yellow,11 so it appears that the Phenomenal Principle must be mistaken.
2.47 It is possible to veridically perceive the location of an object, so it should be possible for the immediate object of your friend's awareness to be located in just the spot that yours is. If the immediate object of your friend's awareness occupies that region and is yellow, yours cannot be white. And yet it must be, according to the Phenomenal Principle.
2.48 Price once suggested that sense data occupy a “special kind of space.” A sense datum that you're aware of might be spatially related to another sense datum that you're aware of, but they don't occupy “public space,” as he puts it.12 Perhaps he'd say that your sense data are confined to your own personal space and your friend's are confined to their personal space. Such a proposal suffers from two problems. First, Michael Huemer argues that this won't do, because sense data stand in temporal relations to physical objects, and anything that stands in such relations to physical objects must also stand in spatial relations to these objects. This, he argues, is a consequence of special relativity.13 Obviously, anything that stands in spatial relations to physical objects must occupy some position in physical space. Second, Price's proposal requires us to say that terms like “spatial location” and “spatial relation” are ambiguous. They can be used to talk about locations and relations in physical space (e.g. relations between your hands and this page) or locations in some alternative space (e.g. relations between sense data). It's not clear that there's good evidence for this ambiguity, but let's suppose Price is right. On Price's view, do sense data appear to have locations in physical space or in this alternative space? If the former, then the Phenomenal Principle requires us to assign them a location in physical space. That's precisely what we want to avoid. If the latter, then the Phenomenal Principle doesn't force us to say that they have some location in physical space, but in some alternative space. Remember that for the naïve realist, the IOAs are physical objects, like tables and chairs. Clever philosophical arguments were needed to show that this identification is mistaken. What this means is that (unbeknown to us) the things that we took to be tables and chairs don't even appear to occupy physical space. Rather, the things that we take to be ordinary tables and chairs appear to occupy some alternative space. If that's so, we wouldn't have any understanding at all of what it would be for a table or chair to occupy a position in physical space. That seems rather implausible.
2.7 The Return of Direct Realism
2.49 As the popularity of indirect realism waned, naïve realism came back into fashion. One nice thing about direct realism is that it allows for the possibility that we're in direct contact with things in the external world. Thus, when we're not hallucinating, it might seem that experience gives us strong support for our beliefs about the external world – when such things are made manifest, we needn't worry that our evidence isn't strong enough to justify belief.
2.50 The challenge for the direct realist is to account for the possibility of hallucination and illusion. This is no trivial task. We can only gesture at some ideas about how the direct realist might address the concerns discussed above.
2.51 Let's start with illusion. In cases of illusion, the things we see appear to have properties that they don't actually have. Think about the bent stick. To the hand, it is straight. To the eye, it is bent or broken. How can we reconcile the fact that the thing we're immediately aware of is both bent and straight (or unbroken and broken)? We can't. Nothing can be both of those ways at once. However, think about Halloween. Suppose you've been invited to a Halloween party and you want to look like a vampire. What do you do? You can't become a vampire. You could at best make yourself look like a vampire, but we'll assume that it's not possible to become a creature of the night. What you do is remain a non‐vampire and then put on a costume, make‐up, etc. to create the appearance of being a vampire. Or a judge. Or a nurse. Or a cat. Or a whatever. There is no problem with some single object having at once these two properties: being a human (who isn't a vampire) and being something that looks like a vampire.
2.52 Although nothing can be both F and non‐F at the same time, something can be F and an apparent non‐F at the same time. To account for the possibility of illusion, the direct realist might say that when we are dealing with a case of perceptual illusion, we are directly aware of an OMO and its properties. These include properties like having a certain shape and color and having a certain apparent shape and apparent color. A wall can be both white (its color) and appear yellow (its apparent color). A coin can be both circular (its shape) and appear elliptical (its apparent shape). You can be both a non‐vampiric human (your kind) and appear to be a vampire (your apparent kind). As long as an object's apparent properties are understood as properties that this object has (given its non‐appearance properties, the conditions under which it is perceived), illusion might be no problem for the direct realist.
2.53 It would seem that this strategy isn't useful for dealing with hallucination. In hallucination, there is supposed to be nothing that can be the object of awareness. With no object, no appeals to the distinction between kinds of properties that an object can have will work. The direct realist might pursue a divide‐and‐conquer strategy and appeal to the disjunctive theory of experience to fend off the Argument from Hallucination. The argument starts from the idea that it can appear just the way things do now; you perceive a book before you even know if you're hallucinating. It then is supposed to convince us that this means that even when we perceive a book, the IOA isn't a book but must be something that's present in the case of hallucination. The disjunctivist thinks that it's a mistake to model our account of perceptual awareness on the case of hallucination. In the case of perception, we see a book. In the case of hallucination, it's as if we do. Sometimes our experience is a matter of something being made manifest and sometimes it is a matter of it being just as if it is. In one case, experience is a relation between you and an ordinary object. In the other, it's as if it is.
2.54 What makes the disjunctivist view disjunctivist is that it offers a disjunctive account of experience:
Disjunctivism about experience: when an individual has an experience as of something being F, it is either a matter of perceiving an F or a matter of seeming to.14
On this approach, we don't think of experience as the core that a perception and some indistinguishable hallucination have in common. True, a subject might not be able to tell one from the other, but the same might hold true for a lemon and a piece of soap. We wouldn't conclude from this that some lemon and some piece of soap share something in common because some subject couldn't tell the one from the other.
2.55 Because disjunctivist denies that the conscious experience a subject has in the case of perception is identical to the experience in a case of hallucination, they're under no pressure to endorse the Phenomenal Principle. Without it, they can block the Argument from Hallucination. Having denied the principle, they are freed from the idea that the thing we're most immediately aware of must always be as it appears. One of the supposed benefits of this approach is that it opens up the possibility that the evidence or rational support that successful perceptual contact provides is better than that provided by hallucination. As we saw above, some are concerned that if the support provided by experience is not better than the support provided by hallucination, that support on its own isn't enough to justify beliefs about things in the external world. To justify believing, say, that we are holding a tomato, we need something that justifies our rejecting the hypothesis that there is no tomato there and that we're just hallucinating. How could experience furnish this justification if, at bottom, the experience is the very same thing in the case of perception and in hallucination? To justify rejecting the hypothesis that the apparent tomato is just a hallucination, it seems we need a better reason to believe that it is real than to believe that it is merely apparent. But this seems to require