Not So Naive Realism

Not So Naïve Realism

We are usually faced with a binary choice between naive realism and sense-data theories, as if naive realism had only one strength or type. But really there is a spectrum of positions aptly so described, according to their degree of naivety. The strongest type—the most naïve—says that the objects of perception are objectively precisely as they seem subjectively: if the object seems to have a certain property, then it objectively has that property, whether it is perceived or not. If it seems elliptical, then it is objectively elliptical; if it seems red, then it is objectively red; if it seems hot, then it is objectively hot, etc. Such a view comes under pressure from visual illusion, quirks of perspective, lighting conditions, and species-specific perceptual biases and projections. It is too naïve (“naïve naïve realism”). We do well to dial it back a notch. Thus, we reach a more sophisticated form of naive realism: exclude anything idiosyncratic to the perceiver and peculiar to a particular viewing situation. This will give us the distinction between primary and secondary qualities, the latter assigned to the perceiving mind not to the mind-independent object. Perceptions of shape and size are objectively veridical, but not perceptions of color and smell. We can be naïve about some perceived properties, but not all. But this position too can come under pressure, on two fronts. First, physics might indicate that even these supposedly objective properties do not really belong to the objects, being rather projections from our innate perceptual geometry. Nothing is really straight or intrinsically heavy, not when you get right down to it. Second, nothing about the object is precisely as it seems—there is always some degree of inaccuracy. We don’t see objective shape exactly as it is, if we see it at all (i.e., the objective spatial properties of the object). Following this line of thought, we conclude that ordinary perception is never completely veridical with respect to the actual object—never perfectly true to it. Some properties may be better perceived than others, but none are perceived exactly as they objectively are. The subject always gets in the way, as it were. This may be called wised-up naïve realism, as opposed to the dumb kind.

But doesn’t that mean that naïve realism is actually false and that we only perceive sense-data (subjective states)? I think not, because there is room for very sophisticated naïve realism, as sophisticated as you like. We never see objects as they are in themselves, in any respect, but we still see them—or we still see them.For we don’t see anything else, certainly not our states of mind. We see real physical things, just not as they really (objectively, intrinsically) are. Their properties line up with perceived properties (they covary) and in virtue of this they are seen objects. The mug on my table lacks all the properties it seems to me to possess—a certain color, shape, weight, warmth, etc.—but I still see it, that physical thing. What else would I be seeing? So, we can still be naïve realists about perception, but highly sophisticated, scientifically well-informed, super-smart naïve realists—just not naive naïve realists. I rather suspect this position is correct: our perceptual experience never accurately represents the actual objective nature of things—it doesn’t need to from a biological point of view—but that doesn’t prevent us from seeing them. I see my mug all right (not my mind or brain), just not as it is in itself, independently of my visual system. Isn’t the same thing true of all seeing, human and animal? It tracks and mirrors, but it doesn’t transparently portray—it isn’t like a high-resolution photo. Even insects can be granted naïve realism, though they never get the object precisely right—as it is sub specie aeternitatus. I therefore advocate what may be called hyper-sophisticated naïve realism. We really do see physical objects but never as they are from their point of view. Our perceptions are all completely false of the things perceived, but we still perceive those things.

I hear a piercing bellow from the back row: “But don’t we really know the properties of the objects of perception?!” Answer: no, we don’t, not as a result of our perceptual experience anyway; but it is okay to talk as if we do. I can say that I know my mug is blue just by looking at it, but it is not really blue, not in the way it seems to me (it can be said to be relationally blue). Error does not preclude efficacy. Pragmatically, we can talk this way, but strictly speaking things are not as they seem, ever. We can even have true beliefs about the true objects of perception, recognizing their divergence from our perceptions of them; the rampant falsity of our perceptions does not prevent that. Perceptual error does not imply cognitive (intellectual) error. I can know the truth concerning what my senses deceive me about, e.g., the location of color. My senses can be riddled with illusion and falsehood, but my mind may be omniscient, or at least undeceived. The senses do their job, but objective accuracy is above their pay grade. They may not even be anywhere close to the truth, but still serve their biological purpose. The point I have been making here is that naïve realism is not committed to uncritical acceptance of the contents of perception, only the most simple-minded version of it is. Naïve realism can be as sophisticated and revisionary as you like; it can even describe physical objects as totally unlike the way they seem. They might be ideas in the mind of God, not material things in physical space with shape and size; but still, we are seeing them. We don’t know we are looking into the mind of God—we think we are looking into an ungodly physical space—but we are. Objects are what they are, however they seem to us, and they may be very different from the way they seem. We could be seeing noumenal objects all the time (ripples in a ten-dimensional continuum) despite the fact that they seem to us like solid bounded things in three-dimensional space. Perception can pierce its own veil. It has the power to transcend its content. It is like a telescope trained on an occluded star.[1]

[1] It is an interesting point that we are ready to accept that our perceptions of distant stars fail to reveal their intrinsic nature, even to be radically misleading, but we naively assume that nearby objects are seen pretty much as they really are. Why is this? Why aren’t nearby objects as underperceived, or misperceived, as distant stars? Yet we can still see them. Seeing is one thing, correctly characterizing is another. After all, you can see a person in disguise as someone he is not—yet you still see him.