September 1970
Note on Behavior and Koffka’s Behavioral Environment
J. J. Gibson, Cornell University
The World Wide Web distribution of James Gibson’s “Purple Perils” is for scholarly use with the understanding that Gibson did not intend them for publication. References to these essays must cite them explicitly as unpublished manuscripts. Copies may be circulated if this statement is included on each copy.
One of the most influential passages in Koffka’s Principles of Gestalt Psychology is his argument that behavior takes place in the behavioral environment of the individual, not only in his geographical environment (pp. 27-51). By the former he meant essentially, the phenomenal environment, experience, and by the latter the physical environment, physical reality, or the “stimulus situation” (p. 29). He was careful to say that he was referring to molar behavior, not molecular behavior; the latter “takes place within the organism” (p. 27). He was convinced that the relation between ordinary behavior and the physical environment “must remain obscure without the mediation of the behavioral environment” (p. 31). He was saying that behavior depended upon perception, including misperception, and not just on the world of physical stimuli (“the stimulus-providing geographical environment,” p. 30.) Is this distinction between two kinds of environment a genuine solution to the problem of molar behavior? Are there in fact two distinct and legitimate meanings of the fundamental notion of an organism-in-an-environment? The contrast between the two environments was illustrated by the story of the rider over the lake of Constance. His behavioral environment was snow-covered terrain but his geographical environment was the uncertain surface of a frozen lake. Koffka asserted that there was a sense of the term environment in which “the horseman did not ride across the lake at all, but across an ordinary snow-covered plain (p. 23). It seems to me an error to assert that this behavior occurred in the behavioral environment. It is equivalent to saying that behavior takes place in phenomenal experience. And Koffka does appear to accept this implication (p. 40 ff). But it gets him into spiraling difficulties of epistemology, in my opinion. He has to deal not only with two kinds of world, phenomenal and real, but also with two kinds of behavior , phenomenal and real. To the difficulty of explaining how we perceive the world is added the difficulty of explaining how we perceive our behavior in the world. And then there are still further difficulties. It is true that the phenomenal experience of the rider was that of a snow-covered plain but it is not true that he rode over a snow-covered plain. To say that he did “in a sense” comes close to saying that he rode over the experience of a snow-covered plain ñ that one can be supported by a phenomenal experience of the ground as well as by the ground itself, and this is nonsense. Koffka defines behavior as follows (p. 32). “Only such movements of organisms are to be called behavior as occur in a behavioral environment.” He could not conceive how the physical movements of any animal in a physical environment could properly be called behavior. But is there no way in which this can be conceived? If we take as “movement” only what particle physics tells us about it then indeed it is very hard to conceive (and Koffka had been reading Eddington on the nature of the physical world). But we need not do so. There is a level of the physical world (“carpenter’s physics”) at which movement becomes meaningful. And this, I suggest is what we should mean by the environment in which behavior occurs.