Twelve Rounds with Kant (Part 6)

Ellsworth Kelly, Grid Lines (1951)¹

Today, in my ongoing battle with Immanuel Kant's Critique of Pure Reason (1781–87), I would like to spend some time examining the I think.

No, that isn't a typo. What Kant calls "the sentence: I think" (der Satz: ich denke) might be—and I suggest this with no authority whatsoever—treated as the hidden thirteenth item in his catalogue of pure concepts of the understanding. The importance of the I think to the Kantian scheme cannot be understated. All of the other categories are predicated on it; Kant calls it "the formal proposition of apperception."

[T]his concept is the vehicle of all concepts in general, and therefore also of transcendental concepts, and...it is therefore always included among them, and so is itself transcendental; but...it has no claim to a special title, inasmuch as it serves only to introduce all thought as belonging to consciousness.

Kant expands on the importance of the I think in the Transcendental Dialectic (excerpted above), but first introduces the concept in the Transcendental Deduction:

It must be possible for the I think to accompany all my representations: for otherwise something would be represented within me that could not be thought at all, in other words, the representation would either be impossible, or at least would be nothing to me. That representation which can be given prior to all thought is called intuition, and all the manifold of intuition has, therefore, a necessary relation to the I think in the same subject in which this manifold of intuition is found. This representation (the I think), however, is an act of spontaneity, that is, it cannot be considered as belonging to sensibility. I call it pure apperception, in order to distinguish it from empirical apperception, or also original apperception, because it is that self-consciousness which, by producing the representation, I think (which must be capable of accompanying all other representations, and which is one and the same in all consciousness), cannot itself be accompanied by any further representations. I also call the unity of appereception the transcendental unity of self-consciousness, in order to indicate that a priori knowledge can be obtained from it. For the manifold representations given in an intuition would not one and all be my representations, if they did not all belong to one self-consciousness. What I mean is that, as my representations (even though I am not conscious of them as that), they must conform to the condition under which alone they can stand together in one universal self-consciousness, because otherwise they would not one and all belong to me.

Well: what are we supposed to make of this?

Ellsworth Kelly, Seine (1951)

For one thing, the I think consolidates subjectivity's position at the substratum of objectivity in Kant's epistemology. We already know that Kant believes that the mediate nature of sensory experience necessitates an organizing principle within the perceiving organism that arranges the incoming data into a coherent structure; it is this organizing principle (or set of principles) from which the appearance of natural order arises (according to Kant). The I think underscores that this organization occurs within the organism, not in the unknowable world-in-itself.

We can think of a cartoon caveman grunting and saying something like "find rock, make spear," and know what he's really saying is "I intend to find a rock and then I will make a spear." Rephrasing it in this way only makes explicit the actor in the situation. The I think does something similar: emphasizes the I in every statement a person can possibly make. 

"The sky is blue" can be understood to mean "I am looking at the sky and I see that is is blue." If it is nighttime, if the speaker is standing indoors, or if the visual stimulus of a blue sky is otherwise absent, the statement may instead be interpreted as "I observe and report that the verbal stimulus 'sky' produces in me a response of conditioned seeing in which I faintly perceive the stimulus quality 'blue'" or "I have been reinforced for saying 'the sky is blue' during occasions in which a blue sky is not present, such as when it is appropriate to state with brevity something of the painfully obvious without regard for context," or "I have been overtly or subtly corrected [punished] for stating that the sky is any color other than blue," etc.

In any case, what Kant wants to draw attention to is that there's an I implied in anything anyone can say about anything, since all communicable knowledge of reality presupposes and depends on the existence of a self-conscious individual experiencing the world. To use Kant's phrasing, our identity is "inevitably present" in every occasion. Even when we're talking about something so impersonally precise as mathematics, we don't exclude ourselves, despite the elided language we use. We say "two plus two equals four" and not some onerous formulation like "I declare that on every occasion in which I have added two of something to two of something, the sum at which I have arrived has always been four, and I understand this to be universal and absolute"—even though that's actually what we're saying.

Four stones on the ground are going to be what they are, regardless of the presence or absence of an observer—but uniting them as members of a class (despite the non-identity of each rock to the other three) and quantifying them to the exclusion of any other objects in the vicinity, or of any other rocks anywhere else in the world, necessitates an actor. Two plus two always and everywhere equals four, but only when there's someone doing the counting.

Ellsworth Kelly, Curve (1951)

A few of Kant's remarks on time from the Transcendental Aesthetic sketch out this concept without getting too technical:

Time is therefore merely a subjective condition of our (human) intuition (which is always sensible, that is, so far as we are affected by objects), but in itself, apart from the subject, it is nothing. Nevertheless, with respect to all appearances, that is, to all things which can enter into our experience, time is necessarily objective. We cannot say that all things are in time, for in the concept of things in general we abstract from all modes of their intuition and therefore from the very condition under which time belongs to our representation of objects. If, therefore, the condition is added to the concept, and if we say that all things as appearances (as objects of sensible intuition) are in time, then the principle has full objective correctness and a priority universality.

Irrespective of whether Kant has succeeded in convincing us that time has no reality as a thing-in-itself, it is undoubtedly true that the reckoning of time depends on the (subjective) experience of an observer perceiving events elapsing in succession. We can say roughly the same thing about mathematical knowledge: two plus two equal four is objectively true, it will always be true, but the reason for its universality is found in the organizing principles of understanding (which Kant assumes are identical for every human subject).

Kant goes deeper into the weeds in the B Deduction, where he distinguishes between the objective and subjective unities of self-consciousness. The latter is the continuous play of sense impressions which operate upon us to produce intuitions and their combination with concepts. The objective unity of self-consciousness is conceptually congruent with transcendental apperception, the continuous, consistent, and ineffable experience of a thinking self as such: we might call it the essential form of consciousness, irrespective of the content of thought and sensation. Whereas what a person might think or do when presented with some particular thing in his environment is a matter of the subjective unity of consciousness and wholly contingent on circumstances; but the bare fact of a conscious person encountering a sense-object and meeting it with thought concerns transcendental apperception and the objective unity of self-consciousness:

[T]he pure form of an intuition in time, merely as intuition in general containing a given manifold, is subject to...this unity simply through the necessary relation of the manifold of intuition to the one, I think, and hence through the original synthesis of the understanding which forms the a priori ground of the empirical synthesis. The original unity alone is valid objectively; the empirical [subjective] unity of self-consciousness, which we do not examine here, and which moreover is only derived from the former under given conditions in concreto, has subjective validity only. One person connects the representation of a certain word with one thing, the other with another; and the unity of consciousness in which is empirical is not, as regards what is given, necessary and universally valid.

Honestly, yeah: the language is awkward to untangle. Objective knowledge assumes a subjective actor, whose subjective unity of self-consciousness occurs against the background of an objective unity of self-consciousness.

Speaking of language.

Given Kant's jaundiced views regarding animal intelligence, (i.e., animals do not possess consciousness, but a "disunified" and "obscure" awareness), he would readily admit that the I think applies only to human experience. To a certain extent, he's probably right. It's hard to imagine an animal without language in its repertoire being capable of responding to the subtleties of its bodied existence and articulating its implications.

Kant doesn't explicitly examine language's role within the understanding, but verbal behavior can certainly be said to underlie any epistemology in which I think occupies a position of central importance.   

BF Skinner speculates that introspective tendencies are the product of social environments that encourage or demand self-descriptive behavior. Reporting one's current state in more detail than "OW" or "HUNGRY" or "SAD" requires a community that (1) that teaches its members the language to convey information about their own feelings (2) teaches its members the language with which to discriminate between degrees and qualities of sensation (i.e., differentiating between being "angry" and "upset"), (3) provides incentives for its members to inquire of and apprise each other of their mental and emotional states, which also encourages self-observation.

The self-consciousness which Kant takes as a given in his writings depends on the self-conscious person being a verbal person. Our experience of the world would be very different otherwise. We're so profoundly language-conditioned that we constantly emit low-level verbal responses to our environment. When we perform a visual sweep of the objects in our bedroom, we're very weakly saying "shelf," "book," "carpet," "wall," etc. because our history is replete with occasions where we've been reinforced for effective communication with regard to these objects. As we "speak" to ourselves, we also "listen" to ourselves—all "in our head," as they say. The constant presence of speaking self to which we are an absolutely captive audience invariably hypostatizes these internal events as a thing called "mind." As we experience our own entities more often as quiet speakers-to-ourselves than as agents acting upon the external world, the emergence of the belief in the "thinking self" as the "true self" or "soul" should not be wondered at.²

In this regard, Kant's I think attests to the preeminence we typically ascribe to the "inner voice." He could have conceivably rephrased the remark about the I think's role in his scheme which we quoted above—"it serves only to introduce all thought as belonging to consciousness"—as something along the lines of "...as all qualia/experience as belonging to consciousness." As currently worded, Kant's statement stresses on the "inner dialogue" which characterizes some—but not all—cognitive behavior.  

If you read closely enough to have noticed that Kant categorizes the I think as a representation, and have been paying attention to the meaning of the term in Kant's employ, you might ask if that means that our perceptions of self-consciousness is a matter of mere appearance, just like our experience of reality. Kant's answer is yes, definitely: we can only know ourselves as we appear to ourselves. Perhaps without intending to, Kant dangles a string in front of the epiphenomalists and invites them to give it a tug.

On the other hand: it's more likely that Kant was fully confident that any number of materialists could give the yarn as many yanks as they'd like, and still fail to unravel the mind.

Which brings us, once again, to behaviorism.

Ellsworth Kelly, Light Reflection on Water (1951)

I tend to be suspicious of what Skinner derided as "mentalism"—the tendency to situate the "mind" or "soul" at the nexus of human events. On the face of it, this makes behaviorism and Kantian epistemology incompatible in the terms in which they are usually articulated. As you can imagine, I come to my reading of Kant with a bit of a chip on my shoulder.

Behaviorism does not deny qualia or subjectivity; it simply does not include them in its universe of discourse.³ The blind spot is as deliberate as orbital mechanics' exclusion of the possibility of a creator deity. This is an empirical science, and "mind" consistently escapes observation (except in cases of an investigator examining his or her own subjectivity). The methodology is made more efficient by dispensing with the unobservable and unquantifiable. But that doesn't mean that the experience of apperception doesn't matter, especially in our relations with others. The word we commonly use for someone who doesn't regard other people as thinking and feeling actors is "sociopath." Any sane behaviorist appreciates the limits of his discipline: the human as an object of study vis-à-vis the physical sciences is not the complete person, fully considered.⁴

I'm sure I've used this metaphor elsewhere, but I think of behavioral science's relation to more introspective disciplines as analogous to the contradiction between general relativity and quantum mechanics. Our behavior is determined by events and variables described in the physical sciences, and we experience the world in terms of conscious selfhood and qualia—but to the best of my knowledge, we have not developed a discourse which can incorporate both without contradiction. Behaviorism excludes from its universe of discourse what is fundamentally human about, well, being human, but in doing so it brings the organism Homo sapiens into the same conceptual sphere as everything else described by the physical sciences. Introspective disciplines center subjectivity and speak to the form of experience, but invariably engenders suppositions of mind/body duality that end up nominating subjectivity qua subjectivity as a causal force in the world.

None of this is a problem if your beliefs admit of things that transcend the material world while coexisting with it. But if you're like me and can't reason you way into such convictions, then you're stuck between a rock and a hard place, and are made aware of your position whenever it occurs to you that your physicalist cogitations leave something to be desired when your cogitations themselves are their own subject.

In the paralogisms of the Transcendental Dialectic, Kant picks apart a few culturally engraved concepts about the soul, a word which we might take to mean "self-essence" and is related to (but not entirely fungible with) consciousness. We'll probably peer at the Transcendental Dialectic in the next (and final) post about the Critique of Pure Reason, but it will suffice to say here that it puts to use the epistemology elaborated in the Transcendental Analytic and analyzes such ideas as are (apparently) inevitable in human culture, such as the temporal and spatial limits of reality, the existence of a supreme being, and the nature of the soul. To make a (very) long story short, Kant deems these concepts outside the horizon of human knowledge: we are not able to say anything about them with certainty, and we're fooling ourselves by trying.⁵

Apperception is something real, and its simplicity [irreducibility] is already contained in its possibility. In space, however, there is nothing real that is simple; for points (the only simple things in space) are merely limits, and not themselves something which, as a part, serves to constitute space. From this it follows the impossibility of explaining the constitution of myself, as merely a thinking subject, in the terms of materialism.

Remember, Kant had it out for scientific materialism on the basis of its potential consequences for public and private ethics.⁶

As, however, in the first proposition [I think] my existence is taken as given, for it is not said in it that every thinking being exists (this too would predicate too much of them, namely, absolute necessity), but only that I exist as thinking, therefore the proposition itself is empirical, and contains determinability of my existence merely with regard to my representations in time.⁷

I'm carping on terminology, but I would amend the statement "only that I exist as thinking" to a formulation of the same idea that connotes less of mentalism and places greater emphasis on qualia, which are the authenticating tokens exchanged between our selves-as-responsive-systems and ourselves-as-responsive-systems. (The difference in meaning between "thought" and "quale" may be subtle, but nuance matters here.)

But as for this purpose I again require first of all something permanent, which, insofar as I think myself, is not given to me at all in inner intuition, it is quite impossible, by means of this simple self-consciousness, to determine the manner in which I exist, whether it be as substance or as accident. Thus, if materialism is inadequate to explain my existence, then spiritualism is equally insufficient for this purpose; the conclusion is that in no way whatsoever can we know anything of the constitution of our soul, as far as the possibility of its separate existence in general is concerned.

Even though Kant's talking about the immortality of the soul here, these remarks pertain to consciousness in general, and to its exasperating unaccountability.

It is easy and accurate enough to say that I am wholly a physical organism possessed of a complex structure in a state of perpetual alteration as my physiology receives and responds to external stimuli and regulates its interior activity on scales ranging between organ systems and the interactions between individuals and groupings of cells. I can say with full confidence that there is no such object in me as a mind; what I experience as moods, mental states, thoughts, and sensations are an intricately layered and labile manifold of relations of relations to relations within my body. And I maintain that this is true—but none of it comes close to articulating or explaining the forms of what I seem to experience, its force and vivacity (to borrow Hume's phrasing), or its clarity. Measurable neural activity accompanies and causes what I experience as thought, but this activity is not thought as it appears to me, the thinker. Why should I have a "mind's eye" in which a neurochemical reaction is transduced into an image only I can "see"? Why should pleasure and pain feel precisely as they do? Why should I be capable of wonder and dread? Why am I not simply a golem that mechanically (but complexly) responds to the world without any awareness of it?

Ellsworth Kelly, Spectrum Colors Arranged by Chance II (1951)

I can and do say that while I am confident that the Cartesian mind is a fallacy and that consciousness is not a thing, I am at a loss to situate qualia in the material scheme of the cosmos. I'm groping after the ineffable, and the effort's futility is made all the more aggravating by the intimate propinquity of what it is I'm trying to pin down.⁸

Years ago, I'd get in arguments with my friend Jen about the existence of the soul. Some of these talks verged on acrimony when I dug in my heels and affirmed my non-belief in it. I think if I were to renew the debate with her today, I'd have to cede some ground.

No, I don't believe that there's some non-physical essence which permeates my corporeal body and identifies me as eternally myself. I don't believe that I'm anything more than an active pattern of meat made of atoms, nor do I believe that this meat moves and changes independently of the immutable laws which dictate the behavior of all matter and energy in the universe. But in the relation of this meat to itself exists something truly peculiar, even miraculous—even if "exists" is a problematic word for something without substance or force, and which is unobservable beyond the bounded event of itself relating to itself. 

I suppose I'd be comfortable metaphorically calling the ineffable experience of oneself the soul. I don't say this as a statement of belief, but rather as an expression of a qualified agnosticism. Probably my time with Kant has helped me to better appreciate the merits (and inevitability) of such a position.

1. Today's abstract visual aids for our abstract disquisitions come from Ellsworth Kelly. He kind of gets on my nerves, but I like him anyway.

2. Remember that the self is also a listener. The double role of the self as both speaker and listening community member is analogous to the superego of Freudian psychology.

3. Aside from Karl Marx, I cannot imagine any thinker has been as uncharitably misinterpreted by people who've never actually read his work as BF Skinner. 

4. Bear in mind: it's another human being who's doing the considering. The way we behave towards each other face-to-face is (usually) very different from the way we handle other people in the abstract, or at a mediated distance. Obviously socialization plays a role here, but there seems to be evidence that empathy is part of our genetic endowment.

5. If you recall, we read an except of Alfred North Whitehead's The Function of Reason (1929) that criticizes Kant's philosophy for exactly this.

6. And on this point, Whitehead is in complete agreement with Kant.

7. The "first proposition" to which Kant refers belongs to a table of statements extrapolated from an analysis of the I think:

I. I think,

II. as subject,

III. as simple subject,

IV. as identical subject,

in every state of my thought.

8. There is no eluding the ineffable. In developing any conceptual model of the world, we can only choose where to stow the black box in which it resides. Kant placed it in the world external to sense; I've given it over to selfhood.