For a breezy introduction, listen to a generated Deep Dive podcast here.
April 9, 2025
Theral:
Hi there, Rodrigo. Thanks for the invitation to engage in dialogue on Two Hands. I’m interested in how an objective realist deals with the old problem of the “veil of perception.” How does one get to the object without transforming it through the very act of subjective access? It seems to me that physics itself had to face this problem in the early 20th century—and we’re still confused by it.
Rodrigo:
It’s always a pleasure to discuss philosophy with you, Theral. And I want to commend you for picking this particular topic, perception and realism, as it is one of the most fruitful points of contact between our respective philosophical outlooks. Assuming you are largely sympathetic to a Kantian view of perception and that I stand for some sort of metaphysical realism, situated within the analytic tradition, I can think of many philosophers who will stand in the middle ground between us and have rich observations and arguments to keep us engaged. Maybe the topic is too big. In addition to perception and realism, we could consider conceptualism, non-conceptual content, representationalism, transparency, epistemic justification, and reasons vs causes. But let’s take it slow. In philosophy (I’ve heard it said), if one isn’t going slow, one isn’t going.
The veil of perception, originating maybe with Berkeley, is the idea that we do not perceive external objects directly but only through intermediaries—sensations, concepts, or mental representations. How does one get to the object? I would say that these intermediaries are the vehicles by which we passively get to the object, not distortions or subjective contributions. Our knowledge of metaphysical reality is finite and selective, and since our cognitive faculties are fallible, what passes for knowledge is sometimes false. But this filtering does not imply subjectivity, uncertainty does not imply distortion, and representation does not imply ignorance of presence. Representationalism and realism are comfortably compatible—provided the representations are grounded in an informational semantics whose structures are causal consequences of their referents.
What of the physics? Certainly those who have followed Bohr, interpreting quantum mechanics phenomenologically, have confused us all. But there are other options. Anyway, the problem was the phenomenology and idealism of the early 20th century physicists, not with the physics itself. More contemporary philosophers of physics have offered less confusing alternative conceptions.
I don’t mean to make quick work of these very complicated questions, however. Consider each of these points an invitation to extend the conversation along some more specific direction.
Theral:
Thank you for your kind comments. It sounds like for you, filtering does not necessarily entail distortion, and that representations—if grounded causally—need not preclude realism. I’d like to press you further on this question of filtering. You write that intermediaries (sensations, concepts, representations) are vehicles, not distortions. Yet doesn’t the very notion of a “filter” suggest some transformation or mediation? It seems to me that once we admit filtering, we’ve admitted a gap—however slight—between the object and our access to it. And once there’s a gap, aren’t we in the terrain where intersubjectivity becomes the best we can hope for?
Karl Popper has been helpful to me here. He insists that while objectivity is an ideal, it’s never fully achieved—it’s something we approximate through communal processes like scientific dialogue and peer review. On my podcast today, a cancer researcher reminded me of that oft-quoted line in science: “All models are wrong, some are useful.” That feels very Popperian—and maybe even Kantian. It’s not that we access things as they are in themselves, but that we get better at testing, revising, and refining our models in a shared space of inquiry.
So my question is: how does your realism maintain a foothold in objectivity without slipping into a kind of idealized metaphysical access? Can you say more about how filtering doesn’t imply subjectivism? Is it a matter of degrees of reliability, or of being causally tethered to the object, or just what?
If I might be bold, something tells me that you’ll say the filtering is passive and therefore a subject is not involved actively. But how can this be? Even with the passive argument, I’m not sure how the filtering doesn’t impact the object. But I’m just not comfortable with a fully passive universe. (Are we really talking about determinism?) This might bring up the question of whether a simpler life form such as bacteria or nematode worms has any subjectivity. I would argue yes. So then we must define subjectivity. Subjectivity = lived, active, irreducible first-person experience, inseparable from consciousness and embodiment.
Of course, I’m aware that I’m speaking of subjectivity as an object here. But it is also something I truly live.
Rodrigo:
We both truly live our experience. I will not be goaded into your erasure by denying your lived experience! There is subjectivity, because there are subjects, perceivers, and persons, such as you and me. The question is whether perception is thereby a form of self-seeing, and ultimately, whether our lived experience, conscious and embodied, is also irreducibly first-personal. But let us take it slow, and continue with the question of the filter.
The intermediaries of perception, such as the eyes, filter what we know by perception. By this I mean that we do not, by looking, see all that there is. We see what is ahead of us, not behind us. We see the opaque surfaces of things, not their hidden interiors. We see colors from red to violet, but not the infra-red or the ultra-violet. We do not see everything, but what we do see passes through the filter to be seen as it is. Normally, appearances are not deceiving. Normally, we see things as they really are.
Let me also assure you that the universe is not fully passive. We are agents and embody activity, an essential aspect of our being. But perception and action remain distinct, even if mutually informed. We actively look then passively see; actively touch then passively feel; actively study then passively learn. You will ask whether the looking shapes the seeing, whether the active mind conditions the passive percept. It certainly does, when it is a distortion, but unless it is systematically distorted, objective knowledge of metaphysical reality is possible. Why should we suppose otherwise?
The very idea of perception already implies a separation, a gap, between perceiver and perceived. But knowledge of reality as it is does not require a collapse of the object into the subject—representation is not presence. Idealized metaphysical access is not omniscience but instead merely faultless representation, which is not only ideal but also commonplace since illusions are the rare exception to the norm. Of every word on this page, for example, how many did you read correctly? Surely, more than most, if our successful communication is any indication. But objectivity isn’t a measure of reliability. Even if only one word were read correctly, then that one word would have been read as it is.
Metaphysical realism is the view that there is a mind-independent way things are, and epistemic objectivity is the view that we can know the way things are. How reliably is questionable, as is how much of reality we perceive. I claim only that under normal circumstances, the model is useful because it’s not wrong.
Theral:
Thank you, Rodrigo. I’m glad to hear you won’t erase me—nor I you. I’d hate to lose my place in the world so early in the thread! I’m also pleased, and somewhat surprised, to hear you recognize subjects and subjectivity without hesitation. I had suspected you of a more uncompromising objectivism, which is partly why I first posed the question: how does the objectivist get to the object?
In fact, I find little here with which to disagree. I’m struck by how closely your account of the “filter” resembles a Kantian intuition. It sounds like you accept both that there is a gap and that we are active agents—does that not already place us closer to Kant than to, say, a Fodorian representationalism?
Your example of reading words on a page is great. Even if most words are read correctly and a few misread, knowledge remains intact. Yet I wonder: what about the words read correctly, yet interpreted differently? Is this not commonplace? We could consult the dictionary, but as you know, no two dictionaries are quite the same. I’m especially fond of Dr. Johnson’s mischievous definitions. Take “lexicographer: a harmless drudge.” Or my favorite: “Oats: a grain, which in England is generally given to horses, but in Scotland supports the people.”
Dr. Johnson is not denying the objective reality of oats. He is playing with the subjective version in a place where he shouldn’t be. How delightful!
I want to push on one persistent point. When you write that epistemic objectivity is the view that we can know the way things are, is this your view? If so, does knowledge require faultless representation, or only sufficient representation for successful navigation? And, to return to the earlier concern, how is it that representation, even if accurate, does not itself shape what is perceived, at least in part?
I am not advancing relativism. What I am exploring is whether epistemic subjectivity may itself be a genuine source of knowledge. Might there be a kind of knowing available only through the irreducible first-person perspective you kindly preserved in not erasing me.
To move forward: Suppose there is both a mind-independent reality and a mind-dependent reality. Are these necessarily in contradiction? Suppose further that we can neither fully access the mind-independent world nor fully share the mind-dependent one. Does this not place us, irretrievably, in an “in-between” space—not quite Kantian, not quite realist? I suspect, given your acknowledgement of subjectivity, that you too inhabit this space.
Of course, the problem is not merely theoretical. Contemporary politics shows us how often these two approaches clash. So the practical question becomes: how do we navigate these tensions without collapsing one into the other, or, worse, simply denying either?
Rodrigo:
You are quick to suppose I have conceded so much. I recognize subjects and subjectivity as nothing more than minds and their points of view. But there remains something uncompromising in that I deny that any truth is epistemically subjective. Ontological subjectivity is not a genuine source of knowledge. There is no epistemic subjectivity. Minds are subjects, but they are also objects, and as such, all truths about subjects are epistemically objective.
Additionally, while I follow Kant—by way of Sellars—in recognizing that perception available for epistemic inference is necessarily conceptual, I consider the process of conceptualization itself passive. We do not choose to see the apple as an apple. So yes, we are active agents, but neither perception nor conceptualization is among the actions. I accept that there is a gap between subject and object, and that we are active agents. So does Fodor. How did you guess that my own view is closest to a Fodorian representationalism?
Speaking of Fodor, it is worth revisiting the implications of his conceptual atomism, which I accept. Are dictionaries—or the usages they report—truly standards for the meanings of words? I would suggest quite the reverse. The meanings of the words are the standards by which we judge the dictionaries. What goes for reading most of the words correctly equally applies to interpreting them correctly. Harmless drudgery, indeed.
We can know the way things are, yes, and this requires faultless representation. A failure of correspondence or reason is either a false or an unjustified belief—hence, not knowledge. I appreciate your invitation to a more permissive pragmatism, but I decline. As for why representations do not shape what they represent, the reason is the order of causation. Cognitive representations are passively caused by what they represent. The cause-and-effect relation is asymmetric and does not reverse—not even in part. Conative representations, by contrast, actively cause what they represent, as is expected from desires and intentions, but conceptualization is not intentional.
I recognize how uncompromising and unusual this view is, affording no room for in-between spaces, no quasi-realisms with moderate anti-realist inclinations. Isn’t philosophy astounding?
Theral:
You are devilish in offering me an apple! And even more sly in saying that I am a subject who can perceive yet cannot subjectively know of the apple—insisting that I do not have any subjective knowledge of the roundness and redness or of mouth watering anticipation to take a bite of its firm flesh and crisp acid. Your seduction is rather kinky in offering me the apple, but not letting me enjoy it. To let me actively look at it, but not let me actively see it.
I am sympathetic, of course, to your argument about passivity. So let’s dive in: where does the active looking give way to the passive seeing? How does this happen? Why does this happen? Why cannot the ontological “subject” have epistemic “subjectivity?”
Sellars is here whispering and MacDowell even louder that the “given” is already entangled with the life we live. Yes, the world causes, but it causes in our complex beings, not on blank slates—on subjects full of experience and social history. So even going with the passive argument, my passive perception is not a mere camera.
Here’s what I propose. The universe has been mostly passive for eons, yet activity has evolved, emerged. The active on the passive. I would guess that C. elegans, the nematode worm, has less active power than us. And the cat more than the worm but less than the human. How much has human perception become active? And even when you can guarantee that a certain portion of perception is passive, does not that happen within an active subject? The passive causation is quickly interpreted. Why is the active part any less epistemic than the passive? I will go out on a limb (depending on your superior knowledge to correct me) and argue that it is really only the active which can become knowledge because knowledge is normative. Normative implies action. There is no normative of the passive except through action. This works not only for justified true belief but also that other kind of knowledge Socrates talks about, perception.
When you deny epistemic subjectivity, am I to understand that you would deny that two people—two subjects—might perceive and conceptualize the same apple differently? Are there not meaningful differences here, without lapsing into distortion or error? You grant agency, you grant subjectivity, yet you will not let the subject enter the domain of knowledge. Does this not risk reducing ourselves to sophisticated cameras?
When you offer the apple, let me partake.
Rodrigo:
We cannot have epistemic subjectivity because there isn’t any to be had. By this, I do not mean that we cannot know subjective reality, but rather that we can know it objectively. I am assuming a solution to the hard problem of consciousness, of course, and perhaps you are with the pessimistic or dualistic majority in supposing that the problem will remain forever unsolved—in which case, let’s openly say so. Savor the apple all you like. I deny you nothing.
Two subjects may conceptualize and savor the apple differently, but disagreement is not evidence of subjectivity unless we also have reason to believe they are in contradiction and both equally valid. I may conceptualize the apple from one side and you from another, but there is no contradiction, as the apple has both sides. I may conceptualize it as acidic and you as sweet, but again, the apple may be both acidic and sweet. But if I say it’s ripe and you say it’s unripe, our claims cannot be equally valid. Epistemic subjectivity is impossible if the subjects of the world are among its objects.
And yet, we agree: there is normativity in perception and knowledge. But not all normativity is action-guiding. Perhaps that is where we differ? Consider the normativity of the heart. Arrhythmia is abnormal, isn’t it? The condition is abnormal and passive. Normativity does not imply action because—contrary to Kant—the good is not grounded in the will. Anything that can be unhealthy can be abnormal or wrong, including the passive vehicles of perception and knowledge.
Theral:
How does a subject have objective knowledge?
Suppose two people differ on the ripeness of a single apple before them. One said it was ripe, the other unripe. Now suppose those two people are scientists studying the ripeness of the apple based on various criteria. One scientist has a lab in room #9 and the other scientist has a lab next door in room #10. The scientist in room #9 uses criteria such as color, sugar content, and texture, among others. The scientist in room #10 uses some overlapping criteria as well as other criteria. With apple A, scientist in room #9 says it is ripe while the scientist in room #10 says it is unripe—the original positions. They each publish papers on their data and conclusions. Other scientists weigh in with comments and secondary publications, and a majority of them concludes that apple A is ripe. Now it certainly could be that another apple B is concluded by both scientists to be ripe and by an overwhelming majority of the scientific readers as well. And it could be that an apple C could be considered unripe by both primary scientists and the overwhelming majority of the scientific readers.
Let’s consider a heart with arrhythmia. Science by cardio researchers in the style mentioned above could be done on this heart. Arrhythmia vs non-arrhythmia. It is important in the discussion of the heart to know whose heart we are discussing and who is discussing it because no knowledge happens in a vacuum. Is it a heart beating on its own (without a subject) in a laboratory? Or is it a heart in a subject, mine or yours? How does one access knowledge of that heart? Is a doctor listening to it through a stethoscope? Is there data from an electrocardiogram hooked up to the chest of the subject? Is general scientific knowledge of the model of a heart being used? When you present questions about knowledge of an arrhythmic heart we need to ask whose knowledge and whose heart. In addition, the concepts of abnormal and passive belong to human subjects. You could say that I’m interested in the plumbing of knowledge as well as of the heart.
We began with the question about how the subject gets to the object. Now let’s turn that question around. How does knowledge of an object happen independent of a subject, first? And if it does, how does it get “in” to the subject, any subject? There is no knowledge independent of a subject, say out in space. Once knowledge is held by the subject, it is, by definition, subjective knowledge.
Perhaps it would be wise at this point for me to ask you for your definition of objective. Let me explain. My notion of objective is that third person—all-seeing eye of god—omniscient view of all things. I get the sense, though, that your definition of objective knowledge is the historical and scientific consensus of subjective knowledge. Please tell me where we differ.
Finally, do you think that only humans can have knowledge?
Rodrigo:
I want to begin with your characterization of objectivity as a kind of God’s Eye View—omniscient, absolute, transcendent. It’s a powerful image, and one that Putnam rightly criticizes as metaphysically incoherent and epistemologically useless. But I think we should be careful to resist this image, and even more indignantly to resist Rorty’s related claim that “the desire for objectivity is just another version of the attempt at contact with the nonhuman”. Rorty reads this desire as a secularized theism—a yearning for finality, for dogmatic closure. Do I seem to you to yearn so?
There is a tendency in this critique to treat objectivity as an authoritarian aspiration—as if the desire to know truly were a disguised refusal to allow others their lived perspectives. Earlier you expressed this very resentment, if only in a friendly and mocking tone. But objectivity, as I understand it, has nothing to do with suppressing difference or claiming total access. It has nothing to do with perspectives at all! Perspective is how a subject encounters the world. But objectivity is not a kind of seeing. It is not a “view from nowhere”, nor is it the consensus of many somewheres. It is not a view, period.
Objectivity, in the context of perception or knowledge, is best understood as the independence of content and justification from the perceiving or knowing subject. A belief is objective when its justification does not depend on who holds it. To think objectively is not to reach outward toward some transcendent tribunal, but to allow the real to determine belief through correct conceptualization. What is objective in knowledge is not its perspective but its form—its fit with the real, under normal conditions of receptivity and rational recognition.
Once this is understood, subjectivity can be reintroduced—not as a threat to truth, but as its necessary medium. Knowledge occurs in subjects. Sensation is registered in the subject. Concepts are formed by subjects through experience and memory. The whole architecture of perception and belief is in the subject. But this does not mean it is of the subject, nor that it is cut off from the world. The “veil” that is so often imagined between subject and object—what you rightly evoke in your question, “how does the subject get to the object?”—is, on my account, a residue of idealism.
Rather than asking how the subject gets to the object, I would reframe the question: how does the subject unself, by rendering perceptual content objective? Not by denying his embodiment or context, but by allowing concepts that apply independently of his position to shape the content of belief. A mature subject learns to apply concepts that, while formed in experience, correspond to real universals instantiated in the world. Under normal conditions, reality literally informs the subject. There is no epistemic gap to be closed because the subject is already in the world, and the world already in the subject, through sensation.
As Iris Murdoch put it, “If quality of consciousness matters, then anything which alters consciousness in the direction of unselfishness, objectivity and realism is to be connected with virtue”. The task, then, is not to transcend subjectivity, but to unself—to allow the object, through proper conceptualization, to take its place in thought without interference from projection or ego. Objectivity is not a mystical detachment, but a kind of humility before what is given.
Normativity, here, is not the product of volition or evaluative stance. It is not a standard we create or choose. It is passive—functional, teleological, and natural. The mind’s recognition of form is not guided by norms in the deliberative sense; rather, it is norm-governed in the way that a heart is governed by what counts as a healthy rhythm. The norm does not direct; it defines. To conceptualize correctly is to do what the mind is for—not by command, but by function.
Finally, your question about animal knowledge. I think you were asking more than whether animals can know. I suspect you were probing whether knowledge can exist without conceptual subjectivity—without reflection, without rational structure. My answer is this: animals, in many cases, do have knowledge, because they interact with the world through perceptual and behavioral norms that track the real. But their knowledge is non-conceptual, non-propositional. It is neither justified nor unjustified, discursively. It is a form of successful orientation, not a cognitive endorsement. As animals, much of our own knowledge, if not all of it, also has this character.
But your deeper point, I think, was to press whether all knowledge must be held by a subject. And here I agree: it must. But the subject is not an ontological exception to the world. He is part of it—an object among objects, whose difference lies not in separateness, but in receptivity and reason. To know is not to leave subjectivity behind, but to let it become the medium through which the world speaks clearly.
So yes, knowledge is in the subject. But it is not about the subject. It is not made by him. It is received, conceptually organized, and judged. The more rightly this is done, the more objective it becomes. Not because the subject disappears, but because he no longer stands in the way.
Theral:
Beautiful response. That last paragraph, where you write that “to know is not to leave subjectivity behind, but to let it become the medium through which the world speaks clearly”—that’s lovely. I think we must be at the end of the conversation, for I am in agreement. Allow me a modest and final reply.
You insist—rightly, I think—that knowledge must occur in the subject, and that the subject is not outside the world, not ontologically exceptional. You reframe the question beautifully, asking not how the subject gets to the object, but how the subject “unselfs,” allowing the object to enter thought without distortion. There is a deep ethic here, an ethic of epistemic humility. And when you quote Murdoch on consciousness and virtue, I feel a resonance: a mature consciousness is one that gives way to the real, that learns to “see justly,” as she says elsewhere.
I see in this that knowledge could be on a spectrum. Perhaps the nematode worm has a “thin” knowledge, the cat “thicker,” and humans the “thickest.”
Doesn’t the very call to “unself” already presuppose a subject with particular capacities and vulnerabilities? Unselfing is not the function of a camera, or a thermometer (human creations). It is the work of a certain kind of subject—an agent who can be wrong, who can grow, who can, in Murdoch’s phrase, turn toward the good. Even to receive knowledge—to let the object “take its place in thought”—is, I would say, not only passive, but receptive in a deeply human way.
A writer who I admire recently admitted that she turns to the “telescopic view” during troubled times “such as these.” She is referring to Carl Sagan’s blue dot idea and image—the earth as an infinitesimally tiny dot in the universe. I’m fascinated by this conception of minimizing the human self to provide new meaning. Surely the self is one of human’s greatest, if not the greatest, creations. I appreciate the epistemic humility, but why must the self be diminished so? Why can I not feel my pleasure or pain when I learn? After all, why is the blue dot important? Why is it worth mentioning? It’s because that’s where we live, us humans. Why are you and I having this dialogue? We are not blank slates, and so the content of thought, even when it corresponds with the real, is shaped by a history of our lived meaning. As McDowell might say, the conceptual is already shot through with the lived. The apple is not mine. But to see it truly—to let it be what it is—requires everything I am.
It appears that the “subject” is not your enemy at all, the “self” is. I’m quite happy to leave it there, for a (future?) discussion of the self and identity will lead us to many new places.
Rodrigo:
Would that all philosophical conversations were so rich and fruitful! We have set a high standard to follow. Certainly I have learned something important which I will here lay out, and multiple avenues for further discussion are wide open and very promising.
We are agreed that knowledge occurs in subjects and that one way or another this fact conditions what passes for knowledge. The contents of thought, when rightly formed, are formed by the world. The more this formation is conditioned by the object, the less it betrays the distorting influence of the self. The distortion is not absent by default, however, and here lies the veil of perception.
That effort of self-removal, which Murdoch called unselfing, you have rightly asked: who can achieve it? Surely not the thermometer, the camera, the passive receptor, however sophisticated. No—only the one who can suffer, who can love. And here, Plato was right. In the Republic, knowledge is inseparable from justice in the soul--to know justice is to be just. And in the Phaedrus, the soul’s ascent toward truth begins in eros, a desire for union with what is. To know is to love the known. This is not in tension with reason or observation. On the contrary, love of the known makes reason possible. It opens the conceptual eye, and disciplines the self. Seeing an apple requires everything one is, yes. Not because the whole self actively contributes to the subject's passive receptivity, but because the self must be brought under control to avoid interference—an effort requiring everything one is. Seeing is passive, but becoming ready to see is not. The apple is there, but we are not ready-made to see it as it is, not as food, not as color, but as seed-bearing fruit.
And now to the veil. The veil of perception is not between subject and object, nor even between subject and self. The veil is the self. Not the minimal subject who sees, believes, and reasons, but the accumulated, historically contingent self—the tangle of enduring beliefs, desires, and identities. The self is an object like any other, but it obscures. It resists correction, guards its conceptions, and so, must be pierced. Piercing the self is a doubled metaphor. It names the epistemic act of countering distortion, and the ethical act of countering selfishness. It is not a rejection of the self, but its refinement. The self must be opened by the subject, using the very knowledge it contains: knowledge of reason, of the senses, and of the epistemic community to which it belongs. The self is both resource and resistance, while the subject, as agent, does the piercing.
Admittedly, that this is possible is a demanding intellectualism, but it is also rational life as we find it. In the concrete and specific, scientific discipline trains the self to stand aside. In the general and abstract, philosophical discipline contributes where science leaves off. Still, until we resolve the mind-body problem, the hard problem of consciousness, with a materialist solution, we remain epistemically bound to the self. That is, we do not yet fully understand how the world forms the contents of thought. In that ignorance, our beliefs about the most abstract questions, those concerning mind, value, or meaning, will be stamped with our biological and cultural situation. But this does not establish subjectivity as an essential feature of knowledge. It only shows that, for now, some of our knowledge remains justified externally, with its objective justifiers internally obscure. This is not epistemic subjectivity, but an externalist metaphysical realism of objective justification, a realism that knows where it stands, even when it cannot see its own ground.
Let me close with a sentiment we share: an unashamed attachment to the pale blue dot, to the inescapable contingency of our circumstances, some shared, some not. Seeing an apple, or anything else, for what it truly is, is demanding work. And I want to stand with you in the recognition that we are all works in progress. Transparent epistemic objectivity is no more given than are pre-conceptualized sense-data. Outside the most ordinary natural truths, the transparency of our external justifiers is an occasional achievement—typically we know, but do not know that we know. In its absence, we share a common condition: provisional epistemic opacity, mistaken for subjectivity.
References:
Adam Becker (2018) What Is Real?
George Berkeley (1710) A Treatise Concerning the Principles of Human Knowledge
Jerry Fodor (1975) The Language of Thought
Immanuel Kant (1785) Groundwork for the Metaphysics of Morals
John McDowell (1994) Mind and World
Iris Murdoch (1970) The Sovereignty of Good Over Other Concepts
Thomas Nagel (1986) The View from Nowhere
John Cooper, Plato (1997) Plato: Complete Works
Karl Popper (1972) Objective Knowledge: An Evolutionary Approach
Hilary Putnam (1981) Two Philosophical Perspectives
Richard Rorty (1985) Solidarity or Objectivity?
Wilfred Sellars (1956) Empiricism and the Philosophy of Mind