For millennia, philosophers have debated the nature of perception and its relation to reality. Their speculations have been shaped by the prevailing concerns and metaphors of their age. The ancient Greeks, with slaves to do their work, were less interested in labor-saving inventions than in abstract concepts and principles. Plato’s allegory of the Cave refers to no technology more sophisticated than fire—harking back, perhaps, to times when people literally lived in caves. (It does refer to the notion of prisoner, long familiar from slavery and military conquest in the ancient world.)
In Plato’s low-tech metaphor, the relationship of the perceiving subject to the objects of perception is like that of someone in solitary confinement. The unfortunate prisoner’s head is even restrained in such a way that he/she is able only to see the shadows cast, on the walls of the cave, by objects passing behind—but never the objects themselves. It was a prescient intuition, anticipating the later discovery that the organ responsible for perception is the brain, confined like a prisoner in the cave of the skull. Plato believed it was possible to escape this imprisonment. In his metaphor, the liberated person could emerge from the cave and see things for what they are in the light of day—which to Plato meant the light of pure reason, freed from dependence on base sensation.
Fast forward about two millennia to Catholic France, where Descartes argues that the perceiving subject could be systematically deceived by some mischievous agent capable of falsifying the sensory input to the brain. Descartes understood that knowledge of the world is crucially dependent on afferent nerves, which could be surgically tampered with. (The modern version of this metaphor is the “brain in a vat,” wired up to a computer that sends all the right signals to the brain to convince it that it is living in a body and engaged in normal perception of the world.) While Descartes was accordingly skeptical about knowledge derived from the senses, he claimed that God would not permit such a deception. In our age, in contrast, we not only know that deception is feasible, but even curry it in the form of virtual entertainments. The film The Matrix is a virtual entertainment about virtual entertainments, expounding on the theme of the brain in a vat.
Fast forward again a century and a half to Immanuel Kant. Without recourse to metaphor or anatomy, he clearly articulated for the first time the perceiving subject’s inescapable isolation from objective reality. (In view of the brain’s isolation within the skull, the nature of the subject’s relation to the outside world is clearly not a transparent window through which things are seen as they “truly” are.) Nevertheless, while even God almighty could do nothing about this unfortunate condition, Kant claimed that the very impossibility of direct knowledge of external reality was reason for faith. In an age when science was encroaching on religion, he contended that it was impossible to decide issues about God, free will, and immortality—precisely because they are beyond reach in the inaccessible realm of things-in-themselves. One is free he insisted, to believe in such things on moral if not epistemological grounds.
Curiously, each of these key figures appeals to morality or religion to resolve the question of reality, in what are essentially early theories of cognition. Plato does not seem to grasp the significance of his own metaphor as a comment on the nature of mind. Rather, it is incidental to his ideas on politics and the moral superiority of the “enlightened.” Descartes—who probably knew better, yet feared the Church—resorts to God to justify the possibility of true knowledge. And Kant, for whom even reason is suspect, had to “deny knowledge in order to make room for faith.” We must fast forward again another century to find a genuinely scientific model of cognition. In Hermann Helmholtz’s notion of unconscious inference, the brain constructs a “theory” of the external world using symbolic representations that are transforms of sensory input. His is a precursor of computational theories of cognition. The metaphor works both ways: one could say that perception is modeled on scientific inference; but one can equally say that science is a cognitive process which recapitulates and extends natural perception.
Given its commitment to an objective view, it is ironic that science shied away from the implications of Kant’s thesis that reality is off-limits to the mind. While computational theories explain cognition as a form of behavior, they fail to address: (1) the brain’s epistemic isolation from the external world; (2) the nature of conscious experience, if it is not a direct revelation of the world; and (3) the insidious circularity involved in accounts of perception.
To put yourself in the brain’s shoes (first point, above), imagine you live permanently underwater in a submarine—with no periscope, port holes, or hatch. You have grown up inside and have never been outside its hull to view the world first-hand. You have only instrument panels and controls to deal with, and initially you have no idea what these are for. Only by lengthy trial and error do you discover correlations between instrument readings and control settings. These correlations give you the idea that you are inside a vessel that can move about under your direction, within an “external” environment that surrounds it. Using sonar, you construct a “picture” of that presumptive world, which you call “seeing.”
This is metaphor, of course, and all metaphors have their limitations. This one does not tell us, for example, exactly what it means to be “having a picture” of the external world (second point), beyond the fact that it enables the submariner to “navigate.” This picture (conscious perception) is evidently a sort of real-time map—but of what? And why is it consciously experienced rather than just quietly running as a program that draws on a data bank to guide the behavior of navigating? (In other words, why is there a submariner at all, as opposed to a fully automated underwater machine?) Furthermore, the brain’s mastery of its situation is not a function of one lifetime only. The “trial and error” takes place in evolutionary time, over many generations of failures that result in wrecked machines.
In the attempt to explain seeing, perhaps the greatest failure of the metaphor is the circularity of presuming someone inside the submarine who already has the ability to see: some inner person who already has a concept of reality outside the hull (skull), and who moves about inside the seemingly real space of the submarine’s interior, aware of instrument panels and control levers as really existing things. It is as though a smaller submarine swims about inside the larger one, trying to learn the ropes, and within that submarine an even smaller one… ad infinitum!
The problem with scientific theories of cognition is that they already presume the real world whose appearance in the mind they are trying to explain. The physical brain, with neurons, is presumed to exist in a physical world as it appears to humans—in order to explain that very appearance, which includes such things as brains and neurons and the atoms of which they are composed. The output of the brain is recycled as its input! To my knowledge, Kant did not venture to discuss this circularity. Yet, it clearly affirms that the world-in-itself is epistemically inaccessible, since there is no way out of this recycling. However, rather than be discouraged by this as a defeat of the quest for knowledge or reality, we should take it as in invitation to understand what “knowledge” can actually mean, and what the concept of “reality” can be for prisoners inside the cave of the skull.
Clearly, for any organism, what is real is what can affect its well-being and survival, and what it can affect in turn. (This is congruent with the epistemology of science: what is real is that with which the observer can causally interact.) The submariner’s picture and knowledge of the world outside the hull is “realistic” to the degree it facilitates successful navigation—that is, survival. The question of whether such knowledge is “true” has little meaning outside this context. Except in these limited terms, you cannot know what is outside your skull—or what is inside it, for that matter. The neurosurgeon can open up a skull to reveal a brain—can even stimulate that brain electrically to make it experience something the surgeon takes to be a hallucination. But even if the surgeon opened her own skull to peek inside, and manipulated her own experience, what she would see is but an image created by her own brain—in this case perhaps altered by her surgical interventions. The submariner’s constructed map is projected as external, real, and even accurate. But it is not the territory. What makes experience veridical or false is hardly as straightforward as the scientific worldview suggests. Science, as an extended or supplementary form of cognition, is as dependent on these caveats as natural perception. Whether scientific knowledge of the external world ultimately qualifies as truth will depend on how well it serves the survival of our species. On that the jury is still out.