Kant on the Self
Critique of Pure Reason, Elements of Transcendentalism, subsection 8:
"The difficulty here lies wholly in this, how a subject can have an internal intuition of itself: but this difficulty is common to every theory. The consciousness of self (apperception) is the simple representation of the ego, and if by it alone all the manifold (representations) in the subject were given spontaneously, the inner intuition would be intellectual. In man this consciousness requires internal perception of the manifold, which is previously given in the subject, and the manner in which this is given in the mind without spontaneity, must, on account of this difference, be called sensibility. If the faculty of self-consciousness is to seek for, that is, to apprehend, what lies in the mind, it must affect the mind, and can thus only produce an intuition of itself. The form of this, which lay antecedently in the mind, determines the manner in which the manifold exists together in the mind, namely, in the representation of time. The intuition of self, therefore, is not, as if it could represent itself immediately and as spontaneously and independently active, but according to the manner in which it is internally affected, consequently as it appears to itself, not as it is."
To parse this out (and oversimplify it probably), I want to say: we have an introspective sense, which is literally a sense, not just analogous to the senses. This introspective sense looks at the mind, and thus affect it, and basically produces an intuition of itself by affecting the mind; the effect of the introspective sense on the mind is, itself, detected by the introspective sense, and this effect is given as consciousness of self when it is so detected.
Now, when Kant says, "the form of this, which lay antecedently in the mind..." I am going to assume that by "this" he means the intuition given when the introspective sense affects the mind. The form of the subject's intuition of itself, therefore, is time, which also determines the character of the manifold -- that is to say, the representation of time is the a priori form in the mind which (in fact) makes possible the intuition of self, and the intuition of self in turn functions as a premise from which to make a transcendental argument for time as an a priori form.
The last sentence is difficult. For Kant, we know nothing of the noumenal realm, only of the manner in which it affects us. The self is not conscious of itself, either, only of the manner in which it affects itself; we are part of the noumenal reality (Kant is not a dualist in the final analysis) but we don't have consciousness of the noumenal reality by that token, because we don't really sense ourselves, either; we sense only the manner in which we affect ourselves. The limits of cognition are thus cleanly drawn, terminating, both externally and internally, at the points at which cognition tries to touch the noumenal. If the eye peers far enough out, or deeply enough in, it beholds a mirror.
In this situation, one understands Faust's frustration: "And here I stand with all my lore, poor fool, no wiser than before."
"The difficulty here lies wholly in this, how a subject can have an internal intuition of itself: but this difficulty is common to every theory. The consciousness of self (apperception) is the simple representation of the ego, and if by it alone all the manifold (representations) in the subject were given spontaneously, the inner intuition would be intellectual. In man this consciousness requires internal perception of the manifold, which is previously given in the subject, and the manner in which this is given in the mind without spontaneity, must, on account of this difference, be called sensibility. If the faculty of self-consciousness is to seek for, that is, to apprehend, what lies in the mind, it must affect the mind, and can thus only produce an intuition of itself. The form of this, which lay antecedently in the mind, determines the manner in which the manifold exists together in the mind, namely, in the representation of time. The intuition of self, therefore, is not, as if it could represent itself immediately and as spontaneously and independently active, but according to the manner in which it is internally affected, consequently as it appears to itself, not as it is."
To parse this out (and oversimplify it probably), I want to say: we have an introspective sense, which is literally a sense, not just analogous to the senses. This introspective sense looks at the mind, and thus affect it, and basically produces an intuition of itself by affecting the mind; the effect of the introspective sense on the mind is, itself, detected by the introspective sense, and this effect is given as consciousness of self when it is so detected.
Now, when Kant says, "the form of this, which lay antecedently in the mind..." I am going to assume that by "this" he means the intuition given when the introspective sense affects the mind. The form of the subject's intuition of itself, therefore, is time, which also determines the character of the manifold -- that is to say, the representation of time is the a priori form in the mind which (in fact) makes possible the intuition of self, and the intuition of self in turn functions as a premise from which to make a transcendental argument for time as an a priori form.
The last sentence is difficult. For Kant, we know nothing of the noumenal realm, only of the manner in which it affects us. The self is not conscious of itself, either, only of the manner in which it affects itself; we are part of the noumenal reality (Kant is not a dualist in the final analysis) but we don't have consciousness of the noumenal reality by that token, because we don't really sense ourselves, either; we sense only the manner in which we affect ourselves. The limits of cognition are thus cleanly drawn, terminating, both externally and internally, at the points at which cognition tries to touch the noumenal. If the eye peers far enough out, or deeply enough in, it beholds a mirror.
In this situation, one understands Faust's frustration: "And here I stand with all my lore, poor fool, no wiser than before."
Comments (21)
Markus Gabriel sets it out in the clearest manner I know: "The self becomes an object among others as soon as it is drawn within the sphere of representation. Kant developed this problem in his First Critique and his argument is as plain as it is striking. If the self was a substance, our cognitive access to it would have to be the grasp of a substance. Yet, our cognitive access to any substance is fallible insofar as it has to represent the substance in question. Even if we represent ourselves, the represented self is not identical with the representing self given that the subject of experiencing is never identical with any possible object of experience. Whatever the object of our scrutiny may be, it has to become an object among others whereby it is determined as such in a wider context." (Gabriel, The Mythological Being of Reflection).
Kant himself famously speaks of "this I or he or it (the thing) that thinks", and in the Second Critique Kant describes what would happen if the self were to access 'itself' in its noumenal dimension: "God and eternity in their awful majesty would stand unceasingly before our eyes... the conduct of man... would be changed into mere mechanism, where, as in a puppet show, everything would gesticulate well but no life would be found in the figures". Another way to flesh this out is that this inability to access itself is not a mere 'impediment' to subjectivity, but a consitutive condition of our being subjects. It's this last implication - I think perhaps the most important - that is often forgotten.
Zizek: "The basic gesture of Kant's transcendental turn is thus to invert the obstacle into a positive condition ... In the standard Leibnizean ontology, we, finite subjects, can act freely in spite of our finitude, since freedom is the spark which unites us with the infinite God; in Kant, this finitude, our separation from the Absolute, is the positive condition of our freedom. In short, the condition of impossibility is the condition of possibility." Elsewhere: "It is therefore not that Kant simply limited causality to the phenomenal domain in order to be able to assert that, at the noumenal level, we are free autonomous agents: we are only free insofar as our horizon is that of the phenomenal, insofar as the noumenal domain remains inaccessible to us" (Zizek, The Ticklish Subject).
To reiterate in a slightly different way: you know that you have an introspective sense because it affects you, and you are cognizant of those effects because you have an introspective sense. But the eye never quite turns around and sees itself, per Wayfarer's reference. This is circular, but if you begin with the premise that we only know the world by the way it impacts is, then you're pretty much guaranteed to end up with something like this.
(For Kierkegaard, our freedom results from our finitude, but the freedom is also the aspect in which we are most like God.)
One is tempted to postulate that Buddhism is basically a method for stopping the cycle by basically sabotaging it. But that's a completely different discussion.
This leads to an understanding that the real identity or 'ground of being' in ourselves (however you want to conceive of it) is unknowable - while at the same time, being right in the middle of things, so to speak. Of course that will temperamentally appeal to some, and annoy many others; we would like to think we know ourselves, and this seems to undercut that.
There's a good discussion of this on Eric Reitan's blog, in the context of a series of posts on Kant and naturalism. Specifically:
Remainder here.
The important thing is to distinguish between treating this inability to 'see oneself' as a mere epistemological limitation, and between treating it instead as an ontological condition of subjectivity as such. Hence: "It is not sufficient to say about the I of pure apperception that "of it, apart from them [the thoughts which are its predicates], we cannot have any concept whatsoever" (CPR, A 346). One has to add that this lack of intuited content is constitutive of the I; the inaccessibility to the I of its own "kernel of being" makes it an I ... I am conscious of myself only insofar as I am out of reach to myself qua the real kernel of my being ("I or he or it (the thing) which thinks")." (Zizek, Tarrying with the Negative).
With respect to the metaphysical tradition - and this feeds into the discussion regarding the 'ground of being' etc, the point is that this constitutive condition is what makes Kant's philosophy properly 'Critical' and not 'dogmatic': "This is what Kant's theory of metaphysics ultimately is about: metaphysics endeavors to heal the wound of the "primordial repression" (the inaccessibility of the "Thing which thinks") by allocating to the subject a place in the 'great chain of being.' What metaphysics fails to notice is the price to be paid for this allocation: the loss of the very capacity it wanted to account for".
At any rate, there is no "I" without others, so in that sense, seeing oneself for what one is cannot be done alone, yet I do think that it can be done, unlike the Noumenal realm and the phenomenal realm. I am not just my body. Rather, I am the sum total of my own thought and belief.
So self-relation is a weird thing, right, because relation implies difference. The most basic form of relation is: there's something here, x, and there's something over here, y, and they relate like this, z. Maybe x (house) is related to y (basement) in a relation of z (being above it).
If you want to relate to yourself, you have to introduce a difference. There's no way around it. To have a relation is to have all 3 elements: the two things related, and the relation itself. If you want to relate to yourself, you have to cleave yourself. 'Relation,' as used here, is cognitive. It's a way of organizing two discrete elements.
What Kant's doing in the passage quoted is treating self-relation as analogous to the relation between that which represents and that which is represented. The subject is less a thing than an organizing principle + invisible locus to which everything appears. If something is (in the mundane sense) it is only insofar as it appears. And if something appears, it must appear to [ ] -- the perspectival space to which things appear. If you put a 'something' in that space, you initiate, immediately, an infinite regress - and here is where Wayfarer (& Sartre but no one likes him anymore) pop up (and are right, I think.) What we have here is a kind of paradox, the 'solution' to which is obvious to almost anyone who hasn't gotten hamstrung, tragically, by anxious philosophical handwringing. The solution: What we are is not an object. We are, in essence, a pure translucence which, if you let the sediment of thought settle, will be as clear as clear water. Or, put differently (Heideggareanly) we are the 'there' itself, in which being takes place.
Yes, but....
So take ourselves as object-- as something that appears, in the perceptual field, to that which perceives.
If we were to appear to ourselves - if we somehow became the sort of thing that could appear - we would have to appear like, well, things that appear - in spacetime, in accordance with the categories etc. In a word, what street said: by allocating to [ourselves] a place in the 'great chain of being.' In other words: We would be trying to render the condition as something conditioned.
But the key thing is, whether you're Buddhist or Zizekian, whether you see a 'wound' or ?tman, no matter how you approach the paradox: We *do* appear. To others and ourselves. We do appear. Profanely, yes, and in a limited way. But, tragically or wonderfully, we take a place in 'the great chain of being'. We get a name, we get a role (first in our parents relating to us, then as 'it' in a game of tag, then as position x in sport y and so forth, up to our job).
That's the thing of growing up in a community - you become object, at least partially.
But it's another question, entirely, of whether and to what degree our innermost selves actually are inaccessible. Is inaccessibility a condition of our subjectivity? Well, idk. We're both who we are, and also able to relate to a web of knowledge and representation in which we play a part. We can relate to ourselves abstractly while being ourselves concretely.
So, again, is the inaccessibility of self a condition of our subjectivity?
It feels right, rhetorically; it feels like something you might like to say, depending on your mood. and so Zizek does, often and everywhere. I think there's an argument for it, somewhere, but it hasn't been presented here, and it's hard to find, precisely, in Zizek.
I guess it all hinges on what it is, exactly, thats inaccessible.
We are both objects in the world and subjects taking an account of it and/or ourselves... I think that the dichotomy of subject/object is fraught...
Now, if we introject the subject/object condition/conditioned dichotomy into our core then sure, the self's access to itself is the paradox of two mirrors facing one another. But that mirror paradox isn't some essential metaphysical feature of the self, such that the self is a wound etc.
The weird regress that arises is more like an epistemological error message popping up saying ' System Crash due to: you're trying to apply your Object Cognition program in a way that doesn't make sense at all'.
So I think the Buddhist tack is closer here. (I'm not as confident with Sartre, but I believe he's making the same point when he talks about 'nonpositional consciousness')
It's been a while since I've read either the second critique or Zizek, so maybe the context bears out the interpretation you've presented.I'm not sure. But I do think it's a step to go from there to the idea of the self as constitutively unknowable. The terms are all very wobbly here. What are we talking about when we talk about the self as unknowable? Do we mean, simply, that we don't know what we'll do in the future? But that isn't even folk wisdom, it's folk being a person, and is entirely compatible with self-acquaintance. I think we do kinda know ourselves, and a lot of times we don't do, but we can learn to know ourselves, and that whole spiritual process is well outside what, I think, Zizek is talking about (which is structural)
I have a somewhat different take on this. I am mostly relying on Kant's discussion of causation in the Third Antinomy.
As an embodied agent, the self is both a phenomenon (more precisely, an empirical substance) and, well, an agent. This duality doesn't correspond to two different concepts that one can subsume oneself under (there is just one relevant concept in the vicinity: that of a human being, or person) but rather two different stances: empirical and agential. So, I would surmise that knowing oneself is, indeed, knowing what one will do. But this can be achieved from two different stances. From the empirical stance, the agent is subsumed under laws of physiology and (empirical) psychology. It's not so much that her behavior is predictable (which it may very well be, given sufficient information about her past and present circumstances and material constitution), but rather that it is determined by the law-governed chains of causation that she is empirically embedded into (necessarily so, according to Kant, because of the a priority of the Categories of substance, causation, etc.)
On the other hand, the agent who deliberates what to do pictures herself to be initiating new chains of causation that don't have necessitating causal antecedents reaching into her past. The reason for that is because the agent who is deciding what to do must reflect on the reasons why she would do it, and this knowledge, grounded in valid principles of practical deliberation, is free from empirical considerations about physiology and psychology. For sure, one can predict that one is likely to behave irrationally, in some circumstances, owing to some merely empirical fact about oneself (e.g. that one is lazy, or a glutton, say). But such empirical considerations can only constitute excuses or grounds for blame, not reasons for doing what one does. When one does know what one will do (as may occasionally happen) because of the reason one can cite as a good (rational) justification for doing it, then one thereby expresses knowledge of what it is that one will do and, also, at the same time, spontaneous knowledge of the intelligible ground of one's decision and action. This is a form of self-knowledge that is irreducible to empirical self-knowledge (such as knowledge on one's psychological tendencies). This intelligible object of self-knowledge (i.e. one's reason for acting) is the ground that was missing when one attempted to predict what one would do on the mere basis of antecedent empirical facts about oneself. (In that case, what was predicted wasn't intelligible behavior at all, but mere bodily motions).
Offering reasons for one's intended actions thus exemplifies what Kant calls the intelligible character of causation, which he contrasts with the empirical character of causation. The former is potentially gained from the agential stance (given an ability to deliberate rationally) while the latter is potentially gained from the empirical stance (given sufficient knowledge of physiology and psychology). Since angential self-knowledge (from spontaneity) is a form of self-knowledge that is free from empirical self-determinations, this might justify the idea that it is knowledge of the 'noumenal self', where the term 'noumenal' may be construed negatively to signal that its ground is free from empirical determinations, and can also be construed positively to point to the intelligible character of causation which is at play in rational agency. (On that view, 'noumenal ground' and 'intelligible ground' are synonymous).
That's an interesting point.
I think it's actually the other way round, the self never was anything other than an object, for others; we introjected the self-as-object, and instead of being the self that we actually are (the self of pure consciousness, pure kenning, pure knowing), we think of the perceived image as ourselves.
IOW the first "home" of the self is the image others have of me, of me as a public, perceivable object. And the link is that we can see bits of ourself (e.g. I can see my hand, torso, or I can see an image of myself in a mirror) - but we can't see the eye that we're seeing out of. But subsequently, that self somehow gets from the outside world into our heads.
Another way of looking at it: in public language, when both A and B are in purview, A is a distinct self from B, this is something anyone can perceive, the relationship between them is flat, horizontal, on the surface, we can see it all at once laid out before us, and if A is conscious of B, that means A is evidently aware of it, can step around it, avoid it or cleave to it.
But in private, in that totally imaginary creature "subjective" space, we turn that schema <A is conscious of B> around 90 degrees and put it on our face, and thereby we create an imaginary A "inside" of us looking out at B (intentionality). We double up the meaning of self - so it's not just our self as others see us now, but also some imaginary point of view at the center of us, looking out of the body in some way.
And then, even worse, we imagine that that inner self perceives not the world directly, but a screen of its own mental material.
So. I am familiar with the distinction you're drawing, and I think the distinction is fair. The space of reasons is different than the space of [bodies clunking into one another, so on and forever]
The distinction is clear and i get it. So two ways to go at this, one more relevant to the one above, and one more relevant to the one i care about
(1). The space of reason, the intelligible waltz of justification and rationality and all that - that could be chalked up to an aesthetic gloss on a more primordial clunk and clatter of bumping bodies. The chain of reason and justification could be an emergent cloak of rationalization to drape above the din and clatter.
Now, I personally don't believe in (1). I think it a fair and justified fear and criticism, but I don't believe it.
(2) The problem with this secondary, autonomous, order is that it offers no incentive, one way or the other, to do this or that. Absent a new form of incentive, it's hard to see anything but an retroactive reframing of material decision in more stately terms. You're balancing a moral mechanics on a mechanical mechanics. And so the thing you're looking for immediately slips away.
A form of rationality, like this, I feel, will inevitably fall back into the lower-level system from which it came from.
What you need to make it actually constitute a system of its own is a value in which it believes.You need a third level. An ethical way of life is, precisely, not knowing one what will do. That is the linchpin of the whole thing. If you conflate knowing what one ought do with one knowing what one will do, you lose humanity in a blink. Ethicality, properly, is valuing something as a value in itself, and knowing that even though you know this or that is the right thing, not knowing what you'll do.
This gap between knowing what you ought do and not knowing what you will do --
That's where all the 'good stuff' ethically and spiritually and even aesthetically speaking comes in
NB - I'm not being romantic. I'm talking about structural differences
I agree. I am very much a neo-Aristotelian about practical reason. Hence, on my view, ethical motivation is grounded in virtue of character and the "space of reason" merely signifies the forms of justification that belong to an ethically informed form of life. I don't view reason and morality to be separable at all, not even notionally. The severely impoverished forms of practical rationality being studied formally by some economists and rational decision theorists only can be called "rationality" by analogy, on my view.
Also, my characterization of knowing what it is that one will do because it is what one ought to do is mostly indebted to Elizabeth Anscombe and, though her, to Aquinas and Aristotle. I am suggesting that there is a reading of Kant (or of some strands within Kant's practical philosophy) under which he is much closer to Aristotle than he is usually made out to be.
There is nothing of the self that is either object or subject... all of it is both, in some way. Thus, the dichotomy is out of it's bound to start with.