What Is Psychology?
JEAN PIAGET Geneva, Switzerland CONSTANCE KAMII University of Illinois at Chicago Circle (Trans.) and University of Geneva
ABSTRACT: Five points are made: (1) Psychology is the science not only of the individual but also of humans in general. For example, mathematics and physics have been created by human beings, and this creation can be understood only in terms of human intelligence in its totality. (2) Psychology is a natural science, and, like every other science, it is built not only with what comes from the object but also with the structures constructed by the subject. (3) Psychology occupies a key position among the sciences because it explains the notions and operations used in the development of all the sciences. (4) It is impossible to dissociate psychology from epistemology. (5) Psychology, like all other sciences, can thrive only on interdisciplinary cooperation.
I am very honored by the confidence you have shown in awarding me the Edward Lee Thorndike Award for 1977. I am also moved by the exception you have made in allowing me to designate someone else to deliver this address, as traveling from Europe has become more and more difficult for me. My principal merit is to have been surrounded by first-rate collaborators. The books I have written over the years have also been signed by others who have contributed to their essence, beginning with B. Inhelder. If I gave a broad title to today's address, it is partly because the public does not fully realize the extraordinary developments that have occurred in psychology since the beginning of this century. The 18th International Congress of Psychology, which was held in Moscow in 1966 with 6,000 participants, ended its work with an address by Paul Fraisse, the This article was an address by Jean Piaget translated and then-new President of the International Union of delivered by Constance Kamii at the annual convention of Scientific Psychology. In this address, Fraisse sum- the American Psychological Association, San Francisco, August 29, 1977. Division 15 (Educational Psychology) marized his impressions with the striking sentence, awarded the author the 1977 Edward Lee Thorndike "Psychology no longer knows any taboo subject." Award for Distinguished Psychological Contribution to To cite only one example [of what he meant], ever Education. Material in brackets, throughout the article, has been since psychology understood that it was a science of added by the translator. 1 conduct and not only one of consciousness ("con[The French words conduite and comportement are duct" referring to behavior, but behavior that in- usually both translated into English as behavior, perhaps because conduct has a moral connotation. As can be seen cludes the act of "becoming aware" of what we in this context, the term conduct does not necessarily have 1 do), we could have worried that it would neglect any moral connotation.] 648 • JULY 1978 • AMERICAN PSYCHOLOGIST
Copyright 1978 by the American Psychological Association, Inc. 0003-066X/78/3307-0648$00.75
consciousness, and this indeed happened to certain extremists. However, the problem of consciousness has come back to the forefront of psychology, with the work of psychoneurologists on "vigilance," or attention, and with the study of development. Everybody pays attention to it today, including Russian psychologists, regardless of their philosophy. I have five remarks to make about psychology. The first is perhaps obvious, but not to everybody. It is that psychology is the science not only of the individual but also of humans in general and notably of "the subject" in a universal sense. Surely, applied psychology can in certain situations be interested in individual cases, and the empirical or theoretical study of individual cases represents an interesting chapter called "differential psychology" [or the psychology of individual differences]. But even in applied psychology the subject in general necessarily intervenes. For example, to reform the teaching of mathematics or physics, we cannot consider only the retardation or difficulties of particular individual pupils. The problem consists first of situating the knowledge of mathematics, physics, and every other science in the totality of the process of intelligence and its development. And this is a problem of human intelligence in its totality, which is related to general problems of knowledge to which I shall return shortly. My second point requires more comment. Psy-
chology is a natural science, and if this is now understood in most countries, it pleases me to point out that this position in the classification of fields of knowledge has been practiced for a long time in the great Genevan tradition since 1890, when the chair I had the honor of occupying until my retirement in 1971 was founded. T. Flournoy wanted to place this chair in the Faculty of Sciences, and this example is often cited. But this situation is sometimes a source of serious misunderstanding, as certain philosophers (I am thinking, among others, of Husserl and phenomenology) oppose the transcendental against the "natural" in order to emphasize the inadequacy of "naturalism." Let us recall that the transcendental, according to philosophers, refers to conditions prior to knowledge and to instruments of knowing that exist before experience, in a sense more logical than chronological—namely, the [cognitive] instruments that are necessary to render experience possible. If I use this term transcendental here (which I do not often use), it is simply to designate what, in knowledge, comes not from the object but from the structures constructed by the subject. I will return to this point shortly in connection with a discussion of actions. These misunderstandings are due also to certain scientists as well as to philosophers. The source of the misunderstanding must be sought in "positivism," which gave too narrow an image to nature and particularly to the sciences of nature, thus making an easy target that does not even require careful aim. Positivism is a doctrine of the limits of science, and positivists wanted to limit science to certain problems and consider others as being "metaphysical" (hence the temptation to have both scientific psychology and "philosophical" psychology). We know that these boundaries have always been violated. Auguste Comte proscribed "causal explanations" in favor only of the search for "laws," but, in fact, scientists are not satisfied with descriptions of phenomena and do not stop looking for explanations. Comte condemned the calculation of probabilities, the study of astrophysics, and the use of the microscope, and the positivists who followed him proscribed the study of the atom as resulting from the search for causes, and we now know what happened to all of these beliefs! It is good to recall these facts from time to time because this allows us to judge certain prohibitions or limitations that no one today would want to impose on science. Neither science nor psychology is positivist. They are "open" to new problems without limit. They live on crises and revolutions as well as on con-
tinuity and tradition (see Kuhn, 1970). They submit to all new facts and must constantly reexamine their principles and methods. They acknowledge the two fundamental principles Flournoy laid out for himself a long time ago: (1) Everything is possible ("There are more things in heaven and earth . . . " ) , but (2) the weight of proof must be proportional to the strangeness (or novelty) of the facts. In such a perspective, "naturalism" in the classical sense (and in the sense that Husserl still wanted to combat) is only a myth, and this is true for two reasons. The first is that nature is inexhaustible, and we can know it only by successive approximation. The danger of naturalism, as understood through positivism, was in its reduction of higher order concepts to lower order ones, as exemplified by the exaggerated reduction of certain higher order mechanisms to conditioned reflexes. But those who believed in this danger did not suspect the possibility of forms of thought such as those of contemporary dialectical currents. In reality, when we try to reduce "higher order" concepts to a lower level, the reduction is only apparent, because sooner or later the "lower order" concept is enriched by higher level conceptualizations. This is what we saw with Einstein when Newtonian gravitation was reduced to geometry. Animal psychology, too, has already enriched biology, as certain authors such as G. G. Simpson interpret animal psychology as a factor in evolution. The second reason is that knowledge, or the sciences of nature, is constantly undergoing reorganization. Indeed, no science can be placed on a single plane, and each one of them involves multiple and distinct epistemological levels. All sciences of nature, therefore, involve transcendental aspects— as defined above in the sense of instruments necessary for structuring—transcendental aspects that are inherent in research itself and that are in constant movement and construction impossible to substantiate or put down on paper once and for all. In fact, there exists a reflexive progress in the sciences (which is indissociable from their progress in extension). It consists of the constant delineation of new conditions of intelligibility, which are transcendental with respect to the content of later experience. To engage in physics or biology, for example, mathematics 2 and logic are necessary, and
2 Even in mathematics itself, the Bourbakis used the method of mapping to construct their matrix structures, and
AMERICAN PSYCHOLOGIST • JULY 1978 • 649
mathematics and logic are built not on facts but on boundless human construction and the very coordination of our actions on reality. This notion of the subject's "actions" often gives rise to misunderstanding because there is a tendency to reduce actions to mere material actions, when they can be interiorized in the form of mental "operations," and the operations themselves can (and even must) be coordinated into "structures" such as seriation, classification, and enumeration. These structures, which are constructed by children themselves (and not received from the outside), result from their actions. Structures consist of "what the children can do" in given situations, and not oi what they think of it (their becoming aware of what they do always remains incomplete). Indeed, elementary actions begin by being material (i.e., moving objects, pulling them, pushing them, turning them over, etc.), but even from the sensorimotor level they become coordinated among themselves into structures such as the "(practical) group of displacements." Later on, the same movements, and many other more complex ones, will be carried out mentally, but these will still be actions, which enrich reality with a set of possible transformations, that is, with structures constructed by the subject, without which knowledge of objects would remain impossible. To stay with examples related to space, one can cite the case of systems of reference (or "coordinates") that allow children 7-10 years of age to understand that to determine the position of a point on a surface, there must be two coordinates, one vertical and the other horizontal. Younger subjects believe that it is enough to use one diagonal measure from a corner of the surface. These young subjects do not realize that if they do this, there are many indeterminate positions possible, depending on the angle of the diagonal line. This is a good example of actions that have not yet become operatory and have not yet attained a "structure." The structure the children later construct expresses what they "know how to do" and what they figure out with their own means. Likewise, in the case of a logical structure such as seriation—for example, A<B<C<D...<N —children 7-8 years old understand and deduce immediately that there are as many elements greater than A (i.e., B . . . N) as there are elements smaller
once these structures were constructed, they became the basis for McLane and Eilenberg to build their theories of morphisms and categories.
than N (i.e., from A to N — 1). Younger children 5-6 years of age, on the other hand, need to count the elements in both directions in order to answer this question! We see clearly in this case that a structure of actions or operations, once constructed by the children themselves (and we know well the successive and laborious phases of this construction of seriation), gives them new "powers" that enrich their knowledge of objects. To these spatial and logico-arithmetical structures must be added causal [or physical] structures that also rest on operatory deductions. For example, young subjects acknowledge that a weight exerts pressure on a piece of foam rubber because they see the weight pressing down on it. But they think that the same object does not have any weight on the table because they do not see the same phenomenon. Later, children deduce that if an object has weight on foam rubber, it has the same weight on the table, but that the table resists the pressure. Finally, at about 11-12 years of age, they deduce that if the body submitting to the action of pressing down succeeds in resisting it, this is because it, too, acts, but in the opposite direction. With no formal instruction, 11-12-year-old children can thus deduce Newton's third law of the equality of action and reaction. This is a beautiful example of a causal structure based on operations, but these operations, although they are constructed by the subjects, are then "attributed" to objects themselves conceived in this case as "operators." Let us further note that these diverse structures are of special interest to the theory of intelligence or of knowledge (epistemology). However, we can also use them as instruments of analysis in observing individual children. Contrary to "tests," which only measure performance, structures permit us to see how an individual subject reasons and, consequently, what we can expect of him or her in the future. But for this it is essential to follow children in their spontaneity and not to transform the method of operatory questions into a standardized test (see Kamii & DeVries, 1978). I return to psychology with a third remark, which complements the preceding one. Psychology presupposes the existence of other natural sciences and, in fact, it is derived from them (from physiology and biology to physico-chemistry and mathematics), but a fundamental fact in interpreting these relationships is that psychology in turn explains the notions and operations that are used in these sciences. It is psychology that enables us to understand the construction of number (and how the solu-
650 • JULY 1978 • AMERICAN PSYCHOLOGIST
tions proposed by the famous authors of Principia Mathematica are, in reality, inadequate),3 algebraic structures (and child psychology shows us how the "matrix structures" of Bourbaki are "natural" and are due to man's intelligence), geometrical structures (and why topological structures precede Euclidean and projective frameworks), elementary kinetic notions (among others, the relationship between time and speed), logical structures in general, etc. It is to this research that I have devoted more than SO years. Psychology thus occupies a key position, and its implications become increasingly clear. The very simple reason for this is that if the sciences of nature explain the human species, humans in turn explain the sciences of nature, and it is up to psychology to show us how. Psychology, in fact, represents the juncture of two opposite directions of scientific thought that are dialectically complementary. It follows that the system of sciences cannot be arranged in a linear order, as many people beginning with Auguste Comte have attempted to arrange them. The form that characterizes the system of sciences is that of a circle, or more precisely, that of a spiral as it becomes ever larger. In fact, objects are known only through the subject, while the subject can know himself or herself only by acting on objects materially and mentally. Indeed, if objects are innumerable and science indefinitely diverse, all knowledge of the subject brings us back to psychology, the science of the subject and the subject's actions. The fourth remark: People may say that I thus engage in philosophy or epistemology and no longer in scientific psychology. But, in the research that we pursue, it is impossible to dissociate psychology from epistemology. Indeed, if we study only one level of development (for example, that of the adult or adolescent), it is easy to distinguish the problems: psychological experience, emotions, intelligence and its functions, etc., on the one hand, and the broad problems of knowledge (epistemology), etc., on the other. But if we want to study cognitive functions and pursue a developmental point of view in order to study the formation and transformations of human intelligence (and this is why I specialized in child psychology), then the problems must be formulated very differently: How is knowledge acquired, how does it increase, and how does it become organized or reorganized? These are the very questions that must be answered. But here, the answers we find, and from which we can only choose by more or less refining them, are neces-
sarily of the following three types: Either knowledge comes exclusively from the object, or it is constructed by the subject alone, or it results from multiple interactions between the subject and the object—but what interactions and in what form? Indeed, we see at once that these are epistemological solutions stemming from empiricism, apriorism, or diverse interactionism, which are more or less static or dialectic. In short, it is impossible to avoid epistemological problems in this kind of research—epistemological problems that concern epistemology in movement, or genetic (.psychogenetic) epistemology. From this follows my fifth and final remark. Psychology, like all other sciences, can live and prosper only in an interdisciplinary atmosphere. Interdisciplinary relationships indeed exist, but they are still insufficient. In the science of human beings, it is clear, for example, that the study of intelligence brings up the problem of the relationship between thought and language—hence a collaboration between psychology and linguistics; and the current work of linguists on transformational grammar and linguistic structure in general is very promising for possible comparisons with the operatory character of intelligence. But this is an immense field to cover, and collaborations are only beginning. Likewise, there exist numerous relationships between data from the science of economics and of "conducts," and game (or decision) theory, which was elaborated by economists, constitutes a very enlightening instrument for the analysis of "strategies" of behavior. But here, too, collaboration is only beginning. The relationship between psychology and sociology is evide.nt but still not sufficiently elaborated, notably in the realm of development. As far as the biological sciences are concerned, the connections among psychology, physiology, and neurology are close, but many relationships between general biology and the theory of intelligence remain untouched. Indeed, many other fields of cooperation remain equally indispensable and are only beginning to take place—-with logic or general algebra and with the epistemology of mathematics and physics, for example. Everything that I have been able to do in more than 20 years of interdisciplinary research is due to
3 In fact, Russell and Whitehead explain the formation of number in terms of one-to-one correspondence between elements of equivalent classes. However, one-to-one correspondence already implies number, and thus there is here an obvious vicious circle.
AMERICAN PSYCHOLOGIST • JULY 1978 • 651
collaboration with specialists from all of these fields, and the International Center of Genetic Epistemology, to which the Faculty of Psychology has extended its kind hospitality, was created in 1955 precisely to facilitate interdisciplinary cooperation. With the present state of knowledge, it would, in fact, be regrettable and visionless to leave such collaboration to chance encounters or individual initialive. Interdisciplinary cooperation is necessary and must be organized. More than 20 years of experi, .I , , , ,,. ,. ,.. ence and the abundance of publications resulting
from these exchanges among researchers who, at
first, had difficulty in understanding each other, have shown us the increasing and unexpected fecundity of these ever closer contacts. I hope to continue the work of the International Center of Genetic Epistemology in the same interdisciplinary manner, REFERENCES Kamii, C., & DeVries, R. Physical knowledge in preschool education: Implications oj Piaget's theory. Englewood
Cliffs, N.J.: Prentice-Hall, 1978.
Kuhll) T s The stmctwe o} scientific revolutions (2nd
ed.). Chicago: University of Chicago Press, 1970.
Editors Sought for New JPSP
In January 1980, the American Psychological Association will publish the Journal of Personality and Social Psychology in a new format consisting of three separately edited and identifiable sections appearing each month as a single bound issue. The sections are Attitudes and Social Cognition, Interpersonal Relations and Group Processes, and Personality and Individual Processes. Articles in the three sections will be selected by three autonomous editors, each of whom will work with separate groups of advisory editors. The number of pages to be published in JPSP each year will be based on the three editors' recommendations to APA's Publications and Communications (P&C) Board. Authors will be asked to submit manuscripts to the appropriate section editor according to a manuscript's major variables and theoretical orientation. When the three section editors are ready to receive articles in 1979, detailed lists of topics appropriate to each section will be published. The plans to restructure JPSP are the result of several years of deliberation by the P&C Board. In 1974 and 1975, after considering the content of submissions, number of submissions, and number of published pages, a task force chaired by Albert Hastorf recommended a sectioned JPSP. Board action on this recommendation was originally postponed; during the past year, however, surveys of individual subscribers and members of APA's Division of Personality and Social Psychology have made it clear that the change to a sectioned journal is desirable and should be implemented as soon as possible. The P&C Board invites immediate nominations for the three editorships of JPSP. Nominations should be received before August 1, 1978. The new editors will be selected by early fall so that they can begin receiving manuscripts early in 1979 for publication beginning in January 1980. Nominations should be sent to David Zeaman, Department of Psychology, Box U-20, University of Connecticut, Storrs, Connecticut 06268. Names of the new editors will be announced later this year.