Do We Look Material? Human Ontology and Perceptual Evidence

Abstract

According to certain views about human ontology, the way we seem is very different from the way we are. The appearances are a threat to such views. Here I take up and defuse the threat to one such view.

Pure immaterialism says that each of us is wholly immaterial. The appearances suggest otherwise. I argue that despite the fact that we might sometimes appear to be at least partly material, and that we can be perceptually justified in believing something solely on the basis of having a perceptual experience as of its being the case, none of us is ever perceptually justified in believing that we are even partly material (or that we’re not). Bottom line: we might be able to know whether we’re material, but we can’t know just by looking.

1. Dualism threatened

My late Yorkshire terrier looked short and shaggy. While most of us look taller and more trimmed, we still seem to have some height and some hair style. More generally and universally, we human beings appear to have all sorts of physical properties: some specific height, weight, shape, and color.

The fact that that’s how we seem might be thought to provide good reason to reject certain views about what we are. I’m thinking primarily about a certain version of dualism. Dualism about human persons comes in at least two flavors: compound and pure. ¹ Both disagree with materialism, since they both say we’re not entirely material. They say we have some immaterial part or parts. They differ, though, over whether we also have some material parts. ‘Compound dualism’ says I have both an immaterial part (it’s almost always assumed to be just one immaterial part) and material parts (it’s almost always assumed to be many material parts), like all the parts of a human animal. ‘Pure dualism’—or, as I prefer to call it, ‘pure immaterialism’—says I am entirely immaterial, with no material parts at all. ²

Compound dualism isn’t threatened by our seeming to have physical properties. For, according to compound dualism, we have parts that have physical properties. And I take it that anything that has parts that have physical properties in turn has physical properties itself. Nor is compound dualism threatened by our seeming to have the particular physical properties that we do. It’s plausible that if compound dualism is true, each of us indeed has the particular physical properties that we seem to have.

But pure immaterialism is threatened by these seemings. For, according to pure immaterialism, you and I are wholly immaterial. We have no physical properties at all. So then it’s not the case that you have some specific height, weight, shape, and color. And yet you seem to.

I should note that while I’m going to focus on the threat to a version of dualism, some versions of materialism might be threatened by the same seemings. For example, according to some versions of materialism (what we might call ‘brainism’), each of us is identical with a brain. ³ But brains generally don’t have the physical features that we seem to have. We don’t seem to each other to be gray, gooey, and weigh three pounds. So brainists should pay heed; their view is also threatened by the appearances.

To be clear about what the threat is: my point isn’t that the claim that we have a height and weight and color just strikes us as true, or that it’s just common sense. Perhaps that claim does strike us as true and is a piece of common sense. It’s indeed a familiar charge against dualism that it doesn’t fit with certain pieces of common sense, or common speech—it’s pretty commonsensical that each of us weighs more than a bowling ball, for example. ⁴ Just as familiar as the charge are the many ways of parrying it. ⁵ In any case, commonsense isn’t what interests me here.

Rather, the “seeming” at issue is perceptual. The claim, at least, is that we have “visual (auditory, olfactory, etc.) seemings” as of one another and as of ourselves having all sorts of physical properties. It looks to me like Abe is over there, and that he’s short and portly; upon waking, it looks to me like I’m disheveled and bleary-eyed.

And this sort of visual seeming is arguably pretty good evidence against any position according to which we human people are wholly immaterial. Likewise, it arguably does confer without further ado at least prima facie justification on the belief that we’re at least partly material. ⁶ For if we’re not, then these experiences, which represent us as having properties that entail our being at least partly material, are systematically misleading. And that would be bad and somewhat surprising news. We would expect there to be some explanation of why our perceptual faculties so badly misfire, and none is obviously forthcoming. ⁷ More simply, one might reasonably think we’re entitled to believe in accordance with how the world appears to us unless we have reason to doubt the appearances.

Of course that might not be the end of the story. There might be good arguments for pure immaterialism that will compel us to live with being systematically misled by our senses. ⁸ And we might be able to develop a decent theory about why we came to be so misled. But at the very least these perceptual seemings are the beginning of the philosophical story, and they tell against pure immaterialism (and brainism). So say certain philosophers, at any rate. ⁹ If they’re right, then a pure immaterialist (and brainist) is at a significant dialectical disadvantage. Before any arguments come in, the needle points in other direction.

2. My view

But I don’t think those philosophers are right. I am happy to grant the general idea that a perceptual experience as of p can be evidence for p. I’m even happy to grant that, at least under the right conditions, if someone has a perceptual experience as of p, and she believes p on the basis of that experience, then she is thereby prima facie justified in so believing. (Let’s say, somewhat stipulatively, that someone in that situation is “perceptually justified in believing p.”)

What do I mean by the “right conditions”? These conditions include at least this: it’s possible to perceive p’s being the case. If it’s not so much as possible to perceive p’s being the case, then whether or not p is true, the perceptual seeming as of p is no indication of p’s truth. ¹⁰ It’s something the subject is bringing to her experience, rather than deriving from the world. I’ll return in section 4 to this condition; in the meantime, I’m going to assume it is indeed necessary in order to be perceptually justified.

Although I a grant that there are many propositions that we are perceptually justified in believing, I maintain that no one is perceptually justified in believing the denial of pure immaterialism. More generally, I maintain that there’s no proposition inconsistent with pure immaterialism such that we are perceptually justified in believing it. This includes propositions of the sort we’ve been attending to, like Abe is short and portly and I am disheveled, both of which entail that some human person has sensible properties, and so isn’t wholly immaterial.

But it also includes “negative” propositions, such as there are no immaterial things, there are no immaterial concrete things, and there are no immaterial things in causal contact or preharmonized with any human organisms/parts. Each of these is inconsistent with dualism, period, and so of course inconsistent with pure immaterialism. I’m committed then to us not being perceptually justified in believing any of them.

Indeed, I’d be willing to go a little further. There are certain propositions that aren’t inconsistent with pure immaterialism, as I’ve defined it, but which no sane pure immaterialist would accept. I’m thinking in particular of the claim that this or that human organism/brain/nervous system thinks, or is conscious, or is a person. Pure immaterialism says that all of us human persons are wholly immaterial. Now, that’s logically compatible with there being material things—even wholly material things—that also think, and also are persons. It’s even logically compatible with there being wholly material parts of human organisms that are thinking people. It might even be logically compatible with there being whole human organisms that think, if there’s some way to weasel out of them thereby being human persons. But no sane dualist (whether compound or pure) would think that there are all of these material people (even if not human people), in addition to all of the immaterial ones. There’d be far too many people around. And no sane dualist would think that there are all of these human-animal-thinkers (even if not human people). There’d be far too many thinkers around. So, for the purposes of my claim about perceptual justification, I’m willing to take the denial of material thinking things as part and parcel of both compound dualism and pure immaterialism.

We can call the extended theses ‘compound dualism+’ and ‘pure immaterialism+’ (and ‘dualism+’ for the disjunction), and take my claim as saying that there’s no proposition inconsistent with pure immaterialism+ that we are perceptually justified in believing.

My view has real teeth. But let me just note two ways in which it’s weaker than you might take it to be. First, I’m not claiming that we can’t get perceptual evidence against dualism in general, or against pure immaterialism in particular. We have built up a pretty big stock of empirical evidence—which ultimately boils down to perceptual evidence—for correlations between our mental properties and the properties of brains. And some philosophers contend that these correlations in turn serve as evidence for materialism, and so against dualism, about human persons. I disagree with them; but even if they’re right, that does nothing to impugn the view I’m defending here. For even if we could be perceptually justified in believing that these correlations obtain, their obtaining isn’t inconsistent with pure immaterialism. (It had better not be because the evidence for at least some of the correlations is pretty overwhelming.) My view is the more limited one—that (roughly) we don’t seem to see (or hear or feel) one another as being (at least partly) material. Or, at least we don’t do so in such a way that could make us perceptually justified in believing that we are, directly on the basis of those experiences.

This last caveat brings me to the second way in which my view is weaker than it might be mistaken to be. I don’t claim that there’s no proposition inconsistent with pure immaterialism that is ever the content of our perceptual experiences. Some of us, sometimes, might well perceptually represent the world in a way that’s inconsistent with pure immaterialism (or at least with pure immaterialism+). For example, some of us might sometimes represent a human animal as itself being in pain. But I claim that none of us is perceptually justified in believing such a thing. And that’s because, as I will argue, none of us can perceive such a thing. So, assuming my proposed necessary condition for perceptual justification, the conditions aren’t right for us to be perceptually justified in believing that a human animal is itself in pain. More generally: any anti-immaterialist-seeming, I claim, is something the subject is bringing to his experience, rather than deriving from the world. So, no such perceptual seeming can justify a belief in the falsity of pure immaterialism.

3. Threat averted

My view is about the contents of experience and what propositions they (can’t) justify. Of course, if experience has no content—or no content assessable for accuracy—then my view is trivially true. ¹¹ Likewise, if the only contents that our experiences ever have are thin—things like bare colors and shapes—then it’s hard to see how we could ever have a perceptual experience as of some proposition that entails the truth or falsity of some view on human ontology. And so again my view would be trivially true. I will assume, at least for the sake of argument, that experiences have content, and that such content is richer than just shapes and colors. Indeed, to be as concessive as is reasonable, I will assume that we can experientially represent objecthood (as in when we look in the direction of an orange and visually represent the presence of a single object that is both orange and spherical), the instantiation of kind-properties (as in when we look in the direction of an orange and visually represent the presence of an orange), and even particular individuals (as in when I visually represent the presence of Abe). ¹²

With that concession, you might wonder how I could even hope to show that all the contents of our experiences are consistent with pure immaterialism. That’s a good question. The answer is that I will begin by taking what I take to be a relatively innocuous step in the direction of my view; and then I will argue that once we’ve taken that step, my conclusion follows not too far behind.

The relatively innocuous step is to say that our perceptual experience is perceptually neutral between materialism and compound dualism. I’ll say more soon about what’s meant by perceptual neutrality, but for now we can think of it as saying that our perceptual experience is insensitive to the difference between them. Even if our senses aren’t always silent on that difference, they’re blind to it. They can’t detect anything that would tell one way or the other. The conclusion follows not too far behind given the further claim that our perceptual experience is also perceptually neutral between compound dualism and pure immaterialism. That is, I will argue for these two claims of perceptual neutrality:

1. 1. Our experience is perceptually neutral between materialism and compound dualism

2. 2. Our experience is perceptually neutral between compound dualism and pure immaterialism

And from these two claims I’ll then infer:

1. 3. Our experience is perceptually neutral between materialism and pure immaterialism

The inference is valid if perceptual neutrality is transitive. More on that later. And while (3) isn’t quite my conclusion, we shall see that it’s but one short step from (3) to my conclusion. More on that later. In the meantime, let’s turn to the two particular claims of perceptual neutrality.

3.a Materialism and compound dualism

Remember that according to compound dualism, we have material parts as well as an immaterial one. As far as compound dualism says, any of the leading materialist candidates for what we are could be parts of us. So, a certain kind of compound dualist could say that I have an entire human animal as a part. To reiterate what this implies: the fact that you visually represent me as being 5’ 6" is no threat to compound dualism. According to the animalist version of compound dualism, I am 5’ 6". Your experience isn’t going to be partial toward materialism (as against compound dualism) just in virtue of the fact that it represents my having sensible properties. ¹³

If it’s not going to be in virtue of representing one another as 5’ 6", how might our experience nonetheless fail to be neutral between materialism and compound dualism? I can think of two basic ways, one negative and one positive. ¹⁴

The negative way is so-called because it’s a matter of our representing an absence—a negative fact. When you look at me and I’m not wearing a cowboy hat, then if all is right with your vision, I won’t look to you like I’m wearing a cowboy hat. That is, your experience won’t have the content Aaron is wearing a cowboy hat. So much is obvious. But even more seems to be true. You’ll seem to see the absence of any cowboy hat atop my head. That is, your experience will have the content Aaron is not wearing a cowboy hat. If it turns out I am wearing a cowboy hat, and the angle at which the light reflected off the hat made it so that it looked to you just like it does when I’ve got no hat, then your experience will have been partially falsidical, rather than merely incomplete. It will have misrepresented the situation as one in which there’s no hat atop my head, when in fact there is.

Thus, one might suggest that our experiences (or some experiences, had by some of us) represent the world as lacking in immaterial things; or, at least as lacking in immaterial concrete things; or, at least as lacking in immaterial spirits, ghosts, and souls. More narrowly, you might suggest, our experience represents situations involving human organisms as lacking in any immaterial things “hooked up” to the relevant organism/part—as being the kind of situation in which there is no immaterial thing in the vicinity of the organism/part, or in causal contact, or harmonized with it.

My response to this is twofold. First, I doubt that we could really represent any of this visually—or any other experiential way. Start with an experience of another human organism that is “silent” on whether there is an immaterial soul “hooked up” to it. (I’m supposing that even someone who thinks that we can experientially represent the absence of an immaterial soul will concede that we can have an experience that is silent on the question.) Now: What can you add to the experience—more exactly, what can you change phenomenally—such that it’ll now represent the absence of an immaterial soul? Is there anything? If so, it’s very mysterious what it would be. Assuming that content supervenes on phenomenology, there doesn’t seem to be any way we could get the experience to represent the absence of an immaterial soul hooked up to the organism.

In addition to such appeals to mystery, we can give a more direct and positive argument that there’s no way for us to represent the absence of the soul—at least not without already representing the presence of something that would all by itself tell in favor of materialism. How do you manage to represent the absence of a cowboy hat atop my head? Why is it that you don’t also represent the absence of loads of subatomic particles whirling around?

A pretty plausible account (developed by Farennikova [2013]) of perceptually representing absence is that you match, or compare, an expected scene (a scene that includes a cowboy hat on top of my head) to the scene you actually meet (a scene in which you can look right above my head and see sky); and in (visually) representing the difference between the two scenes, you thereby (visually) represent the absence of a cowboy hat. If that account is right, in order to detect the absence of something, the absence of that thing has to make a discernible difference to something else in the scene, something that isn’t an absence (else the account will be circular). That is, representations of absences are parasitic on representations of differences between presences. But then this “negative way” for our experience to fail to be neutral between materialism and compound dualism doesn’t stand on its own; it requires a “positive way” for our experience to fail to be neutral. And we’ll get to that way next. For the moment, then, we can set this possibility aside.

But let’s grant that we can experientially represent the absence of a soul. Remember, though, that in order for such an experience to make us perceptually justified in believing that there’s no soul, it has to be the case that it’s possible to perceive the absence of a soul (more exactly, its being the case that there is no soul “hooked up” to this organism). (At least that’s what I’ve been assuming until now. I will return to a justification of this assumption.) And I think it’s pretty clear that we can’t possibly perceive that absence—not just given the perceptual faculties we have, but given any perceptual faculties we could have had.

It’s hard enough to understand how we can perceive (as opposed to merely experientially represent) absences in general (if we can). After all, absences don’t seem to be the kinds of things that could cause anything. ¹⁵ And it’s relatively uncontroversial that in order to perceive X or p’s being the case, X or p has to stand in some causal relation to the perceiver. But whatever solution we propose to the general problem is not going to help us understand how we could perceive an absence of something that, were it present, would be imperceptible.

Suppose, for example, that we take our cue from Jonathan Schaffer’s solution to the puzzle of absence causation. Partly to address the puzzle of how there could be any such thing—given that absences, even if there are such things, don’t have any causal oomph—Schaffer (2005) puts forward a contrastive model of causation: causation “is a quaternary, contrastive relation: c rather than C* causes e rather than E*, where C* and E* are nonempty sets of contrast event.” In the case of causation by absences, C* is nonactual. Thus, the truth that the gardener’s failure to water the plant caused it to wilt is to be understood as follows (suppose he was napping instead): the gardener’s napping (c) rather than watering the plant (C*) caused the plant’s wilting (e) rather than blossoming (E*). And that’s true because c and e occurred, but if C* had occurred instead of c, then E* would have occurred instead of e. Thus, we can speak correctly of absences being causes without having to assume that absences really have any oomph.

We might extend this to account for perception of absences. Thus, we might say that you perceive the absence of (some nonactual state of affairs) just in case (a) you perceptually represent the absence of that state of affairs, and (b) were that state of affairs present, you would perceive it. Thus, we can speak correctly of absences being perceived, without having to assume that absences are really perceptible.

But this account will only allow us to speak correctly of the perceptible absence of things, like cowboy hats, which if they were present would be perceived in the straightforward sense. Souls are not like cowboy hats in this regard. You wouldn’t see or hear or smell them if they were present. Their absence literally makes no perceptible difference, and so their absence is imperceptible. ¹⁶ That’s true given the account we just suggested. But it seems like any account of absence perception should respect that truism: you can perceive the absence only of something you could perceive were it present. The soul’s absence is no more perceptible than its presence.

Concluding this discussion of the negative way, and putting matters in terms most relevant to representational neutrality: our perceptual experience is insensitive to this particular difference between materialism and compound dualism—the difference between the absence and presence of a soul—whether or not it is silent on that difference.

So much for the negative way. Now for the positive way. The positive way is so-called because it’s a matter of representing a presence—a positive fact—presumably, some fact having to do with thinking, consciousness, or personhood. But we need to be precise here. Remember, the fact that when looking “in my direction” you seem to see that there’s something that’s both 5’ 6" and conscious introduces no partiality toward materialism over compound dualism. According to compound dualism (at least those versions of compound dualism that says I have a whole human animal as a part) I am both 5’ 6" and conscious. So the experiential representation of material things as in some way mental has to be more specific in order to be partial toward materialism.

I can think of two ways our experiential content might be more specific in this respect. One way would be if our experiences represent some event, or activity, or process as both material (consisting in the activities of material objects) and mental (consisting in the activity of thinking). Thus, we might look at a brain and visually represent that there is thinking going on in that brain. This claim—that there is thinking going on in the brain—is inconsistent with dualism+. Again, no sane dualist of any stripe would think that that there are material processes that are themselves episodes of consciousness. There’d be too much consciousness about.

But I don’t think our sensory experiences represent that claim as being true. We need to keep in mind the distinction between genuine consciousness and functional indistinguishability from consciousness. (It goes without saying that a functionalist will deny that there’s any distinction here. I’m relying here on the falsity of functionalism. This is a legitimate assumption to make in this context: if functionalism is true, then evidently material things do think and are conscious, and so dualism+ is false. That is, if functionalism is true, there’s no need to rely on perceptual evidence to support materialism, since it just follows from the truth about what mentality is. ¹⁷ So my whole discussion can be seen as conducted under the supposition that functionalism is false.) It doesn’t seem to be the case that our sensory experience represents the presence of consciousness (as opposed to them representing the presence of either consciousness or functional indistinguishability from consciousness). It’s in fact commonplace in the philosophy of mind literature that we don’t seem to see consciousness when we look at things like brains, central nervous systems, and organisms. I suspect we don’t seem to see consciousness because we can’t, and that we can’t because of the chasm between the view from inside (which is how we can become acquainted with consciousness) and the view from outside (which is the only view that sensory experience can give us). ¹⁸ Thus, even when we look at ourselves—in the mirror, or even on a screen during brain surgery—we don’t seem to see the subjective awareness as much as see a thing, and be separately acquainted with its being a subjectively aware thing. This, of course, is the heart of Leibniz’s Mill argument. ¹⁹

There is another way that our experiential content might be more specific in mental respects, and in a way that is partial toward materialism. Suppose we experientially represent a certain material thing as itself having some mental feature or other. Thus, perhaps some of our sensory experiences have as content that dog over there is in pain. Maybe that’s just abstract enough that we can sensorily represent it as being the case, even though we can’t, as I’ve just argued, sensorily represent the undergoing of that pain, the subjective painful awareness.

The content in question—that the dog over there is in pain—may, or may not, be consistent with dualism+. It depends on what the content is exactly. If the content is that the canine animal over there—the organism over there, which has nothing but material parts—is in pain, then it’s indeed inconsistent with dualism+. As before, no sane dualist of any stripe would think that some wholly material object like a canine animal has genuine mental properties. If a canine animal has genuine mental properties, then so, presumably, does a human animal, and there’d be too many thinkers about. If, on the other hand, the content is more neutral about the bearer (how far it extends and what its parts are), then it might well be consistent with dualism+. So our experience will be partial toward materialism over compound dualism on these grounds only if we can be perceptually justified in believing a very specific claim about property bearers, such as that the canine animal itself feels pain—that the thing that instantiates the property of being in pain is a canine animal.

Now, I’m willing to grant that we can experientially represent such a thing. And I’m even willing to grant that we sometimes do experientially represent such a thing—although I doubt such a specific sensory representation is very common. But even if we sometimes do so, I don’t think that specific content is perceptible. It’s too specific to make any causal difference—and a fortiori any perceptible difference.

Suppose that as a matter of fact the thing that instantiates the property of being in pain is a canine animal. And now ask, under that supposition, if it hadn’t been true that the thing that instantiates the property of being in pain is a canine animal, what would have been the case? I don’t know. But I very much doubt that what would have instead been the case is something different enough that it would have looked or smelled different. Presumably the metaphysics of mental property bearers would have been slightly different, so that something else, bigger or smaller, would be the property bearer. And that situation would have looked and smelled just the same. ²⁰

Note that this is very different from our perception of far less specific content, such as there’s a tree out there. That’s sufficiently nonspecific so that in order to get to a situation in which it’s false with as little departure from actuality as possible, you’d have to go to a situation that is perceptibly different. In order to get back to a scenario in which things look and feel the same, we’d have to go much further—to some sort of skeptical scenario involving evil geniuses or BIVs. So nothing I’ve said rules out our perceiving that there’s a tree out there. ²¹

So much then for the positive way for our experience to be sensitive to the difference between materialism and compound dualism. This wraps up my case for the claim that our experience is perceptually neutral between materialism and compound dualism.

3.b Compound dualism and pure immaterialism

While it took some work to argue for that claim about materialism and compound dualism, I think it’s a relatively innocuous claim. But, as I said earlier, once we’ve taken that claim on board, my conclusion follows not too far behind. And that’s because, as I see it, our experience is perceptually neutral between compound dualism and pure immaterialism.

The difference between compound dualism and pure immaterialism, as you’ll recall, is a difference about which things we are. This difference leads to a number of other differences: especially regarding which things are conscious, and which are persons. But it’s important to note just how much common ground there is, or at least can be, between the two sides.

For one thing, they need not differ over any ‘ontological’ question—any question about what there is or how many things there are of a given kind. They will both agree about the existence of immaterial souls, and how many there are. They can both agree about the existence of material things, and how many and which ones there are. And, what may not be as obvious but is still true, they can even agree about which material/immaterial amalgams there are. Compound dualism is committed to there being such amalgams, because it identifies each of us with some one of them. But pure immaterialism is also consistent with there being such amalgams—and even with there being exactly the same ones as there must be according to compound dualism. Pure immaterialism says that you and I are wholly immaterial; it doesn’t say that everything is either wholly material or wholly immaterial, and it’s consistent with me and my body composing something further.

For another thing, they’ll agree about all of the basic qualitative facts: this includes facts about the size and shape and weight and color of human bodies and brains, and it includes facts about where (or, in which things) there is mental activity going on. ²² And for one final thing, they can agree about all of the causal facts: this includes facts about the interaction between the immaterial thinking things and material things such as human brains.

So the facts over which the two views will differ are rather limited, and seem to be of just two kinds: ‘conceptual’ facts and ‘haecceitistic’ facts. The conceptual facts are things like: given that the basic qualities are distributed over things in such a pattern, and there are mental processes going on over here and there, and these things stand in causal relations to those things, then such-and-such counts as a conscious being, and a person. The haecceitistic facts are things like: given that the basic qualities are distributed over things in such a pattern, and there are mental processes going on over here and there, and these things stand in causal relations to those things, then so-and-so is Abe, or you.

Now, I’m willing to grant that we can experientially represent such things. And I’m willing to grant that we sometimes do experientially represent such things. ²³ But I don’t think that these facts are perceptible. They’re too metaphysical to make any perceptible difference. They’re just not the kinds of things that our perceptual faculties can detect.

Suppose, per impossibile, God tells you he’s created two planets as much like Earth as the following stipulations allow. One is inhabited by human people whose nature is the same as what compound immaterialism says our nature is, and the other is inhabited by human people whose nature is what pure immaterialism says our nature is. (If materialism is true, neither of these planets is exactly like Earth. And I say “per impossibile” because I assume that whatever the truths are about the concept person won’t vary from world to world, let alone planet to planet.) On both of these planets, whenever and wherever there is a human person, there is an immaterial soul and a body that are intimately connected with one another: there is a two-way, direct, and immediate causal connection between the soul and the body. And the causal connection between them is exactly the same on the two planets. Moreover, on both planets the same number of total objects exist, located in the same places; and basic qualities, both physical and mental, are distributed in the very same way on both.

And then suppose God tells you that you and your friend Abe are human persons who inhabit one of these two planets, but He doesn’t tell you which one. Would you be able to tell by looking—whether at Abe or at yourself—which planet you’re on? I don’t see how you possibly could. No difference between the two scenarios would be picked up by your visual, or olfactory, or whatnot perceptual system.

This wraps up my case for the claim that our experience is perceptually neutral between compound dualism and pure immaterialism.

3.c Transitivity of perceptual neutrality

As I already noted in section 3, the final step of the argument is to bring these two claims of perceptual neutrality together so as to yield the conclusion that our experience is perceptually neutral as between materialism and pure immaterialism as well.

1. 1. Our experience is perceptually neutral between materialism and compound dualism.

2. 2. Our experience is perceptually neutral between compound dualism and pure immaterialism.

Therefore,

1. 3. Our experience is perceptually neutral between materialism and pure immaterialism.

From which premises and conclusion it follows that: if our experience represents matters in a perceptible way that’s inconsistent with pure immaterialism, that can only be because it represents it in a perceptible way that’s inconsistent with our existing. So we’re perceptually justified in believing something incompatible with pure immaterialism only if we’re perceptually justified in denying our existence. It’s fairly obvious that we’re not perceptually justified in denying our existence. So we get my conclusion that we’re not perceptually justified in believing anything incompatible with pure immaterialism. ²⁴

But the inference of (3) from (1) and (2) is valid only if perceptual neutrality (relative to some set of experiences) is transitive. Or, at least I see no way to vindicate the validity of that inference unless we can assume that it’s transitive. But is it? The intuitive gloss on perceptual neutrality—as a matter of being insensitive to the difference between the two claims—might initially make it seem obvious that it’s transitive. But for those of us familiar with sorites series of phenomenal indiscernibility, the gloss might instead give us pause. After all, isn’t a failure of transitivity exactly what we have regarding the relation of phenomenal indiscernibility? (The claims that the truck looks red_1,238,987 and that it looks red_1,238,988 are phenomenally indiscernible; and more generally, for any n, the claims that the truck looks red _n and that it looks red _n+1 are phenomenally indiscernible; and yet the claims that the truck looks red₁ and that it looks red_1,238,988 are not phenomenally indiscernible.) Isn’t that at least one of the lessons of the phenomenal sorites? And isn’t phenomenal indiscernibility just a matter of our perceptual experience being insensitive to the difference between the claims?

Fair question. First, it’s worth noting that not everyone agrees that the failure of transitivity of looking the same, or phenomenal indiscernibility, is the right lesson to draw. Some philosophers (Fara 2001; Raffman 2000) take its transitivity as a truism, or at least as true, and so deduce that there can’t really be any genuine cases of a phenomenal sorites (only ones that we mistake for a phenomenal sorites).

Second, the relation could be restrictedly transitive. That is, it could be that there is some rather natural/simple condition φ such that if p and q are phenomenally indiscernible, and q and r are phenomenally indiscernible, and φ (p, q, r), then p and r are indiscernible. Perhaps φ is a condition that rules out vagueness as an explanation of the twin phenomenal indiscernibilities. Since vagueness seems not to be part of the explanation for the perceptual neutrality (of our experience) between materialism and compound dualism, and seems not to be part of the explanation for the perceptual neutrality (of our experience) between compound dualism and pure immaterialism, we could rely on such a restricted transitivity principle to license the inference in the above argument.

But third and most importantly, even opponents of the transitivity of phenomenal indiscernibility acknowledge that there are several relations in the vicinity, some of which are demonstrably transitive. And the relation I have in mind is demonstrably transitive. It’s a matter of our perceptual experience failing to pick up on any fact that speaks in favor, so to speak, of one over the other. Here’s a more careful definition:

Experience (or set of experiences) E is perceptually neutral between p and q (we can write ‘p ~_E q’) = _df for any perceptible proposition r such that E represents r as being true:

1. 1. r entails p iff r entails q, and

2. 2. r is consistent with p iff r is consistent with q, and

3. 3. r is inconsistent with p iff r is inconsistent with q.

As can be easily verified, this relation is transitive. ²⁵

In the case where p and q are incompatible (as in our hypotheses about human ontology), p ~_E q implies that for any perceptible proposition r such that E represents r as being true, (a) r entails neither p nor q, and (b) either r is consistent with both p and q or it’s inconsistent with both p and q. This is exactly why I said that from the premises and conclusion of my argument it follows that: if our experience represents matters in a perceptible way that’s inconsistent with pure immaterialism, that can only be because it represents it in a perceptible way that’s inconsistent with our existing. If there is some perceptible proposition r such that our experience represents r as being true, and r is inconsistent with pure immaterialism, then it follows (given the premises and conclusion of perceptual neutrality) that there is a perceptible proposition (i.e., r itself) that is inconsistent with pure immaterialism, materialism, and compound dualism. But it is a logical truth that if we exist, then either we are wholly material, or wholly immaterial, or some amalgam of material and immaterial parts. So then there is a perceptible proposition that is inconsistent with our existing.

One last point on this issue: one can in principle dispense with transitivity, and just argue directly that our experience is perceptually neutral between materialism and pure immaterialism. It’s not that the arguments above for each of the premises won’t come in handy. They will, since they’ll help us see, regarding various candidate contents that might be partial to materialism over pure immaterialism, either that they are not perceptible, or that our experience doesn’t represent them as being the case. Indeed, we could in principle dispense entirely with the two premises, seeing them as ladders to be kicked away when we get to the end. That is, we could just examine a slew of candidate propositions, which are inconsistent with pure immaterialism, and argue that none is both perceptible and such that we represent it in experience as being true. One could give either one of those more piecemeal arguments. To my mind they’re less compelling, and less illuminating, than the original two-step argument. But they’re available if needed.

4. A debt to be paid

Along the way I incurred an important debt. My conclusion is that there is no perceptible proposition, inconsistent with pure immaterialism+ (or compound dualism+), that our experience represents as being the case. And I assumed that to be enough for our not being perceptually justified in believing any proposition inconsistent with pure immaterialism+ (or compound dualism+)—even if it turns out that there are propositions, period, which are inconsistent with pure immaterialism+ (or compound dualism+), and that our experience represents as being the case.

Of course, this amounts to a substantive assumption only if the notion of perceptual justification in play is a nontechnical pretheoretical concept, whose conditions of application we can sensibly look to discover, rather than merely stipulate. But I’m assuming it is such a concept. So my assumption is indeed substantive. Why should we think it’s true? Am I right that having a perceptual experience as of p, and believing p on the basis of that experience, are insufficient by themselves for being prima facie justified in believing p—and that a further necessary condition is that p be perceptible? If one accepts the view (variously called ‘dogmatism’ [Pryor 2000], or ‘phenomenal conservatism’ [Huemer 2001]) that having a perceptual experience as of p can all by itself justify believing p—without that justification resting on the subject’s having justification to believe some other proposition about the connection between one’s experience and the truth of p—then might the added condition I laid down be unnecessary?

A full answer to this question requires a longer discussion of that view than I have space for. But I think enough can be said to make it quite plausible. If we ask why perceptual experience all by itself confers justification (assuming it does), I can think at least two possible answers, one externalist in flavor, the other internalist.

One answer appeals to the very same (explanatory or probabilistic) connection between the content of experience and the truth, which a more stringent theorist of justification might have required the subject to have justification to believe (or even to justifiably believe)—but simply doesn’t require the subject to have any such further justification. (Neither the connection, nor a belief in the connection is doing any justifying; but its truth is what allows the perceptual experience to justify the belief held on the basis of that experience.) Putting the connection very generally, the idea would be that of the two competing hypotheses—one, that things are as our experience represents them, and the other, that things are not as our experiences represents them—only the first explains why our experiences are the way they are. The other leaves it unexplained. So we are prima facie justified in believing the one that provides an explanation. Or, if we want to put it in Bayesian terms: the likelihood of our experience having the content it does given the first hypothesis is (substantially) higher than given the second hypothesis, so our experience representing the world being some way (substantially) confirms the hypothesis that things are that way. And then assuming in the background some sort of indifference principle, which gives them roughly equal priors, the rational posterior credence is going to be quite high. ²⁶

But this sort of answer only works in cases where the following is true: were things as they seem, their being so would explain why they seem that way, or make more probable that they’d seem that way. But that won’t be the case—or will far from obviously be the case—when the seemings are of things that are imperceptible. The seemings will then be just as inexplicable, and just as unlikely, given the claim that they do accurately represent reality, as given the claim that they don’t. For the imperceptible propositions, even if true, won’t explain why we seem to perceive them to be true.

A second (internalist-flavored) answer to the question why perceptual experience confers prima facie justification is something like this: you’re entitled, epistemically speaking, to start with how things seem to you; you’re epistemically blameless for doing so. You have to start somewhere, after all, and how could it be wrong or blameworthy to start where you in fact are?

This answer is rather vague, and so it’s hard to say anything very clear about it or its implication for whether perceptibility is a necessary condition. But I think we can at least say the following. First, if it were right then it would support the highly dubious doxastic conservatism just as much as it would support phenomenal conservatism. ²⁷ Second, why think we “start” only once we’ve got the perceptual representation in hand? If we push things back a step, we can intelligibly ask whether we are entitled (in some broad sense) to perceptually represent things as though p is true; and in cases where p is imperceptible, it seems that the answer should be “no.”

Third and most importantly, everyone agrees that while this sort of thing might buy you prima facie justification, that justification can of course be defeated. And one very plausible defeater is coming to learn that the proposition that you’re representing as being true is imperceptible. (Pryor [2000, 534] in fact explicitly acknowledges this as a defeater.) If you seem to see the future—and then you come to the conclusion that the future cannot be seen—then even if you were entitled at the very outset to take on board what your perceptual experience represents as being true, you’d no longer be entitled to do so once you come to that conclusion about the imperceptibility of the future. The reason is that the conclusion you’ve drawn implies that your experience would have the same content whether or not that content was true. ²⁸ So at worst I’d have to slightly dial back my central claim. Rather than claiming that no one is perceptually justified in believing the denial of pure immaterialism, I’d claim that no one apprised of the considerations I adduced in this paper is perceptually justified in believing the denial of pure immaterialism. But that’s good enough for my purposes since I just want to figure out what to believe about what I am; I don’t care much about whether the hoi palloi materialists are justified.

To be sure, in my quest to figure out what I am, the question of whether we are prima facie perceptually justified in denying pure immaterialism is hardly the only relevant one. There are very many arguments for and against materialism, compound dualism, and pure immaterialism (along with more specific versions of each, and for and against our existence). ²⁹ But I hope to have shown that, contrary to what some philosophers have claimed, none of the options can be ruled out just by looking.