Fog Over Sesame Street
by Samuel Kliger - 1970
In this article Mr. Kliger adds his voice to the discussion of the popular children's program, Sesame Street. A consultant on early childhood education, he suggests several more effective techniques for teaching young children that Sesame Street might employ.
In this article Mr. Kliger adds his voice to the discussion of the popular childrens program, Sesame Street. A consultant on early childhood education, he suggests several more effective techniques for teaching young children that Sesame Street might employ.
The child says, I throwed the ball, and the fond mother smiles indulgently, for she is confident that with maturation the child will discover the correct form of the verb. The primary question for Sesame Street, televisions most ambitious program to produce massive infusions of language learning experience in the ghetto, is not whether the mother should or should not correct immediately the unacceptable grammatical form. Rather, the question is: What is the source of the childs error? Children learn language by imitation of what they hear. This is a proposition so clear and so self-evident that no one questions it. It must be questioned, however. The real point is that the child learns language not merely by imitating, but by creating actively, young as he is, by forming hypotheses of his own far beyond anything which is taught to him, by reaching generalizations of his own, by establishing a learning-to-learn set of his own.
The very plain fact of the case is that regular verbs (their past signaled by ed) predominate in the English language. By saying throwed instead of threw the child is regularizing, categorizing, conceptualizing, seeking for consistencies in the English language as if the child were a full-fledged grammarian. The human use of language at all levels from childhood on goes far beyond formal teaching. Consider carefully that if the adult, being adult, knows better, then why is it that when new verbs are added to the language, they are always cast in regular verb form? We say the dairy pasteurized the milk, the doctor x-rayed the broken arm, the sinking ship radioed for assistance. The child and the adult both are regularizing, creatively innovating by an infinitely complex internal process far beyond what is taught. An adult will say, or accept when he hears it spoken, the sentence:
1) I bought five round red fluffy pillows.
What is the formal rule which prohibits:
2) I bought round five red fluffy pillows.
3) I bought red round five fluffy pillows,
4) I bought red round fluffy five pillows.
The adult obeys the rule of (1) without knowing what the rule is. Professor Noam Chomsky at MIT has been acclaimed at home and abroad for his insights into the internal processes at work in language acquisition, independent of and far beyond what is taught from the outside and far beyond the formal rules of traditional grammar. Theodore Roethke, a distinguished poet, expresses the thought more succinctly: Words are not formed but forming. The question I am raising is whether the producers of Sesame Street have a clear understanding of the language-learning experience.
The child does indeed mature; but in language processing, it is far more helpful, if we wish to understand how a child acquires language and syntax, to stress the continuities rather than the discontinuities between a childs and an adults internalizing of language and syntax. For one thing, to suppose that children and adults learn sentences by memorizing all sentence forms possible is to suppose the absurd since no one could memorize all the sentences possible. For a second argument against the naive idea that the chief way that children learn language is by imitation, by response to stimuli from the outside, consider:
1. The advantaged child who says, I throwed the ball. That child certainly has never heard that form of the verb in that particular household.
2. The mother holds up an apple and says to the child, Say apple. Has anyone ever heard a child repeat, Say apple? The child confines his statement to apple, as if he were a well-taught grammarian who knows that say is a verb which has a noun apple as its object; that is, he would seem to understand what an imperative sentence is and how it differs from declarative and interrogative sentences. Likely, not even the parent knows what verbs, nouns, and imperative sentences are.
Doing Well Feebly
President Theodore Roosevelt said of William Howard Taft: He meant well feebly. Sesame Street, judged on the basis of programming already broadcast to the public, does well feebly if we take the goal of Sesame Street to be first, last, and always the shaping of the language learning experience of the preschool child so that he enters the elementary grades unburdened of language deficits. Beyond a shadow of doubt, Sesame Street is a beamish program which entrances and delights millions of children. What child wont squeal with delight to see a baby deer, a squirrel drinking milk from a baby bottle, to watch the antics of the Get Together People whose scene in Lesson XV may just be the funniest puppet show ever put together, to puzzle over the knuckleheaded behavior of Buddy and Jim? The answer is no child will fail to respond joyously to Sesame Street. But the question itself is based on a different conception of precisely what constitutes a learning experience. The truth is that we see only intermittent signs on Sesame Street that the producers really understand and apply rigorously and systematically what we now know as a result of the researches of Chomsky, George A. Miller, Jerome Bruner, Roger Brown, Eric Lenneberg, Susan Ervin-Tripp, Jean Berko, and many other psycholinguists about how to arrange, or at least understand how to arrange, a genuine language-learning experience. Given the inspired inventiveness of the Sesame Street staff, it is almost certain that with a clearer theoretic model a finer educational result would have been accomplished. It is perplexing in the extreme to a well-wisher of Sesame Street to see the script wander almost by blunder (one supposes) from time to time into creating true learning-to-learn sets unaware apparently that the program is proceeding in a different and more educational direction without at the same time diminishing the sheer delight of Sesame Street.
No one has discovered how to make an omelet without breaking eggs, and neither has Sesame Street. There is an ancient explanation of the universe in which the world is pictured as supported by four elephants which themselves stand on the back of a tortoise. A legitimate question is: On what does the tortoise stand? On what basis does a reading methodology for the pre-literate child stand? I propose to argue not for a new and revolutionary theory, but for the most conservative theory possible to which any teacher can readily assent without reservation. I propose that any theory for teaching a subject-matter skill which does not lean hard on an explanation of how humans learn anything whatsoever does not deserve to be taken seriously. I find it ironic that we never hear from our schoolmen regarding general human learning capacity except when:
1. The reading readiness question is raised, and some teachers who fear damage to the childs ego raise questions about the childs general capacity for learning.
2. The parent brings a retarded or brain-damaged child to the school, and is asked whether she has observed any capacity in her child to learn.
Learning cannot and does not bring something out of nothing. It is not automatic, and on this point Professor B. F. Skinner has argued well in his Saturday Review article (Oct. 16, 1965) entitled Why Teachers Fail. Teachers fail because of their naive belief that the teachers task consists of nothing more than showing and telling the pupil. Actually, the teaching task has not even begun. Making Sesame Street enjoyable merely reinforces the childs desire to have fun. Learning does not begin until it is caused to happen. Learning is defined as shaped behavior when stimuli are brought under control by means of what Skinner calls operants. Properly reinforced, response-cued, the learner gets to know. (Limitations in Skinners very oversimplified stimulus-response theories are overcome by rival theories stemming from mediated response theories urged by Osgood. For example, Chomskys phrase structure explanations are most valuable, especially in describing language behavior, since they totally repudiate Skinners stimulus-response model. But this is a more complicated question.)
The main point is that the question How can we teach? is a real and answerable one. The history of reading instruction casts strong doubts on any hope that one system is the final answer. Perhaps so, but even common sensical demands on reducing the surplus meanings of terms frequently used in reading theory will help to close the information gap between learning theory and reading theory. Instead, therefore, of belaboring the point let us proceed to actual examples taken from Sesame Street.
We are told, correctly, that beginning handwriting should not be cursive but manuscript (handwriting conceived, of course, not as an end in itself but as reinforcement of visual discrimination of printed letters). Manuscript writing is printlike, resembling the print which the child sees on the printed page. As the child writes in manuscript style, he is matching the sample of the printed letter, a task which he can perform or be trained to perform, and a task which it is desirable for him to perform since it reinforces his knowledge of the printed letters. Cursive writing, on the other hand, is a distraction from the reading task, since it is, in effect, a second alphabet. So far, so good.
There does not exist, however, a single handwriting manual (either alone or in conjunction with a reader) which calls for the child to write the letter g as he sees it on the printed page. The child writes
but he is never taught to write
(g is always pictured as g on Sesame Street)
Admittedly, this is an objection to only one letter out of twenty-six, but for consistencys sake alone is there a single reason why a child should not be taught to write g as he sees it? When our children are systematically taught to discover the concept that what they write bears a relationship to what they see on the printed page, perception and muscular activity fusing, so that matching to printed sample always produces the correct result, why is the concept thwarted in the case of the g? The same teacher who painstakingly shows the minute difference between closing and breaking as significant for discriminating o, e, and c is the same teacher who frustrates the logic of matching to sample whenever g is the case in point. Clearly, incorrect performance, both in reading and in writing, is being reinforced. Entirely too close for comfort, the child is being asked to hallucinate the correct response. Is the matching to sample of g too difficult for the child to perform? But how does one know if it is not tried, and do not Chinese children learn to control far more complicated handwriting?
The silence in handwriting manuals is deafening on this point. The crux of the matter is hardly that a certain percentage of pupils get it right anyway. The crux of the matter is that any bit or any large segment of curriculum design can show improved instructional effect when the behaviors which go into the student performance are completely under control of a specified teaching procedure based on a sound theoretic model drawn from learning theory. The operative word here is completely.
As all reading teachers know, a problem exists with respect to the so called reversibles, b and d, for example. Both advantaged and disadvantaged children make the reversible error, and this fact alone should have alerted teachers to a more plausible explanation of the phenomenon and to hints on how to correct the situation. What teachers need is a stronger appreciation of the continuity between what the infant in the crib sees and what the school-child sees.
b and d require finer discrimination than w and k, for example. The teacher can say: In b the bump is this way, in d the bump faces the other way. This procedure is no doubt helpful but ignorant nevertheless of the deep-seated cause of the problem. Visual stimuli pour in on the infant in the crib, the crawling child, the mobile child. As objects come within the field of the childs eyes, it is by no means clear to the child that an object when it is reversed is a different object. The stuffed teddybear the child plays with is exactly the same teddybear whether it is in right profile or left profile. The nurturing mother who is, of course, the center of the childs entire universe is the same mother in right and left profile. Because the child is visually dependent on this conservation of identity, he in the beginning confuses b and d. The real point is that the child is only temporarily dependent on visual perception. In critical circumstances he begins to ignore his perceptual capacities. He drops them and becomes more logical, i.e. he learns to play the game of recognizing letters which can only mean a conceptualizing, inner processing of what the environment offers him. The task for the teacher, therefore, is to not inject transmodal redundancy (do pictures interfere or do they strengthen the verbal message)? Perhaps the child is being flooded with too many inputs unless they are so coordinated as to allow the system to function efficiently. The task for the teachers is to find stimuli-properties that are equally appropriate as pictures and as sounds. Perhaps the cross-modal transfer can best be effected by a vanishing procedure well known and studied by behavioral psychologists. Present the picture, sound, and printed form of the word simultaneously but gradually vanish portions of the picture until the child is no longer dependent on the picture for a cue to sound and word, i.e. he knows. As a matter of fact, in recognition of the splendidly talented cartoonists on the show, vanishing portions of the picture a bit at a time ought to stimulate the creation of hilarious cartoons also serving a predictable and measurable terminal behavior: reading with reasonable confidence and speed. For instance, there is much better reason to believe that the cross-modal match between the picture and the printed word duck will be made if the sole remaining leg of the duck vanishes with an accompanying outcry, Whoops: There goes the poor duck!
Sensations pour in on the child. If all sensations were received, pattern or form perception would be impossible. There has to be selection and suppression. According to Miller, Galanter and Pribram, humans have internalized schema, a TOTE System, they call it. Stimuli which fit into these schema are easier to process as compared with stimuli which do not fit in. The teacher must not wait for the appearance of the cognitive drive, but she must accept it as her basic responsibility to foster it first, last, and always. Instruction is a switching mechanism or else it produces rote learning stuck in cement. When the child is acting as his own control, we know (and all teachers pay lip-service to the idea) that learning becomes most pleasurable and enduring.
The letter X is presented on Sesame Street and a cartoon character turns it upside-down while remarking: See, its the same letter when its turned upside-down! A worse procedure can hardly be imagined. Does Sesame Street really wish to implant the idea that letters, any letter, can be turned upside-down and be read the same? How will the program prevent the child from forming this dangerous generalization?
Russian is one example of an alphabet in which, with one exception, capital letters and lower-case letters are the same. Nowhere in the programs seen thus far on Sesame Street is an awareness shown of the problem presented by our alphabet and its different capitals and lower-case graphemes (12 alike, 14 different). If the child must be taught or must somehow learn how to discriminate carefully between break and closure in differentiating find stimuli which in fact vary depending on right or left orientation. Then the child facing the graphemes b and d will be neither over nor under perceptually dependent, but dependent enough to discriminate, i.e. to hypothesize, to guess at, to regularize, to differentiate, in sum, to conceptualize the differences between b and d.
Jean Piaget is justly famous for his developmental theories of a childs mental growth, although he is not very clear in recognizing the jump from visual dependency to logical understanding. Nevertheless, his experiments with conservation tell us a great deal. Piaget arranges marbles this way:
Ask the child which is more, and he will usually reply that A is more. This is true for the child because what he generally sees around him is that things spaced out are more.
Take vessel (1) and pour the water into vessel (2). Even when the pouring is done with the child looking on, the child replies that vessel (2) is more. His reasoning is the same as before because in his visual experiences things which are stacked higher are generally more. These are the empirical facts of a childs cognitive development and ignorance of them imperils any teaching procedure. Sesame Street replicates the Piaget conservation experiment with an opaque container of milk which hardly permits the observing child to notice the full container phenomenon; in any case, the child is given the answer too soon for him to formulate a hypothesis of his own. Greater instructional skill is required.
The isolated instance of Sesame Streets failure to supply the correct match of printed g and the grapheme g may be taken as a leading example of what happens when teaching ignores cognitive development. Are the producers of Sesame Street certain that their undeniably attractive cartoons and film-shots do o, e, and c, then how can Sesame Street justify the casual treatment of B and b as if they were alike? Is it a gross exaggeration to reflect that the much discussed credibility gap begins with a childs growing awareness that teacher says one thing and the text says another? The whole point of postponing cursive writing to the second or third grade is to avoid teaching the child what is in effect two different alphabets. Moreover, since about 99 percent of what a child sees on a page is in lower-case letters, what possible objection can be raised to postponing capital letters until the time when children learn that capitals are used as the initial letter of proper names, the pronoun I, and the first letter of the first word in the sentence? Nor is Sesame Street consistent in its practiceB is presented in one frame as in Butterfly but as b in bee in the next frame.
Alphabetizing has been banished from the schools for several decades for the very good reason that knowing the alphabet is not important until the child is taught how to use the dictionary. It is true, of course, that the names of letters must be eventually memorized, since we spell, by naming letters rather than sounding them. But reciting the alphabet is no kind of preparation for good reading. As Professor Eleanor J. Gibson has pointed out, learning the name of each letter is associative, but a secondary stage of recognition. The primary stage is not associative, but a process of isolating and focusing on the distinctive features of a letter that make it unique and invariant. Pseudo-words (words without referential meaning), she points out, are easier to learn when they are pronounceable. The learner, in other words, gets to recognize higher units, not a single letter, a consideration which should put an end at once to the reading argle-bargle, the bone-wearying trivialities of the phonics vs. whole-word discussion. Have either the phonics or the whole-word proponents asked why it is that in words like statement, the break between state and ment is greater within a word than between words, whereas in the phrase in a moment there is no break at all (insofar is actually written as one word)? Chomskys phrase-structure protocols leave single letters even further behind than Gibsons higher units. Chomsky stressespredictably from his stress on inner dynamic processing of languagewhat he calls deep grammar, phrase boundaries as a formal statement of the way in which we read:
(1) They fed her dog || biscuits.
(2) They fed her || dog biscuits.
The words are exactly the same and in the same order but differ as different phrase boundaries or restraints are drawn: in (1) a pause after dog; in (2) a pause after her.
Impressive evidence supports Chomskys profound insights into language behavior; and the main problem for Sesame Street, and for that matter, for all reading teachers, is to shape a reading discipline which leaves letters and single words behind in favor of phrase structures, even though the conventional wisdom is that one learns words by accumulating one word at a time. Even such simple sentences as:
This is a bee.
These are cats.
are infinitely superior to lists of single words. Two experiments from a vast supply can be cited. Children were taught to reverse pairs of words:
The experimenter found the greatest difficulty (predictably from the phrase-structure interpretation of reading) in getting the children to reverse these pairs. The only other equal difficulty was in inducing the children to reverse semantic absurdity:
table - goes
A second experiment reveals a great deal about a childs acquisition of syntax. A tiny plastic cat and a tiny plastic dog are placed one in each hand of the child. The child is then instructed to act out The dog chases the cat. The child acts out the sentence by moving the dog in pursuit of the cat in the other hand. Now, when the child is given the passive form to act out, The cat is chased by the dog, despite the introduction of a by-phrase following a passive verb, despite the transposition of object to subject, the child acts out the situation exactly as beforeas if he is indeed a grammarian who knows the difference between the active verb and the passive verb. The conclusion seems obvious: The great advantage of a behavioristic-linguistic account of reading behavior is that a small number of strategies, compared to a much larger number of specific items to learn, results first in an economy of description and second in an economy in learning. Unless the curriculum designer understands thoroughly the childs internalized capacities for responding, making guesses, hypothesizing, selecting and repressing, categorizing, conceptualizing, his is a Sisyphean labor, an unending, futile task.
Performance and Competence
Performance is all too often taken as evidence of learning accomplished when performance is nothing more than echoic, basically mindless behavior. Competence, evidenced by skill in hypothesizing, forming and then utilizing concepts, shaping categories, going beyond the information given, as Bruner expresses it, is true learning; and it begins from infancy on. The question is how much training does Sesame Street offer which leads to competence.
A teacher asks, How much are 5 plus 2? and the student replies, 7. The student has performed correctly, but this is merely a way of stating that 7 is a better answer than 6 or 8. The question to the student is better put when he is asked:
5 plus ? = 7
Merely to ask the student:
5 plus 2 = ?
is not teaching at all but testing whether any learning has previously taken place. A correct performance tells nothing because it does not tell whether the student really has computational skillshe might have them but he might also not. Some students seem to be learning by memorizing myriad collections of facts. But to the extent that they cannot recast and recast again the facts in a fresh situation
10 divided by 5 = 8 divided by 4
to that extent they remain that constantly exasperating product of the schools: They know much but understand nothing. What is truly dreadful about rote knowledge is not only that it precludes discovery, as Professor Bruner always stresses, but one never knows whether the student repeating the rote fact is revealing knowledge or concealing ignorance. The performing student is not greatly different from Pavlovs slavering dog. The competent student is found on the margin for grace, utilizing his margin for grace for moving through cognition into that always mysterious universe we call consciousness.
The arithmetic materials presented on Sesame Street are a mixed bag. The hapless cook bearing a tray of 5 cakes, 4 sodas, 3 pies, etc., who constantly falls down stairs may be reinforcing nothing more than the childs desire for fun. On the other hand, the child might be receiving the idea that numbers are being taught. The sing-song recital by the children, voices over the display of the numerals 1, 2, 3, etc., may just possibly be teaching children the number system, but the whole exercise looks a great deal more like mere word-calling without real understanding. Glip, flor, lun, etc., would be equally efficient and just as easily learned as one, two, three. In short, the child is performing but not displaying number competence.
On the other hand, however, Sesame Street introduces a brilliantly conceived scheme to induce the child to conceptualize the number system, particularly the basic concept of one-to-one correspondence and the concept of the set. A frame of the program shows an empty cardboard egg carton. We see the childs hand placing one egg after another in each compartment, and we hear the child say voice over: one, two, three, etc. When the child reaches the eighth compartment and says eight, we see an adult hand move into the picture and place a cookie in the ninth compartment. After a moments hesitation, the child continues to place an egg in each remaining compartment, but this time the child intones, nine eggs and one cookie, ten eggs and one cookie. This is a genuine learning experience, reinforcing not only the idea of one-to-one correspondence but also, as every arithmetic teacher knows, the concept of the set. The actual term set can easily be introduced later, a prime example of sequential learning in which previously learned material is recast either in fresh applicability or in elicitation of the act of verbalizing the concept itself.
It would have been very simple for Sesame Street to follow through the concept of set:
There is a Zen riddle which says: We know the source of two hands clapping but what is the sound of one hand clapping? Teaching a child to call out numbers is like expecting him to hear the sound of one hand clapping. Numbers qua numbers must yield to their conceptualization, to notions of one-to-one correspondence, to set, to equality. Some sort of multi-stage mediation process must be triggered in the childs mind. And the teacher should not fear the increase in abstractness, for with abstractness there always goes, paradoxically, an increase in differentiation.
From the inside of arithmetic, the egg and cookie lesson has drawn on the concepts of one-to-one correspondence and set. From the outside of arithmetic, the egg and cookie lesson, drawing on behavioristic postulates, has introduced the mediating effect of the concept as an intervention between stimulus and response; in other words, call it discovery or call it freedom from the Skinnerian bondage of learning one thing at a time.
In a second sequence, Sesame Street exhibits an even surer grasp of the importance of direct teaching of concept-formation to the young child. This effect comes out in the exercise accompanied by the song, One of these things is not like the others. Four objects are displayed, three of which are in a class along one dimension of one sort or another. A tractor, an auto, a motorcycle are pictured. The fourth item is a horse, the one thing which is unlike the others in that it is not man-made. However, no one thing is unlike the others if all four are conceived as means of moving from one place to another. As Sesame Street has perhaps already discovered from the hesitation of the children to respond, great care must be taken in designing such exercises. The children may or may not know the distinction between natural and man-made, and if they dont, it must be taught in gradual steps and then applied to the criterional frame. A paradox is at the heart of the exercise, and some thought must be given to the paradox that the more one generalizes, the more one differentiates. A whale can be described as a fish since like the latter it swims. Generalizing across classes and within a class, it turns out that the superordinate classification is the differentiating information that a whale is a mammal, i.e. unlike fishes in this regard. In other words, an attribute of an object can turn out eventually to be nondefining. An object can have several discriminate attributes, and as a result, the task of concept attainment is much more difficult than the designers of Sesame Street know. Bruner, Goodnow, and Austins A Study of Thinking (New York: John Wiley & Sons, 1962) supplies excellent guidance. For now, it is enough to stress that without freedom of choice, concept development is not aided. Learning theory is the very gateway to the world of readinganything else is mere show and tell. The hesitance of the children to reply to the question, Which one thing is unlike the others? reflects both the strength and weakness of a teaching strategy which undertakes to teach concept formation directly. On the one hand, if the child utilizes completely his capacity for registering the differences of things, he will be overwhelmed by the number of things in his environment. On the other hand, discrimination training provides lead time for adjusting ones response to objects: The child wants to find out as early as possible what is the identity of something. This anticipation or exploratory scanning of objects provides the time required for adjusting ones response to the object. There must be freedom of choice or else there is no learning. The monumental silliness of the Bloomfield schema in reading (mat-pat-fat-rat), ITA, UNIFON, and similar artificial alphabets are all cases in point. All these schema fail precisely because the pupil is not taught to avoid choosing redundant hypotheses, not taught to eliminate some hypotheses in favor of the crucial one. A child is not an ideal thinking machine (neither is a computer), although it is easy to be persuaded that this is so considering the childs genius in accumulating vast vocabularies and massive syntactic control long before the child steps into a classroom. At that, however, the child is more of an ideal thinking machine than most educators know. Countless experiments in the Bruner laboratory at Harvard show that if there exists a creative type of child he is one whose actual performance runs well ahead of his ability to state a verbal justification of it. What we do know is that the ability to state the concept verbally greatly improves learning:
(2) UNITED STATES
If compelled to .do so, the child can be taught to memorize and recall (1). But given the concept of the name of a nation (and its easier pronounceability) the child in a flash can recall (2) with minimum strain. A teaching strategy which directly teaches the child to construct a response, to transform the given by going beyond the information given is a teaching procedure which frees the child from the bondage of performance and leads him on the road towards competence.
Words Are Not Formed But Forming
Dogs bark, cows moo, ducks quack, and dolphins whistle, but only language is human-specific (Eric Lennebergs magnificent Biological Foundations of Language should be read by all who dont know the book). A dog can be taught to obey the command sit, but it responds only to the acoustic configuration of the sound as any dog owner can test if in the hearing of the dog he says, I think that I shall sit down. The dog very likely will not respond because the one thing that the dog cannot do is to understand language combinably. Children combine phrase structures exactly like a highly trained linguist. They are always studying, as it were, some finite set of utterances in search of recurring units and patterns of combination. The child does not know the rules but he obeys them just the same. The child says asked (askt) but he also says grabbed (grabd). When the child is told, Here is one shoe, there is another, later the child will say, Here is one toy, there is another. In a teaching procedure, the child can be asked to create linguistic forms: Today I glur. I glur every day. Yesterday I------? to see if he says glurred.
These instances seem trivial but they are of basic importance to the curriculum designer. The teacher who declares, as frequently is the case, that children will learn to read from any kind of textbook is really saying that children learn somehow in spite of the teacher and in spite of the textbook. I cannot believe that the producers of Sesame Street will so willingly abdicate their teaching role, especially as the nine eggs and one cookie episode and the one thing unlike the others episode show that they are attuned to the basic cognition problem. One hopes that forthcoming Sesame Street programs will indicate more awareness of the cognition and psycholinguistic problems.
As Gertrude Stein lay dying, she turned to those around her and asked: What is the answer? Hearing no reply, she then asked: If so, what is the question? What is the question regarding reading in early childhood education? We are only beginning to find answers. This we do know: Perception and cognition must come to be understood in terms of information processing. The reading mandarins have learned to bandy two new terms: encoding and decoding. Recoding, however, deserves much more explicit attention than it has received. Chomskys transformational, phrase-structure formulations supply a credible teaching model. Reading is defined as a low capacity, single channel but multi-stage decision process in which decisions are made sequentially and decisions made early in the process are recoded, i.e. they affect decisions made subsequently. Reading theory spread from educationist quarters is an uncharted wilderness. Stimulus-response theories merely lead to habit acquisition, performance, that is. A reconciliationif it can at all be effectedbetween stimulus-response theories and the Chomskian insight (but not his intimidating, somewhat arcane terminology) provides hope for a breakthrough at long last in reading theory. The capital question remains not only how reading is learned, but how any learning at all is carried on. Before this capital question is answered, no policy of teaching reading to the preschool child can be planned intelligently. What we see now is the ludicrous spectacle of the reading teacher walking up an escalator that is going down. Reading theory emanating from the teacher training college commands: Dont raise the bridge, lower the water. The real point is do not diminish the content of the program but raise the ingenuity of the teaching techniques.
Sesame Street and Compensatory Education
Television has an isolating and hypnotic effect on young childrenthe stimuli pour in on the child. On a cost-basis alone, the eight million dollars spent on Sesame Street is very cheap if videotapes remain permanently available, if hands on materials are placed in the hands of each child to reinforce the screen. But the vexing problem of compensatory education remains, I fear, unsolved, and largely, I believe, for the reasons I have already given regarding the dynamics of learning. We know a great deal about welfare techniques but considerably less about teaching techniques. Everyone has had his say on compensatory education and this is my say, repeating largely from my article in Teachers College Record, May 1968.
My effort was to try to turn around thinking on compensatory education. As long as compensatory education is thought of as mere therapeutics, learning cannot and does not take place. Therapeutics does not produce learning but learning is of itself therapeutic.
A learning set produces a character set but a character set has very little or nothing to do with producing a learning set. A New York Times report on conditions in Harlem tells the whole story in a capsule. The author reports on a little boy called Dennis:
This is his teachers observation in the first grade: Dennis behavior is impossible. He is a terrible fighter. He cant be put in line with another child. He seems to resent being told what to do. He just looks at the teacher with a blank stare . . . What did the teacher say in the second grade? Dennis wont listen. He wont behave without supervision. He cant count above three. He recognizes the number 1 and is delighted when he is called to identify the numeral 1 for his class (the italics are the authors).
New York Times Magazine, Jan. 31, 1965
Can anyone seriously doubt now that it is learning which is therapeutic? The hollowness and shallowness of conventional educations claim that all which is necessary for teaching is to show and tell the learner are the same hollowness and shallowness responsible for the assertion that therapeutics induces learning. For learning to take place, there must be an interaction and feedback from an environment which changes constantly as it feeds back to the learner. The environment has altered precisely because the learner has interacted with it. Self-esteem, self-awareness, the impulse to self-improvement, as in the case of Dennis, result from a learning process so contrived that the learner interacts with materials on a schedule which permits him easy successes at first, proceeds towards more difficult materials, reinforces him at every step. Both primary and secondary reinforcers are working with the predictable result that a learning set and a character set are created adaptively together. This and this alone is truly compensatory education. There is indeed an inter-relatedness between welfare planning and educational planning but it does not show itself until welfare planning is subordinated to what we now know from experimental psychology how to plan for learning to take place and what does take place in terms of character change when learning is going on.
The theory of compensatory education today is not sufficiently explicit with respect to the learning outcomes expected from compensatory education. Compensatory education merely leads into broad scale discussion which can neither be refuted nor verified. Behavioristic psychology, on the other hand, offers a considerable body of knowledge, principles, and facts labelled learning theory. For progress to be made in understanding the requirements of compensatory education, the gap must be closed between education and behavioristically-oriented learning theory. The failure of many antipoverty projects to achieve the stated goal of improvement in learning among slum-damaged children is a bitter epitaph to the failure of conventional education to come to grips with experimental psychology as one of the possible strategies in the classroom. Precisely because Sesame Street is so fresh, so new and inventive, a higher degree of instructional sophistication is expected.