College and Research Libraries Machines That Teach Better Than Books? BY J A Y W. S T E I N OT H E R T H A N T H E P E R S O N of the teacher, most readers would probably con- sider the book itself as the best guide to learning. It is unlikely that they have given thought to machines that teach. At a time of serious concern for the teacher shortage in schools and colleges, some civic leaders and educational ad- ministrators are ready to turn in any di- rection that offers a panacea. Federal and foundation moneys are readily available for experimentation with "automated teaching." T h e r e is a danger, however, of widespread adoption before adequate experimentation. Unless some of the claims are carefully examined and the book's role upheld, numberless students will soon be punching the buttons of a "spoon-fed" machine instead of turning the pages of books in selected variety. Machine teaching is prelude to a book- less generation. At meetings of professional associa- tions, educators are hearing and discuss- ing assertions that machines not only can teach, but that they can do it better than "live" teachers, "who are over-burdened and subject to human foibles of inef- ficiency and bias." A panel speaker at the spring 1960 meeting of the Association for Higher Education of the National Edu- cation Association continued: " T h e y are certainly better teachers than books, which do not teach at all." T h e incredible totality of functions claimed for the machines is summarized in the following paragraph: " T h e teach- ing machine may present specific infor- mation to the student; examine the stu- dent on each piece of information as it is presented; correct the student's errors; provide additional explanation on points where the student has erred; verify the correctness of an answer when it is cor- Dr. Stein is Administrative Associate to the Vice President, Syracuse University. rect; proceed automatically to the next point when the student has mastered the preceding point, keep detailed records of the progress of each student; and, per- form all the stated functions as a con- trolled monitor with infinite patience and completely without bias."1 T h e machine that purports to perform this superhuman combination of teach- ing functions comes in several varieties. More than a half dozen firms from coast to coast have begun manufacture, under such trade names as the Programs Scan- ner of the Dyna-Slide Co. (Chicago) and the Visitutor of Hamilton Research As- sociates, Inc. (New Hartford, New York). Other firms include Foringer 8c Company (Rockville, Maryland), Rheem Califone Corporation (Hollywood, California), and U. S. Industries, Inc. Western Design Di- vision (Santa Barbara, California). T h e i r advertising is slick and colorful; the lan- guage is unequivocal. T h e y have strong backing from certain psychologists and school of education defenders of favorite versions of learning theories. A typical machine is the approximate size of a portable typewriter and bears some similarity in appearance to a com- bination calculator and vendor. A cam- pus newspaper likens it to a slot machine. T h e student presses a button or turns a crank or dial for the question to appear on a small screen or in a "scanning win- dow." He writes his answer on a paper record or sheet. He retracts a lever, presses a button again, or turns the dial farther for the machine answer to ap- pear, so that he may grade himself. If his 1 Automated Teaching Bulletin, 1 ( 1 9 5 9 ) , 25. M A Y 1 9 6 1 195 answer is correct, he presses another but- ton to proceed to the next question. If he is wrong he presses the key to score a wrong response. He then begins the next item. Books of no kind enter the scene or come into the picture during the "auto- mated learning experience." T h e teachers' presentation through the machine of questions and answers, prob- lems to be solved, or exercises to be per- formed is called "programming." T h e full set of questions covering the material to be studied, together with any supple- mentary panels of pictures, graphs or di- agrams is called the program. Students at their machines form a new classroom pattern, states a message from one firm, but the concept remains that of "the time-honored tutor-pupil relationship, combining the Cartesian idea of break- ing down a problem to its smallest parts, and the Socratic method of teaching through enlightened questioning." Widely acclaimed methods of program- ming information are associated with two teaching machines pioneers, Dr. B. F. Skinner of Harvard and Dr. Sidney Pres- sey of Ohio State University. T h e Skin- ner method requires the student to re- call the correct answer suggested by the text material, construct the answer, and write it down before being presented the correct response. T h e Pressey principle requires a student to recognize the cor- rect answer to a statement by choosing from several multiple-choice alternatives. T h e program material may merely in- form the student whether he is right or wrong in his selection, or it may include with the choice selected an explanation for being right or wrong. A "vanishing" principle, especially useful for memoriza- tion, involves presentation of a complete item, such as a poem, and the subsequent removal of increasing portions of it until the student is able to reproduce the en- tire item without a prompt. T h e secrets of success for the teaching machines dwell upon certain accumulat- ing evidence from psychological research. Learning is more effective when the stu- dent receives immediate knowledge oi the results. Students learn what they want to learn and have difficulty in learn- ing material which does not interest them. They must be motivated. "Auto- matic feedback" of the correct responses immediately informs the student of his progress, gives him a basis for revising his errors, and provides a "built-in moti- vation" to learn more. Promotion of the teaching machine ranges between extremes of criticizing and endorsing other modern "mechani- zation" as suits the needs for professional and commercial adoption. T h e recipient of the advertising risks falling into con- fusion and then, from its sheer weight, succumbing to the adoption of machines on his campus. It is important that he listen carefully to the claims. In objec- tion to the often raised criticism that ma- chines are lacking in human understand- ing, the representative of one firm as- serted that school buses, washroom facili- ties, and thermostatic heating systems are mechanical but no one objects to them on antihuman grounds. Similarly, the mechanical structure of aids for strength- ening the eye movements for faster read- ing, of movie projectors, radios, televi- sion, and phonographs, of the printing press are noted as hardly contributing to an inhuman society. T h e teaching ma- chine, states one sales pitch, is really "an- other mechanical or automated aid, along with many others, although the best." On the other hand, according to A. A. Lumsdaine of the American Institute for Research, Pittsburgh, writing in Audio- Visual Communication Review,2 teaching machines differ from films, television, and other audio-visual media in three ways: (1) the student responds continuously and actively, with practice and testing of each step to be learned; (2) the ma- chine informs the student promptly alter 3 A. A. Lumsdaine, " T e a c h i n g Machines and Self- instructional M a t e r i a l s , " Audio-Visual Communication Review, V I I ( 1 9 5 9 ) , 163-181. 196 C O L L E G E A N D R E S E A R C H L I B R A R I E S each response whether it is correct, al- lowing him to correct his errors; and (3) the student proceeds on an individual basis at his own rate. Research aiming at further verification and confirmation of claims made for au- tomated teaching is underway at a num- ber of institutions, including Earlham and Hamilton colleges and Harvard, In- diana, Illinois, and Arizona State univer- sities, and under various commercial and military auspices. According to a report of the Fund for the Advancement of Ed- ucation Committee on the Utilization of College Teaching Resources, the experi- ments suggest that effective learning can take place without the presence of the teacher who has initially planned for and helped guide the student's learning ex- perience. T h e r e is little doubt that the machines can be used for teaching. T h e very important question which mechanizers have been hesitant to face remains: Can and do the machines teach and communicate better than books? Un- like the book, properly used, does the ma- chine unduly fragmentize the student's learning, make him dependent on mech- anized programming, and discourage in- dependent thinking and questioning? Does it eliminate appreciation for the humanizing warmth of linking author and reader in a good book. Except for machine gadgetry and fanfare, one skep- tic asked, is the student really receiving anything that he should not rather be receiving from a well-written book em- bodying, where pedagogically necessary, Cartesian and Socratic principles of learning? Research has not yet answered these questions. Book-minded people, however, can think of answers which of- fer satisfaction to anyone who learns and likes to read. Teaching machines can never substi- tute for the teaching book because of their dependence on the book. T h e ma- chines depend on printed (book) instruc- tions for their invention, assemblage, maintenance, and improvement. T h e y de- pend on teachers who read books in great numbers in order to get the necessary background for curriculum building, course presentation, test-making, and "program construction." T h e y depend on the ability of the learner to read and un- derstand the questions presented which, unless the result is to be sheer rote and verbalism, must relate to wide reading of books and other printed matter. More important are the positions of the reviewer, historian, and librarian who can show that books have been effective "teachers" for generations and the insist- ence of the book reader and scholar that the book can do whatever is claimed for the machine and may do it better. T h e book can inform, stimulate, present ques- tions and answers, explain the errors and verify the correct response, proceed grad- ually in step-by-step fashion, embody Cartesian and Socratic principles of learning, and do so "with infinite pa- tience" and as "completely without bias" as any machine. Books, it is said, fail to teach because they are "passive and unprogrammatic and dead." They do not communicate in a machine-lively way. Thus, students do not give them the respect necessary to as- sist learning. Such statements as these are highly unfair and presumptuous from the lips of a machine promoter. For, apart from flashes of light, clicks of sound and movement of rollers, tapes, and sheets, the machine is certainly quite "passive." These, no less than the "auto- matic" corrective measures in the pro- gramming, still depend upon an unpas- sive human being who pulls the levers and turns the dials for the activity. Any discerning reader, writer, scholar, or teacher who has captured a motiva- tion and purpose for reading also knows the spirit of its being active and alive. Books have, in fact, "moved mountains," transformed personalities, and "changed the world." Teaching machine promoters may not concede these achievements to books, but they do admit that books can M A Y 1 9 6 1 197 be programmed and that "teaching ma- chines" broadly defined include books written programmatically, "scrambled" books, punch boards, and various printed paper devices. John A. Barlow, coordina- tor, Self-instruction Project, Earlham College, stated that "Even a specially de- signed envelope in some cases fulfills the function." Without motivation and purpose, any psychologist will admit, no machine use can result in effective learning any more than can book reading. Moreover, these come best from example. In listening to some "book-loving" objections to machine use, a psychology professor from an east- ern university affirmed vigorously how much he, too, would like more students to read more books, how vital book read- ing was. But when asked whether he read books, he implied in a stammering nega- tive that he read only books about teach- ing machines. While these so far hardly number a half dozen, the quantity in preparation is probably large. But he is hardly an example to motivate pupils to read in any "general" or "liberal" areas. Perhaps if he did read widely he, too, would know that books can teach. T a l k of machine substitution for books is all the more foolish when one asks the question, " W h a t function does learning serve?" As much as any, it serves to help people read books, reports, newspapers, and magazines, and to read their contents with understanding. From proper under- standing, it assists clarity of expression in speaking and writing. T h e machine pro- grammer may help certain students to improve in these communication skills, but it appears to be an indirect and di- verting means to the end of literacy. T h a t the book is still not doing for knowledge and education as well and as much as it should is doubtless true. But this is no reason to say, "therefore the campus must mechanize and use its money for machines." It provides no basis for comments that in five or ten years our open shelves of library books on all topics, and the rooms for browsing and reserved reading will give way to "classrooms and laboratories" containing rows of booths or compartments, each tabling a machine before which the student sits in his very own private learning situation. It is no reason to say that a publisher's listings or bookseller's wares lack "teaching" values. It is rather a reason for improving the book still further and teaching more stu- dents how to read intelligently. It is necessary to stress that while books may wear out after centuries, depending upon their use and care and the quality of their paper and binding, they never "break down," as the best of machines admittedly do. T o paraphrase slightly the well known statement of Carlyle, "A true university library remains a collec- tion of books and shows little possibility of becoming a collection of machines." New A C R L Committee Edmon Low, President of A C R L , announces the appointment of a special Ad- visory Committee to the President on Federal Legislation with the following mem- bership: Lewis C. Branscomb, director of libraries. Ohio State University, Colum- bus; Robert B. Downs, dean of library administration, University of Illinois, Urbana; Frederick G. Kilgour, librarian, Yale Medical Library, New Haven, Conn.; Richard H. Logsdon, director of libraries, Columbia University, New York; Stephen A. McCarthy, director of libraries, Cornell University, Ithaca, N. Y. President-elect Ralph E. Ellsworth will serve as an ex-officio member of the committee. 198 C O L L E G E A N D R E S E A R C H L I B R A R I E S