The Atlas of Reality

by Robert C. Koons,Timothy Pickavance

  13.3T Veil of Perception. Wholly mental things are the only possible objects of sense perception.

  What is the difference between Phenomenalism and the Veil of Perception? Phenomenalism entails that we know or are familiar with nothing but mental phenomena, while the Veil of Perception entails only that we can perceive only mental phenomena. To move from the Veil to Phenomenalism depends upon the assumption that our knowledge is limited to the objects of perception.

  Let's try to reconstruct an argument from the Veil of Perception to Phenomenalism, from Phenomenalism to Idealism, and from Idealism to Solipsism. First, suppose we accept the Veil of Perception. In that case, all of our perceptual evidence provides direct evidence only for the Existence of Mental Phenomena, since both the subjects and the objects of sense perception are mental in nature (according to the Veil). If we perceive nothing but mental things, if there is a Veil, Ockham's Razor (PMeth 1) justifies the further claim that there are no non-mental things. Phenomenalism is, therefore, plausible given the Veil of Perception.

  The move from Phenomenalism to Idealism also involves Ockham's Razor. If the world as we know it might consist of nothing but mental things, and we know that mental things exist, then Idealism would seem to be the simplest and most economical account of reality. This presupposes, of course, both that mental things exist and that they cannot be reduced to extra-mental things. In addition, we must assume that the existence of any mental thing is metaphysically independent of any non-mental thing. That is, we must assume that the simplest theory about what mental phenomena are does not by itself entail the existence of things other than minds and their experiences.

  Existence of Mental Phenomena. We are justified in believing that there are mental entities.

  The Irreducibility and Self-Sufficiency of the Mental. The simplest account of the truthmakers of true propositions about the existence of mental entities does not entail the existence of anything non-mental.

  Given Phenomenalism, the Existence of Mental Phenomena, and the Irreducibility and Self-Sufficiency of the Mental, Idealism is the simplest account of the world as we know it. The Existence of Mental Phenomena is hard to deny, as René Descartes pointed out. Even if I am deceived about everything else, there must be something mental going on, namely, my false thoughts. The Irreducibility and Self-Sufficiency of the Mental is far from obvious, in contrast. At this point, the most we can say is that Idealism is prima facie simpler than other theories, since it posits both fewer entities and fewer fundamental kinds of entities. We'll consider in Sections 13.5 and 13.6 some arguments for doubting the Irreducibility and Self-Sufficiency of the Mental.

  If we accept Idealism on the grounds of its greater simplicity, similar considerations should push us even further, toward Solipsism. It is obvious that each of us cannot directly perceive the experiences of others, nor is one subject of experience the direct object of the experience of another subject.2 Hence, we can express the Veil of Perception more forcefully:

  13.3T.1 Veil of Perception (Solipsistic Version). The only possible objects of sense perception for each subject are the wholly mental parts and properties of that very subject.

  Given the Solipsistic Veil, we can make suitable adjustments to the rest of the argument.

  13.2T.1 Phenomenalism (Solipsistic Version). For each subject of experience, the world as that subject knows it consists only in that very subject and its wholly mental parts and properties.

  Cartesian Justification. Each subject of experience is justified in believing in its own existence and that of its experiences.

  The Irreducibility and Self-Sufficiency of the Isolated Mind. The simplest account of the truthmakers of true propositions about the existence of a subject and its mental parts and properties does not entail the existence of anything apart from that subject.

  Given these three principles, the most economical theory for each subject to adopt is the one according to which only that subject and its actual, conscious experiences exist. This simplest theory is just Solipsism. In fact, we can go even farther. The simplest theory of all, consistent with the data of Solipsistic Phenomenalism, would be that nothing exists at all except a bare subject of experience and the bare states of consciousness experienced by that subject. In other words, the simplest account of the world is one with no hidden facts or causes, one according to which there are no truths except those made true by aspects of the self and its experiences that are completely and explicitly “given” in that subject's conscious life.

  In addition, as Santayana (1955) noted, we could press this argument all the way to the Solipsism of the present moment, the conclusion that nothing exists except the present state of one's own consciousness. How can one know that one's memory, even of one's own past states, is reliable, when it is impossible for one to perceive again the past events that one remembers having experienced?

  On the other hand, surely one can know more than one's own present state of consciousness. Solipsism of the present moment is an unattractive theory, to say the least, even if we can recognize the initial plausibility of the argument on its behalf. We don't want to find ourselves in the position of the American who once wrote to Bertrand Russell, stating that she was a Solipsist herself, and that she was quite surprised that there weren't many more Solipsists!

  Each of the steps that got us to Solipsism can be challenged. We will examine the Veil of Perception in the next section, and then turn in the following sections to a series of direct arguments against the Solipsistic Veil (Section 13.4), Phenomenalism (13.5), and Solipsism itself (13.5).

  13.3 Theories of Perception

  Why believe in the Veil of Perception? On its face it seems incredible, since we seem to perceive many physical, non-mental things, and events. We perceive lakes, trees, dogs, rocks, and flashes of lightning. None of these seems to be mere modifications of our minds.

  Thomas Reid (1785) proposed a principle of common sense according to which we are justified in taking our experience and common sense more generally to be reliable. This is in the spirit of the Presumption of Reliable Perception:

  PEpist 4 Reliable Perception Presumption. It is prima facie plausible to suppose that human perception and memory are reliable.

  In addition, recall Appearance of Bodies and Minds:

  PEpist 3 Appearance of Bodies and Minds. Perception and memory present us with what are apparently distinct physical things, including some embodying apparently distinct minds.

  Given these Reidian facts, why believe that we really perceive only events and processes going on within our own minds? We look first at two obviously inadequate arguments for the Veil, namely, Berkeley's inconceivability argument and the argument from causal mediation. Then, we will look at two much more interesting and important arguments, namely, the argument from hallucination and illusion (Section 13.3.1) and the argument from color and other secondary qualities (Section 13.3.2).

  TWO INADEQUATE ARGUMENTS FOR THE VEIL We already discussed (Section 12.1.5) Berkeley's argument for his claim that it is inconceivable that something sensible exist that is not being perceived. Berkeley gives a similar argument for the conclusion that it is inconceivable that anything should exist that is not being thought of. The argument begins with a challenge: try to conceive of something that is not being thought of. As soon as you succeed in thinking of something, you also make it true that the thing is being thought of. Hence, you can never conceive of anything that is not being thought of. Since conceivability, like imagination, is a guide to possibility, inconceivability should provide some evidence for impossibility. Therefore, we have some reason to think that nothing can exist without being thought of.

  Now suppose we try to conceive of a sensible thing, that is, a thing with sensible qualities like color or shape or texture. Berkeley claimed that the only way to conceive of such a thing is to imagine it, and to imagine a sensible thing is to imagine sensing it. Hence, if one succeeds in conceiving of a sensible thing, one must conceive of it as being se
nsed by oneself. Therefore, one has good reason to think that it is impossible that any sensible object should exist unsensed. Note that Berkeley's style of argument should lead to an even stronger conclusion: the necessity of Solipsism. If one cannot imagine a sensible object without imagining that one is sensing it, then one cannot conceive of a sensible object that isn't being sensed by oneself. Therefore, one should conclude that it is impossible for a sensible object to exist that isn't being sensed by oneself (and right now)!

  The fallacy of Berkeley's argument lies in his failure to distinguish between features of the act of thinking or representing and features of the object of that act. If we cannot conceive of an object with a certain feature F, that inconceivability does indeed provide evidence for the impossibility of F's. However, we cannot project necessary features of the act of thinking onto the object of thought. Just because one cannot think of an object without thinking of it, this does not make one's thinking of it a feature of the object as one thinks of it. When one thinks of a right triangle, the object of one's thought is simply a right triangle, not a right triangle being thought of now by oneself.

  The second inadequate argument for the Veil of Perception appeals to the fact of causal mediation between physical objects and our internal, mental ideas of them. This argument seems to be implicit in the early philosophical tradition of Idealism, beginning with Descartes.

  We perceive sensible objects directly, not indirectly. That is, we do not perceive them by perceiving something else.

  However, our sensory perception of external or physical objects is causally mediated by processes involving the transmission of light, sound waves or other processes, as well as internal neural processes.

  If an act of perceiving x is causally mediated, then we cannot perceive x directly in that act.

  Therefore, sensible objects are wholly mental (and not physical).

  The error in this argument lies in step 3. The existence of a direct or immediate relation of perception between a subject and an object can be causally mediated, in the sense that the first fact can be grounded in or caused by a complex series of causal links. The series of causal links from the surface of the physical object to the mind of the perceiver simply extends the perceiver's power to perceive objects directly. This can be seen to be true in two ways.

  The first way is phenomenological. When we perceive a causally distal object, like a distant star, we are not aware of the intervening causal links between the star and our mental state. If we were perceiving the star indirectly, we should be aware of our mental state or the state of our retina directly, and aware of the star only be being aware of some proximal stimulus. However, this is just not how our perceptual system works. When one sees a star, one is normally not even aware that one has a retina, much less that it is being stimulated in a certain way. It's even less plausible to suppose that one is aware of one's related brain states. So to perceive something directly does not, phenomenologically, require that there be no causal intermediaries between the object and the perceiver.

  The second way is epistemological. As Fred Dretske (1988) makes clear, we are more reliable at tracking distal stimuli than we are at proximate stimuli. For example, we can reliably locate objects in our environment or identify their colors, but we are not at all reliable at identifying the patterns of stimulation of our retina or our optical nerve. There is a good biological or evolutionary explanation for this: it was essential to the survival of our ancestors that they should be sensitive to the existence and states of objects in their environment, but not at all important that they should know much about the state of their own sensory organs. So, epistemologically, perception seems to relate us directly to the external environment.

  13.3.1 The argument from hallucination and illusion

  One popular argument for the Veil of Perception draws on the phenomena of sensory illusions and hallucinations. When we are having a hallucination, as when Shakespeare's Macbeth hallucinates the presence of a bloody knife floating before him, we seem to be perceiving something. In fact, the thing we perceive when we are hallucinating seems to have the very same types of characteristics, like shape, volume, color, and location, that we attribute to the supposedly physical things that we perceive when not hallucinating.

  There are also cases of sensory illusion that share this feature with hallucinations. For example, I might think that I am seeing a bear in the dark, when in fact I am reacting only to a pattern of light and shadows. Examples of illusory objects abound in everyday life. We regularly experience reflections, mirages, rainbows, and objects whose shapes and sizes are so distorted by refraction so as to constitute unreal entities. Even something as common as a rainbow could count as an illusory object, since there is in fact nothing located in the relevant part of the sky with any of the color-like features that we seem to see there. In all of these cases of hallucination and illusion, we seem to be perceiving something when there is nothing physical of the right sort to be an object of genuine perception.

  Hallucination. There are cases in which no non-mental thing is perceived that are introspectively indistinguishable from cases in which some physical object is perceived.

  Def D13.7 Veridical Perception. S perceives x veridically only if x is in fact as it appears to S to be.

  Some philosophers have suggested that during hallucinations we perceive such things as properties (like colors and shapes), complexes of properties (like redness-plus-triangularity), or propositions or possible states of affairs (like the possible existence of a red triangle). In this context, we want to preserve the word ‘perceive’ for those cases where the thing perceived is something concrete, like a table or a flash of lightning. Let's use the word ‘apprehend’ for any relation to more abstract objects, like properties or states of affairs. Thus, we will insist that in hallucination we are not perceiving any non-mental thing, even if we are apprehending things like properties, propositions or possible states of affairs.

  One might suggest that during a hallucination we are perceiving a region of space, which we misperceive as being occupied by a body of some kind. There are several problems with this suggestion. First, in a hallucination one might not be receiving any information from the region of space in which the hallucinated object seems to be occupied, so this would be a problematic case of perceptual knowledge. Second, the hallucinated object might not occupy any definite region of space—when for example, it does not occur with a perception of depth, or when we hallucinate an image in a mirror. Third, a hallucinated object can seem to move through space over time, which is something that a region of space cannot do.

  Advocates of the Veil of Perception could argue that their view is the simplest and most elegant explanation of the phenomena of perception and hallucination. Nonetheless, this has to be weighed against the fact that it entails rejecting many common-sense views about what it is we perceive when we are not hallucinating.

  Let's try to organize the alternatives to the Veil of Perception logically. Perceptual Realism is the denial of the Veil:

  13.3A Perceptual Realism. It is possible to perceive something other than wholly mental things.

  13.3A.1T Indirect Realism. It is possible to perceive something other than wholly mental things, but only indirectly, by virtue of perceiving wholly mental things.

  13.3A.1A Direct Realism. It is possible to directly perceive things other than wholly mental things (i.e., not by virtue of perceiving wholly mental things).

  Note that we are using the term ‘Realism’ in two quite different senses in this book. In Chapters 7–10, ‘Realism’ refers to views that accept universals as real and as grounds for similarity. In this chapter, we use ‘Realism’ to mean simply the negation of Idealism.

  What are direct and indirect perception? It is important not to confuse this distinction with the distinction between direct (perceptual) and indirect (inferential) evidence. Direct and indirect perception both provide direct, perceptual evidence of some fact. Indirect perception do
es not involve any inference. Instead, indirect perception occurs whenever one perceives one thing by way of perceiving another. For example, a rabbit can perceive the approaching hawk by perceiving the hawk's shadow. The rabbit perceives the shadow directly (one might think), the hawk indirectly. Perception via reflection would be another example. A fisherman can perceive a fish by perceiving the disturbance of the water in its wake, as a physicist can perceive an electron by perceiving a trail in a cloud chamber. A child can perceive his mother's presence by hearing her voice. He perceives the sound directly, the source of the sound indirectly. We ordinarily take ourselves to perceive opaque objects by perceiving the colors on the object's facing surface.

  Indirect Realists claim that we indirectly perceive all physical phenomena, by virtue of directly perceiving some mental phenomena. Direct Realists insist, to the contrary, that we can sometimes directly perceive physical objects and processes.

  Indirect Realism implies Sense Datum Theory. Sense data, on this view, are the mental phenomena that we directly perceive. Sense Datum Theory came under heavy fire in the twentieth century (see, for example, Barnes 1945, Austin 1962), but many of the criticisms focused on features that are not essential to Indirect Realism, such as the so-called ‘Phenomenal Principle’ (see Crane and French 2015).

  Phenomenal Principle. If something appears to someone to be F, then there must be something that really is F.

  This implies that our perception of sense data is infallible. We can't perceive a sense datum to be F unless it really is F. Otherwise, we risk an infinite regress by making our perception of sense data indirect. However, Indirect Realists need not accept the Phenomenal Principle. To resolve the problem of hallucination, all the Indirect Realist needs is the much weaker Exportation Principle:

  Phenomenal Exportation Principle. If it appears to someone that something is F, then there is something that appears to be F (to him or her).3