Ped. Sem., Vol. V, No. 4. \ 444 BIBLIOGRAPHY HYGIENE OF READING, AND REQUIREMENTS OF THE PRINTER Allport, Frank. Tests for Defective Vision in School Children. Educational Review, New York, 1897. Blasius und Ludicke. Deutsche Vierteljahrschrift f. off. Gesund- heitspflege, Bd. XIII, p. 432. Bravais, V. Du Mouvement des Yeux dans la Lecture. Lyon Medical, Nov. 29, 1891. Burgerstein und Netolitzky. Handbuch der Schulhygiene, zweite umgearbeitete auflage. Jena, 1902. Clouston. Neuroses of Development. Cohn, H. Hygiene of the Eye. Midland Educational Co., Ltd., Birmingham, England, i886. Lehrbuch der Hygiene des Auges. Wien und Leipzig, 1892. L'Ecriture, La T3rpographie, et le Progres de la Myopie. Rev. Scientifique, 1881, p. 297 ff. Dodge, Raymond. Act of Vision. Harpers, May, 1902. Bonders. Anomalien der Refraction und Accommodation. Griffing and Franz. Conditions of Fatigue in Reading. Psych. Review, Vol. Ill, 1896. Hartwell, E. M. Report of the Director of Physical Trainmg, Boston, 1894. School Document, No. 8, 1894, pp. 69-85. Javal, Emile. Myopia. Rev. Sci., 1879, p. 494 ff. L'Evolution de la Typographie. Rev. Sci., June 25, 1881. Bulletins et Memoires de la Socidtd franjaise d'Ophtalmologie, Vols. I-VI. Paris, 1883-1888. Mosso, A. Fatigue. Translated by Drummond. Putnam's, 1904, PP- 334- Sack, Dr. N. Die Ausseren Eigenschaften unserer Schulbucher vom Standpunkte der Hygiene des Auges. Reviewed by Erisman in Zeitschrijt fur Schtdgesundheitspflege, Nos. 4 and 5, 1898. BIBLIOGRAPHY 445 Sanford, Edmund C. Relative Legibility of the Small Letters. American Journal of Psychology, Vol. I, May, 1888. Shaw, Edward R. School Hygiene. The Macmillan Co. 1902. Weber, A. Ueber die Augenuntersuchungen in den hoheren Schulen zu Darmstadt. Referat und Memorial, erstattet der gross- herzoge MinisterialrAbtheilung fiir Gesimdheitspflege. 1881. INDEX Abell, Adelaide M., 171. Abnormalities of reading, 182. Abstract ideas represented in pic- tures, 196. Abstract thinking causes fatigue in reading, 403. Abundance pictured, 196. Accadian characters, 198-199. Accent placed on first part of word in English, 98. Accommodation, length of eyeball sometimes changed in, 391; con- stant tension of, in reading is fatiguing, 393; varies constantly in ordinary vision, 393. Accuracy of fixation fatigues in reading, 388. Aerology principle, 206; helped in word-analysis, 215 £f. Adams, President John, 244. Adams' "Agricultural Reader," 253. Adler, Felix, 335. Adolescence, brings craze for exten- sive reading, 368; over-analysis especially hurtful during, 368, 374. Adolescents, choice of reading mat- ter for. Chap. XIX. Adult methods used in teaching to read, 281. Advertisements suggest improve- ments in page, 427. ^gean civilization almost as old as Egyptian, 218. After-images, from reading, 18, 99; method of measuring rate by, 22 ; persistence of, after reading-pauses, 39; experiment with, seems to show greater importance of upper half of line, 99. Afternoon sessions advised, 393-394. Age of reading, i, 2. Ahrens, Dr. A., 20. Alertness, mental, increases rate of reading, 174. Alger, Edith G., 343. Alphabet, age of, 187; not attained in New World, 202 ; evolution of an. Chap. XI; importance of, 203; came by degrees, 204; a difficult acquisition, 211, 314; lack of, in cuneiform writing, 212; older than the pyramids, 213; transmission of, by Greeks and Romans, 217 ff., 222-223; gingerbread method of teaching the, 241, 275, 314. Set Letters. "Alphabetic Reform Print," 271. Alphabet method, 240 £f. ; 2 54 ff., 265 £F. See Method. Alphabets, list of world's, 221. American Philological Association, 355- American Spelling Reform Associa- tion, 355. Analysis, formerly carried to extreme in reading, 255; not desirable in early reading, 297; of sound, should be only partial at first, 355; over-analysis especially hurt- ful at adolescence, 368, 374; analy- sis of speech, unnatural, 353; must not come early, 398-399. Anglo-Saxon derivation of reading, i . Animals, drawings of extinct, 191. Aphasia suggested letter-reading, 71. Apperception, retouching of retinal image by, 67-68; contribution of, in reading, 79-88; as preceding assimilation, 83; activity of, in re- lation to memory-image, 105; of 447 448 INDEX dominant parts, 1 13-1 14. 5ce Per- ception. Arabs, paper introduced into Europe by, 235- Arnold, Sarah Louise, 343. Art, pedagogic, is often the removal of rubbish, 378. Articulation, mechanism of, 134; and elocution taught in early readers, 252; trained, in phonic methods, 266; correct habits in, not learned by sound analysis, 352; faulty, may be brought to consciousness, 353. See Pronun- ciation, Speech, etc. Artistic side emphasized in modern primers and readers, 276 ff. Assimilation, process of, 83-go. Association-habits, cause larger read- ings to the right, 61 ; of languages, 142 ff. ; condition extent of Sprach Umfang, 144. Associations, work best in original order, 98; few suggested by rela- tional words, 154 ff. ; flow with fatiguing rapidity in reading, 402. Associative expectancy, "forward push" of, 150. Assyrian cuneiform writing difficult to read, 212. Astigmatism may lead to myopia, 392-393. Asymmetrical accommodation, causes fatigue, 388; required by long lines, 409. Attention, expands at certain reading- pauses, 45 ; wandering of, during a reading-pause, 48-50, 69-70, 88-90 ; relative to fixation point, 60 ; attention-acts during a reading- pause, 68-70 ; to objective factors in brief exposures, 83; to dominating complexes, 88-go; eye movements and pauses in relation to move- ments of, 149 ff. ; behavior of, abnormal in exposure-experiments, 250; field of, in spider story, 157 ff.; to meanings assisted by presence of images, 162; to sub- stantive places in thinking, 166; concentration of, in reading pro- duces lip-movement, 173; power of concentration conditions rate of reading, 174; wider span of, a condition of rapid reading, 178, 179", direction of, in picture-reading, 226; direction of, in the Vcirious reading methods, 273; habits of, injured by too early study, 310; a normal activity, 400-401; atten- tion-acts in reading are so nu- merous as to be causes of fatigue, 401. Auditory elements, aroused at sight of determining letters, 81 ; in read- ing, 120 ff. ; and motor elements exceed the visual in reading, rS4 ff. Automatic reading, 149. Aztecs, writing of proper names in rebuses by, 205—206; analysis of words into syllables by, 210. B Babylonia, age of writing in, 187; libraries of, 236. Bagley, Dr. W. C, 182. Baker advocated for schools, 241. Barber's Shop lesson in "Orbis Pictus," 257. Barnum, Edith C, 290 ff. Basedow, J. B., 241. Battledore paddles as primers, 247. Beard, Frank, 326. Beech and book^ i. Bell, A. Melville, 397. Bible, the material for an ethnic, 377- 378; Bible stories for children, 335, 372- "Bible Symbols, or the Bible in Pictures," 326 ff. Bingham's Readers, 251. Blackboard, use of, in the sentence method, 273; and script, preferred by American teachers, 276; draw- ings on, as basis for reading-lessons, 342; requirements for legible writing, 415. Blake's "Historical Reader,'' 253. INDEX 449 Blending of sounds, taught in phonic method, 266, 284 ff. Blindness from myopia, 392. Blind persons, reading of, 82; may become progressively myopic, 391. Book, reputed connection of, with beech, i ; probable derivation of word, 188; development of mod- ern form of, 234. "Book of Puzzles," 326. Books, cost of, prevented extensive reading in early times, 237-238; superb illustrations of children's, 276 ff. ; for children's reading, 291 ff. ; bad effects of too early use of, 302 ff., 305; not needed in the elementary school, 309; inter- est in, comes spontaneously with children, 329; use of, in learning to read, 329 ff. ; progress in learn- ing to know, 331; originals pre- ferred to adaptations, 374; large, should have two columns per page, 411; optimum line-length for, 411; small, are preferable, 412; badly printed, should be rejected from schools, 417-418. Boston children, stuttering among, 352, 396: Boys, stuttering among, more prev- alent than among girls, 397. Brain, seat of consciousness, hypo- theticaUy, 106. Brazil, natives of, 190. Breath-groups, 136 ff. Bridgman, Laura B., method of teaching to read, 315 ff. Broca and Sulzer, 422. Brown, Alexander, 16. Bryan, E. B., 337. Budge's " Egyptian Language " quoted, 230. Buffalo-hunt pictured, 194. Bumstead Readers, word-method ad- vocated in, 252, 258-259. Burgerstein and Netolitzky, 407 ff. Burk, Professor, quoted, 340. Burnham, Professor, 304. Burroughs, John, 374. Burt, Mary E., 347, 372. Calkins, Mary W., I7r. Capitals, percentage of, per page, 95. Carlyle, Thomas, 423. Carnegie Institution, 414. Cartoons, and advertisements as examples of picture-writing, 227; suggest means of improving the page, 427- Cattell, J. McK., 36, 37, 42, 54, 62, 72-73. 84, 89-90. Cattell Fall Apparatus described, 55. Central America, pictography in, 202. Chadwick, Professor, 338. Chaldeans, reading of, 7; education of, compulsory, 236. Chart, use of, in teaching reading, 292 ff. Charts and maps, requirements of type for, 414-415. Chicago, primary curriculum in schools of, 307; methods in, of teaching to read, 339, 371. Chicago Institute, reading in the, 297. Children, reading of, 48; perception of reading-units by, 93; normally use lip-movement in reading, 121— 122; characteristics of child-lan- guage, 123-124; articulate in thinking and reading, 163; imag- ing of, in reading, 163; conscious- ness of meaning, 166; rate of reading, 180; reading of, should be studied, 181, 416; develop together the spoken and gesture languages, 189; drawings parallel those of primitive man, 194; analysis of word-sounds, 207 ; draw their thought of an object, 227; at first care little for order of words, 227; picture-stories of, 228; expe- riences of, used in teaching by sentence method, 273 ff. ; illustra- tions of books for, 278; natural language of, not used in primers, 279; want an outcome to discus- sions, 280; Readers for, 290- 292 ff. ; originality of, atrophied by too early reading, 302-303; 450 INDEX cannot respect primer-langutige, 305; proper school activities of young, 306-308; language of, is of ear and tongue, 308; need not read much until eight years of age, 311; learning to talk, 330; choice of reading-matter for, 334— 335; not ready for logical forms of most reading-matter, 337; natural rate of thinking and read- ing! 35°~35i; '^s^'i but little in school, 367 ; should read real litera- ture, 372; type-requirements for, 415 ff. ' "Children's Hour, The," 379. Children's librarians, as teachers of reading, 366. Chinese, picture characters of, 198 ft.; determinants in language of, 200 ff. ; language of, has many homophones and is monosyllabic, 208; words of, are not analyzed to letters, 208; use of tone or accent by, 208 ; writing of, 208 ff. ; language of, remained in rebus- phonograph stage, 208; never arrived at an alphabet, 209; ver- tical reading of, and of Japanese, 425. Chivalry, ideals of, 377. Choroid, straining of, in near work, .391 ff-. Civilization, reading and writing always possessed by, 187; grows eye-minded, 309. Classical languages, discipline of study of, compared with that of formal reading, 364. Classical study, slow Reading may be caused by, 179. Classics, really preferred by young child, 335; child's early reading of, lays foundations for correct tastes and ideals, 335; Old Mother Goose and other classics for chil- dren, 372. Clinical method of study, 182. Clodd, Edward, 187, 200 ff., 228 ff. Clouston, work by, quoted, 397. Cobb's Readers, 251. Cocaine, 25. Codices, 234. Cohn, H., 391 ff., 407 ff. College entrance requirements, should not be over-standardized, 368) should not dominate high schools, 369- College records correlated with read- ing rate, 174. Colored papers, illegibility of, 414. Columns per page, two, for large books, 411. Combat pictured, 195. Comenius, 256, 258, 272. Committee of Fifteen, 307. Committees of specialists to determine ideals for printing, 430-431. Common Prayer, book of, and primer had common origin, 243. "Comprehensive Method" of teach- ing to read, 286. Concept, meaning of, lies in fringe of consciousness, 165-166. Concept words, imagery from, 159 ff. Concert, learning to read in, 240, 274; reading in, prevents natural expression, 343. Concrete thinking should lead to abstract, 363. Congestion in the eye caused by near work, 391 ff. Consciousness, not in retina, 106; a stream of processes, 128 ff. ; diagram representing the sentence- consciousness, 131; characteristics of, in reading, r49 fi. Consonant, meaning of, 214. Consonants, not the determining letters exclusively, 80; character- ize word, 81; how-produced, 137 ff. Context, as affecting reading range, 56; exposure of words in, gave characteristic associations, 155 ff. ; used to suggest what new words are, 293, 296^ 333. Control, premature attempts at, cause strain, 398. Conventionalizing of pictures, 195- 196 ff., 328. Conversation, without images, i6oj INDEX 45' impossible in the dark with some tribes, 189; pictured, 198. Cooke, Flora J., 297 £F., 339. Correlation of reading with other subjects, 296. Counters, relational words serve as, in reading, 156. Counting eye-movements, 18. Crete, age of writing in, 187; exca- vations of Sir Arthur Evans in, throw doubt on Egypto-Phoeni- cian origin of alphabet, 217-218; place of origin of ^gean civiliza- tion, 218; Cretan signs similar to Egyptian and Hittite, 219. Criticism, tendencies to, often devel- oped before powers of appreciation, 375-376. Cross pictured, 196 and Fig. 35. Culture, progress of, measured by its interpretations of mythic tra- dition, 378. Cuneiform writing, 211, 212. Curriculum, of Chicago primary schools, 307 ; of elementary school, 310 ff., 336 £f. ; history of pictog- raphy has suggestions for new primary, 338-339- Cursive script of Romans, 223. Cypriote or Cyprus syllabary similar to the Cretan, 219. D, origin of letter, 213. Dearborn, Dr. Walter F., 43-50, 176, 411 ff. Delabarre, E. B., 20. Deniker, 226 ff. Dentition, period of .second, 332, 396 ff- De Quincey, Thomas, on discovery of printing, 238. Determining letters, 79-81. See Dominating Lettr^rs and Letters. Determinants, in picture-languages, 200 ff., 209. Devices, for teaching the alphabet, 341; for using various 'methods. 275; and technique not used in learning to talk, 330. See Meth- ods. Dewey, John, ideal of, used in Horace Mann School, 290; article by, cited, 304 ff. Diacritical marks, need of, shows deficiency of alphabet, 222; used to supply deficiencies of alphabet, 269 ff, ; systems of, 269 ff., 354 ff. ; in Pollard system, 282 ff.; in Ward method, 285 ff. ; not used in Gordon method, 287, 294; learning of, postponed, 299; page should ordinarily be free from, 350-351; needed for use of dic- tionary, 351; requirements of a good system of, 357. Dictation habit modifies style, 429. Dictionary, may be made by children, 299; use of, requires phonics and diacritical marks, 351-354 ff ' notoriously small print in, 417. Differential psychology, 182. Set Individual differences. DiscipUne, formal reading as, 10, zi; from reading compared with dis- cipline from study of classical languages, 364. Distance of reading-matter as affect- ing number of eye-movementa and pauses, 29. Distraction in turning to next line, 409. Dodge, Raymond, 17, 20-22, 34, 37. 39. 121, 389-390- Dominating letters and complexes, 82-91; not spatially arranged at first, 108-109; ^^ ff^st factors in recognition, 113-116; their place in the reading consciousness, 149. See Letters and Determining letters. Donders, Professor, 392. Dorians, Mycenas conquered by, 218. Dramatization in reading English, 37°- Drawing, as related to gestures, 189- 191; of children, parallels that of primitive man, 194; primi- 452 INDEX tive, "hits off" the essential, 195; used in teaching to read, 273, 319 ff., 328 ff. ; in correlation with early reading, 338 ff. ; inter- pretative, 344-345. See Pictures. Dualism, 106. E Ear, the nearest gateway to child- soul, 334. See Auditory. Easter Islanders, reading and writing of, 231. Ealing pictured, 196. Education, aided by differential psychology, 182; compulsory, of free Chaldeans, 236. Educators, views of representative, on early reading. Chap. XV. Egger, v., 123, 132, 164. Egypt, age of writing in, 187; sys- tem of writing as a type, 188; picture-metaphors of, 200; deter- minants in language of, 200; development and characteristics of writing of, 213 ff. ; alphabet early attained in, 213 ff. ; analysis of syllable in, 213; writing of vow- els and consonants, 214-215; rela- tions of, with early Greece, 218;- hieroglyphs arranged vertically and horizontally, 230; vast literature and great libraries of, 237; use of pictures and words together, 328; wasted characters as in modern spelling, 422. Elementary school may largely dis- pense with books, 329 ff. Elements, visual, exceeded by audi- tory and motor in reading, 154 ff. Elephant, drawings of woolly-haired, 192. Eliot, President, 253, 367. Elizabethan style, modern writing shows improvement upon, 428- 429. Emphasis, dependence of, on famil- iarity with granamatical construc- tion, 130 ff. English printers, running Italian script selected by, 224. English study, materials for, 376 ff. English teaching, too much analysis in. 353. 367-368. Environment, language, of young child, 330-331. Equilibrium of eye-muscles, main- tenance of, during a reading-pause, 39°- Erdmann and Dodge, 20 ff., 37, 52, 61-64, 73. 7S. 9°. 151- Esquimau writing, 228. Essential, "hit off" in primitive drawings, 195; practice in choos- ing the, trains the judgment, 364. Evans, Sir Arthur, 218 ff. Everett, Cora E., 275. Evolution, did not prepare eye and mind for reading, 8, 387; of letter- forms, 225; of the printed page. Chap. XII; will adapt eye to modern conditions, 395-396. Exchange, picture-character for, 226. Expectancy, aroused by relational words, 154 ff. Expectation-habits, in language, 142 ff. ; in reading, 157-158. Experiences, use of children's, for reading lessons, 273 ff., 297 ff., 339- "Experimental Phonetics, Elements of," quoted, 134 ff. Experiments, upon interpretation, imagery, etc., 152 ff. ; by Ribot, on imagery, 159; needed to deter- mine norms tor type, 406. Experiment schools, a better pri- mary course will be worked out by, 306. Exposure, optimum time of, 54, 90 ff. ; advantages of brief, 82-83; of unrelated words caused inner pronunciation, ii8; experiments, 150, 152 ff. Expression, teaching to read with, 324. Eye, work of, in reading. Chap. II; requires a reconstruction of pri- mary course, 395; will gradually adapt itself to reading, 395-396; INDEX 453 does not traverse whole line in reading, 410. Eyeball, length of, sometimes changed in accommodation, 391. Eye-fatigue, lessening of, by improv- ing page, 11; as related to mental fatigue, 399 S. See Fatigue. Eye-movements, counting of, 18; unconscious, 23; graphic record of, 26, 28; guidance of return, 29; and pauses per line, 29 £f. ; relation of attention to, and to fixa- 1 tions, 47 ; unconscious, produce illusions as to reading range, 53; and pauses in relation to move- ments of the attention, 149 £F. ; habits of, 176; excessive number of, in reading causes fatigue, 388; in traveling, 389; and pauses condition a state of mental ten- sion, 400. Eye-muscles not at rest during read- ing-pauses, 390. Eye-pauses, location of, 21, 30, 31, 47, 48; duration of, 32; persis- tence of after-image after, 39; attention expands at certain, 45; attention acts during, 68-70; at- tention wandering during, 69-70, 88-go; initial, are the longest, 178, 411-412; fatigue in reading caused by, as well as by eye-move- ments, 388; secondary long pause near end of line, 412. Eye-Strain related to nerve-strain, 395 ff- Eye-voice separation, 145 ff. Fairy tales, value of, for children, 372 ff. Familiarity with reading matter, as determining eye-voice separation, 145- Families of syllables, etc., 283, 287. Family, reading aloud as means of uniting the, 334- Famham, 272-273. Fatigue from reading, 19, 29, 48, 49, Chap. XX; causes of, 387- 388; from constant tension of accommodation, 393; is usually psychic, 399; as conditioned by attention-acts, 400, 401; causes of , 401-409. See Eye-fatigue. Favorite stories and poems, reading of, 295. Fechner, H., 275. Feelings, of relation, 130; vague, representing forgotten words and phrases, 153; "forward feeling" in reading, 156 ff. ; include the consciousness of meaning, 163; unanaJyzable, 163 ; training of, by reading, 377. Fence, fatigue from counting pickets on, 39. Fetich, the book a, 2 ; reading a, 303, 304 ff.; of Greek is passing, 304. Feudalism literature advised for adolescents, 376 ff. Fibel, German word for primer, meant Bible, 242. Field of clear vision, 52. See Read- ing range. Finnic language, few consonants in, 221. Fish, drawing of, in pictograph for Nineveh, 199. Fixation, fluctuations of, 53, 60; relation of, to attention, 477. Su Eye-pauses. Fixation point in reading, 30-31; relation of attention to, 60. Flavor of a great book elevates the personality, 365. Flax and linen not used among Greeks and Romans, 238. Flechsig, 398. Flournoy, 165. Fluent reading, how developed, 292. Folk-soul appealed to by books, 4. Foreigners, diacritical marks an aid to, 358. Foreign language, reading of, 19, 21, 147-148; learning of, in pri- mary school, 309 ; what constitutes real reading of, 428. 454 INDEX Form in reading as in swimming, 179. Forsyth, W., 238-239. "Forward push" of associative ex- pectancy in reading, 150, 155 ff. Fovea, size of, 66. Francis W. Parker School, 297 ff. French, sentence division in, 138. Fringe of consciousness, 161; carries the meaning, 182. Fuller's "Illustrated Primer,'' 319 ff. Fundamental movements should be learned before accessory, 316. Fundamental muscles need develop- ment first, 398. Funk and WagnaUs' Readers, 268 ff., 288-289, 355. Fusion, 35, 36, 39-43; of meaning with word-sound, 164. Gallon, Francis, 24. Games and plays, use of, in teaching to read, 299, 322 ff. See Play. Gedike's primer, 258. German literature edited for adoles- cents, 378-379. German printers, illegible Gothic script chosen by, 224. Gesture-languages, 189-190. Gestures, were the first pictures, 189; much the same for all chil- dren and all peoples, 191; pictures representing, 196 ff. Gingerbread method of teaching the alphabet, 241. Glasses, are often improperly pre- scribed, 395. Glides in breath-groups, 136-138. God Rock, Indian, 193. Golden Age of race not devoted to reading, 303. Goldscheider and Mviller, 54, 75-82, 102-103, 146, 354- Gordon, Emma K'., 286. Government publications, hygienic printing should be enforced for, 406. Government supervision, of research, 430; of school practice in reading and printing, 431. Graduate students, reading rate o^ 174, 177- Grammar, the old age of language, 367- Graphic methods, reader's time saved by, 426. Graphophone books, possibility of, 429. Graphophone experiments on sen- tence-utterance, 133 ff. Greece, far older than was supposed, 218; an inscription dating from early, 232; amount of reading in, and in Rome, 237. Greek alphabet, recognition of, 103; derivation of, 219. Grefeks, analysis of word-sounds by, 223; early, read from right to left, 231; music and gymnastics made the chief primary subjects by, 307; were relatively ear- minded, 309. Griffing and Franz, 406 ff. Growth will bring much that we strive for, 303. H Habit, hierarchy of, 112; of reading increased with invention of print- ing, 239, 297; of eye-movement prevented by broken lines, 412. See Reading. Habit-forming epoch, 308. Hall, President G. Stanley, 255, 304, 363. 365. 374 ff- Halves of words, relative importance of right and left, 96-98. Hartwell, Dr. E. M., 396 ff. "Heart of Oak" Readers, 254, 345. Heating the eye tends to myopia, 394- I Helmholtz, 53, 66. Henry the Eighth's primers, 242-243. Hereditary degeneration from eye- strain, 395. Hierarchy of habits, 112. INDEX 455 Hieroglyphic alphabet of Egypt, 216. High schools should not be domi- nated by college requirements, 369. Hirn, Yrjo, igo-191. History, of reading, needed for deter- mination of methods, 9-10, 328; of writing has suggestions for improving page, 424. See Evolu- tion, etc. Hittite arrangement of written char- acters, 231. Hittite signs similar to Egyptian and Cretan, 219. Hoffman, A. H., 195 ff., 231. Hogarth, 218. Holocain, 25. Holt, E. B., 38, 41. Home, learning to read at, 274-275, Chap. XVI, 311, 332, 379; read- ing of children who learn at, 330; question of school or, for young children, 336. Home reading for school children, 291 ff., 297-298. Home tasks and eye-strain, 394 ff. Homer, tablets in time of, 234. Homophones, in Chinese language, 208; in cuneiform writing, 212. Horace Mann School, reading meth- ods in, 290 ff. Horn Book, described, 244 ff. ; illus- trated, 246. Humboldt, W. von, 190. Hunter, record of starving, 192-193. Hygiene, requirements of, in read- ing. Chap. XX ; in printing. Chap. XXI. Hyphen, when used with compounds, lis; came late, 233. Ickelsamer, alphabet method modi- fied by, 255. Idealism, 106. Idealization of illustrations in chil- dren's primers, 276 ff. Tdeals, need of determining, for reading and printing, 430-431. Ideas, too numerous for picture symbols, 201. Ideograms developed to phonograms, 205. Illumination, effect of insufficient, less marked with large type, 407 See Light. Illusion, as to field of dear vision, 51-53; of simultaneity in per- ception, 88-90. Illustrations, for children's books, 278; of reading lessons by children, 298-299; economy from use of, 426. See Pictures. Imagery, at beginning of sentence- utterance, 130; from words and phrases when exposed, 154 ff, ; from concept words, 159 ff. Images, presence of, aids attention to meanings, 162. Imagination of children, 372-373. Imitation, language learned by, 124- 126; sometimes determines rate of reading, 179; of visible qualities by the picture-language and ges- ture-language, 190; fidelity in, not essential in primitive and in children's drawings, 194; em- ployed in picturings of letter- sounds, 255, 283; method of, in reading, 274; teaching of letter- sounds by, 281 ff., 286-287, 326, 352-353; method of, in learning to read and learning to talk, 330; gives us most of our habits, 365; discipline of reading partially due to, 365. Implicit apprehension, 160 ff. Incidental learning to read, 297 ff. Indentation of line, 21, 23, 30, 410; greater in second readings, 178; slight variations in line-length may assist the reader, 412. Indians derived picttires from ges- tures, 191. "Indifferent" letters, 79. "Indifferent" word-form, 80, 85. Indirect vision, first and last parts of line read in, 410. Individual differences, in perceiving 456 INDEX 92, 103, no, 114, 1 1 5-1 16; in imagery, 154 ff., 156 ff.; in rate of reading, 170, 361; in reading, 181-184, 403; in the consciousness of meaning, 183-184; in teacher's use of reading methods, 295. Individuality of words, 95. Inhalation pauses in speech, 136-137. Inner speech, 24; usually present in reading, 117; characteristics of, 121; slurred in rapid reading, 140-141 ; follows behind the eye, 144 ff.; enlarges visual range, 144; span of, 144; in reading words and phrases, 153 ff. ; accu- rate control of, 162-163 ; used to get meanings, 164; as hindering speed of reading, 172; only par- tially present in very rapid reading, 181. See Language and Speech. Inscriptions, Roman letters used for, 222. Interpretation and meaning. Chap. VIII. See Meaning. Introspection method of studying reading consciousness, 150. Irrelevant stimuli, ignored in read- ing. 39. 40. 41. Irving, Sir Henry, 370. Itzcoatl, rebus for, 205. Jacotot, Jean Joseph, 258. James, William, 106, 128-133, ^4^. 163, 165. 168, 350. Jansenists, phonetic system of, 258, 266. Japanese, writing of, 210^ have two syllabaries, 211; did not arrive at an alphabet, 211. Japanese and Chinese methods of learning to read, 315 ff. Javal, Emile, 16, 18, 27, 49, 72, 99, Chaps. XX-XXI. *' Johnny Story," 281 ff. Johnson, Clifton, 246, 256 ff., 426. Joseph, Story of, pictured, 228. Joubert, 132. fudd, Charles H., 207, 222. 223, 224. Keagy's Pestalozzian Primer, B49- 250. Kehr, K., 242. Kindergartens have less stuttering than primary schools, 397. "King Sent" inscription, 213, 424. Lamansky, S., eye-measurements by, 22. Lamare, 18. Landolt, 19, 20, 49. Lang, Andrew, 372. Language, differentiation of parts of speech, 125 ff. ; ancient and modern, compared, 125-126; dif- ferences of written from spoken, 126, 428-429; as a "movement of fixation," 162; spoken, arose with gesture-language, 189 ff. ; spoken, kept pace with needs of civilization, 203; American pro- nunciation made uniform by Web- ster's Spelling Book, 248; of children is of ear and tongue, 308; and literature as they should be taught in the elementary school, 309; interests at adolescence, 368. See Sentence, Speech, eic. Languages, associative habits of, 142 ff . ; differ in completeness of analy- sis to vowels and consonants, 221. Latin study miy be replaced by study of classics in the mother- tongue, 369. Lead, tickets of, for gladiatorial shows, 239- Leading, requirements as to, 408 ff. ; may be sacrificed in favor of larger type, 416. Learning of foreign languages, 428. Learning to read, very difficult in China, 209; mechanical devices for, 242; wastes in, 427; at home. Chap. XVI. See Methods. "Legato" and "staccato" reading, 181; "legato'' reading taught by sentence-method, 274. INDEX 457 Legible writing, requirements for,4i5. I^egibility, of upper and lower halves of line, i8; improvements in, needed in systems of diacritical marks, 355-357 ; requirements for, 406 S. ; lessened when book is supported at an angle, 412; may be increased by changes in letter- forms, 413; of letters, not the same in isolation and in context, 413; depends on black-white con- trast, 414; studies of, incomplete and little used, 422. Leigh, Edwin, 266 ff. ; "Pronounc- ing Orthography" of, 260. Leland Stanford requirements in English, 375. Lessons in reading, length of, 292. Letters, reading by, 68, 81-82; theory of reading by, suggested by apha- sia, 71, and founded on assump- tion of uninterrupted eye-move- ment, 71-72; differ in recogni- tion-time and in legibility, 84; projecting above and below the line, 91, 99; number of small, in page, 93; analysis of, 94-96; terminal, more legible, 97; rate of reading not proportional to number of, loo-ioi; part of, in perception, 109-116; factors producing consciousness of, 1 1 1 ; derivation of, 188; over-conscious- ness of, 206; final letter sometimes attached to the following word, 233; should be taught, 313; do not fully indicate pronunciation of words, 354 ff. ; minimal re- quirements of type for, 406 ff. ; unnecessary silent, used in spell- ing, 422; German use of initial capital, 423. Letter-forms, as characterizing word, 93-96; originally varied with writers, 224; letters contain rudi- ments of their stages of grovrth, 225; changes needed in, 413 ff. ; simplicity of, may facilitate recog- nition, 422; have never been adapted to reader's needs, 422. Letter-reading, suggested by apha sia, 71. Letter-sound, depends on context of letter, 138; represented by constant mark in Shearer system, 270-271; plays a part in German reading, 353; need not come to consciousness in reading, 354. Letter-writing in learning to read, 315- 318-319. 338, 371- Levels of nervous system develop at different times, 398. Librarians, will advise about choice of reading for children, 335, 379; as teachers of reading, 366. Libraries, Babylonian, 236; Egyp- tian, 237. Library, derivation of word, 188; growth of child's acquaintance with, 331; readers should be taught the effective use of, 365, 427; is the reading laboratory, 365-366; provides for story-telling and readings to children, 366; books of, should be school Readers, 374- Light pictured, 199. Light, insufficient, causes myopia, 39 1 ; apt to be bad at home, 394 ; conditions eye-strain, 394; white, gives greatest legibility, 414. See Illumination. Line, losing the, 29; indented more at right, 30; length of, 44-46; reading upper and lower halves of, 99; eye does not traverse whole, in reading, 410. Lines, rate of reading increased by short, and by uniform length of, 177; development of arrangement into, 229 ff. ; arrangement of, into columns, 234; asymmetry of accommodation lessened by short, 388; requirements as to length of, 409 ff. ; arrangement of, rests on tradition, 424. Lip-movement, 121-122, 173, 175, See Inner speech. Literary wholes should be read, 253, 373- 4S8 INDEX Literature, has become the main subject-matter of Readers, 253; study of, should begin very early, 345; brilliant, when people wrote as they talked, 369; importance of hearing as well as reading, 370; first readings in, 372 ff.; old Readers gave little acquaintance with, 373; for adolescents, needsi editing, 378. Livingstone, David, -;. Location, consciousness of, in read- ing, 158, "Looking" at line should be auto- matic, 401. LougH, Dr., 20. Lukens, H. T., 227-328. Luther's primer, 243. M 1(1, derivation of letter, 219 ff. Macula lutea, size of, 66. Magazine article, number of pauses per line in reading, 30. Mann, Horace, 253 ; argued for word- method, 259. Maps preferably do not present physical and political features together, 415. March, Francis A., 301. Marks. See Diacritical marks. Materials for writing, ' determined character of letters, 223; tended to develop serial arrangement of characters, 229. Mathematics ill-fitted to children, 308. Mayas, wrote words in rebuses, 207; analyzed syllables, 210; developed a few alphabetic characters, 221. McGuffey's Readers, 252. Meaning, first welds letters together in perception, 87, 161; usually in upper parts of objects, 99; domi- nates perception, 116; inheres in spoken language, 123; as guide to sentence-utterance, 123 ff., 140; as expressed by pitch, accent, modulation of voice, etc., 123 ff. ; as felt in different parts of sen- tence, 131-133; guides move- ments in throwing, 139; felt com- pletely only in sentence utterance, 146-147; of words and sentences, 158 ff. ; not in the image, 160; attention to, assisted by presence of images, 162 ; lies in feeling and in motor reaction's, 165-166; "mean- ing in itseU," 166; body of, may lie in bodily attitudes, etc., 167- 168; Professor Titchener's discus- sion of, 182; of concepts, 182-183; of picture characters, 199 ff.; learner should be concerned with total, 348 ff.; rate of getting, 359; grasped better when we read in large units, 360-361; appreciations of, may require acts of the atten- tion, 400; meaningful complexes apperceived preferably, 84. Mechanical reading, developed by un- natural subject-matter of primers, 279 ff., 319; causes of, 302. Mediterranean peoples alone at- tained to a true alphabet, 221. Melancthon, Philip, first Protestant primer made by, 243. Melody of speech appears in inner speech, 140. Memory, apperceptive activity in relation to memory image, 105; phrases and sentences existing as memory- wholes, 142; primary, of inner speech, 148 ; of rapid readers is more exact, 173; affective ty^e of, I S3; good among primitive peoples, 200; strengthened by substituting oral work for book methods, 310; memorizing of poems, stories, songs, etc., to be used in learning to read, 332-333. Mental discipline from reading, 427-428. Merry, Robert, 326. Messmer, Oskar, 61, 90 ff., 99, 109, 121, 182, 422-423. Metaphor in pictures, 195 ff. Methods, no rationalization of, 9; review of, needed, 9; of learning to read determine individual differ- ences in perceiving, 116; history INDEX 459 of alphabet method, 240 ff.; of ieaxning to read began with rise of syllabaries and alphabets, 240; and texts in elementary reading, Chap. XIV; classification of, 265 ; use of blaikboard in ' sen- tence method, 273; directions of attention in the vatious, 273; adult, used in teaching to read, 281 ; in Chicago schools, 339, 371. Metonymy, used in pictures, 195. Mexican writing, 205 ff. Microphone experiment, ig. Middle Ages, reading in the, 237. Minuscule letters chosen for Anglo- Saxon readers, 224. ^ Mirror observation of eye-move- ments, 15, 20. Monogrammatic linking of final letters, 233. Moon pictured, 198. Moral purpose of adolescent reading, 375- Morning pictured, 197. Morning sessions should be shorter, 394- Mosso, Professor A., 304. Mother Goose, 299, 334, 372. Motion, representation of, in read- ing, 157- • Motive in reading, 299, 305. Motor elements, and auditory ele- ments more stably knitted together than visual elements, 144; exceed the visual in reading, 154 ff. Motor habits in reading, 44, 176 ff. Motor training should dominate primary school, 398. Motorizing, experiments which re- vealed, 119-122; of words in isolation, 146. Miinsterberg, Hugo, 79. Miuray's Readers, 251. Muscae volitantes, show character of eye- movements, 23. Mycense's relations with Egypt and Assyria, 218. Myopia, 19, 390; astigmatism may lead to, 382, 383; caused by reading and other near work of the school, 390 ff. ; blind persons may become progressively myopic, 391; incurable, 392; progressive, in youth, 392. Mystery of reading, 2. Myth, the mother of poetry, religion, and art, 377. Myths, give the essential, 195; and folk tales for children's reading, 334-335. 372 ff-; of North and South compared, 376-377; inter- pretation of, shows progress of culture, 378. N N, origin of letter, 213, 220. Names, proper, written in rebuses by Aztecs, 205; child's learning of, 314 ff.; learned from pictures, 315. See Words. National Educational Association Report, 268. National Readers, 252. Nature Study in primary school, 336. Neapolitans proficient in gesture- language, 189. Near work, troubles of vision caused by, 388 ff. Nerve exhaustion, and painful vision, 399; in persons who read actively, 403. Nervous system disturbed at second dentition, 396 ff. Neural functionings in reading, 151. New England Primer, 244 ff. Newspaper, number of pauses per line in reading an article in, 30; has felt reaction of readers, 410- 411; line-length in, near the pres- ent optimum, 411; characteriza- tion of important words and phrases by, 423- Nibelungen mythology as basis for adolescent literature, 376 ff. Nineveh pictured, 199. Nonsense-matter, reading range for, 63-65. • "Normal Word Method" of Gm mans, 315. 460 INDEX Norms, minimal, for printing, 408 ff. See Type-size. Norton, Charles Eliot, 254. Novel, read experimentally, 118; rate of reading selections from a, 174. "Now I lay me," prayer in New England Primer, 244. Nystagmus, 53. O Objective type of reader, 92. Object-lessons, in Keagy's Primer, 256; in "Orbis Pictus," 256. Objects labeled to teach names, 316 £f. Observation should precede reading, 362. Ojibwa pictures, 197. "Old-Time Schools and School Books," 245, 246, 247, 256, 257. Oral language work, preferable from a hygienic point of view, 395. See Speech, etc. Oral methods in the primary school, 336 ff- Oral reading, overemphasized, 302; an accomplishment, 342. See Read- ing aloud. Oral teaching best for young children, 308 ff., 334. " Orbis Pictus" started word-method, 256, 257. Organic sensations as determining types of individuals, 183. Organism taxed severely in reading, 387- Orient, reading in the, 240, 274. Ostracize, the word, suggests the poverty of writing materials, 239. Over-reading, dangers of, 362. Pace, of judging, as affected by clas- sical study, 365; in reading, 404- 405. See Rate of reading. Page, changes proposed, 11; evolu- tion of printed. Chap. XII; im- provement of, means a great ser- vice to human race, 42 r; not rationalized, 421; history of writ- ing has suggestions for improving, 424; improvements of, suggested by advertisements, 427; possi- bilities of improving and beautify- ing, 427. Pantomimic language, 190. Paper, absence of, among Greeks and Romans, 238; early known to Chinese, 234; different from papyrus, 234; brought to Europe by Arabs, 235; requirements as to, for printing, 414. Papyrus, 238; and veUum used in books, 234 ff. Paragraph-divisions, development of, 233- Paraphrasing, 147-148. Parents, instruction of, by the schools, 312. Parker, Francis W., 302, 342, 371; Francis W. Parker School, 279 ff. Pater, Walter, 378. Paternoster, Aitec rebus for, 206. Patrick, G. T. W., 306 ff., 307. Pauses. See Eye-pauses and Fixa- tion. Pedagogical conclusions summarized, 379 ff- Perception, of total form, 74-75, 114-116; as facilitated by inter- relation of units perceived, 75-79; of general forms, 77; theory of, by dominant complexes, 85-89; not simultaneous, 88-90; of total form by subjective type of readers, 92 ; of details by objective readers, 92; simultaneous and successive, 92-93 ; by dominant letters, 92-96 ; general features of, 104 ff. ; of word length, total form, domi- nant parts, etc., 109-116; of a letter, 112-113; by letters, iii- 116; sketchiness of, 194; ad- vantages of, in large units, 423. Peripheral parts of retinal image, 67. Personification, in Pollard method, 283; of letter-sounds, 287. Pessimism concerning bad effects of / INDEX 461 reading not warranted if final outcome is regarded, 396. Philanthropinists, use of alphabet method by, 255. Philological Association, American, 355- Philologists, commend Scientific Al- phabet, 355; point of view of, needed in studying the problems of reading, 430. Phoenicians, supposed to have adopted Egyptian alphabet, 217 ff. ; con- tribution of, to our alphabet, 219. Phonetic method in Funk and Wag- nails' Readers, 288. Phonetics may result in word read- ing, 294. Phonetic spelling, a goal that will be reached sooner or later, 356- 357; systems of, too often con- structed from a single point of view, 430. Phonetic system, reached by Chinese, 202; attained by nations about the Mediterranean, 202; signs and ideographic keys required in Chinese, 209; of Jansenists, 25S; in the Horace Mann School, 294- 296. Phonetic systems need revision from psychological and pedagogical view- points, 358. Phonetic writing of early peoples tended to start the alphabet method, 240. Phonics, always involved in alphabet methods, 266; taught with reading in the primers, 280; as taught in Pollard method, 281; kept apart from reading, 284, 299; in the Gordon method, 287; tends to mechanical reading, 302; primer- directions about, should ordinarily be ignored, 325; should be taught in due time, 334; has double pur- pose, 351; is dangerous in early years, 352; method of teaching, 354 ff. Phonic method, described, 260; developed to phonetic method, 260; as training articulation, 266; used by Jansenists in Port Royal Schools, 266. Phonograms, development of, 204 ff. ; of Ward method, 284 ff. Photographic measurements of rate, 22, 34. 43- Phrases, perception of, 114-116. Phrasing, differences in, 181. Physicians read effectively, 366. Physiological objections to early reading, 304-305. Pictographs, used in learning to read, 328; became ideographs, 195; un- realized possibilities of, 201. Pictorial origin of Egyptian letters, 216. Picture atlas, 315; pictiu-e-books forbidden in school, 426. Picture dictionaries, 328. Picture-languages, advantages of, over phonetic lajiguages, 201. Picture letter, of a Mandan Indian, 226; of Chippeway Indians, 227. Picture letters in teaming to read, 326. Picture reading, direction of atten- tion in, 226. Picture readings, 326-327. Pictures, reading and writing began with, i88; materials used in making, 192 ff. ; contained the germs of alphabets, 192; repre- senting abstract ideas, 196; tended to become symbols, 196, 198 ff. ; came to represent names, 204; sounds represented by, 204 ; origin of letters in, 220; use of, in the "Orbis Pictus," 256 ff. ; arrange- ment of, in pictography, 266 ff.; reading of, 288-289; used to teach words and sentences, 319 ff.; illustrative, call attention to sen- tence-meanings, 323-324; of sen- tences and stories, 324-325; use of, in learning to read, 333. Picture stories, children's, 228. Picture writing, much the same foi all primitive peoples and for all children, 191; very ancient, 191- 462 INDEX 192; of early Greeks, 218; and perceiving, 227; read by all, 236,' 322 ff. Pierpont's Readers, 250. Play as means of learniijg to read, 313 ff., 241. Plus distance between seat and desk causes eye-strain, 394. Poems, reading of, by school chil- dren, 292; memorization of, 296; poems and stories for children, 335. Pointing results in word-reading, 292. Pollard, Rebecca S., 281 ff. Practice and rate of reading, 174, 179. Preposition, often attached to related word, 233. Fresentative states other than images, i6r. Preyer, W., 123. Priests came to be the readers, 236. Primary reading, dissatisfaction with, 301. Primers, should usually be avoided, 319; contents of early, 242; were religious books at first, 242-243; New England Primer described, 244 ff. ; battledore paddles as, 247 ; contents of early American, 249; and Readers emphasize artistic side, 276 S. ; natural language of children not used in, 278, 279, 305- Primitive life, reading lessons based on, 290 ff. Primitive man, proficiency of, in use of gestures, 189; mind -of, works similarly in all times and places, 191. Primitive readers, acuteness of, 201. Print, Leigh's, 261; used before script at the Horace Mann School, 292; disjointedness of, gives a hobble to reading, 360; must not show on reverse side of paper, 414. "Print, Alphabetical Reform," 271. Printers, the first, selected types from handwritings that pleased them, 224; are inconvenienced by very short lines, 411. Printing, separates parts which belong together in speech, 115; stereo- typed handwritings, 235; inven- tion of, increased habit of reading, 239, 297; came soon after intro- duction of paper, 239; future of, Chap. XXII; committees of specialists should determine ideals for, 430-431. / Printing of reading lessons, by chil- dren themselves, 298; large char- acters, at first, 316. Printing-press, habit of reading pro- duced by, 7. Prior, Matthew, 241. Problems, specific, should be studied, 43°- Professors of English too analytical, 367- Projection in perceiving, 105-106. Promptness, taught in Ward method, 284 S.; in judging, may be trained by reading, 363-364. Pronunciation, as taught by Shearer system, 270-271; of words, the goal of most primers, 280 S.; how learned,. 351. Proof-reading, 21. Prose, printed as poetry, r37 ; mod- ern, compared with modern speech, 428-429. Proto-Medic tribes reduced the As- syrian cuneiform to a syllabary, 212. Psychic economies and pace of read- ing, 404-405. Psychologist's point of view must be synthesized with those of philol- ogist and educator, 430. Psychology and pedagogy of picture- printing, 426. , Psycho-educational departments, im- portant problems for, 358. Publishers, as influencing methods, 9; courtesies of, 276; forced by competition to develop artistic side of Readers, 276; will print hygienically if schools require it, 4J8. Punctuation marks, development o!, 233- INDEX 463 Puritans' Catechism succeeded by New England Primer, 243. , Putnam's Readers, definitions of words introduced by, 250. Quantz, Dr. J. O., 54, 121, 145 0., 172-174. Quintilian, 24, 376. R's, the three, percentage of time given to, in different grades, 307. Race degeneration from reading habit, 8. Radical empiricism, James', 106 u. Rapidity of thought in reading causes fatigue, 403. Rapid reading, advantages of, 170 ff., 360 ; accompanied by longer initial pauses, 178; from the first, 350. See Rate of reading. Rate of eye-movement. See Eye- movements. Rate of reading. Chap, IX; ignored in teaching, 10, 33, 43, 49; slow, for unrelated forms, 73; by various assigned methods, 119-121; cor- related with eye-voice separation, 146; how increased, 172-179; conditioned by power of concentra- ting attention, 174; as dependent on habits of eye-movement, 176, 177; natural rate of children, 350-351; importance of rapid rate, 359; speed drills advised, 381. Rationalization of letter-forms has been slight, 225. "Rational Method" of teaching reading, 284 ff. 1 Readers, first grading of, 250; articu- lation and elocution taught in early, 252; used in Horace Mann School, 295, 296; composed by children, 339 ff.; value of the old, 373; formal, may not be neces- sary, 373-374; "Das Deutsche Lesebuch," 378. Reading, derivation of word, i; age of, I, 2; long controlled by priesthood, 3; as identical with education, 3; habit of, 7-9; in larger units, 72; wandering of attention in, 148-150; simulta- neous or successive, 150 ff. ; rate of, 180; abnormalities of, 182; amount of, at various times in world's history, 236; difficulty of, in early systems of writing, 236; in modern sense came with devel- opment of syllabaries and alphabets, 240; and pronouncing, confused in primers, 280;' lesson in, about visit to farm, 298; not an end in itself, 300; mechanics of, and of spelling, consume too much time, 301; should be secondary until at least the eighth year, 303 ; place in curriculum, was acquired in a different age, 304; time wasted in teaching, too early, 309; may be learned incidentally, 334; by acting, 343; reading directions for occupation-work, etc., 344; learning to read without technique of method, 345; as translation, 349; as thought-getting rather than thought-expressing, 350; as mental discipline, 361-364, Chap. XVIII; may habituate to inaction and indecision, 362 ; by ear, 370 ; craze for, in adolescents, 375; is always neurally expensive, 404; not rationalized, 421; possible displacement of, by other means of communicating, 429; future of. Chap. XXII. Reading and writing, are very old, 187; development of, much the same in all countries, 188. Reading aloud, characteristics of, 120— 121; means by which proper emphasis is given, 130-131; rate of, 173, 175; determines rate ol reading, 179; to chUdren, 331-332; at home, 334; need not be ir exact words of book, 342, 349; is natural in the early years, 349; at the expense of reading foi thought, 359. 464 INDEX Reading machines, 275, 283. Reading matter, for children, 334- 335; should be of two classes, 371; the real and the unreal in children's readings, 372 ff. Reading-pauses, habit of making fixed number of, per line, 176 ff. See Eye-pauses. Reading range, measured, 55, Chap. Ill; varies with matter exposed, 58, 65; to right and left, 58-61; conditions determining, 68-70 ; increased in vertical reading, 426. See Field of clear vision. Rebus-writing, as transition stage from picture-writing to word- writing, 204, 205 ; use of, in learn- ing to read, 326. Recapitulation, theory of, in relation to the primary school course, 308, 309; in learning to read, 326; re- capitulation interest in pictography, 338-339; in reading, 372 ff. Recognition as a physiological pro- cess, 183. Recognitions, possible number of, during a reading-pause, 69-70. Reader, R. R., Chap. XIII. Reference books, small type found in, 417- Reforms can be hastened under modern conditions, 431. Reindeer Period, drawings from, 192. Relation, feelings of, r63 ff. Relational words, suggest few asso- ciations, r54 ff. Renan, Ernst, 240. Research, departments for, should carry out studies upon legibility, 4r4; needed to determine ideals for reading and printing, 430. Rest, periods of, 393-394; rest- periods needed in reading, 405; how to rest the eyes while travel- ing, 389-390. Retina, sensitive during eye-move- ment, 39; peripheral discrimina- tion of brightness, 51; detach- ment of, in myopia, 392. Retinal fatigue, 387-388. Retinal image, characteristics of, 65-68; retouched by apperception, 67-68. Retinal periphery more available in vertical reading, 426. Retracal movements of eye in read- ing, 27. Retracal pauses, 32. Return movements in reading, 27, 29, 44. See Eye-movements. Reverence, for reading and readers, 2-3; the premature, for books, 302; for written words, 415. Rhetoric, was oratory in Greece, 369. Rhymes for children, 372. Rhythm, as factor in determining rate of reading, 46, 177, 179; of inner speech, 140-141; as factor in recall, 153; in reading, 175. Rhythmic succession in perception, 88-89. " Rhythmizing," of word by its long letters, 94; of words by dominant letters, 96. Ribot, Professor, 159, 189. Rock drawings, 193 ff. Romanes, G. J., 170-171. Romans, word for reading in lan- guage of, i; capital letters of, 222; cursive script of, 223; uncial letters of, 224. Rousseau, J. J., 362. Russell, Principal, 181. Sack, Dr. N., 407 ff. Samplers as primers, 247. Sanford, Edmund C, 413. Sanskrit, fine analysis of sounds in, 221. Santa Rosa, Cal., schools, 339 ff. School, learning to read at, Chap. XVII. School activities of young children, 306-308 ff. School-books, hygienic printing of, should be enforced, 406, 409; often badly printed, 417-418; "Orbis Pictus" the first of illus- trated, 256. INDEX 465 Schools, elementary, curriculum of, 310 ff.; the nurseries of stuttering, 3S2. 397- Schulte's primer, 243. Schumann, Dr. F., 151. "Scientific alphabet," illustrated and described, 268-269; ^ used in Funk and Wagnalls' Readers, 288, 3SS ff. "Scrappy" Readers denounced by Horace Mann and President EHot, 253- Script, used at first in Ward method, 286; transition to print is easy, 317; and blackboard, preferred by American teachers, 276. Scripture, Edward W,, 133 ff. Scripture, Mrs. E. W., 316 ff. Scudder, Horace, 248, 334. Search, Professor, 337. Secor, ir7. Secret languages of children, 328. Selective reading, 360, 423, 427; fatiguing at first, but becomes automatic, 404, 405. Semites, supposed to have borrowed Egyptian alphabet, 217; wrote from right to left like modern Hebrew, 231. Semitic alphabet mainly consonantal, 221. Sentence, is the unit in language, 123 ff., 273; not a mere sequence of word-sounds and word-names, 125, 273; formation of, analytical but also synthetic, 127-128; a "vol- untary act," 128; physical utter- ance of, 133 ff. ; welded together by unity of its physical utterance, 139; unitized by variations in stress, pitch, rhythm, melody, 140; written in rebuses, 207-208; anal- ysis of, in sentence method, 274; learning to recognize sen- tences, 318. Sentence divisions, development of, 233- "Sentence hash" in primers, 280. Sentence method, 293, 298 ff. ; sug- gested by Comenius, popularized by Farnham, 272-273; with the use of the blackboard, 273 ff. ; usually contrasted with phonic method, 274; begins with mean- ing-wholes, 274. See Method. Sentence-reading is more than word- reading, 317 ff. Sentence utterance, simultaneity and succession in consciousness of, 128; cues to, 142 ff. Sentence words, 123-124. Serial writing, development of, 228 ff. "Set" in perceiving, 108, 113; in reading, 15s ff. Shaw's requirements as to type for children's books, 416-417. Shearer, J. W., 270 ff., 357. Shells as material for writing, 239. Sherman, L. A., 428-429. Shifting eye-movements, 46. Sight- words of Ward method, 284 ff. Signs, indicative of distance, position, size, etc., 107-108. Silent letters, retention of, resembles Egyptian failure to rationalize writing, 217; in Leigh's system, 268; as marked in Shearer system, 357- Silent reading, ignored in teaching, 10, 359; simpler and faster than reading aloud, 120-12 1; in second year's course, 296; importance of, 342 ff. ; in games and plays, 344- Simultaneous action of cues to per- ception, 110-116. Singing pictured, 198. Skimming, facilitation of, by short lines, 411, 423. Sky pictured, 199. Slates, writing upon, less legible than upon paper, 415. Slow reading injurious, 350, Smith, E. Louise, 341. Smith, Jessie R., 278, 339. Siunv pictured, 197. [Song pictured, 199. Songs, singing of, in learning to read, 333, 335. Sound, as drawn, 194; and speech 466 INDEX pictured, 197-198; analysis of initial, by aerology, 206. Sound-blending, taught in phonic method, 266. Sounds, selection of signs to repre- sent, 203 ; pictured in rebus, 205 ff. ; blending of letter sounds, 284 ff.; of letters should not be attended to in reading, 350-351. Spacing, development of, between words, 232; of letters and words, 407 ff. Spears and Augsburg's "Preparing to Read," 319. Specialists, committees of, should determine norms for printing, 413. Speech, mechanism of, 133 ff.; not broken into letters, syllables, and words, 134 ff.; rhythm and melody of, 137 ff.; habits of, more deeply founded than visual language habits, 143 ff. ; psycho-physics of, needed in peda- gogy, 141; not inherited, 330; conscioiLsness of, causes stuttering, 352; disturbance at second den- tition, 352; analysis of, 353; organs affected in stuttering, 397; dis- turbances from reading aloud, 396 ff.; analysis of, must not come too early, 398-399. See Inner Speech, Language, eic. "Speller, Combination," Shearer's, 270. Spellers, displacement of, by First Readers as introductory to read- ing, 252. Spelling, an old lesson in, 249; pre- ceded pronunciation in alphabet method, 255; opposed by Come- nius, 258; taught by alphabet method, 266; irrational, wastes time, 301 ff.; psychology of, 356; habits of, confused by reform agitation, 356-357; in Egyptian characters, 422. Spelling Board, Simplified, 302. Spelling Reform, 301, 355-357. Spelling Reform Association, Ameri- can, 355. Spider story, reading of, 152 ff. Spoken language the real language 122. See Speech. Sprach Umjang, extent of, as con- ditioned by association habits, 144. Standard Dictionary, 268, 270. Standard methods, 265. Standardizing of college entrance requirements, 375 ff. Starving hunter, record of a, drawn by an Indian, 192. Stewart, Dugald, 189. Stone Age, drawings from, 192. Stories, reading of true, 373. Story interests of children, 296. Stout, G. F., n6, 160 ff. Strabismus caused by near work of reading, 393. Stress, alternations of, as factors in unitizing sentence, 140. Students, inability of, to use books effectively, 306; unable to read selectively, 364. Study, bad mental habits acquired by too early attempts at, 309- 310. Stutterers, census of, 396. Stuttering among Boston children, 352; treatment of, 398. Style, need of studying the psychology of, 429. Subjective type of reader, 92. Subject-matter as conditioning rate of reading, 177. Su£6xes, preponderance of, makes beginning of words more impor- tant, 98. Summers, Maud, "The Thought Reader" by, 324, 343. Sun pictured, 198. Supplementary Readers, first appear- ance of, 253; in Ward method, 286. Survey of line taken at the initial pause, 412. Sweet, Henry, 136 ff. Syllabary, of Cypriote or Cyprus, similar to Cretan, 219. Syllable analysis, by aerology, 206; aSDEX 467 sane with difficulty, 207, 214; by Egyptians, 213-214. -^liable division, determination of, 138. syllables, number of, that can be uttered during single exhalations, 136; differentiated by variations in stress, 137; represented by pictures, 204 ff.; analysis of words into, by Aztecs, 210. oynthetic method of reading and spelling, 281 £F. Systems of teaching to read, 281 S. Tablets, wooden, the first books, 234. Talk, learning to read like learning to, 297; devices and technique not used in learning to, 330. See Speech, etc. Tappan, Eva March, 379. Tattooing, 194. Taylor, F. Lilian, 322 £F., 333, 344- 34S. 346. Taylor, I., 204 ff,, 208, 232. Teachers, the over-strenuous, 350. Technique, of learning to read, 275, 285, 300. Telegraph reading by ear, 370, 429. Telescope observations of eye-move- ments, 21. Texts and manuals in elementary reading. Chaps. XIII-XIV. Thickness, of letters, 407 ff.; of paper, 414. Thompson, 232-234 ff. Thornton, Dr., 266; phonetic system of, 259. Thoroughness misunderstood, 367- 368- Thought, of what is read, means of interesting in, 292; reading for, 302. See Consciousness, Meaning, etc. Time, waste of, in learning to read, 301. Titchener, Professor, 182-184. Tomb-board of Indian chief, 193. Tone-variations, use of, in Chinese language, 208. Total idea, as guide to sentence utterance, and as expressed in parts of speech, 124 ff. ; total ideas require acts of the attention, 400. See Sentence. Tradition maintains reading as main study of primary school, 303-304. Train, fatigue from reading on, 389- 39°- Transitive places in consciousness, 129. Translation, reading as, into speech, 123. Translation English, at adolescence, 368. Travehng, pictured, 197; eye-move- ments in, 389. Tylor, E. B., 189 ff. Types of perception in reading, 92. See Individual differences. Type-size, as affecting number of eye-movements and eye-pauses in reading, 29-30; the most impor- tant factor in legibility, 406 ff. See LegibiUty, etc. Typography, history of, should be consulted before changing letter- forms, 413. U Uncial letters of Romans, 223-224. Uniformity of line-length urged, 412. Verbal associations in reading, 155 S. Verbal fringe of consciousness, 183. Verbal fringe type, 183. Vertical reading, without lateral movement of eyes, 410; experi- ments upon, 425; of Chinese and Japanese, 425; advantages of, 425-426. Vision, during eye-movement, 36-42; 46. See Perception, etc. a68 INDEX Visual field, size of, 37. See Read- ing range, etc. Visual form, characteristic feeling aroused by, 168. (Visualization, static, in reading, 157. Visual perception, quickness of, a condition of rapid reading, 174. Visual reading, 10, 117, 180-181. Visual types and auditory types differ in rate of reading, 1,73. Vocal organs studied in learning to read, 283. See Speech. " Volker-Psychologie," Wundt's, quo- tation from, 125-126. Vowels, give the clew to the number of syllables, 80; awaken memory of rhythm and accent, 80; mis- read oftenest, 83; vary constandy in pitch, 135; how produced, 137; assist in syllable division, 137; writing of, in Semitic, 222. See Letters, Speech, etc. Vowel sounds, that are of determin- ing significance, 81; analysis of words to, easier than to conso- nants, 214. W Wallin, Dr. J. E. W., 133 £f. Ward, E. G., 284 ff., 350. Ward Reader quotation, 285. Warning pictured, 195. Washington, story of, 340. Waste, ehmination of. Chap. XXII; elimination of, by changing the rate of reading, 180; wastes from failure to rationalize reading and writing, among ancient peoples, 427; in modern times, 427 ff.; elimination of, from improve- ments in style, 428-429. Waterloo, Stanley, 290. Webb's Normal Readers, 259. Webster's Reader, 249. Webster's Spelling Book, enormous sales of, 248; contents of, 24S. Werner Primer, 323 ff., 344. Whately, Dr. Richard, 238. Wheaton, Margaret, 342. Whispering, all consonants can be whispered, 138. Whitney, W. D., 122. Williams, Sherman, 345. Willson's Readers used science as subject-matter, 253. Winthrop, Dr., 366. Woodworth, R. S., 38, 41. Woolen garments worn by Greets and Romans, 238-239. Worcester Mass., State Normal School, reading at, 181. Worcester's Primer advocated word method, 258. Worcester's Readers, 250. Word-analysis, helped by principle of aerology, 215 ff. Word as continuum of sounds, 134 ff. Word-division, determination of, 138. Word-reading theory, 72. Words, reading range for, 64; recog- nition and naming times for, and for letters, 72; recognized far from fixation-point, 73-74; domi- nant, in perception of sentences, 84-85; form of, 92-96, 423-424; height of, measured by the small letters, 94; length of, a minor factor in perception of chil- dren, 96; root of, in first part of word, 98; upper half of, more important, 98, 99; recognized as wholes, loo-ioi; as printed, written, typewritten, etc., iio-iii; as products of analysis of sentence, i25ff.; sound diflferently alone, 139; subexcited in reading, 142 ff. ; read when letters could not be recognized, 151; visual appearance of, 153-154; connective and rela- tional, give little imagery, 154 ff.; different appearance of, when seen in isolation, 155; meanings of, 158; as instruments for fixing attention upon meanings, 162; we think in, 163; as names of feelings, 163-164; meaning of, felt in a perspective of total mean- ing, x66; look like their meaning, 168; sounds necessary to express. INDEX 469 203-204 ; analyzed by aerology, 206 ; words of like sound but unlike meaning, 208; not separated by spaces in early writing, 232; division of, at end of lines, 233; word-names learned as quickly as letter-names, 272; word not a sum of letter-names, letter-sounds, or letter-shapes, 272; pronunciation of, well taught in primers, 281; learning to pronounce new words, 287; learned from context in read- ing,> 291, 299, 322 S., 326-327, 348; need not aU be known to get general meanings, 333-334; niay be learned with help of diacritical marks, 351; sub-arousal of, uses energy in reading, 402; give nicety of control, 402 ; word-thinking is near work for the mind, 402 ; more read per fixation in shorter lines, 410; legibility of, increased by adding to number of dominant letters, 423 ; temporary charac- terization of the important words and phrases, 423. See Speech, etc. Word-method, described, 272 ; as used in the "Orbis Pictus," 272; does not give power over new words, 272. Word-sound, 81, 146, 164, 203, 207. Word-syllabary of Chinese, 210. Work-periods should be compara- tively short, 393-394- Writing, eye-movements in, 21; in Babylonia, 187; systems of, de- velop similarly among all peo- ples, 188; as communication of sounds, 203; of Chaldea, Baby- lonia, and Assyria, 211; arrange- ment of characters in, 231 ff. ; compulsory among Chaldeans, 236; material for, not plentiful in early times, 238; too slow to form natu- ral habits of speech, 369; as affect- ing eye-fatigue, 394; on black- board should be in large, plain hand, 415. .See Drawing, Pic- tures, etc. Written language, fails to record vital parts of speech, 140; fails to record transitional sounds, 140. Written letters, represent but a por- tion of sounds uttered, 138. Written words, perception of, iio- III. Wundt, Wilhelm, 124-128, 163, 189, 226 S. Yellowish paper not injurious, 414. Yucatan, Mayas of, 207. Zeitler, 64-65, 82-90, 108-109. Zodiac, signs of, are very old, 213. Printed in tht; United States of America.