Inconsistencies in Keith Frankish’s Case Against Panpsychism (Part I)

I’m not a Panpsychist, or, rather, I’m not a Panpsychist in any contemporary sense of the term. Nevertheless, Panpsychism has a rich and rigorous history. It deserves to be taken very seriously. With that being said, I stumbled across an article written by the Illusionist philosopher, Keith Frankish—an article that criticizes Panpsychism severely. I wanted to come to Panpsychism’s defense. I also wanted to examine several inconsistencies in Frankish’s own arguments against Panpsychism.    

“Panpsychism’s popularity stems from the fact that it promises to solve two deep problems simultaneously. The first is the famous “hard problem” of consciousness. How does the brain produce conscious experience? How can neurons firing give rise to experiences of color, sound, taste, pain and so on? In principle, scientists could map my brain processes in complete detail but, it seems, they could never detect my experiences themselves—the way colors look, pain feels and so on: the phenomenal properties of the brain states involved. Somehow, it seems, brain processes acquire a subjective aspect, which is invisible to science. How can we possibly explain this?” (Frankish, Why Panpsychism is Probably Wrong, 1)

Keith Frankish insists that colors, sounds, tastes, and pains are real, and that our experiences of colors, sounds, tastes, and pains are also real (he simply qualifies this by saying that we mischaracterize our experiences (or objects of those experiences) as having properties which they don’t actually instantiate but only “seem” to). If we take the above passage at face value, then Frankish seems to be suggesting that my experience of a color, sound, taste, or pain is the way in which the color, sound, taste, or pain looks, sounds, tastes, or feels to me. Or, to simplify, “my-experience-of-a-color” is “the-way-in-which-the-color-looks-to-me,” “my-experience-of-a-pain” is “the-way-in-which-the-pain-feels-to-me,” and so on. 

Frankish goes on to say that the ways in which colors, sounds, tastes, and pains look, sound, taste, and feel to me are “phenomenal properties.” For example, he says, “the way colors look, pain feels and so on: the phenomenal properties of the brain states involved,” (Why Panpsychism is Probably Wrong) “It seems obvious that phenomenal properties, such as the feel of pain,” (Panpsychism and the Depsychologization of Consciousness), etc. To Frankish, “phenomenal properties…seem completely inaccessible to science. They are wholly subjective features, which simply do not show up on the scientific radar.” (Panpsychism and the Depsychologization of Consciousness). Now, if “phenomenal properties” were real, and were instantiated in the world, then what would these “wholly subjective features” be properties of? Frankish states that these “phenomenal properties” (if they were indeed instantiated in the world) appear to be properties of the brain states involved in our experiences of colors, sounds, tastes, pains, and so on (Why Panpsychism is Probably Wrong).

Confusion ensues. If “my-experience-of-a-color” is just “the-way-in-which-the-color-looks-to-me,” and “the-way-in-which-the-color-looks-to-me” is a “phenomenal property,” then we are faced with several inconsistencies.

(I) Frankish denies that “phenomenal properties” are “real” and instantiated in the world. Since Frankish holds that “my-experience-of-a-color” is just “the-way-in-which-the-color-looks-to-me” (cf. “my experiences themselves—the way colors look, pain feels and so on...” (Why Panpsychism is Probably Wrong)), and that “the-way-in-which-the-color-looks-to-me” are “phenomenal properties” (cf. “the way colors look, pain feels and so on: the phenomenal properties…” (Why Panpsychism is Probably Wrong)), it follows that Frankish is committed to the unreality of my experiences of colors, sounds, tastes, and pains. However, this conflicts with his insistence that “our lives are filled with conscious experiences—episodes of seeing, hearing, smelling, tasting, feeling, and of having bodily sensations of various kinds” and that these experiences are indeed real (Cf. The Demystification of Consciousness). 

(II) Furthermore, Frankish, being an Identity Theorist, is committed to the view that “my-experience-of-a-color” is identical to a process in my brain. However, Frankish asserts that “phenomenal properties” (if they were instantiated in the world) would be properties of the brain states involved in our experiences of colors, sounds, tastes, pains, and so on. (cf. “the way colors look, pain feels and so on: the phenomenal properties of the brain states involved.” (Why Panpsychism is Probably Wrong)) This is a manifest contradiction. Let me illustrate this. If “my-experience-of-a-color” is identical to “the-way-in-which-the-color-looks-to-me,” and “the-way-in-which-the-color-looks-to-me” is a “phenomenal property,” then it cannot be the case that the “phenomenal property” is a property of the brain processes involved in “my-experience-of-a-color.” Indeed, “my-experience-of-a-color” is ex hypothesi identical to that very brain process. A property (“phenomenal” or “non-phenomenal”) of the brain process cannot be identical to the brain process of which it is a property—an adjective cannot have itself as its own substantive!

That Which Knows All Things and Is Known by None: An Argument Against the Identity Theory of Mind

 Abstract

“That which knows all things and is known by none is the subject.”¹

“He sees, but he can’t be seen, he hears, but he can’t be heard; he perceives, but he can’t be perceived.”²

This essay presents an argument against the Identity Theory of Mind”³ that does not depend upon the apparent epistemic gap between physical and phenomenal truths, the existence of qualia, or the conceivability of philosophical zombies. The argument begins by establishing three principles: the Principle of Perceptual Determinacy, the Principle of Perceptual Self-Exclusion, and the Principle of Perceptual Non-Loopability—each identifying a necessary condition underlying the existence, or occurrence, of perceptual states. Next, a contradiction is shown to follow from the three principles and the Identity Theory being jointly true. Since the three principles can be demonstrated a priori, it follows that the Identity Theory is false. Potential objections to the argument are then raised and addressed. 

The Principle of Perceptual Determinacy

“When you perceive you perceive something, and…what you do perceive cannot be the same as the perception of it.”⁴

The world unfolds itself for us conscious beings through communions of perceivings and perceiveds: perceptual states and their percepta. These distinguishable factors march two abreast, side-by-side, hand-in-hand, as mutually-implying correlates within a subject’s experience. For a subject to “perceive” a tree is for him to “have” a perceptual state that is a “perception of a tree;”⁵ thus, if a subject has a perceptual state, P, and P is a “perception of a tree,” then P’s perceptum is that tree which the subject perceives—that particular “somewhat” which P is a “perception of.”⁶

Whenever a subject perceives, the perceptum of his perceptual state is an existing, or occurring, determinate “somewhat.” Moreover, this perceived “somewhat” appears to him, the perceiving subject. If the perceptum of a subject’s perceptual state was not an existing, or occurring, determinate “somewhat,” then the perceptual state in question (i.e., the “perceiving”) would be a “perception of nothing,” and this is absurd. For, no existing, or occurring, perceptual state is a “perception of nothing.” Furthermore, if the perceptum of a subject’s perceptual state was not itself determinate, then the perceptum would be incapable of appearing to and for the perceiving subject—and this, of course, is absurd. Indeed, unlike an existing, or occurring, determinate “somewhat,” an indeterminate would lack any definite characteristics of its own, and, as such, would not admit to being perceived in any particular way, let alone a variety of ways.⁷ However, a perceptual state is determinate only insofar as it terminates in an existing, or occurring, determinate perceptum. Indeed, a perceptual state is what it is (i.e., is determinate and individuated) in virtue of possessing characteristics that are fixed or determined by what the perceptual state is a “perception of.” If this were not so, then it should be possible to render a perceptual state determinable independent of the perceptual state’s objective reference (i.e., without any reference to what it is a “perception of”); however, such a task is impossible not only in practice but in principle. Since no perceptual state can either exist, or occur, unless it is determinate, and a perceptual state is determinate only insofar as its perceptum either exists, or occurs, and is determinate, we are driven to the following result: The Principle of Perceptual Determinacy. According to this principle, for any perceptual state, P, P can neither exist, nor occur, unless P’s perceptum, O, either exists, or occurs, and is determinate. 

The Principle of Perceptual Self-Exclusion

“That the subject should become object for itself is the most monstrous contradiction ever thought of...”⁸

Using the Principle of Perceptual Determinacy, it can be demonstrated that no existing, or occurring, perceptual state either has itself as its own perceptum, or is identical to its own perceptum. In other words, for any existing, or occurring, perceptual state, P, P is not a “perception of P.” I call this the Principle of Perceptual Self-Exclusion. The following argument illustrates how the Principle of Perceptual Self-Exclusion follows a priori from the Principle of Perceptual Determinacy:

If a perceptual state, A, was a “perception of A,” then A could neither exist, nor occur, nor be determinate, unless it was the perceptum of an existing, or occurring, perceptual state, B, that was a “perception of a perception of A.” But B could neither exist, nor occur, nor be determinate, unless it was the perceptum of an existing, or occurring, perceptual state, C, that was a “perception of a perception of a perception of A.” But C could neither exist, nor occur, nor be determinate, unless it was the perceptum of an existing, or occurring, perceptual state, D, that was a “perception of a perception of a perception of A.” But D could neither exist, nor occur, nor be determinate, unless it was the perceptum of an existing, or occurring, perceptual state, E, that was a “perception of a perception of a perception of a perception of A.” But E could neither exist, nor occur, nor be determinate, unless it was the perceptum of an existing, or occurring, perceptual state, F, that was a “perception of a perception of a perception of a perception of a perception of A.” But F could neither exist, nor occur, nor be determinate, unless it was the perceptum of an existing, or occurring, perceptual state, G, that was a “perception of a perception of a perception of a perception of a perception of a perception of A.” But G could neither exist, nor occur, nor be determinate, unless it was the perceptum of an existing, or occurring, perceptual state, H, that was a “perception of a perception of a perception of a perception of a perception of a perception of a perception of A.” And so on, ad infinitum. We find that neither A nor any of the other aforementioned perceptual states, could either exist, or occur, or be determinate, unless we had reached the end of a vicious regress of perceptual states—and this, of course, is impossible. Therefore, no existing, or occurring, perceptual state either has itself as its own perceptum, or is identical to its own perceptum. Therefore, the Principle of Perceptual Self-Exclusion is true.

The Principle of Perceptual Non-Loopability

“Knowledge that turns inward upon itself creates a current so hot that it burns out the fuse.”⁹

In a similar vein to the Principle of Perceptual Determinacy and the Principle of Perceptual Self-Exclusion, I will introduce what I call the Principle of Perceptual Non-Loopability. This principle maintains that no n-number (n ≥ 2) of existing, or occurring, perceptual states form a series of perceptual states and percepta wherein the first state is a “perception of the second,” the second is a “perception of the third,” the third is a “perception of the fourth,” and so on—the last perceptual state in the series being a “perception of the first.” For clarity, I have provided a list of eight examples of perceptual “loops.”

A Digonal Perceptual “Loop”

Any situation wherein two perceptual states are such that the first is a “perception of the second,” and the second is a “perception of the first.”

A Trigonal Perceptual “Loop”

Any situation wherein three perceptual states are such that the first is a “perception of the second,” the second is a “perception of the third,” and the third is a “perception of the first.”

A Tetragonal Perceptual “Loop”

Any situation wherein four perceptual states are such that the first is a “perception of the second,” the second is a “perception of the third,” the third is a “perception of the fourth,” and the fourth is a “perception of the first.”

A Pentagonal Perceptual “Loop”

Any situation wherein five perceptual states are such that the first is a “perception of the second,” the second is a “perception of the third,” the third is a “perception of the fourth,” the fourth is a “perception of the fifth,” and the fifth is a “perception of the first.”

A Hexagonal Perceptual “Loop”

Any situation wherein six perceptual states are such that the first is a “perception of the second,” the second is a “perception of the third,” the third is a “perception of the fourth,” the fourth is a “perception of the fifth,” the fifth is a “perception of the sixth,” and the sixth is a “perception of the first.” 

A Heptagonal Perceptual “Loop”

Any situation wherein seven perceptual states are such that the first is a “perception of the second,” the second is a “perception of the third,” the third is a “perception of the fourth,” the fourth is a “perception of the fifth,” the fifth is a “perception of the sixth,” the sixth is a “perception of the seventh,” and the seventh is a “perception of the first.”

An Octagonal Perceptual “Loop”

Any situation wherein eight perceptual states are such that the first is a “perception of the second,” the second is a “perception of the third,” the third is a “perception of the fourth,” the fourth is a “perception of the fifth,” the fifth is a “perception of the sixth,” the sixth is a “perception of the seventh,” the seventh is a “perception of the eighth,” and the eighth is a “perception of the first.”

And so on, ad indefinitum.

If the Principle of Perceptual Non-Loopability is true, it would follow that no n-number (n ≥ 2) of existing, or occurring, perceptual states form a series of perceptual states and percepta wherein the first perceptual state in the series is a “perception of the second,” the second is a “perception of the third,” the third is a “perception of the fourth,” and so on—the last perceptual state in the series being a “perception of the first.” Like the Principle of Perceptual Self-Exclusion, the Principle of Perceptual Non-Loopability follows a priori from the Principle of Perceptual Determinacy. This can be demonstrated by means of the following argument:

If two perceptual states, A and B, were such that A was a “perception of B,” and B was a “perception of A,” then A and B could neither exist, nor occur, nor be determinate, unless A was the perceptum of an existing, or occurring, perceptual state, C, that was a “perception of a perception of B,” and B was the perceptum of an existing, or occurring, perceptual state, D, that was a “perception of a perception of A.” However, C and D could neither exist, nor occur, nor be determinate, unless C was the perceptum of an existing, or occurring, perceptual state, E, that was a “perception of a perception of a perception of B,” and D was the perceptum of an existing, or occurring, perceptual state, F, that was a “perception of a perception of a perception of A.” However, E and F could neither exist, nor occur, nor be determinate, unless E was the perceptum of an existing, or occurring, perceptual state, G, that was a “perception of a perception of a perception of a perception of B,” and F was the perceptum of an existing, or occurring, perceptual state, H, that was a “perception of a perception of perception of a perception of A.” And so on, ad infinitum. We find that neither A, nor B, nor any of the aforementioned perceptual states, could either exist, or occur, or be determinate, unless we had reached the end of a vicious regress of perceptual states—and this, of course, is impossible. Therefore, no two existing, or occurring, perceptual states are such that the first is a “perception of the second,” and the second is a “perception of the first.” Therefore, the Principle of Perceptual Non-Loopability is true.

One might object by saying that the above argument only demonstrates that a digonal perceptual “loop” can neither exist, nor occur. However, this objection rests on an overlooking the fact that the same conclusion follows if we were to substitute the digonal perceptual “loop” for a trigonal, tetragonal, pentagonal, hexagonal, heptagonal, octagonal (and so on, ad indefinitum) perceptual “loop.” Thus, we can affirm that the Principle of Perceptual Non-Loopability has been proved.

An Argument Against the Identity Theory of Mind

Let’s imagine two neighbors: John and Thomas. If John were to have a perceptual state, A, and A was a “perception of Thomas’s perceptual state, B,” and if Thomas’s perceptual state, B, was a “perception of John’s perceptual state, A,” then John’s and Thomas’s perceptual states would form a digonal perceptual “loop.” However, since we have proved the Principle of Perceptual Non-Loopability, it follows that such a digonal perceptual “loop” can neither exist, nor occur. Therefore, if John were to have an existing, or occurring, perceptual state, A, and A was a “perception of Thomas’s existing, or occurring, perceptual state, B,” then Thomas’s existing, or occurring, perceptual state, B, cannot—on pain of contradiction—be a “perception of John’s existing, or occurring, perceptual state, A.”

According to the Identity Theory, all of John’s perceptual states are identical to some of John’s brain states, and all of Thomas’s perceptual states are identical to some of Thomas’s brain states. In other words, if John were to have an existing, or occurring, perceptual state, A, then A would be identical to John’s existing, or occurring, brain state, θ, and if Thomas were to have an existing, or occurring, perceptual state B, then B would be identical to Thomas’s existing, or occurring, brain state, ψ. Furthermore, since the Principle of Perceptual Non-Loopability is true, we arrive at the following result: 

If the Identity Theory is true, then if John has an existing, or occurring, perceptual state, A, and A is a “perception of Thomas’s existing, or occurring, brain state, ψ,” then Thomas’s existing, or occurring, brain state, ψ, cannot—on pain of contradiction—be a “perception of John’s existing, or occurring, brain state, θ.”

Now, there is nothing absurd or contradictory in the idea that John can perceive (i) any of his own existing, or occurring, brain states, and (ii) any existing, or occurring, state of Thomas’s brain whilst Thomas perceives any existing, or occurring, state of John’s brain. Likewise, there is nothing unintelligible or self-contradictory in the idea that Thomas can perceive (i) any of his own existing, or occurring, brain states, and (ii) any existing, or occurring, state of John’s brain whilst John perceives any existing, or occurring, state of Thomas’s brain. Indeed, we must admit that it is possible—in principle—for John to perceive (i) any of his own existing, or occurring, brain states, and (ii) any existing, or occurring, state of Thomas’s brain whilst Thomas perceives any existing, or occurring, state of John’s brain. Again, we must also admit that it is possible—in principle—for Thomas to perceive (i) any of his own existing, or occurring, brain states, and (ii) any existing, or occurring, state of John’s brain whilst John perceives any existing, or occurring, state of Thomas’s brain. However, if the Identity Theory is true, then it would be a contradiction for John to perceive a certain existing, or occurring, state, ψ, of Thomas’s brain whilst Thomas is perceiving a certain existing, or occurring, state, θ, of John’s brain. Indeed, such a state of affairs would be impossible—in principle. Therefore, since it is possible—in principle—for John to perceive any existing, or occurring, state of Thomas’s brain whilst Thomas perceives any existing, or occurring, state of John’s brain, it follows that the Identity Theory is false. We may put the general argument in the following form:

P1)  If the Identity Theory is true, then John cannot—in principle—perceive any existing, or occurring, state of Thomas’s brain whilst Thomas perceives any existing, or occurring, state of John’s brain.

P2)  John can—in principle—perceive any existing, or occurring, state of Thomas’s brain whilst Thomas perceives any existing, or occurring, state of John’s brain.

C1)  Therefore, the Identity Theory is false. [From P1 and P2]

We may transpose the above argument in a way that focuses solely on John. This variation of the argument relies solely on the Principle of Perceptual Self-Exclusion without relying on the Principle of Perceptual Non-Loopability. The Identity Theory, as we have seen, is committed to John’s existing, or occurring, perceptual state, P, being identical to one of John’s existing, or occurring, brain states (at the time of John’s having of the perceptual state, P). Since the Principle of Perceptual Self-Exclusion is true, it follows that whenever John has an existing, or occurring, perceptual state, P, P’s perceptum, O, cannot—in principle—be John’s existing, or occurring, perceptual state, P. However, it follows from this that if it is possible—in principle—for John to have an existing, or occurring, perceptual state, P, that is a “perception of” any one of his existing, or occurring, brain states (at the time of John’s having of the perceptual state, P), then John’s existing, or occurring, perceptual state, P, is not identical to any one of his existing, or occurring, brain states (at the time of John’s having of the perceptual state, P). But if John’s existing, or occurring, perceptual state, P, is not identical to any one of his existing, or occurring, brain states (at the time of John’s having of the perceptual state, P), it follows that the Identity Theory is false; for, such a theory is committed to John’s existing, or occurring, perceptual state, P, being identical to one of John’s existing, or occurring, brain states (at the time of John’s having of the perceptual state, P). Since it is possible—in principle—for John to have an existing, or occurring, perceptual state, P, that is a “perception of” any one of his existing, or occurring, brain states (at the time of John’s having of the perceptual state, P), it follows that the Identity Theory is false. We may standardize the argument as follows:

P1)  If the Identity Theory is true, then all of John’s existing, or occurring, perceptual states are identical to some of John’s existing, or occurring, brain states.

P2)  If all of John’s existing, or occurring, perceptual states are identical to some of John’s existing, or occurring, brain states, then some of John’s existing, or occurring, brain states cannot—in principle—be the percepta of John’s existing, or occurring, perceptual states. 

P3)  All of John’s existing, or occurring, brain states can—in principle—be the percepta of John’s existing, or occurring, perceptual states. 

C1)  Therefore, the Identity Theory is false. [From P1—P3]

The Identity Theorist cannot reject the argument’s first premise without giving up the Identity Theory, and the Identity Theorist cannot challenge the second premise without first refuting the Principle of Perceptual Self-Exclusion—something which I take to have been sufficiently proved. Furthermore, the Principle of Perceptual Self-Exclusion follows a priori from the Principle of Perceptual Determinacy—another principle which I take to have been sufficiently proved. Therefore, if the Identity Theorist is to reject the Principle of Perceptual Self-Exclusion, he must also reject the Principle of Perceptual Determinacy. With respect to the third premise, there is clearly nothing absurd or contradictory involved in the idea that John can—in principle—perceive any given existing, or occurring, state of his own brain whilst having an existing, or occurring, perceptual state. Indeed, if John’s optic nerves were lengthened and his eyes positioned correctly, he would certainly be able to perceive his own brain states (Not to mention the possibility of John using an autocerebroscope to perceive his brain states).¹⁰ Now, suppose the Identity Theorist were to argue that whenever a perceptual state (or a brain state) is perceived, it cannot be perceived in its entirety, but only partially, thus a perceptual state (or a brain state) having itself as its own perceptum would present no difficulty. Such an objection fails because it would lead to the following contradiction: a proper part of a whole (i.e., the perceived part of the whole perceptual state) would be both less than, and identical to, the whole (i.e., the whole perceptual state) of which it was a proper part.

Now, there is a third argument that we may employ against the Identity Theory that makes use of the Principle of Perceptual Self-Exclusion and the Principle of Perceptual Non-Loopability—both of which follow a priori from the Principle of Perceptual Determinacy. The argument takes the following form:

P1)  If the Identity Theory is true, then there are some brain states which are—in principle—imperceptible. 

P2)  There are no brain states which are—in principle—imperceptible. 

C1)  Therefore, the Identity Theory is false. [From P1 and P2]

As mentioned in the preceding paragraphs, there is nothing at all incoherent or contradictory in the idea that all existing, or occurring, brain states are—in principle—perceptible; indeed, it seems that it is not only possible in theory, but also—eventually—in practice.¹⁰ As the force of the argument lies in the first premise, I will concentrate my attention on defending it the most. The first premise is based on the fact that if the Identity Theory is true, then all existing, or occurring, perceptual states are identical to some existing, or occurring, brain states. Supposing that all existing, or occurring, perceptual states are identical to some existing, or occurring, brain states, it can be demonstrated that there are some existing, or occurring, brain states which are—in principle—imperceptible. The argument is as follows:

If an existing, or occurring, perceptual state, A, is a “perception of an existing, or occurring, perceptual state, B,” then A is neither the perceptum of B, nor the perceptum of A; because if A was the perceptum of either A or B, then either the Principle of Perceptual Self-Exclusion, or the Principle of Perceptual Non-Loopability would be violated. However, this entails that A is imperceptible—in principle—unless A was the perceptum of an existing, or occurring, perceptual state, C,—a perceptual state that was different from A and B.

Indeed, if A was the perceptum of A, then we would have an existing, or occurring, perceptual state that had itself as its own perceptum—a clear violation of the Principle of Perceptual Self-Exclusion. Furthermore, if A was the perceptum of B, then we would have a digonal perceptual “loop”—a clear violation of the Principle of Perceptual Non-Loopability. 

However, this entails that C is imperceptible—in principle—unless C was the perceptum of an existing, or occurring, perceptual state, D,—a perceptual state that was different from A, B, and C.

Indeed, if C was the perceptum of C, then we would have an existing, or occurring, perceptual state that had itself as its own perceptum—a clear violation of the Principle of Perceptual Self-Exclusion. Furthermore, if C was the perceptum of either A or B, then we would have either a digonal, or trigonal, perceptual “loop”—a clear violation of the Principle of Perceptual Non-Loopability. 

However, this entails that D is imperceptible—in principle—unless D was the perceptum of an existing, or occurring, perceptual state, E,—a perceptual state that was different from A, B, C, and D.

Indeed, if D was the perceptum of D, then we would have an existing or occurring, perceptual state that had itself as its own perceptum—a clear violation of the Principle of Perceptual Self-Exclusion. Furthermore, if D was the perceptum of either A, B, or C, then we would have either a digonal, trigonal, or tetragonal perceptual “loop”—a clear violation of the Principle of Perceptual Non-Loopability. 

However, this entails that E is imperceptible—in principle—unless E was the perceptum of an existing, or occurring, perceptual state, F,—a perceptual state that was different from A, B, C, D, and E.

Indeed, if E was the perceptum of E, then we would have an existing, or occurring, perceptual state that had itself its own perceptum—a clear violation of the Principle of Perceptual Self-Exclusion. Furthermore, if E was the perceptum of either A, B, C, or D, then we would have either a digonal, trigonal, tetragonal, or pentagonal perceptual “loop”—a clear violation of the Principle of Perceptual Non-Loopability. 

However, this entails that F is imperceptible—in principle—unless F was the perceptum of an existing, or occurring, perceptual state, G,—a perceptual state that was different from A, B, C, D, E, and F.

Indeed, if F was the perceptum of F, then we would have an existing, or occurring, perceptual state that had itself as its own perceptum—a clear violation of the Principle of Perceptual Self-Exclusion. Furthermore, if F was the perceptum of either A, B, C, D, or E, then we would have either a digonal, trigonal, tetragonal, pentagonal, or hexagonal perceptual “loop”—a clear violation of the Principle of Perceptual Non-Loopability.

And so on, ad infinitum.

We are forced into a vicious, infinite regress. Every new perceptual state being posited (as existing, or occurring) to account for another perceptual state’s imperceptibility would itself be imperceptible—in principle—unless an additional perceptual state was posited (as existing, or occurring); however, this additional perceptual state (posited as existing, or occurring) would itself be imperceptible—in principle—unless a further perceptual state was posited (as existing, or occurring), and so on, ad infinitum. Therefore, since the regress is vicious, it follows that there are some existing, or occurring, perceptual states which are—in principle—imperceptible. However, this means that if all existing, or occurring, perceptual states are identical to some existing, or occurring, brain states, then some existing, or occurring, brain states are—in principle—imperceptible. But, as we saw above, all existing, or occurring, brain states are—in principle—perceptible. Therefore, it follows that some existing, or occurring, perceptual states are not brain states. Since the truth of the Identity Theory rests upon all existing, or occurring, perceptual states being identical to some existing, or occurring, brain states, it follows that the Identity Theory is false.

 

I have provided a diagram illustrating the vicious regress and the two-horned dilemma which follows from it. Each of the circles represent a particular perceptual state (e.g., Perceptual State A, Perceptual State B, Perceptual State C, Perceptual State D, Perceptual State E, and Perceptual State F). If a circle is green, then said perceptual state is the perceptum of another perceptual state (i.e., It is perceived). If a circle is red, then said perceptual state is not the perceptum of another perceptual state (i.e., The perceptual state is imperceptible). A black arrow pointing from one perceptual state to another perceptual state signifies that the former perceptual state is a perception of the latter perceptual state (e.g., The black arrow pointing from Perceptual State A to Perceptual State B signifies that Perceptual State A is a perception of Perceptual State B). A red arrow pointing from one perceptual state to another perceptual state signifies that the former perceptual state is not, and cannot be, a perception of the latter perceptual state—on pain of violating the Principle of Perceptual Self-Exclusion (e.g., The red arrow pointing from Perceptual State B to Perceptual State A signifies that Perceptual State B is not, and cannot be, a perception of Perceptual State A—on pain of violating the Principle of Perceptual Self-Exclusion). Perceptual situations contained in red boxes violate either the Principle of Perceptual Self-Exclusion (e.g., Fig. 1) or the Principle of Perceptual Non-Loopability (e.g., Fig. 3, Fig. 5, and Fig. 7). Perceptual situations contained in green boxes neither violate the Principle of Perceptual Self-Exclusion, nor the Principle of Perceptual Non-Loopability; however, in such perceptual situations, there is always a remaining perceptual state—an imperceptible one—that is left unaccounted for (e.g., Fig. 2, Fig. 4, Fig. 6, and Fig. 8).

Addressing Two Objections

First Objection

“I wondered why one couldn’t perceive the contents of perception at different times. Perhaps when a tree is the perceptum of a perceptual state (brain state B), I could later have a new perceptual state to perceive brain state B. Then all the brain states would still be perceptible in principle.”

My concern with the proposed solution is that it faces the same dilemma that would arise if the perceptual states were simultaneous. Let me try to illustrate the issue:

If an existing, or occurring, perceptual state A was a “perception of a perceptual state, B,” and B had either existed, or occurred, prior to A, then A could have neither been the perceptum of B (since A hadn’t even existed yet to be the perceptum of B), nor could A have been the perceptum of A (without violating the Principle of Perceptual Self-Exclusion). However, it follows from this that A would be imperceptible—in principle—unless it was the perceptum of another existing, or occurring, perceptual state, C,—a perceptual state that was itself different from A and B. Now, it is clear that if C either existed, or occurred, later than A and B, then C could have neither been the perceptum of A, nor the perceptum of B (since C hadn’t even existed yet to be the perceptum of either A or B). However, if C was the perceptum of C, then the Principle of Perceptual Self-Exclusion would have been violated. Thus, C would be imperceptible—in principle—unless it was the perceptum of another existing, or occurring, perceptual state, D,—a perceptual state that was different from A, B, and C. And so on, ad infinitum.

Just as before, every new existing, or occurring, perceptual state needing to be posited to account for another perceptual state’s imperceptibility would itself be imperceptible—in principle—unless another existing, or occurring, perceptual state were posited; however, this existing, or occurring, perceptual state would itself be imperceptible—in principle—unless another existing, or occurring, perceptual state were posited, and so on, ad infinitum. In short, no matter how many perceptual states we posit, we will always get stuck with a “imperceptible” remainder left unaccounted for.

Second Objection

If I follow, the core point is that a perceptual state’s character is determined by its perceptum—the object of the perceptual state (q.v. The Principle of Perceptual Determinacy); and since a perceptual state’s character is “fixed” or rendered determinate by its perceptum, it follows that the perceptum of a perceptual state must have a character independent of that perceptual state. From this it seems to follow that (i) a perceptual state cannot be a “perception of itself” (q.v. The Principle of Perceptual Self-Exclusion), (ii) no two perceptual states can be of one another, and (iii) no n-number (n ≥ 2) of perceptual states can form a series of perceptual states and percepta wherein the first perceptual state in the series is a “perception of the second,” the second is a “perception of the third,” the third is a “perception of the fourth,” and so on—the last perceptual state in the series being a “perception of the first.”  (q.v. The Principle of Perceptual Non-Loopability). However, if the Identity Theory is true, and all perceptual states are identical to some brain states, then it is possible—in principle—for a perceptual state to be a “perception of itself”; indeed, it is possible—in principle—for someone to have a perceptual state of any of their brain states. For example, a person can perceive any of their brain states (a) in a mirror, (b) by having their optic nerves surgically extended—a medical operation which, if not capable of being performed with today’s technology, is by no means incapable of being accomplished in the near future, or even (c) by means of Feigl’s hypothetical autocerebroscope. Furthermore, if the Identity Theory is true, and all perceptual states are identical to some brain states, then it is possible—in principle—for two perceptual states to be of one another; indeed, it is possible—in principle—for two craniotomy patients to be positioned in such a way so that they perceive one another’s brain states simultaneously. Likewise, if the Identity Theory is true, then it is possible—in principle—for n-number (n ≥ 2) of perceptual states to form a series of perceptual states and percepta wherein the first perceptual state in the series is a “perception of the second,” the second is a “perception of the third,” the third is a “perception of the fourth,” and so on—the last perceptual state in the series being a “perception of the first.” Indeed, like before with two craniotomy patients, it is possible—in principle—for n-number (n ≥ 2) of craniotomy patients to be arranged in such a way that the first patient in the group perceives any one of the brain states of a second patient, the second perceives any one of the brain states of the third, the third perceives any one of the brain states of the fourth, and so on—the last patient in the group perceiving any one of the first patient’s brain states. Therefore, the Identity Theory is false. It seems that the best reply for an Identity Theorist would be to say that it is not true that a perceptual state’s character is determined by its perceptum. Rather, a perceptual state represents an object/character that may or may not exist. And so, there is no problem with a perceptual state being of itself, any more than there is a problem with other cases of reflective representation (e.g., the novel, Don Quixote, is, in part, about the novel, Don Quixote).”

There appear to be a couple of difficulties entailed by this suggested response. My first concern is that there seems to be a serious dilemma when it comes to the self-representation involved in a perceptual state having itself as its own perceptum. Coincidentally, A.E. Taylor’s criticism of Josiah Royce’s conception of the Absolute as a self-representing system parallels the issues involved in the actual implementation of a perceptual state that had itself as its own perceptum:

“Professor Royce’s own illustration of the map of England executed upon a portion of the surface of the country is really a typical instance of a self-contradictory purpose. He argues that such a map, to be theoretically perfect, must contain a reduced facsimile of itself as part of the country mapped, and this again another, and so on indefinitely. But the whole force of the reasoning depends on overlooking the distinction between the surface of England as it is before the map is made, and the surface of England as altered by the presence of the map. Prof. Royce assumes that you set out to represent in the map a state of things which can in fact have no existence until after the map is made. The previous existence of the map at a certain spot is falsely taken to be one of the conditions to which the map-maker is to conform in executing it. Every one of the supposed “maps within the map” will thus involve distortion and misrepresentation of the district it proposes to map. It is as if Hamlet had chosen “Hamlet” as the subject of the “play within the play.” The professor’s illustration thus does less than justice to his theory.”¹¹

My second concern is that the proposed objection (i.e., that no existing, or occurring, perceptual state has any of its characteristics determined by the characteristics of its perceptum) ends up vindicating the Principle of Perceptual Self-Exclusion—the very principle the objection sought to dissolve. The self-undermining character of the proposed objection can be made more explicit with the help from the following argument (an argument inspired by J.M.E. McTaggart’s conception of “Intrinsic Determination”¹² as a relation between characteristics corresponding to (the relation of) implication between propositions):

Let X represent the characteristic of “having the characteristic of ‘neither existing, nor occurring, if perceptual state, P, neither exists, nor occurs’.”  

Let Y represent the characteristic of “neither existing, nor occurring, if perceptual state, P, neither exists, nor occurs.” 

Now, the occurrence of X as a characteristic in a given thing “intrinsically determines” the occurrence of Y as a characteristic in a given thing because the proposition “something has the characteristic X” implies the proposition “something has the characteristic Y.” As a matter of fact, the occurrence of X as a characteristic in a given thing determines the occurrence of Y as a characteristic in that very same thing which possesses X as a characteristic.  

Now, if an existing, or occurring, perceptual state, P, has none of its characteristics determined by any of the characteristics of its perceptum, then it immediately follows that no existing, or occurring, perceptual state, P, has itself as its own perceptum. This can be demonstrated in the following manner: 

If an existing, or occurring, perceptual state, P, had itself as its own perceptum, then P’s perceptum would have the characteristic of “having the characteristic of ‘neither existing, nor occurring, if P neither exists, nor occurs’.” However, the “having” of this characteristic by P’s perceptum would thereby determine the occurrence of the characteristic of “neither existing, nor occurring, if P neither exists, nor occurs,” in P’s perceptum. However, ex hypothesi, P’s perceptum is P itself. Thus, X being a characteristic of P’s perceptum would intrinsically determine Y as a characteristic of P. In other words, if no existing, or occurring, perceptual state, P, has any of its characteristics determined by the characteristics of its perceptum, it follows that no existing, or occurring, perceptual state, P, has itself as its own perceptum. And since the Principle of Perceptual Self-Exclusion holds that no existing, or occurring, perceptual state, P, has itself as its own perceptum, it follows that the Principle of Perceptual Self-Exclusion is true.

The proposed objection to our argument (i.e., that no existing or occurring perceptual state has any of its characteristics determined by the characteristics of its perceptum) has thus ended up vindicating the Principle of Perceptual Self-Exclusion.

Concluding Remarks

“I am not what I see, / And other than the things I touch.”⁹

If the preceding arguments are sound, then the Identity Theory of Mind has been refuted—not by an appeal to the apparent qualitative character of our subjective life, but through an exploration of the necessary conditions underlying the existence and occurrence of perceptual states and their percepta.

Footnotes

^[1] Arthur Schopenhauer, The World as Will and Idea, trans. Richard Burdon Haldane and John Kemp, vol. I, III vols. (London, England: Trübner & Co., 1883), 5.

^[2] Patrick Olivelle, ed., “Bṛhadāraṇyaka Upaniṣad,” in The Early Upaniṣads: Annotated Text and Translation, trans. Patrick Olivelle (New York, New York: Oxford University Press, 1998), 29–165, 3.7.23.

^[3] John Jamieson Carswell Smart, “The Mind/Brain Identity Theory,” in The Stanford Encyclopedia of Philosophy, ed. Edward Nouri Zalta and Uri Nodelman, Winter 2022 Edition (Stanford, CA: Metaphysics Research Lab, Stanford University, 2022).

^[4] Charlie Dunbar Broad, Perception, Physics, and Reality: An Enquiry Into the Information That Physical Science Can Supply about the Real (London, England: Cambridge University Press, 1914), 5.

^[5] The italicized “is” should be interpreted as the “is” of identity. Thus, for any perceptual state, P, P is identical to a “perception of a perceptum, O.”

^[6] Just as every perceptual state has a correlative perceptum, so too does every perceptum have a corresponding perceptual state. Thus, for any perceptum, O, there is a particular “somewhat” which is identical to the “perception of” that perceptum, O.

^[7] It is vital to insist upon two important points. Firstly, it is for a perceptual state’s perceptum needn’t be perceived accurately for it to be the perceptum of the aforementioned perception state. Suppose that (i) a perceptual state, A, is a “perception of a “somewhat” as having property f,” and (ii) there just so happens to be no existent, or occurrent, “somewhat,” that has property f. Although A’s perceptum is perceived as having properties that it does not, in fact, have (i.e., the property f), it does not follow that A’s perceptum (i.e., the particular “somewhat” which is erroneously perceived as having property f) neither exists, nor occurs, nor is determinate. On the contrary, just as perceptual state A exists, or occurs, and is determinate, A’s perceptum exists, or occurs, and is determinate—the perceptum is simply misperceived or erroneously perceived. Secondly, one may broaden the definition of “perceptum” so that it would be the following: “For any existent, or occurrent, perceptual state, P, a perceptum is whatever serves to render P existent (or occurrent) and determinate as “this” perceptual token rather than indeterminate nothing. In cases of veridical perception, the perceptum of P would be something in the percipient subject’s environment. In cases of illusion or hallucination, the role (of perceptum) is still filled (albeit not necessarily by a worldly object external to the subject’s body), but something must serve to determine P as of something. Moreover, a perceptual state is determinate if and only if it possesses its individuating character sufficient to (type-)identify it as “this” very perceptual (token).

^[8] Arthur Schopenhauer, Critical Debates: (1809 – 1818), ed. Arthur Hübscher, trans. Eric Francis John Payne, vol. II, IV vols., of Manuscript Remains (Oxford: Berg, 1988), 349.

^[9] Charles Henry Rieber, Footnotes to Formal Logic (Berkeley, California: University of California Press, 1918), 10.

^[10] Now, “perceptible-in-principle” and “imperceptible-in-principle,” like “knowable-in-principle,” are, in fact, genuine or “extensional” properties, rather than merely “intensional” properties. The fact that “perceptible-in-principle” and “imperceptible-in-principle” are extensional properties renders the preceding argument immune from accusations of committing a “masked-man” or “intensional” fallacy.

^[11] Alfred Edward Taylor, Elements of Metaphysics, 12th ed. (London, England: Methuen & Co. Ltd, 1961), 150.

^[12] Cf. John McTaggart Ellis McTaggart, The Nature of Existence, vol. I, II vols. (Cambridge, England: Cambridge University Press, 1921), 111.

^[13] Alfred Tennyson, In Memoriam (London, United Kingdom: Henry S. King & Co., 1876), sec. 45, lines 7–8.

Bibliography

Broad, Charlie Dunbar. Perception, Physics, and Reality: An Enquiry Into the Information that Physical Science Can Supply about the Real. London, England: Cambridge University Press, 1914.

McTaggart, John McTaggart Ellis. The Nature of Existence. Vol. I. II vols. Cambridge, England: Cambridge University Press, 1921.

Olivelle, Patrick, ed. “Bṛhadāraṇyaka Upaniṣad.” In The Early Upaniṣads: Annotated Text and Translation, 29–165. New York, New York: Oxford University Press, 1998.

Rieber, Charles Henry. Footnotes to Formal Logic. Berkeley, California: University of California Press, 1918.

Schopenhauer, Arthur. Critical Debates: (1809 – 1818). Edited by Arthur Hübscher. Translated by Eric Francis John Payne. Vol. II. IV vols. of Manuscript Remains. Oxford: Berg, 1988.

Schopenhauer, Arthur. The World as Will and Idea. Translated by Richard Burdon Haldane and John Kemp. Vol. I. III vols. London, England: Trübner & Co., 1883.

Smart, James Jamieson Carswell. “The Mind/Brain Identity Theory.” Edited by Uri Nodelman and Edward Nouri Zalta. Stanford Encyclopedia of Philosophy, Winter 2022. https://plato.stanford.edu/archives/win2022/entries/mind-identity/.

Taylor, Alfred Edward. Elements of Metaphysics. 12th ed. London, England: Methuen & Co. Ltd, 1961.

Tennyson, Alfred. In Memoriam. London, United Kingdom: Henry S. King & Co., 1876. 

A Critical Examination of Illusionism

Whether or not Illusionism is growing in acceptance, I do not know; however, it does appear that Illusionism is receiving more and more attention in the contemporary philosophical scene. The following essay is a brief analysis of an assumption that lies at the heart of Illusionism. When discussing their views, Illusionists usually invoke the terms “qualia,” “phenomenal qualities,” “phenomenal characters,” “phenomenal feels,” “phenomenal properties,” “what-it’s-likenesses” and “sensory-qualities.” For the sake of clarity and uniformity, I am going to use “sensory-qualities” as a catch-all for these terms. The “programme and platform” of Illusionism is summarized in the following passages by its most active proponent, Keith Frankish:

“Illusionists believe that consciousness [and/or objects of consciousness] involve no properties that are not detectable and fully describable by third-person science. Any other properties we think are involved are illusory….[Sensory-qualities] are illusory.” (Frankish, Illusion or Identity, 1-2)

“[Illusionism] holds that [sensory-qualities are] an illusion and aims to explain why it seems [that sensory-qualities] exist….I shall refer to the problem of explaining why experiences [and/or objects of experiences] seem to have [sensory-qualities] as the illusion problem….Illusionists deny that experiences [and/or objects of experiences] have [sensory-qualities] and focus on explaining why they seem to have them. They typically allow that we are introspectively aware of our sensory states but argue that this awareness is partial and distorted, leading us to misrepresent the states [and/or objects of those states] as having [sensory-qualities].” (Frankish, Illusionism as a Theory of Consciousness, 1-3)

Nothing—whether it be an object or an experience—has sensory-qualities. If an object (or experience) seems to have sensory-qualities, it is because our brain’s introspective mechanisms have “misrepresented” the object (or experience) as having them. So much then for Illusionism.

In defending Illusionism, Keith Frankish makes heavy use of “representations.” But what is a “representation”? In cognitive science, a representation is understood as being a structure that stands for something by virtue of relations such as similarity, causal history, and connections with other representations. Frankish uses the term “representation” in a similar way:

“First, how do I use the term “representation”? I use it in its standard sense in cognitive science for a brain state (a pattern of neuron firing) that serves as a proxy for something else, producing psychological effects appropriate to the presence of the thing....[They are] part of the neural machinery that makes us aware of the things they represent.” (Frankish, Illusionism and its Place in Contemporary Philosophy of Mind, 304)

I’ll accept this account of “representations” for the sake of argument. 

Frankish’s Illusionism cannot get off the ground unless it can establish the truth or plausibility of the following proposition: There exist “representations of sensory-qualities” (e.g., a “representation of a red sensory-quality”) that not only do not have the particular determinate (or determinable) sensory-qualities they represent, but are themselves totally devoid of determinate (or determinable) sensory-qualities of any determinable kind. This is a very strong claim, and Frankish must adequately defend it without recourse to faith, intuition, or revelation. This is how Frankish defends Illusionism’s foundation stone: 

“It is true that cognitive scientists talk of there being representations in the brain. Representations in this sense are...parts of the machinery that makes us aware of things…Such representations need not share properties with the things they represent. The mental representation of redness need not be red, any more than the word ‘red’ or the numeral that denotes red in a painting-by-numbers kit…These representations [of sensory-qualities] don’t have [sensory-qualities], of course…Rather, they are patterns of neuron-firing that signal the presence of [sensory-qualities] in much the same way that written words signal the presence of things.” (Frankish, What if your Consciousness is an Illusion Created by your Brain?, 5)¹

“When illusionists say that [sensory-qualities] are illusory, they mean that we have introspective representations like those that we would have if our experiences had [sensory-qualities]. And we can have such representations even if our experiences don’t have [sensory-qualities]. Of course, [Illusionism requires] that the [introspective] representations [of sensory-qualities] don’t have [sensory-qualities]. But, as I noted, representations needn’t possess the properties they represent. Representations of redness needn’t be red, and representations of [sensory-qualities] needn’t [have sensory-qualities].” (Frankish, What if your Consciousness is an Illusion Created by your Brain?, 10) 

“A representation of a [sensory-quality] needn’t be qualitative, just as the word ‘red’ needn’t be red.” (Frankish, Twitter/X, Sep 27th, 2019)

“Representations needn’t possess the properties they represent. A representation of redness needn’t be red, and a representation of phenomenal redness needn’t be phenomenally red.” (Frankish, Twitter/X, Jan 2^nd, 2020)

“It implies some representation of the illusory thing—[sensory-qualities] in this case. But that representation needn’t itself have the properties of the illusory thing. A representation of [sensory-qualities] needn’t itself have [sensory-qualities].” (Frankish, Twitter/X, Feb 8^th, 2021)

“A representation of F needn’t itself be F. A representation of a square needn’t be a square, a representation of worldly green needn’t be green, and a representation of phenomenal green needn’t be phenomenally green.” (Frankish, Twitter/X, Mar 15^th, 2018)

Frankish’s line of reasoning can be summarized as follows:

P1) If there have been instances where we have observed “representations of  particular determinate (or determinable) sensory-qualities” as not having the particular determinate (or determinable) sensory-qualities they represent, then “representations of particular determinate (or determinable) sensory-qualities” can exist without needing to have the particular determinate (or determinable) sensory-qualities they represent. 

P2) If  “representations of particular determinate (or determinable) sensory-qualities” can exist without needing to have the particular determinate (or determinable) sensory-qualities they represent, then there can exist “representations of particular determinate (or determinable) sensory-qualities” that not only do not have the particular determinate (or determinable) sensory-qualities they represent, but are also themselves totally devoid of determinate (or determinable) sensory-qualities of any determinable kind. 

C1) Therefore, if there have been instances where we have observed “representations of  particular determinate (or determinable) sensory-qualities” as not having the particular determinate (or determinable) sensory-qualities they represent, then there can exist “representations of particular determinate (or determinable) sensory-qualities” that not only do not have the particular determinate (or determinable) sensory-qualities they represent, but are also themselves totally devoid of determinate (or determinable) sensory-qualities of any determinable kind. [From P1—P2] 

P3) There have been instances where we have observed “representations of  particular determinate (or determinable) sensory-qualities” as not having the particular determinate (or determinable) sensory-qualities they represent. 

C2) Therefore, there can exist “representations of particular determinate (or determinable) sensory-qualities” that not only do not have the particular determinate (or determinable) sensory-qualities they represent, but are also themselves totally devoid of determinate (or determinable) sensory-qualities of any determinable kind. [From P1—C1, P3]

Such reasoning is clearly fallacious. The consequent in P2 does not at all follow from the antecedent. Indeed, it is a complete non sequitur—as well as an example of hasty generalization. Before I dive into a thorough analysis of Frankish’s reasoning, I think it will be most helpful if we do a bit of phenomenology. 

In many (if not all) of our observational activities, we direct our attention to an object, and this object is immersed within a multeity of objects. We observe the object of our attention as having or as consisting of determinate (or determinable) sensory-qualities of one or more determinable kinds; likewise, the multeity of objects (in the periphery) are observed as having or as consisting of determinate (or determinable) sensory-qualities of one or more determinable kinds. Not only do these sensory-qualities constitute the “what” (i.e., the “content” or characteristica) we appeal to whenever we say that This observed (or observable) object is different from That observed (or observable) object, they also differentiate the object of our attention from the multeity of objects (in the periphery) and the multeity of objects (in the periphery) from the object of our attention. Objects (“representations” included) totally devoid of sensory-qualities would be nothing to us (observationally speaking); it is only by means of their apparent sensory-qualities that said observable objects have actual or possible significance for us.^2,3,4 Regardless of whether or not the sensory-qualities that the objects are observed as having or as consisting of are “real,” it is clear that it is only by means of them that we are able to  single out an object (in observation) from a field of objects—singling it out as a particular This among other Thises.

Suppose that we were to observe an object, X, and X is a “representation of a red sensory-quality.” Now, it does not necessarily follow that we will observe X as having or consisting of the particular determinate (or determinable) sensory-quality which it represents (i.e., the red sensory-quality); indeed, we may observe X as having or consisting of determinate (or determinable) sensory-qualities that do not even fall under the heading of “color”—e.g., X may be observed as having or consisting of determinate (or determinable) sensory-qualities of a tactile, gustatory, auditory, or olfactory kind. Nevertheless—and this point is essential—we have never observed objects (“representations” included) as neither having nor consisting of sensory-qualities; on the contrary, we have always observed objects (“representations” included) as having or consisting of determinate (or determinable) sensory-qualities of one or more determinable kinds.⁴ The following scenario will help make this point more concrete:

Suppose that John is observing a cup on a table, and he observes the cup as having a determinate (or determinable) blue sensory-quality. Whenever John observes the cup as having a blue sensory-quality, he has a “representation of the cup as having a blue sensory-quality.” John’s “representation of the cup as having a blue sensory-quality” is localized to a particular chunk or region of his brain, and we can observe this particular chunk or region of John’s brain whenever he observes the cup as having a blue sensory-quality. Now, if we were to observe John’s “representation of the cup as having a blue sensory-quality” (i.e., a particular chunk or region of John’s brain) as he observes the cup as having a blue sensory-quality, it is almost certain that we will not observe it as having a blue sensory-quality; however, we can be certain that we would observe it as having or consisting of determinate (or determinable) sensory-qualities of one or more determinable kinds; indeed, we would observe it as having or consisting of sensory-qualities, e.g., pinkness, warmth, smoothness, sliminess, etc. Although John’s “representation of the cup as having a blue sensory-quality” is not observed as having a blue sensory-quality, it is nevertheless observed as having or consisting of determinate (or determinable) sensory-qualities of one or more determinable kinds. Indeed, both the cup (i.e., the cup which John observes as having a blue sensory-quality) and John’s “representation of the cup as having a blue sensory-quality” are both observed as having or consisting of determinate (or determinable) sensory-qualities of one or more determinable kinds. In light of this illustration, what evidence is there for the following cornerstone of Illusionism:

There can exist “representations of particular determinate (or determinable) sensory-qualities” that not only do not have the particular determinate (or determinable) sensory-qualities they represent, but are also themselves totally devoid of determinate (or determinable) sensory-qualities of any determinable kind.^{5, 6}

If Illusionism is to be held to its own standards, it cannot be allowed to defend the above claim by appealing to ‘intuitions’. Illusionism must appeal to the ‘empirical’. But what ‘empirical’ evidence is there? Or, rather, what is sample of available evidence is there for the Illusionist to appeal to in order to support the above-mentioned premise? (Whether or not the evidence does in fact support it). The following:

• We have had many instances of observation wherein we have observed a “representation of a particular sensory-quality” (e.g., a representation of a red sensory-quality) as not having the particular determinate (or determinable) sensory-quality it represents.

• We have never had any instances of observation wherein a “representation of a particular sensory-quality” (e.g., a representation of a red sensory-quality) is observed as being totally devoid of determinate (or determinable) sensory-qualities of any determinable kind.

• We have always observed objects (“representations” included) as having or consisting of determinate (or determinable) sensory-qualities of one or more determinable kinds.

• We have never had any instances of observation wherein the observed objects (“representations” included) are observed as neither having nor consisting of determinate (or determinable) sensory-qualities of one or more determinable kinds.

Does the ‘empirical’ evidence support the consequent featured in P2 of Illusionism’s argument? Indeed, based on the bullet-points, can the Illusionist he validly conclude that there can exist (let alone that there does exist) “representations of particular determinate (or determinable) sensory-qualities” that not only do not have the particular determinate (or determinable) sensory-qualities they represent, but are also themselves totally devoid of determinate (or determinable) sensory-qualities of any determinable? No, he cannot do so. From these facts, it is clear that the Illusionist cannot validly infer (or posit) that there can exist (let alone that there does exist) “representations of particular determinate (or determinable) sensory-qualities” that not only do not have the particular determinate (or determinable) sensory-qualities they represent, but are also themselves totally devoid of determinate (or determinable) sensory-qualities of any determinable kind from instances of empirical observation.

I'll conclude the first part of this shrill shout of mine with a brief argument:

P1) If we cannot validly infer (or empirically verify) that “there can exist “representations of particular determinate (or determinable) sensory-qualities” that not only do not have the particular determinate (or determinable) sensory-qualities they represent, but are also themselves totally devoid of determinate (or determinable) sensory-qualities of any determinable kind” from the fact of there being instances where we have observed “representations of  particular determinate (or determinable) sensory-qualities” as not having the particular determinate (or determinable) sensory-qualities they represent,” then we are not justified in believing an essential premise in all arguments for Illusionism.

P2) If we are not justified in believing an essential premise in all arguments for Illusionism, then we are not justified in believing Illusionism to be true. 

C1) Therefore, if we cannot validly infer (or empirically verify) that “there can exist “representations of particular determinate (or determinable) sensory-qualities” that not only do not have the particular determinate (or determinable) sensory-qualities they represent, but are also themselves totally devoid of determinate (or determinable) sensory-qualities of any determinable kind” from the fact of there being instances where we have observed “representations of  particular determinate (or determinable) sensory-qualities” as not having the particular determinate (or determinable) sensory-qualities they represent,” then we are not justified in believing Illusionism to be true.  [From P1 and P2]  

P3) We cannot validly infer (or empirically verify) that “there can exist “representations of particular determinate (or determinable) sensory-qualities” that not only do not have the particular determinate (or determinable) sensory-qualities they represent, but are also themselves totally devoid of determinate (or determinable) sensory-qualities of any determinable kind” from the fact of there being instances where we have observed “representations of  particular determinate (or determinable) sensory-qualities” as not having the particular determinate (or determinable) sensory-qualities they represent.” 

C2) Therefore, we are not justified in believing Illusionism to be true. [From P1 and P2] 

Thus, Illusionism’s foundation-stone is an arbitrary and unwarranted assumption. Far from being a viable option, the “coherent research programme” of Illusionism has the structural integrity of a castle in the air...

Addressing a Possible Objection:

I will attempt to address a potential objection, but before doing so, I think it would be best to outline the Illusionist’s argument in full and highlight the fallacy at its heart:

P1) If there have been instances where we have observed “representations of  particular determinate (or determinable) sensory-qualities” as not having the particular determinate (or determinable) sensory-qualities they represent, then “representations of particular determinate (or determinable) sensory-qualities” can exist without needing to have the particular determinate (or determinable) sensory-qualities they represent. 

P2) If  “representations of particular determinate (or determinable) sensory-qualities” can exist without needing to have the particular determinate (or determinable) sensory-qualities they represent, then there can exist “representations of particular determinate (or determinable) sensory-qualities” that not only do not have the particular determinate (or determinable) sensory-qualities they represent, but are also themselves totally devoid of determinate (or determinable) sensory-qualities of any determinable kind. 

C1) Therefore, if there have been instances where we have observed “representations of  particular determinate (or determinable) sensory-qualities” as not having the particular determinate (or determinable) sensory-qualities they represent, then there can exist “representations of particular determinate (or determinable) sensory-qualities” that not only do not have the particular determinate (or determinable) sensory-qualities they represent, but are also themselves totally devoid of determinate (or determinable) sensory-qualities of any determinable kind. [From P1 and P2]

P3) If there can exist “representations of particular determinate (or determinable) sensory-qualities” that not only do not have the particular determinate (or determinable) sensory-qualities they represent, but are also themselves totally devoid of determinate (or determinable) sensory-qualities of any determinable kind, then it is possible for us to “misrepresent” objects (“representations” included) as having or consisting of sensory-qualities even though sensory-qualities do not exist and are not instantiated in the world (i.e., sensory-qualities only “seem” to exist and only “seem” to be instantiated in the world). 

C2) Therefore, if there can exist “representations of particular determinate (or determinable) sensory-qualities” that not only do not have the particular determinate (or determinable) sensory-qualities they represent, but are also themselves totally devoid of determinate (or determinable) sensory-qualities of any determinable kind, then it is possible that sensory-qualities do not exist (i.e., it is possible that they are not instantiated in the world) even though we observe objects (“representations” included) as having or consisting of sensory-qualities. [From C1 and P3] 

P4) If it is possible that sensory-qualities do not exist (i.e., if it is possible that they are not instantiated in the world) even though we observe objects (“representations” included) as having or consisting of sensory-qualities, then it is possible that Illusionism is true. 

C3) Therefore, if there can exist “representations of particular determinate (or determinable) sensory-qualities” that not only do not have the particular determinate (or determinable) sensory-qualities they represent, but are also themselves totally devoid of determinate (or determinable) sensory-qualities of any determinable kind, then it is possible that Illusionism is true. [From C2 and P4] 

P5) There have been instances where we have observed “representations of  particular determinate (or determinable) sensory-qualities” as not having the particular determinate (or determinable) sensory-qualities they represent.  

C4) Therefore, it is possible that Illusionism is true. [From P1—C3, P5]

Now that we have standardized the Illusionist’s argument, we can examine it in detail. The move from P1 to P2 is fallacious—it is, as I have said, a non sequitur and involves an instance of hasty generalization (among other fallacies). As I have shown, the Illusionist cannot validly infer (or posit) that there can exist (let alone that there does exist) “representations of particular determinate (or determinable) sensory-qualities” that not only do not have the particular determinate (or determinable) sensory-qualities they represent, but are also themselves totally devoid of determinate (or determinable) sensory-qualities of any determinable kind, from the fact that we have had many instances of observation wherein we have observed a “representation of a particular sensory-quality” (e.g., a representation of a red sensory-quality) as not having the particular determinate (or determinable) sensory-quality it represents; the reason being that we have always observed objects (“representations” included) as having or consisting of determinate (or determinable) sensory-qualities of one or more determinable kinds. 

Now, throughout his work, Frankish continually addresses common criticisms of Illusionism. I think it would be helpful to include one of these criticisms (and Frankish’s response) here: 

“Another common objection to Illusionism is that in the case of [sensory-qualities] there is no gap between illusion and reality. [According to this objection] something can look like a Penrose triangle without being a Penrose triangle, but an experience that seems to have a greenish [sensory-quality] really does have a greenish [sensory-quality]….This is often presented as a crushing objection to Illusionism, but it is far from compelling. It turns on what we mean by [an experience] seeming to have a greenish [sensory-quality]. If we mean having an introspective experience with the same [sensory-quality] as [an experience that has] a greenish [sensory-quality], then, trivially, there is no distinction between seeming and reality. But of course this is not what Illusionists mean. [Illusionists] mean introspectively representing oneself as having [an experience with] a greenish [sensory-quality], and one can do this without having [an experience with] a greenish [sensory-quality]. The objector may reply that, in order to create the illusion of [an experience with] a greenish [sensory-quality], the introspective representation would have to employ a greenish [sensory-quality] mode of presentation, which would itself have an introspectable greenish [sensory-quality]. However, Illusionists will simply deny this, arguing that the content of introspective representations is determined by [non-sensory-qualities], causal or functional factors. [Footnote: Alternatively, Illusionists might concede that introspection employs modes of presentation that appear to have [sensory-qualities], but argue that this too is an illusion—that introspection misrepresents the modes of presentation as having [sensory-qualities] they lack. For a defense of this option and an argument that it does not generate an infinite regress, see Pereboom (Consciousness and the Prospects of Physicalism, 27-28)]….The Illusionist claims that when we think we are having [an experience with] a greenish [sensory-quality (or when we think we are observing an object with a greenish sensory-quality] we are in fact merely misrepresenting ourselves as having [an experience with a greenish sensory-quality (or misrepresenting ourselves as observing an object with a greenish sensory-quality)]. This claim may be false, but the no-gap objection does not anything to the case against it.” (Frankish, Illusionism as a Theory of Consciousness, 16-17)

In light of Frankish’s response to his hypothetical objector, suppose Frankish was to reply to our argument by saying:

“John’s “representation of the blue sensory-quality” and my “representation of John’s “representation of the blue sensory-quality” as having sensory-qualities” can both be totally devoid of determinate (or determinable) sensory-qualities of any determinable kind, and still represent their respective content as having or consisting of sensory-qualities because the content of a representation is fixed by the representation’s functional and relational features (i.e., non-sensory-qualities) and not intrinsic ones (e.g., sensory-qualities—if they existed).”

Such a reply would beg the question at issue because it presupposes that there can exist “representations of particular determinate (or determinable) sensory-qualities” that not only do not have the particular determinate (or determinable) sensory-qualities they represent, but are also themselves totally devoid of determinate (or determinable) sensory-qualities of any determinable kind—the very thing being disputed; and, as I have shown, the Illusionist has yet to present a valid argument for such a premise. Furthermore, he also begs the question because this premise plays an essential part in the Illusionist’s reasons (and attempted rationalization) for his denial of the existence of sensory-qualities. Unless the Illusionist can establish the premise, he can neither defend his theory, nor his denial of the existence of sensory-qualities, by invoking ideas that presuppose the premise’s truth (i.e., ideas which rest upon the very premise that the Illusionist has yet to establish or render plausible).

Furthermore, the response would be a red herring. It is beside the point to say that “the content of a representation is fixed by the functional and relational features (i.e., non-sensory-qualities) of the representation and not intrinsic ones (e.g., sensory-qualities—if they existed),” because my criticism does not concern how introspective “representations of particular determinate (or determinable) sensory-qualities” get their content “fixed” in the absence of any determinate (or determinable) sensory-qualities being instantiated in the world; rather, my criticism targets Illusionist’s fallacious, non-sequiturial inference from instances where we have observed “representations of  particular determinate (or determinable) sensory-qualities” as not having the particular determinate (or determinable) sensory-qualities they represent, to the conclusion that there can exist (or that there does exist)  “representations of particular determinate (or determinable) sensory-qualities” that not only do not have the particular determinate (or determinable) sensory-qualities they represent, but are also themselves totally devoid of determinate (or determinable) sensory-qualities of any determinable kind. Without this premise, Illusionism cannot get off the ground. 

Secondly, it may be true that the content of a representation is fixed by functional and relational features (i.e., non-sensory-qualities), but that doesn’t change the fact that the Illusionist has yet to prove that there can exist (or that there does exist) “representations of particular determinate (or determinable) sensory-qualities” that not only do not have the particular determinate (or determinable) sensory-qualities they represent, but are also themselves totally devoid of determinate (or determinable) sensory-qualities of any determinable kind. Until the Illusionist is able to provide a valid and sound argument for that premise, Illusionism succumbs to “vicious circle” and “vicious regress” objections.

But what are these “vicious circle” and “vicious regress” objections? They take the following form:

“Many people think of illusions as involving [sensory-qualities]….If [sensory-qualities] were an illusion in that sense, then Illusionism would be circular. There would have to be another set of [sensory-qualities] in which the [sensory-quality] illusion was rendered, and we’d be no better off.” (Frankish, Illusionism and its Place in Contemporary Philosophy of Mind, 301-302) (Cf. Frankish, Like a Rainbow, 1)

In other words, if the “introspective representations” which the Illusionist infers (or posits)—i.e., “representations of experiences/objects as having or consisting of particular determinate (or determinable) sensory-qualities”—had or consisted of determinate (or determinable) sensory-qualities of one or more determinable kinds (regardless of whether or not these inferred or (posited) representations had or consisted of, the particular determinate (or determinable) sensory-qualities they represent), then Illusionism would be viciously circular. In order to avoid falling into a vicious circle or a vicious regress, the Illusionist must infer (or posit) the existence of introspective “representations of particular determinate (or determinable) sensory-qualities” that not only do not have the particular determinate (or determinable) sensory-qualities they represent, but are also themselves totally devoid of determinate (or determinable) sensory-qualities of any determinable kind. Yet, as I have shown, such an inference (or positing) cannot be made (without committing several fallacies). The Illusionist cannot appeal to the premise as being an ‘open possibility’ (i.e., that it is an ‘open possibility’ that there can exist (or that there does exist) “representations of particular determinate (or determinable) sensory-qualities” that not only do not have the particular determinate (or determinable) sensory-qualities they represent, but are also themselves totally devoid of determinate (or determinable) sensory-qualities of any determinable kind). On the contrary, it does not tally (epistemically speaking) with what we have observed and with what we currently observe. The only kinds of representations evidenced by ‘empirical’ observation (i.e., the only available and relevant sample of representations that the Illusionist can validly infer (or posit)—without fallacy) are “representations of  particular determinate (or determinable) sensory-qualities” that have been perceived as having or consisting of determinate (or determinable) sensory-qualities of one or more determinable kinds—i.e., instances that do not permit the Illusionist to infer (or posit) “representations of particular determinate (or determinable) sensory-qualities” that not only do not have the particular determinate (or determinable) sensory-qualities they represent, but are also themselves totally devoid of determinate (or determinable) sensory-qualities of any determinable kind).

Footnotes:

^[1] Contra the Early Wittgenstein, who holds that X cannot be a “representation of Y,” unless X has “something in common” (e.g., a “shared” logical form) with Y. Thus: “The picture is a model of reality.” (Tractatus Logico-Philosophicus, 2.12) “The picture consists in the fact that its elements are combined with one in a definite way.” (2.14) “The picture is a fact.” (2.141) “That the elements of the picture are combined with one another in a definite way, represents that the things are some combined with one another.” (2.15) “According to this view the representing relation which makes it a picture, also belongs to the picture.” (2.1513) “The representing relation consists of the coordinations of the elements of the picture and the things.” (2.1514) “In order to be a picture a fact must have something in common with what it pictures.” (2.16) “In the picture and the pictured there must be something identical in order that the one can be a picture of the other at all.” (2.161) “What the picture must have in common with reality in order to be able to represent it after its manner—rightly or falsely—is its form of representation.” (2.17) “The picture can represent every reality whose form it has. The spatial picture, everything spatial, the coloured, everything coloured, etc.” (2.171) “The picture has the logical form of representation in common with what it pictures.” (2.2) “The gramophone record, the musical thought, the score, the waves of sound, all stand to one another in that pictorial internal relation, which holds between language and the world. To all of them the logical structure is common. (Like the two youths, their two horses and their lilies in the story. They are all in a certain sense one).” (4.014) We find something similar in Nettleship’s Lectures on Logic: “In the first place, we must ask what is a sign or symbol? To say that A is a symbol of B implies, of course, that A is something different from B, but (and this is often forgotten) it also implies that it is in some respect the same as B. The most remote, far-fetched symbol in the world must have something in common with that which it symbolizes; i.e., the person to whom it has the symbolic meaning must have some (however little it may be) of the same feeling or experience when he experiences the symbol, as he has when he experiences that which it symbolizes. In other words, there are such things as symbols just because the most different things in the world have something in common.” (Nettleship, Philosophical Lectures and Remains, Vol. I, 23-24)

^[2] Indeed, in another article, Frankish himself agrees when he says, “When I look at a ripe banana, I seem to be aware of a rich yellow [sensory-quality]….Similarly, with smells, sounds, tastes, pains, and other sensory-qualities.” Likewise, he says, “I agree…that it seems as if [objects of] experiences have [sensory-qualities].” (Frankish, Why Panpsychism Fails to Solve the Mystery of Consciousness, 8). However, he goes on to say, “These [sensory-qualities] are not more present in the world around us than they are in our brains.” (Frankish, What is Illusionism?, 12). 

^[3] In a similar vein, he says, “We experience objects as having a distinctive but ineffable significance for us, which is a marker of their affordances. This is what we call their quality or feel.” (Frankish, Bright Shiny Colours, 1) And again, “As subjects of experience, our interest is in how things seem to us introspectively—the illusion [of sensory-qualities] itself, not the mechanisms that cause it.” (Frankish, Illusionism as a Theory of Consciousness, 8) And yet again, “We perceive things as having a certain impact on us, as infused with potent [sensory-qualities]….[Sensory-qualities] are not really anywhere. Thanks to the combination of perceptual and introspective processes in our brains, we experience objects as qualitatively charged. But the [sensory-qualities] don’t correspond to real, intrinsic features either of objects or us. They are potent illusions, an imperative commentary, generated by our brains to express the significance things have for us.” (Frankish, Illusionism and its Place in Contemporary Philosophy of Mind, 304)

^[4] In a different context, Frankish discusses this when he describes possible variations of Illusionism (e.g., inward-looking Illusionism and outward-looking Illusionism): “I characterized Illusionism as the view that [sensory-qualities are] an introspective illusion, reflecting the widely held view that [sensory-qualities] are properties of experience. This may be too restrictive, however. Some theorists hold that experience is transparent: when we attend to our experiences, we are aware only of properties of their objects. Thus, redness is experienced as a property of surfaces, pain as a property of parts of our bodies, and so on (e.g., Harman, 1990; Tye, 1995; 2000). This points to the possibility of an outward-looking Illusionism, on which experience misrepresents distal stimuli as having [sensory-qualities]. Vision, for example, would represent objects as having illusory [sensory-qualities of colour] as well as real physical colours (for a view of this kind, see Hall, 2007). This view can be regarded as a variant of standard, inward-looking Illusionism, differing principally on where the illusory [sensory-qualities] are represented as being located. And, like the inward-looking version, it may posit processes of internal monitoring. The illusion of phenomenality may involve a combination of introspection and projection, in which we both misrepresent features of experience as [sensory-qualities] and then re-represent these illusory [sensory-qualities] as properties of the external world, mistaking complex physical properties of our sensory states for simple [sensory-qualities] of external objects (Humphrey, 2011, chapter 7).” (Frankish, Illusionism as a Theory of Consciousness, 7)

^[5] Observation is a relation between a person (or a mind) and an object (or a multeity of objects)—not a relation between an object (or a multeity of objects) and a part of a person (e.g., a visual system, an introspective mechanism, a sense-organ, or a brain). Instances of “blind-sight” give us no reason to suppose that we have instances of observation wherein an object is observed as neither having nor consisting of determinate (or determinable) sensory-qualities of one or more determinable kinds. For a “blind-sighted” person to correctly guess the identity of some object placed in front of their eyes is not for him to observe it. Indeed, the “blind-sighted” person does not observe the object at all. Similarly, if I am seated in a room and I happen to correctly guess that there is a man lurking in another room a few doors down, I do not thereby observe him. As C.D. Broad says, “For the relation to the organ, whatever it may be, is not the relation of being perceived, since that is a relation to the mind and not to the body.” (Broad, Perception, Physics, and Reality, 189)

^[6] Frankish endorses the view, advanced by Derk Pereboom, that it is an ‘open possibility’ that there exist “representations of particular determinate (or determinable) sensory-qualities” that not only do not have the particular determinate (or determinable) sensory-qualities they represent, but are also themselves totally devoid of determinate (or determinable) sensory-qualities of any determinable kind. (Cf. Frankish, Illusionism as a Theory of Consciousness, 6). This ‘open possibility’ is subject to devastating criticisms in Torin Atler’s review of Pereboom’s Consciousness and the Prospects of Physicalism. Pereboom’s arguments for this ‘open possibility’ assume a form of Kantianism (regarding the nature of introspection and first-person access]. Furthermore, he attempts to substantiate this ‘open possibility’ by relying upon fragments from John Locke’s epistemology (a ghost laid to rest ages ago by Berkeley, Bradley, etc.). Unfortunately, Pereboom doesn’t take the opportunity to either re-examine the coherence of these appeals to Locke in light of Berkeley’s and Bradley’s crushing objections, or to take a step back and ask himself whether his implicit Kantianism is consistent with his knowledge of the fact of his having such views at all. (Cf. Pereboom, Consciousness and the Prospects of Physicalism, 15)