[Representationalism] is the view that we are directly aware only of internal sense contents, from which we must infer external objects. (And this view is adopted because the representationalist shares with the traditional realist the assumption that direct awareness must be diaphanous [i.e., awareness must be passive and transparent, and there must exist no difference between the physical object in reality and how it appears in consciousness]. Every argument for representationalism reduces in one way or another to the core argument: Direct awareness is diaphanous; the perception of physical objects is not diaphanous; therefore perception is not direct.)
Kelley is guilty here of overstating his case. What he says about every argument for representationalism may be true of Descartes and Locke; but it’s hardly true of the American critical realists, of philosophers such as Arthur Lovejoy and George Santayana. Representationalism, to such philosophers, is merely an obvious corollary to the basic notions of realism. If both consciousness and matter exist, the perception of matter by consciousness must be achieved through some kind of medium, since a physical object cannot literally enter into consciousness. And since through experience we know that people have percepts, concepts, ideas and other mental datum that are obviously not identical with whatever material objects they may be associated with, representationalism, for the realist, becomes de rigueur. Objectivism, despite its rejection of representationalism, is, implicitly, a representationalist philosophy. Nearly every philosophy is at least implicitly representationalist. The only exception would be a philosophy that preached the solipsism of the present moment. As long as the philosopher believes that one idea or concept can serve as a symbol or a sign for something other than that idea or concept, he is, at least implicitly, a representationalist. Rand believes in concepts that are not identical to their object in reality. Therefore, she is a representationalist. Most idealists believe in time, that is, in a past and a future, and in a memory that can know the past. But what is memory but a representation of what is not present in consciousness? Therefore, idealists are representationalists as well, at least in terms of the past.
Why does Objectivism reject representationalism? According to Rand's chief disciples, representationalism (allegedly) holds that individuals can only know their representations—that is, the sense data perceived by the mind. The objects in reality that this sense data “represents” are “unknowable” (as Kant argued) and at best can only be inferred. In short, representationalism cuts the mind off from reality, leaving the mind only with the representations.
Now this is a caricature of representationalism inspired by the confusions of Immanuel Kant. Philosophers such as Kant believed that knowledge should be certain, and so they were driven to ask the question: If I know things only by representations, are not the representations the only things I know?
George Santayana answered this question as follows:
This challenge is fundamental, and so long as the assumptions which it makes are not challenged in turn, it drives critics of knowledge inexorably to scepticism of a dogmatic sort, I mean to the assertion that the very notion of knowledge is absurd…. Plato and many other philosophers … have identified science with certitude, and consequently entirely condemned what I call knowledge (which is a form of animal faith) or relegated it to an inferior position, as something merely necessary for life…. Knowledge is no such thing…. Knowledge ... is belief: belief in a world of events, and especially of those parts of it which are near the self, tempting or threatening it. This belief is native to animals, and precedes all deliberate use of intuitions as signs or descriptions of things... Furthermore, knowledge is true belief. It is such an enlightening of the self by intuitions arising there, that what the self imagines and asserts of the collateral thing, with which it wrestles in action, is actually true of that thing. Truth in such presumptions or conceptions does not imply adequacy, nor a pictorial identity between the essence in intuition [i.e, the representation] and the constitution of the object. Discourse is a language, not a mirror. The images in sense are parts of discourse, not parts of nature : they are the babble of our innocent organs under the stimulus of things ; but these spontaneous images, like the sounds of the voice, may acquire the function of names ; they may become signs, if discourse is intelligent and can recapitulate its phases, for the things sought or encountered in the world. The truth which discourse can achieve is truth in its own terms, appropriate description: it is no incorporation or reproduction of the object in the mind. The mind notices and intends ; it cannot incorporate or reproduce anything not an intention or an intuition…. Therefore any degree of inadequacy and originality is tolerable in discourse, or even requisite, when the constitution of the objects which the animal encounters is out of scale with his organs, or quite heterogeneous from his possible images. A sensation or a theory, no matter how arbitrary its terms (and all language is perfectly arbitrary), will be true of the object, if it expresses some true relation in which that object stands to the self, so that these terms are not misleading as signs, however poetical they may be as sounds or as pictures. [Scepticism and Animal Faith, 170-180]
Rand and her disciples take an entirely different approach to the question If I know things only by representations, are not the representations the only things I know? They simply evade it. Objectivism claims that human beings have direct awareness through perception of external objects. For an Objectivist, direct means: without conceptual processing. This, however, merely evades the problem by redefining the term direct. What is at issue is whether perception uses representations: and all the evidence available to us suggests very strongly that perception is every bit as symbolic and representative as is conception. The very fact that a percept is not identical to its object establishes that the percept is a representation. Visual perception clearly brings forth, within the mind, a field of images. If there exist any doubt on this subject, one merely has to observe that the lens and cornea of the human eye actually invert or turn the image displayed on the retina upside down. The brain then proceeds to turn the image field right side up for viewing in the field of consciousness. Hence what we “see” “directly” in consciousness are images, not the physical objects. After all, the mind, when it flips the image field right-side up, does not flip right side up the real world. It flips over a field of representations. Moreover, if individuals are told to wear special glasses that reverse the field of vision, so that the image is displayed on the retina right-side up, while they at first perceive the world as upside down, if the wear the glasses long enough, the brain automatically reorientates the field of vision right-side up. So visual perception can hardly be seen as “direct.” Consciousness, before it can become aware of material objects, must translate the data read by the senses into a language that the consciousness can read and interpret. The point of perception is not to duplicate or mirror physical objects, but, as Santayana put it, to express “some true relation in which that object stands to the self.” Representationalism, far from rendering knowledge impossible, provides the explanation of how it is that one mode of existence (i.e., consciousness) can become cognizant of a very different mode (i.e., the physical world). Those who equate representationalism with “knowing only our representations” are guilty of making a false demand of knowledge. They demand that knowledge be certain and direct. Knowledge is no such thing. Knowledge attempts to make us aware of objects existing outside of consciousness. It is transitive and symbolic and conjectural right from the very start. Whether described as “perceptual” or “conceptual,” knowledge is indirect, mediated by symbols, which help reduce physical objects to a scale amenable to human intelligence. No logical proof or foundationalist rationalization can validate this indirect, symbolic knowledge. It involves a conjectural leap. We must take nature by the hand, and assume the symbols which arise in consciousness represent objects or events in a physical world. Once we take this leap, we will find our faith justified, not by logic, but by praxis. It is the practical necessities of life, the urgent biological needs of the human organism, that force us to take this conjectural leapand become tacit and practical realists, despite whatever foundationalist or idealist misgivings we may entertain.