March 1974
An Insoluble Puzzle of Epistemology
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 References to these essays must cite them explicitly as unpublished manuscripts.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.
Sensation-based theories of perception seem to face a dilemma which can be stated in the following way: Is it the senses that are to be trusted and ideas not? Or is it only ideas that can be trusted and the senses not? All previous theories of what is ordinarily called sense perception take for granted that separate contributions are made to perception by sensations and ideas, an objective contribution and a subjectivecontribution, one coming from outside and one from inside. If so, is it the objective contribution that makes perception veridical and adaptive and the subjective contribution that makes it distorted and susceptible to illusion? Or is it the other way round, the objective contribution (concepts or accumulated past experience) makes it adequate and trustworthy?
It seems to me that no theory of perception, either any old or new form of empiricism on the one hand, or any form of rationalism, nativism, or Gestalt theory on the other, has been able to resolve this perplexity. Even Piaget who attempts in The Mechanisms of Perception to combine what he considers to be the best features of all these theories is forced in the end to conclude that the objectivity of perception depends on the subjective contribution, that is to say on concepts. This seems to expose the contradiction.
The only solution, I suggest, is to reject the sensation-based theories of perception with their underlying assumption of separate contributions from the external object and the perceiving subject. Perception isof the object (environment) by a subject (observer). These sources of available information are not cognate and the information picked up is not combined or mixed. There is direct perception of the world accompanied by direct proprioception of the self. Errors, of course, can occur for both, but that is another matter.
This suggests, in the last analysis, that both the perceiving and the remembering of the environment are modes of apprehending it, i.e. that they cannot be dichotomized. Hence percepts are not mixtures of sensations and memories and, equally, memories are not the rearousal of “old” percepts. Apprehension is not “time-binding” in the sense that the past is stored, but “time-free” in the sense that what does not change is perceived along with what does change. The radical implications of this hypothesis remain to be worked out.