In response to:

A Little Learning from the April 14, 1966 issue

To the Editors:

There are adages to tell what happens to him who intercedes in a quarrel of his friends, but I risk it. John Holt’s review (April 14, 1966) of Jerome Bruner’s book, Toward a Theory of Instruction, expresses a variety of reactions, mainly negative. These are directed toward the purpose of the book (it ought to be for an audience that would include teachers), its style and organization (loose and weak), and its content, especially the presentation of Bruner’s very important observations and interpretations of the human modes of knowing which he calls the enactive, the iconic, and the symbolic. Bruner’s rebuttal is bad-tempered in a way the review is not, and mistakes as invective (as I did on first reading) a two-way jab about invariance and Dean Rusk. It was the propagandized child, not Bruner, who was the object of the remark. Authors have a sacred right to be angry at reviewers, but not in Bruner’s ad hominem style of argument, a mild example of which is his assertion that Holt has no particular qualifications in the field of child development. I have no particular qualifications in that field myself, but surely a philosopher (myself) as well as a teacher who is reflective and literate (Why Children Fail) can look at a psychologist, or even manage to be right about a psychological issue.

The issue I want to refocus some attention on, which Bruner says is the book’s major theme, is the “special glory of Man’s mind, that he has three approaches to grasping things”—related to action, imagination, and symbolization. I agree with Bruner that Holt’s critique does not get this straight, Bruner does not regard the enactive, the iconic, and the symbolic as three successive stages of development. He regards them as three ways of knowing, potentially reinforcing each other. This multiplicity is not only of importance for the understanding of human knowledge, but also for the examination of the style of our schools, which by tradition and current emphasis are grossly overconcentrated on verbal and symbolic skills unrelated to the more constructive and esthetic modes. I agree with Holt here, but the attack should not be on Bruner. It should be on overwhelmingly dominant stereotypes built into the fabric of our educational system: into teaching, grading, I. Q. tests, and much else. The stereotype identifies knowing with saying things right, a part of the iceberg above the surface. It identifies teaching with what Piaget called the “transfer of verbal structures” from book or teacher to child without seriously examining the child’s own enactive and imaginative contribution, which opens the channels of significant communication and symbolic thought. Not only do I think Holt misfires here, but my philosophic conscience is aroused by his readiness to explain historic error as the result of the besetting sin of verbalism, particularly by his hacking on poor old Aristotle. Aristotle doesn’t say, in any extant work, that heavy bodies fall faster than light ones. That was his medieval following. Moreover, it happens to be a true statement, except in odd places like the vacuum. Both our motor and visual experience conspire to tell us it’s true. It was not by verbal overflow that Aristotle erred, and not by experiment that Galileo straightened things out. Galileo’s achievement was surely in the rich interplay of intuition and experience with the symbolic techniques assimilated from a long tradition.

Holt’s justified negativism about wordiness in educational practice is certainly not sufficient for theory. We are all of us trying to build symbolic structures to match our intuition and experience about teaching and learning, Holt included. But if Holt doesn’t say it right his error hardly warrants calling him bad names. I don’t think Bruner says it right either, nor perhaps will I. But there is one theoretical point, implicit in Holt’s objection to the style of certain experiments of Piaget and of Bruner which I wish to elaborate. It is not a pro-point or antipoint, but I don’t find it in Bruner’s thinking and urge it should be there. Holt objects to the experiments on children’s developing grasp of “conservation”—of number, of liquid volume, of length, etc.—on the ground that their thinking is interpreted through their verbal response to interview. Instead of getting at their conceptual schemes, the experimenter may be getting at something more complex and in a sense less interesting—the child’s failure or lack of concern to cope with the preconceptions of the adult questioner. This is amusingly illustrated by an astute observation of Margaret Donaldson, who has worked with Bruner and who one summer did some interviews for us with six-year-olds in a summer school we operated in Cambridge. The children had been working with cans of hot and cold water and observing what happened to them over time. A consistent error in several descriptions was the report that both kinds of cans, hot and cold, got warmer Clearly the children were lacking the necessary schema, the rank-ordering of temperatures. Not at all! I said “error” without the inverted commas. What Dr. Donaldson found was a perfectly good schema which, so to say, put the arbitrary zero at body temperature, with “warm” as its label and “warmer” as signifying approach to it, contrasting with “hotter” and “colder.” Should I put it pretentiously and say these children had an algebraic rather than a merely arithmetic schema? Holt suggests, and I heartily agree, that children should be observed more in the course of their absorbing work and play and interviewed mainly in relation to it. There is a language of Behavior which the skilled may read. Can deaf and wordless children “think conservation”? Yes, and by enaction! This language includes verbal communication but puts it in a far richer context than most of the experimental designs of the psychological laboratory, which tends to subordinate the interests of children to the interests of experimenters, and thus to diminish both.

The theoretical point behind all this is that the three modes of cognition are not only not developmental stages, but they are not three “approaches” to knowing at all. The verbally symbolic mode, in particular, is not a mode of knowing or representing except in dependance upon the other modes. It can extend them, it can help organize and reorganize them, yet of itself, it is deaf as well as blind. And on the other side, I cannot accept Bruner’s sharp distinction between “enactive” and “iconic.” Visual perception, like tactile, is always a two-way interaction. The gates, filters, and scanning mechanisms by which we discriminate what and how the eye will see, and above all its coupling to the motions of body and hand, establish the enactive character of perception. The visual store of “icons” which dominate Bruner’s second mode are not the sense qualities of things, but their dispositional properties, their ways of responding to selective action by the percipient; they are not cases of “whatness” but of a more complex “what-if-ness.” Indeed it seems plausible that it is not the icon, the image, that functions in non-verbal thought but the image as linked to action contemplated. Only thus does it function as a sign, representing hidden objects as present or non-existent objects as real. If I am right, it is not verbal development which precipitates in the mind of the child such a schema as that of number constancy or volume constancy, but abstraction within a region crisscrossed many times by the child’s experience. The schema is a generalized what-if. What if the “more” water is poured back from the thin tall jar to the wide low one? What if the blocks are put back in the box? From very early the child has means of expressing the achieved constancies of his experience. There comes a time when the smile that was born in the vision of a human face will broaden in a game of peek-a-boo. The schema—the invariance of mama—and the smile will be linked later in wider families of constancies. The tokens of speech will group into associated types because there is something to associate them with. The essential schema is not a consequence of speech but a precondition of it. Bruner seems to be slighting this and thus after all to justify Holt’s criticism. When our working concepts are extended into the range of communication, they then permit a new transfer of information including much that is stored in the culture, current in speech, or coded in books. Perplexities and excitements of learning will prompt the giving and receiving, sometimes in controversies and expostulations.

Good teachers may enactively know more about learning than good psychologists can yet formulate. It has often been true in the history of science that the insight of practitioners was ahead of the formulations of investigators. I do not know Holt as a teacher, so I will not here commend him specifically. But I think all of us theorists ought to be respectful and studious of the insights of teachers, especially when they seem to find us wanting.

David Hawkins

Professor of Philosophy

Center for Education in the Sciences

University of Colorado


This Issue

June 23, 1966