13.3.1.5 Meinongian direct realism.
Meinongian Direct Realists agree with Intentionalists that we don't perceive any real object in cases of hallucination. However, Meinongian Direct Realists insist that we do perceive something in hallucination: something unreal or non-existent. Thus, Meinongian Direct Realists do not need to deny the act-object structure of perceptual states. There is always something perceived in each perceptual state. Nor do they need to posit purely mental entities to serve as the objects of hallucination. Instead, they can insist that all perceptual states have physical, extra-mental objects. These objects exist in the case of veridical perception but do not exist in the case of hallucination.
We considered some of the pros and cons of Meinongianism (12.1A.1A) in Chapter 12. Here we note briefly a worry about the adequacy of this account as a theory of hallucination.
Assuming that we adopt Meinongianism without the Characterization Principle (as Graham Priest (2005) has advocated), we cannot say that perceived non-existent objects really have the sensible qualities that they seem to have. So, we cannot combine Meinongian Direct Realism with the Phenomenal Principle. However, we do get the Phenomenal Exportation Principle: if it appears that something is F, then something appears to be F. Consequently, a single non-existent thing can appear to move or change over time. So, there is some advantage, in terms of Sensory Error Minimization, of Meinongian Direct Realism over Intentionalism.
We want veridical perception to give us evidence for the existence of the things we are perceiving. We might even say that in veridical perception we perceive the existence of physical objects. What happens then in cases of hallucination? Do we perceive a non-existent existence? Does Macbeth perceive the existence of the bloody knife, an existence that happens to be unreal?
There is a danger here of falling into an infinite regress, moving from existence, existing existence, existing exististing-existence, and so on. Such an infinite regress would seem to lead to a global skepticism. I know that I exist, but do I know that my existence is an existing existence? Might it be a non-existing existence, like that of Macbeth's knife? Even if I know that my existence itself exists, can I know that the existence of my existence really exists? It seems that I would have to know an infinite number of things in order to know that any one thing really exists.
Suppose, then, that Meinongian Direct Realists reject all non-existent existences. Then there is a problem about supplying a fully adequate object for hallucinatory states. If Macbeth doesn't perceive the existence of the bloody knife (there being no such existence), won't he be able to notice by introspection its absence from his perceptual act? Meinongian Direct Realists should respond by denying that acts of existence are ever objects of perception. We perceive objects, but we don't perceive their existence. If they make this move, though, what is the connection between perceiving something and knowing that it exists? We seem to have lost our perceptual justification for believing that physical objects exist.
Meinongian Direct Realists could respond that every act of perceiving something prima facie justifies us in believing in the existence of the perceived object. It is normal for us to perceive an existing object. Perceiving the non-existent (as opposed to thinking about or hoping for the non-existent) is abnormal and exceptional.
Still, this leaves Meinongian Direct Realists with an odd dichotomy between existence and other properties.
UPSHOT OF THE ARGUMENT FROM HALLUCINATION As we argued at the outset, the Veil of Perception can only win by eliminating all of the alternative explanations of hallucination and illusion. This has proved difficult. Indirect Realism is the most vulnerable of the alternatives because it concedes so much to the Veil, by admitting that we always perceive wholly mental things. One is left wondering why one should assume that we sometimes also perceive physical things. A simple appeal to Reidian common sense would seem to be the best response on behalf of the Indirect Realists.
Intentionalism is the most popular alternative, but it does seem vulnerable to a challenge based on Sensory Error Minimization. Intentionalism implies that we get absolutely nothing right about the actual world when we experience a hallucination, while the alternative explanations attribute significantly less error to us in such cases.
We have, however, found little fault in either Perceptual Dualism or Meinongian Direct Realism. Of course, Actualists will object to Meinongian non-existent objects, but they could embrace the mental entities required by Perceptual Dualism.
13.3.2 The argument from colors
In the seventeenth century, scientists discovered that colors and other secondary qualities, like smells, tastes, sounds, and felt textures or temperatures, were not among the essential characteristics of physically fundamental entities, as ancient Greeks like Democritus had long suspected. Only the primary qualities, including position and velocity, volume, shape, and relative orientation, along with theoretical properties like mass and charge, are needed in a complete physical description of the world. The Idealist philosopher George Berkeley realized very quickly that these scientific discoveries provided a new basis for the Veil of Perception.
The argument from colors is an argument for the non-solipsistic version of the Veil of Perception, unlike the argument from hallucinations. The argument gives one no reason to suppose that the only sensible objects are parts of one's mind, although it may give one reason to suppose that they are all parts of some mind or other.
A simple version of Berkeley's argument for the Veil of Perception goes as follows:
Everything we can perceive has or can have color (or other secondary qualities).
Nothing physical has or can have color (or other secondary qualities).
Necessarily, everything we perceive is either physical or wholly mental.
Therefore, we cannot perceive physical objects, and everything perceivable must be wholly mental.
Anti-phenomenalists might resist this argument by supposing that although physical objects in fact lack color, they might have been colored, for example, had the laws of physics been different. A second argument for the Veil would appeal to the natural assumption that our sensory perception of the world is not radically and systematically in error:
We typically perceive objects as having color.
Nothing physical has color.
Our perceptions are not typically in error.
So, we don't typically perceive physical things. (From 1–3)
If we perceive physical things at all, then we typically perceive physical things.
So, we never perceive physical things.
Everything we perceive is either physical or wholly mental.
Therefore, we perceive only wholly mental things.
After all, if we discovered that our sense perception of color is constantly in error, this would provide us with grounds for undermining our confidence in any of the information deriving from the senses, leading to a global form of skepticism.
ARGUMENTS AGAINST PHYSICAL COLOR The crux of both of these arguments is premise 2, the claim that no physical entity has color. Many philosophers suppose that the only properties that physical objects have are physical properties. Let's assume for the moment that this is so. That is, assume the Pure Physicality of Physical Things:
Pure Physicality of Physical Things (PPPT). If a physical thing x has an intrinsic property P, then P is a physical property.
Assuming the Pure Physicality of Physical Things (PPPT), we can prove that no physical thing is colored simply by proving that color is not a physical property. We have assumed that physical properties are (1) non-mental (i.e., instantiating one does not entail being a mind or part of a mind), (2) natural, and (3) similar to the properties posited by modern physics. We could define a physical thing as something with at least one intrinsic physical property. PPPT insists that anything with a physical property must have only physical properties. As we shall see, there are some views that deny this, including Panpsychism, according to which every physical
thing is conscious.
What is it to be a physical property? Why think color isn't physical? Let's start by defining a core physical property.
Def D13.8 Core Physical Property. P is a core physical property if and only if P is (1) fundamental or perfectly natural, (2) causally efficacious, and (3) non-mental.
We're going to assume that core physical properties can be mapped one-to-one onto the basic quantitative and qualitative predicates of our current physical theory, at least to a fairly good approximation. That is, we are going to assume that core physical properties are something like quantities of mass or charge, baryon number, volume, and so on. This is the Physics Carves Nature at the Joints Hypothesis:
Physics Carves Nature at the Joints (Joints Hypothesis). The core physical properties are something like the basic properties posited by modern science: mass, charge, baryon number, and so on.
We can now define a physical property in terms of core physical properties:
Def D13.9 Physical Property. P is a physical property if and only if P is either:
(1) a core physical property,
(2) a conjunction of core physical properties,
(3) a disjunction of conjunctions of core physical properties, or
(4) a power or disposition whose defining stimuli and responses satisfy one of condi-tions (1) through (3).
We will not assume that the conjunctions, disjunctions, or definitions of dispositions involved in cases (2)–(4) are finite. Infinitely long real definitions are permitted.
We are now in a position to consider whether colors and other secondary qualities are physical properties.
Here are six observations about color, developed in their most sophisticated form by Adam Pautz (2006), which can be used in an argument for the non-physicality of color.5
1. No one-to-one correspondence with fundamental physical properties. The fundamental properties of physical bodies are those identified by theoretical particle physics, properties having to do with mass, energy, charge, and so on. There is no simple or unitary physical property of this kind that corresponds one-to-one with any color. However, color is a simple and unitary property of colored things.
2. Unique patterns of similarity and dissimilarity. Our color perception, if it is at all accurate, reveals a system of similarities and dissimilarities among the things we perceive. We perceive, for example, all blue things as genuinely similar to one another in color. However, physics and the physiology of color perception reveal that the physical objects that we perceive as blue have no natural similarity with one another. A bewildering variety of physical surfaces can appear to have exactly the same hue.
3. Features of colors without physical counterparts. There are features of color, including the binarity of binary colors (like orange and purple) that correspond to the way in which color properties are processed by the brain, not to any way the perceived physical object is in itself. Orange things seem to have a property that is somehow a blend of red and yellow. Red and yellow things, in contrast, seem to have a pure or unitary quality, one that is constituted by no such blending. This apparent contrast between things with binary and things with unitary color corresponds to no consistent difference at the physical level. It is to be explained, instead, by the details of color-vision processing in the human brain, namely, the fact that signals from the optic nerve are directed into two separate channels, a yellow-blue channel and a red-green channel. Binary colors correspond to the simultaneous stimulation of both channels, and unitary colors to the stimulation of one and not the other. The difference has nothing to do with any duality or lack of duality in the objects themselves.
4. Variation in the perception of color within a species. Perfectly normal observers in normal circumstances do not perfectly agree in their assignment of colors to perceived objects. The human sexes (male and female), for example, vary in their average perception of color. If color were a property of physical objects, then normal observers in normal circumstances would perceive exactly the same color, and yet they do not.
5. Variation in the perception of color across species. Perfectly normal organisms in different species do not perceive the same colors because their eyes are sensitive to different ranges of the electromagnetic spectrum and (more importantly) because their brains process visual information in radically different ways. Pigeons, for example, perceive objects as colored that human beings perceive as gray, lacking hue. Who is right? Physicalists about color have only three options for explaining this variation, assuming that our color perception is mostly accurate: (1) human beings are the only species with correct color perception, (2) physical objects really have a very large number (perhaps an infinitely large number) of distinct colors simultaneously (one corresponding to each possible species), or (3) colors are powers or dispositions to cause a variety of perceptual states in different species. We'll discuss option (3) below. Option (1) is problematic, since there is no reason to suppose that just one species gets colors right, and even less reason to think that our species is so favored.
Option (2) is contrary to the intuition that nothing can have two distinct colors at the same time, much less an infinite number. In addition, it seems plausible that colors can be identified with certain mixtures of “colorish” qualities. For example, orange is a roughly equal mixture of reddishness and yellowishness. Yellow, in contrast, involves the exclusive presence of yellowishness. If any other kind of “ishness” were present, the thing would no longer be yellow. Yet, if there are “ishnesses” corresponding to other species, these will almost always be present in cases of apparently yellow surfaces.6 If we don't perceive these other ishnesses, we will be systematically misperceiving the actual colors of things.
6. Phenomenologically, color is natural, intrinsic, and categorical. If we consider how color appears to us, it seems to be a natural property, the basis of real similarities, and not a disjunctive or gerrymandered property (like grue). Moreover, the color of colored objects seems to be intrinsic to them. Colored objects seem to be colored in and of themselves. Color does not seem to be relational in nature. Finally, color seems to be categorical, a property that concerns simply how things actually are, without reference to how they would be or what they would do under merely hypothetical suppositions. Given Sensory Error Minimization, we should prefer a theory that respects these apparent facts about color.
CAN COLOR BE PHYSICAL? FOUR CASES In light of these six observations, can color be a physical property? There are four cases to consider. Colors might be core physical properties, conjunctions of core physical properties, disjunctions (of conjunctions) of core physical properties, or powers.
1. Are colors core physical properties? Colors don't stand in a one-to-one relationship to the basic vocabulary of physics, so they cannot be core physical properties, given the Joints Hypothesis. But why accept the Joints Hypothesis? Why not think that colors and other secondary qualities are causally powerful, non-physical properties? The main reason for this is an assumption about the causal closure of the physical world. There is some inductive, empirical support for the hypothesis that only core physical properties have fundamental causal powers. There is also a special reason in this case. It doesn't seem likely that ordinary physical objects, including inanimate objects, have any non-physical powers. The facts about light reflection and sound-wave generation, for example, seem to be completely explainable in terms of core physical properties. It just isn't plausible that colors and other secondary qualities have any additional causal oomph of their own.
2. Are colors conjunctions of core physical properties? This view is unlikely because colors can be realized in an extremely large number of ways, with no physical similarities among the realizers (see observation #2 above). In addition, this view cannot explain the appearance of binary colors, since there would be no common denominator between the conjunctions defining orange and its two constituent colors, red and yellow (see observation #3).
3. Are c
olors disjunctions? This view is also unlikely, because disjunctive properties are not causally relevant, and we have already seen that we have good reason for thinking that colors are powerful. In addition, disjunctive properties are not natural, and they cannot ground the kind of systematic similarities and dissimilarities that we observe among the colors (observation #2). Moreover, if colors were disjunctions, there would have to be some basis for deciding which physical-property conjunctions to include in the disjunction. If the list is not to be arbitrary, we would have to treat color as a response-dependent property. That is, we would have to include a conjunction on the list for color C if and only if such a property conjunction is perceived as having color C by normal observers in standard conditions. However, such a definition cannot accommodate the variability of color sensitivity (see observations #4 and #5), since the inclusion or exclusion of any disjunct would have to be determined by whether a physical surface of that kind produces the right color sensation in a normal observer in standard circumstances.
4. Are colors powers? This view has been a popular response by anti-Phenomenalists since John Locke. However, it is subject to two devastating objections. First, there is a phenomenological objection. Colored objects don't appear in our sensory experience to be powers or disposition. When I see something as red, I certainly don't see it as something having the power to cause red sensations in me and subjects like me. I see it as having some intrinsic, categorical, and non-relational property (observation #6). Further, it is plausible to suppose that ‘red’ is ambiguous, sometimes referring to the intrinsic property appearing to us in visual experience, sometimes to the power in things to produce visual experiences of that kind in standard circumstances. It is the first kind of color, the intrinsic property, that we perceive. We do not perceive the power. The defender of the Veil need only argue that this perceived, intrinsic property is non-physical. Second, there is a circularity objection. What appearance is it that the colored objects have the power or disposition to cause? If redness is just a property to cause certain sensations, then the sensations they cause cannot themselves be red. One is left wondering what sort of sensations red objects cause essentially.
The Atlas of Reality Page 49
