1. Boyle on Self-Knowledge and Transparency
In the glory days of Descartes and Hume, philosophers were introverts. Stuck behind a veil of ideas, we never just straightaway perceived the external world, not directly. Instead, what we were directly aware of in perception, so the thinking went, were just ideas in our own minds.
Whatever its attractions, this introversion raised a notorious skeptical problem. Given only awareness of what’s in here, how can I know what is happening out there? Suppose my evidence—what I am directly aware of—is just that there is an experience of a cat on a mat. If I am to know that there is a real cat on a real mat in the real world, this must derive from a seemingly dubious inference from the psychological evidence about my own experience.
Nowadays, philosophers are mostly extroverts, living in a world of mind-independent objects. When we perceive these external objects, what we are directly aware of is nothing more or less than the objects themselves.
If defensible, our modern extroversion might dispel traditional external world skepticism. However, it leaves us with a different problem. Given just awareness of what’s out there, how can I know what is happening in here? If, in perception, all I am aware of is the cat itself, then ascribing to myself an experience of a cat might seem inevitably to proceed by an even more dubious inference from immediately known evidence about a cat.
This latter skeptical challenge, known as the puzzle of transparency, has drawn considerable recent attention.Footnote 1 According to some extroverts, most clearly Alex Byrne (Reference Byrne2018), the right response is simply to embrace inferences from premises about the external world to conclusions about our own minds. From something resembling the evidence that there is a cat, I can infer something resembling the conclusion that I am perceptually aware of a cat. To other authors, however, an inference from some premise that I am aware of, to the conclusion that I am aware of that premise, seems like madness.Footnote 2 The fact that there is a cat certainly does not entail that I am aware of it, and depending on the situation, it might not even be strong evidence of it.
This is where Matthew Boyle’s masterful new book (Reference Boyle2024) comes in. Drawing on a remarkably diverse range of influences, perhaps most especially Sartre, Boyle seeks to explain our self-knowledge by traveling a middle path between introversion and extroversion. When I see a cat on the mat, Boyle follows Sartre in saying, I am positionally aware of a cat. However, at the same time, and in the very same mental act, I am nonpositionally aware of my own awareness of a cat. If I go on to reflect on my own thinking, I might then become positionally aware of my awareness of the cat. But in attaining this reflective knowledge, there is no need to get from an external item of awareness to an internal one by means of a dubious inference, since direct knowledge of both is already present from the outset, albeit in different forms. Reflective (i.e., positional) knowledge of my own mind rests not on positional knowledge of a distinct worldly subject matter, but on a distinct nonpositional knowledge of the same mental subject matter.
Boyle’s early chapters are devoted to explaining the Sartrean distinction between positional and nonpositional awareness, and using it to address the problems affecting other transparency accounts of self-knowledge, such as Moran’s and Byrne’s. Later chapters then connect the discussion to issues of refreshingly broad interest, as witnessed by chapter titles like “The Examined Life.” It is a book brimming with ideas, from beginning to end. However, I am going to focus on just one chapter right in the middle, Chapter 5.
Without getting too deep in the weeds about nonpositional awareness, it’s kind of a big deal for Boyle that when I am initially nonpositionally aware of thinking about a cat, the object of my awareness is not the fact or proposition that I am thinking about a cat. It is, rather, just thinking about a cat—something like a property or event. However, if this is the object of my nonpositional awareness, then when I transition to positional knowledge of a proposition, what proposition am I going to end up knowing? We might expect it to be something like the proposition that thinking about a cat is occurring, or maybe that this particular episode of thinking about a cat is occurring. But usually when I reflectively come to know about my thinking, or assert what I know to others, I do not just judge or say that. I say that I am thinking about a cat—an assertion that seems to implicate not only the episode of thinking itself, but a subject to whom it belongs.
So to explain our ordinary self-knowledge, Boyle needs to say how we get from something resembling the evidence that thinking is occurring, to something resembling the belief that there is a subject doing the thinking. And here he runs into a longstanding controversy between philosophers he terms “Egoists” and “Anti-Egoists,” exemplified by our old friends Descartes and Hume. Despite its pedigree, what the controversy is even about is somewhat nebulous, but Boyle does a splendid job in sorting it out, and in attempting to unify and systematize multiple conflicting inclinations that a lot of us have. But ultimately, I am going to say that there are just too many conflicts, and our inclinations cannot all be saved.
2. Egoism versus Anti-Egoism
As Boyle understands it, the debate between Egoists and Anti-Egoists concerns what can be known by reflection in a particular “Cartesian moment,” dramatized in the Second Meditation. It is there that, having temporarily set aside perception of external objects as doubtful, arch-Egoist Descartes nevertheless claims reflective knowledge of his own thinking. Crucially, this is said to include not only knowledge of the particular thoughts occurring, but of himself, the subject of these thoughts. Descartes elaborates:
I am, then, in the strict sense only a thing that thinks; that is, I am a mind, or intelligence, or intellect, or reason—words whose meanings I have been ignorant of until now (CSM II 18).
Descartes goes on to provide a positive metaphysics of this thinking thing. It is, he says, a substance, capable of existing without other things existing. It is simple, having no parts. And it is immaterial, having no size, shape, or location. However, the crucial thing for Boyle is not these well-known further claims, but an apparently more basic one that, as Boyle puts it, “My awareness of thinking supplies me with an awareness of a being whom I call ‘I’.”
Cue the Anti-Egoists. One canonical example is Lichtenberg, at least as often interpreted.Footnote 3 According to him, what Descartes really should have claimed to know is not I think, but rather There is thinking, understood analogously to There is lightning.
Another canonical Anti-Egoist is Hume, who says:
For my part, when I enter most intimately into what I call ‘myself”, I always stumble on some particular perception or other, of heat or cold, light or shade, love or hatred, pain or pleasure. I never can catch myself at any time without a perception, and never can observe any thing but the perception (T 1.4.6).
One more canonical Anti-Egoist is Wittgenstein. In a passage from the Tractatus often cited alongside Hume and Lichtenberg,Footnote 4 he says:
If I wrote a book called The World as I found it, I should have to include a report on my body, and should have to say which parts were subordinate to my will, and which were not, etc., this being a method of isolating the subject, or rather of showing that in an important sense there is no subject; for it alone could not be mentioned in that book (TLP 5.631).
What should we make of these rather cryptic remarks? For Hume, at least, the issue seems bound up with broader empiricist misgivings about metaphysics, including the notion of substance. In place of simple Cartesian substances, whose thoughts are their modes, Hume proposes that there are merely the thoughts themselves, bundled together. Sticking closely to this historical example, the debate between Egoists and Anti-Egoists might be seen as a metaphysical debate of a kind with debates about composition, causation, and so forth.
But for reasons I return to in Section 5, I do not think that really captures the spirit of Wittgenstein’s remarks. For Wittgenstein, and I think for Boyle as well, the suggestion is that when I reflect on my awareness of a cat on the mat, there is some more fundamental difficulty involved in recognizing myself, the subject of this awareness, out there among the cat, the mat, and the other objects of my awareness. It is not just that I am unaware of being a Cartesian substance, as opposed to some other kind of thing. It is, rather, a difficulty in recognizing myself as appearing in any capacity among the inhabitants of the objective world.
According to Wittgenstein, the upshot is that “There is no such thing as the subject that thinks or entertains ideas.” Or at least, that is what he said before kicking the ladder away. But whatever Wittgenstein’s ultimate attitude toward this statement may have been, Boyle unequivocally rejects it. He thinks I am, at the end of the day, just that human body I was aware of all along. Since I am in fact this object, reflecting on my own thinking makes me aware of thinking which belongs to what is in fact an object, along with the other objects I am aware of.
Despite this, Boyle sees the Anti-Egoists as raising an important insight. While I am in fact an object, I am not aware of being one just by reflecting on my thinking; this takes the kind of bodily awareness Boyle goes on to discuss in a subsequent chapter. So the real challenge raised by Anti-Egoists is not a challenge to some metaphysical theory about my nature. It is a skeptical challenge to my ordinary claims to self-knowledge. Because I am not reflectively aware of myself as an object, Boyle’s Anti-Egoist says:
(Anti-Egoism) Reflecting on my thinking about a cat only gives me knowledge expressible as “Thinking about a cat is occurring.”
However, Boyle thinks it is premature to jettison our ordinary reflective self-knowledge. He sets out to defend:
(Boyle’s Egoism) Reflecting on my thinking about a cat gives me knowledge expressible as “I am thinking about a cat.”
Where both the Anti-Egoist and the Cartesian Egoist go wrong, Boyle thinks, is in assuming that my ordinary self-knowledge must involve awareness of myself as an object, the way I am aware of cats and mats as objects. So while Boyle endorses a minimal form of Egoism, he retreats from the stronger claim that:
(Cartesian Egoism) Reflecting on my thinking about a cat gives me knowledge that an object is thinking about a cat.
Boyle’s strategy is thus to retreat from the weightier commitments of Cartesian Egoism to a kind of middle-ground position. The aim is to vindicate our ordinary reflective self-knowledge, yet also do justice to the Anti-Egoist denial that our self-knowledge involves awareness of ourselves as objects. However, despite a lot of sympathy for Boyle’s attempted compromise, here I will reluctantly argue that it does not work. I mean this not so much as a knock against Boyle’s proposal as an expression of pessimism that any compromise is available. So far as I can tell, there just is not any stable middle ground to occupy.
Here is the plan. In the next section, I summarize Boyle’s response to Anti-Egoism and why the retreat from Cartesian Egoism to a middle ground view is essential to it. Then in Sections 4 and 5, I defend the following argument that no middle ground is available:
(Subjects) If reflecting on my thinking gives me knowledge expressible as “I am thinking about a cat,” then it gives me knowledge that a subject is thinking about a cat.
(Objects) If reflecting on my thinking gives me knowledge that a subject is thinking about a cat, then it gives me knowledge that an object is thinking about a cat.
Therefore,
(No Middle) If reflecting on my thinking gives me knowledge expressible as “I am thinking about a cat,” then it gives me knowledge that an object is thinking about a cat.
3. Why Boyle’s Retreat Matters
Anti-Egoism, as Boyle understands it, raises a skeptical challenge for our ordinary self-knowledge. Meeting the challenge means explaining the justification not for some philosopher’s metaphysical theory but for our ordinary beliefs and assertions about our own thinking. Take my belief that I might express by saying “I am thinking about a cat.” This belief seems more metaphysically committal than a mere belief that some thinking about a cat is occurring. So it might be tempting to say, along with the Cartesian Egoists, that what I believe entails:
Heavyweight Belief: An object is thinking about a cat.
But Anti-Egoists say that all I am really aware of is:
Lightweight Evidence: Thinking about a cat is occurring.
Maybe one possible response is to say, contra Hume, that reflecting on my thinking does give me awareness of myself as an object. If so, maybe my heavyweight belief can be justified by equally heavyweight introspective evidence. However, this fits uncomfortably with Boyle’s overall view, which says that whatever I am directly aware of in reflection is already present implicitly in nonpositional awareness, whose object is merely the episode of thinking about a cat.
This seems to stick Boyle with the skeptical problem, in the form of an apparent gap between my evidence and what I believe. If my ordinary reflective belief really goes so far beyond what I am directly aware of, then arguably it is unjustified, and does not amount to knowledge.
Another possible response to this problem might be to bridge the gap via a metaphysical claim like:
(Single Thought) Necessarily, if thinking about a cat is occurring, then an object is thinking about a cat.
Boyle considers this response, but does not buy it:
It seems that nothing else is needed to warrant me in judging I am thinking about whether it will rain besides my consciousness of thinking about whether it will rain. For mustn’t any thought have a thinker? … On further reflection, however, the inclusion of “I” in “I am thinking of whether it will rain” can seem to assert too much. By hypothesis, what I am aware of is: thinking about whether it will rain. But to ascribe this activity to a subject is to assert more than the mere existence of thinking. It is, seemingly, to assert the existence of a being to whom this thinking belongs, an agent of this activity or bearer of these states. … Yet how could sheer awareness of thinking (or any other mental state or process) inform me of the existence of such a unified, enduring being? (105–106)
Boyle’s response to the skeptical challenge, in contrast, is to retreat from the Cartesian Egoist’s claim that awareness of thinking informs me of the existence of a being (or object). This does not represent a retreat from ordinary claims to self-knowledge, however, so long as we are careful not to overstate just how heavyweight our ordinary claims and judgments really are. When I judge that I am thinking about a cat, this judgment does not really commit me to anything so metaphysically extravagant as the existence of a being (or object) who is thinking about a cat. Instead, there is something more neutral expressed by “I am thinking about a cat,” something Boyle is usually happy to state as follows:
Middleweight Belief: A subject is thinking about a cat.
However, I am more pessimistic than Boyle about an available middle ground, weighty enough to vindicate our ordinary self-knowledge, but less weighty than the allegedly extravagant claims of Cartesian Egoists. In Section 4, I am going to argue for taking at face value the apparent commitment of Middleweight Belief to the existence of something thinking about a cat. And in Section 5, I will question whether there is any interesting sense in which something that is thinking about a cat could fail to be an object. But first, I should highlight another element of Boyle’s anti-skeptical response, to say why it is insufficient on its own.
Boyle takes this second element to be necessary because the Lightweight Evidence allowed by Anti-Egoist Hume is still too weak to justify even Middleweight Belief. As Boyle puts it:
The second element of Boyle’s response is that the Anti-Egoists have undersold our nonpositional evidence. We are not only aware of individual thoughts, but of a unified standpoint these thoughts belong to:
[A] nonpositional awareness of the unity of this standpoint is presupposed in our believing. When we recognize a factual question as open or closed, we implicitly recognize it as undecided or decided from the epistemic standpoint in which this question is posed. The standpoint is such as to involve response to many distinct factual questions, and our responsiveness to each of them shows that we understand these responses to be rationally related to one another. … In such ways, our capacity to consider factual questions presupposes a nonpositional awareness of our own epistemic standpoint as constituting a single overall outlook on the world, one that is determinate in some respects and indeterminate in others. (129)
So instead of Lightweight Evidence, our self-ascriptions of thinking are based on nonpositional awareness of something more like:
Middleweight Evidence: Thinking about a cat is occurring along with other thinking in a unified standpoint.
However, while this second element of Boyle’s response may be necessary to effectively rebut the skeptical challenge, it is arguably not sufficient on its own. The Anti-Egoist’s skeptical challenge cites a gap between Lightweight Evidence and Heavyweight Belief. But if our response were solely to replace Lightweight Evidence with Middleweight Evidence, without also retreating from Heavyweight Belief, we would still have a gap between our evidence and our beliefs, albeit a smaller one. If the larger gap needs filling, the smaller one does too, with something like:
(Standpoint) Necessarily, if thinking about a cat is occurring along with other thinking in a unified standpoint, then an object is thinking about a cat.
Now I do not want to cast aspersions on Standpoint, a plausible if not obvious metaphysical claim. It’s just that I am not sure how much of an improvement it is over Single Thought, which Boyle already conceded to the Anti-Egoist, saying that awareness of mere thinking does not get us all the way to “the existence of a being to whom this thinking belongs.” If this much is conceded, how much further does awareness of a unified bundle of thoughts get us? Presumably, Hume should have granted awareness not only of particular thoughts, but of relations like resemblance and causation that unify them. And Boyle counts him as an Anti-Egoist, who denies that even unified bundles of thoughts must belong to an object.
Maybe there are subtle metaphysical reasons why single thoughts might not belong to an object, but unified bundles of thoughts must. But Boyle himself does not argue this. Instead, his defense of our ordinary reflective beliefs rests on a retreat from heavyweight Cartesian claims about beings to something less weighty, yet still weighty enough to vindicate my ordinary judgment that I am thinking about a cat. This is why Boyle’s middle ground between Cartesian Egoism and Anti-Egoism is important. In the next two sections, I argue that no middle ground is available.
4. The Commitments of Ordinary Self-Knowledge
According to Boyle’s middle ground view, reflecting on our thinking yields self-knowledge that is less weighty than knowledge that an object is thinking, yet still weightier than mere knowledge that thinking is occurring. So far, I have been assuming that this middleweight self-knowledge is at least existentially committal; when I know that I am thinking about a cat, I know that there is something—call it a “subject”—which is thinking about a cat. If this is still less weighty than the heavyweight claims of Cartesian Egoists, the difference lies in their commitments about what kind of thing is thinking about a cat. The middleweight commitment is just that a subject of some kind or other thinking about a cat, while the heavyweight Cartesian commitment is that it is some more particular kind of thing, namely an object.
A lot of the time, Boyle himself puts things this way. While he rejects Cartesian Egoism, Boyle’s view still breaks from Anti-Egoism on the question of what we know just by reflecting on our thinking. While Anti-Egoists claim that all we know is that thinking is occurring, Boyle says we furthermore know that there is a subject doing the thinking. But sometimes, he has a different way of characterizing the question on which he differs from Anti-Egoists, as one of how our knowledge can be expressed. On this way of putting things, the Anti-Egoist says our self-knowledge is expressible only as “There is thinking about a cat occurring,” while Boyle says it is expressible as “I am thinking about a cat.” Before turning to the distinction between subjects and objects in Section 5, I want to consider whether this opens up some other way for Boyle to retreat from Cartesian Egoism, by rejecting:
(Subjects) If reflecting on my thinking gives me knowledge expressible as “I am thinking about a cat,” then it gives me knowledge that there is a subject thinking about a cat.
Not to be dense here, but I cannot understand what it would be. If my self-knowledge is expressible as “I am thinking about a cat,” then what I know is that I am thinking about a cat. But if so, what I know sure seems to entail that something is thinking about a cat, even if it is neutral on what kind of thing it is. If by stipulation anything thinking about a cat qualifies as a “subject,” then my ordinary self-knowledge is committal on there being a subject thinking about a cat.
If it is not obvious that my self-knowledge has this commitment, there is an additional dialectical reason for Boyle to accept it. For he wants my middleweight self-knowledge to be weightier than mere lightweight knowledge expressible as “Thinking about a cat is occurring.” But if these two states of knowledge are different, there must be some difference in what is known. If I had only lightweight self-knowledge, what I would know was just that some thinking about a cat is occurring. So if instead I really have middleweight self-knowledge, there must be something else I know, in addition to that. It cannot be just that the subject in question is me rather than someone else, since either way a subject is thinking. Nor can it be just that the thinking about a cat is unified with other episodes of thinking, at least not in any way that a Humean Anti-Egoist already acknowledges. But then it is hard to see what else I am supposed to know, if not that there is something thinking about a cat.
This dialectical motivation assumes that if my self-knowledge is weightier than knowledge expressible as “Thinking about a cat is occurring,” then I must know something more than just the fact that thinking about a cat is occurring. Could this be rejected? If so, it is unclear how to understand the skeptical challenge Boyle’s Egoism is designed to solve.
As I presented it above, the Anti-Egoist’s skeptical challenge is straightforward. It alleges that I do not really know that I am thinking about a cat, because I do not have justification to believe anything more than just that thinking about a cat is occurring. And Boyle’s response, as I characterized it, also is straightforward. It says I do have justification for a weightier belief, because my evidence supports not only that thinking is occurring, but also that I, a subject, am thinking.
But I am not sure I have not oversimplified things. Boyle himself does not use the word “evidence,” and indeed, he characterizes the target of justification not as a belief at all, but as a kind of cognitive practice. Here are the main passages where he speaks of justification or warrant:
Where exactly does this “I” come from, and how is its introduction justified? (102, emphasis added)
I take the cognitive step that a subject takes when she learns to think in the first person to be significant, inasmuch as it constitutes the cornerstone of all explicit self-representation. It is the warrant for this step that I explore in the present chapter. (105)
My purpose in doing so, however, is not to join the Anti-Egoists in disputing our entitlement to include “I” in our ascriptions of self-awareness, but to sharpen our sense of what the nature of this entitlement must be. (105)
In making these observations about the cognitive significance of the first person, we have not yet answered the question of what justifies the introduction of such a representation. (125)
Yet if the inclusion of “I” in the statement of self-knowledge is not warranted by our apperceptive awareness of our own thinking…
“To present consciousness-as-subject in such a subject-predicate form is, in effect, to impose the structure of positional knowledge on the analysis of subjectivity: it is to understand consciousness-as-subject as presenting manifold determinations of a single subject in the logical sense: a single bearer of various properties. But what can justify the imposition of this structure?” (126)
[W]hat is now in question is our warrant, not for ascribing particular psychological predicates to ourselves, but for thinking in the first person at all. (127)
[Kant] takes our warrant for accompanying our several representations with “I think” to lie in our awareness that they essentially belong to such a unity. (142)
At the risk of being dense again, I am not really sure what it would mean to justify the practice of including “I” in my reflective judgments, if not just to justify particular reflective judgments which include “I,” such as my judgment that I am thinking about a cat. And as with all judgments and beliefs, the justification for this one arguably must take the form of evidence supporting what I believe, in this case that I am thinking about a cat.
Look at it this way. If I have sufficient evidence supporting that I am thinking about a cat, then presumably no other sort of justification is needed for me to include “I” in my reflective judgment that I am thinking about a cat. And without evidence for this, no other justification will suffice. Suppose my evidence leaves open whether I am thinking about a cat or whether instead some thinking about a cat is occurring without anything thinking about a cat. In that case, it is unclear what other kind of justification I might have for including “I” in my reflective judgments. Maybe there could be prudential reason for engaging in a kind of fiction, as with a nominalist philosopher who thinks and speaks as if there are numbers. But if the inclusion of “I” were merely a kind of pretense or fiction, then it seems I would not really know that I am thinking about a cat, even if my fiction happens to be true.
So if we want to avoid Anti-Egoism, and vindicate self-knowledge weightier than what I express by “Thinking about a cat is occurring,” the appropriate target of justification must be a belief or judgment in something more than just that thinking about a cat is occurring. And I do not know what that something more would be, if not that there is something doing the thinking.
5. Subjects and Objects
Boyle’s sees his response to the skeptical challenge as a victory for Egoism, and yet:
It is only a qualified victory … inasmuch as we have rejected the further claim, endorsed by classical Egoists like Augustine and Descartes, that the awareness expressed in “I think that …” necessarily implies the existence of a being who is the possessor of the relevant point of view. On the contrary, we have maintained that the awareness expressed by “I think” does not per se imply the existence of any possible object of conscious awareness who thinks. It implies only the existence of a subject who thinks, which is to say, a conscious point of view that this act of thinking partly characterizes. (142, emphasis added)
In positing herself as such a subject, [a reflective subject] would not assert her identity with some object in the knowable world; she would merely posit herself as a single consciousness of objects. (130)
To be sure, Boyle thinks we are objects, specifically “space-occupying, flesh-and-blood” ones. His point is that we cannot know this about ourselves simply by reflecting on our own thinking. Hence, the retreat from Cartesian Egoism.
The tricky thing is, I do know by reflection that I am thinking about a cat, and this knowledge is committal on there being a subject thinking about a cat. If I still do not know that an object is thinking about a cat, we must deny:
(Objects) If reflecting on my thinking gives me knowledge that a subject is thinking about a cat, then it gives me knowledge that an object is thinking about a cat.
But what is an object, and how might a subject fail to be one?
5.1. Epistemic objects
One way to understand “object” is epistemic; to be an object is to be a potential object of awareness. To deny Objects is then to deny that when I reflectively know that I am thinking about a cat, I also know that I am a potential object of awareness. For all I can tell just by reflecting on my thinking, I might instead be an elusive subject, something capable of being aware of a cat, but which cannot itself be the object of my awareness.
Now I have no general argument that elusive subjects are impossible. There are plenty of things I am contingently unaware of. Maybe there also are some things that I simply could not be aware of, and maybe some of them include subjects. What I do have an argument for is that I am not myself an elusive subject if I know that I am thinking about a cat. For if I know that I am thinking about a cat, then I know that I exist, and I know what some of my properties are. And that is enough to count as being an object of my awareness.
Given this, it is not plausible that reflection on my thinking cannot inform me that I am a potential (indeed, actual) object of awareness. By reflecting, I can know that I am thinking about a cat, and indeed, know that I am aware of a cat. Maybe that does not necessarily mean I am also aware by reflection that I am aware that I am aware of a cat. But it is hard to see what is stopping me from reflecting further and gaining this further knowledge. However, this means nothing is stopping me from knowing that I am an object of awareness, just by reflecting on my thinking.
A possible response on Boyle’s behalf is to restrict “objects of awareness” to things I have a special type of awareness of—a type of awareness I have of cats and mats, but not of myself when I reflect on my thinking. But what type is that?
For Boyle, the answer cannot be positional awareness. Reflecting on my thinking is supposed to give me positional awareness that I am thinking, awareness that, Boyle says, “posits a subject.”
What about perceptual awareness? When I reflect on my thinking, I am not perceptually aware of myself. I do not see myself, or hear myself, or anything like that. And even if I am the kind of thing—a flesh and blood human—that I could be perceptually aware of, I plausibly do not know that just by reflecting on my thinking. So I might reflectively know that I am aware of a cat without knowing that I am, like the cat, the kind of thing that I could be perceptually aware of. In this sense, I might, for all I know, be an elusive subject.
The problem for Boyle is that this hardly amounts to a qualified victory for Egoism. Not only does Arch-Egoist Descartes positively claim that I am not a potential object of perceptual awareness, but it is how he has the Meditator characterize his immateriality in the Second Meditation, before the nature of extension is discussed in the Fifth Meditation. I am, he says, not the sort of thing that could be perceived with the senses, or even depicted by the imagination. If Boyle’s anti-skeptical strategy involves a retreat from Cartesian Egoism, it cannot be to the view that, for all we know by reflection, we might be immaterial substances.
Maybe there are other types of awareness that I do not have of myself in reflection. But as with perceptual awareness, it is unclear why, if being reflectively aware of myself does not tell me whether I can be the object of some other type of awareness, this represents only a qualified victory for Egoism.
5.2. Metaphysical objects
In any case, Boyle seems to have something else in mind. For he also denies that our reflective self-knowledge speaks to “the existence of a being who is the possessor of the relevant point of view.” He continues:
This conscious point of view is posited through mere reflection on our [nonpositional consciousness] of thinking, and so does not imply the existence of any objective entity who possesses this point of view, though it also does not rule out the existence of such an entity.
This suggests a metaphysical reading of “object,” rather than an epistemic one. When I reflectively know that I am thinking about a cat, what I do not know is whether the subject thinking about a cat is an insubstantial subject, lacking the metaphysical heft of a genuine being.
What would it mean for a subject not to be a being? In one way of speaking, to be a being is just to exist. But understood this way, everything is a being, and so the only way for there not to be a being thinking about a cat is for there not to be anything thinking about a cat. That is what I claimed in Section 4 is incompatible with my ordinary self-knowledge. If I know that I am thinking about a cat, then I know that something is thinking about a cat—something that is, in this manner of speaking, a being.
In another way of speaking, a being is just a certain kind of thing. Understood this way, the Cartesian Egoist says that I know by reflection not just that something is thinking about a cat, but that a particular kind of thing is. Some opponents, like the Anti-Egoist Hume, might deny outright that I am the kind of thing the Cartesian has in mind. Others, like Boyle, just say I cannot tell by reflecting on thinking.
This way of construing the Egoist/Anti-Egoist debate, as a metaphysical debate about what kind of thing I am, might be faithful to its historical roots in Descartes and Hume. Without question, Descartes was committal on what kind of thing I am aware of when I am aware that I am thinking. He said I am simple, having no parts; immaterial, having no size or shape; and finally a substance, capable of existing independently of this episode of thinking about a cat—in contrast to the episode of thinking itself, which cannot exist without me. And what is more, Descartes said I know all this about myself, more or less just by reflecting on my thinking.
Hume, for his part, denied my simplicity, claiming instead that I am a composite (or bundle), whose parts are thoughts. And he might also have denied that I am a substance, regarding instead particular thoughts as substances, and me (a mere bundle) as incapable of existing without them. It is less clear whether Hume denied my immateriality, but in any case, we might deny it, or at least deny that I can be aware of it just by reflecting on my thinking.
It is uncomfortable for Boyle, however, to understand a “being” simply as the kind of thing Descartes said we are aware of in reflection. For one thing, Boyle says that I am in fact a being, but the kind shaped like a human body, and composed of body parts. So beings had better include material, composite things.
More importantly, Anti-Egoism is supposed to raise a skeptical challenge to my ordinary self-knowledge. If the Anti-Egoist merely questions whether I am a simple, immaterial substance, then it is not clear how this might even appear to raise any such challenge. Philosophical theories like Descartes’s are committed to my simplicity and immateriality, but my ordinary belief that I am thinking about a cat does not seem to be. More generally, the claim that I am thinking about a cat seems neutral about what kind of thing I am, beyond my being a subject, the kind of thing that can think. So it is unclear how further ignorance about my nature was ever supposed to impugn my ordinary self-knowledge.
Could my ordinary self-knowledge at least be committal on my being a substance? I doubt it, if a substance is just something with the right modal properties (i.e., being capable of independent existence). Maybe it is incompatible with my broader common sense beliefs about myself to claim that I could not have existed without engaging in this very episode of thinking about a cat. But on the face of it, this still seems consistent with my belief that I am thinking about a cat.
But sometimes the denial that we are substances is taken to be more radical, implying that we are not particular things. If, for example, Hume’s view is that there are no composite things, then there really is no such thing as a bundle of perceptions, as opposed to particular perceptions, bundled together.Footnote 5 Even more radically, sometimes “bundle theory” is taken to be the view that there are no particular things at all, but only compresent universals.Footnote 6
Now, if those are the alternative scenarios to my thinking belonging to a being, then it is no wonder they challenge my self-knowledge that I am thinking about a cat. What is less clear is how this skeptical challenge could be met. Unless these scenarios are consistent with there being something thinking about a cat, I cannot know by reflection that I am thinking about a cat without ruling them out (assuming knowledge is closed under entailment). It is also unclear how I am supposed to ever come to know that I am a being. If reflecting on my thinking in the Cartesian moment cannot rule out these scenarios, it is hard to see why bodily awareness can do better.
In any case, there is a more general problem with tying Anti-Egoism to any of these metaphysical scenarios, at least if we hope to do justice to the Wittgensteinian idea that there is some special difficulty in recognizing the subject of awareness among the objects of awareness. Whatever the difficulty is supposed to be, it is a particular difficulty in recognizing myself, not a more general problem in knowing about composite objects or bare particulars. While Wittgenstein’s metaphysical subject is allegedly prevented from appearing in the World as I find it, the same does not go for my body, nor presumably for cats and mats.Footnote 7
This raises a difficulty not just for Boyle, but for any of us who hope to find some common theme among the various authors he terms “Anti-Egoist”. Hume’s point in denying awareness of a Cartesian substance was presumably not to deny that we are parts of the objective world altogether. Individual thoughts, as well as bundles of thoughts, are perfectly objective phenomena, construed in the broadest possible sense to include whatever might be found in the world. If so, Humean Anti-Egoism might after all just be a metaphysical view about what kind of thing I am, which contrasts with the Cartesian substance view. However, if the Wittgensteinian Anti-Egoist instead finds some problem in recognizing ourselves as objects in even this inclusive sense, it is not just a problem of knowing that we are one sort of thing rather than another.Footnote 8
David Barnett is an Associate Professor of Philosophy at the University of Toronto. He works mainly in epistemology and philosophy of the mind, and is currently writing a book about the self and self-consciousness.
Open access
