The "I think" must be able to accompany all my representations

Kant's "I think", the unity of consciousness (transcendental apperception)

Immanuel Kant 1724 · 1804Set the limits of reason; act from duty.

What it claims

The unity of a single consciousness is indeed the condition for my representations to be mine, but we cannot conclude that the "I" is a known substance: neither that it is a simple soul, nor that it is the same person through time, nor that it is immortal.

Why

My representations are mine only if one consciousness unites them, but that unity does not let me know what I am.

In short

"The "I think" must be able to accompany all my representations" is an argument by Immanuel Kant (Critique of Pure Reason, 1787, § 16 and Paralogisms) on the question "Who am I, really?" (answer: my consciousness, my memory). Its claim: The unity of a single consciousness is indeed the condition for my representations to be mine, but we cannot conclude that the "I" is a known substance: neither that it is a simple soul, nor that it is the same person through time, nor that it is immortal. According to AskPhilosophy's logic check (Z3 solver), the argument is valid only if one grants a premise the author leaves unstated (P4). AskPhilosophy presents 3 objections, 2 of them with a documented reply, and the original text behind each premise, checked word for word.

Critique of Pure Reason, 1787, § 16 and Paralogisms

Valid with an implicit premise

Your thoughts, your sensations, your memories: what makes them all yours, and not ten different people's? Kant answers: they are tied together in a single consciousness, and an "I think" must be able to accompany every one of them. Without this unity you would have "as many selves as representations". But note: Kant does not say this "I" is a thing we know, a simple and immortal soul. For him the "I" is a form of thought, not an object we observe. So he keeps from Locke the idea that consciousness makes identity, but he refuses the metaphysics of the soul.

click a premise to reject it

If you accept every premise, logic forces you to accept the conclusion. So the debate is about the premises.

Checked by a logic solver: the premises cannot all be true while the conclusion is false.
Without the implicit premise: Without P4, the computer finds a case where the unity of consciousness is a form of thought, where I nevertheless have an intuition of myself, and where I therefore know a substance. The conclusion is then false.

The argument is valid with P4. Without it, it no longer follows: one can imagine that the "I think" is a mere form of thought and that I nevertheless have an intuition of myself, and then nothing would prevent knowing the soul as a substance. That is why the Paralogisms insist: the "I" contains no intuition, it is "empty of content".

Analogy. Think of the lines in a notebook: they let you arrange everything you write, and nothing would be readable without them. But the lines are not one more page of writing: you cannot read them as text. The "I think" is like those lines: it organises everything, but it is not itself a content we could study as an object.

Objections

Each objection is a conversation: click to see who replies and where the debate stands.

Hume 1739 · attacks premise 2

When I look into myself, I never find a "self": only one perception or another, heat, cold, pleasure, pain. I can never catch myself without a perception. So there is no unity to presuppose: the "self" is only a bundle of perceptions that follow one another.

Kant replies

see the argument

Kant Reply drawn from the Deduction (1787)

The consciousness that accompanies different representations is, in itself, scattered and unrelated to the identity of the subject. This relation does not come from each representation being accompanied by consciousness, but from my connecting them and being conscious of doing so. Otherwise I would have as many selves, of as many colours, as I have representations.

The debate goes on (our summary) Kant does not name Hume in this passage: it is the answer his text allows. It grants Hume that the self is not found as one more perception, but it holds that the very idea of a bundle supposes a single subject gathering the perceptions. A Humean can answer that this unity is explained by links among the perceptions, without a separate subject.

Where the debate stands (our summary)Disagreement of substance: is the unity of consciousness a condition that must be presupposed (Kant), or a result to be explained by the links among perceptions (Hume)? Still debated.

Descartes 1641 · attacks the conclusion

I am certain I exist as long as I think. And I find that thinking is the one thing that cannot be separated from me. So I am, precisely speaking, only a thing that thinks: a substance, a mind, and not a body.

Kant replies

see the argument

Kant Paralogisms, B404

The "I" represents nothing more than a transcendental subject of thoughts, called x. It is known only through the thoughts that are its predicates, and apart from them we have not the least conception of it. So we turn in a circle.

Kant Paralogisms, B407

That the "I which thinks" must always be a subject in thought is an evident proposition, but it does not mean that I, as an object, am a self-subsistent being, a substance. That would require data that are not found in thought.

The debate goes on (our summary) Kant here targets the reasoning that moves from the "I think" to a thing that thinks (the "Paralogisms"), and Descartes is the great example of that move. A Cartesian can answer that I know myself immediately as thinking, without the senses, and that this knowledge is precisely that of a substance.

Where the debate stands (our summary)Disagreement of substance: can we know a thinking substance without sensible intuition, through the sole awareness of thinking? Kant says no, Descartes says yes.

Peter F. Strawson 1966 · attacks premise 3

Strawson keeps from Kant the idea that an experience supposes a subject who can say "this experience is mine". But he judges that the story of acts of synthesis performed by the mind, which Kant calls "transcendental psychology", is not needed for the argument and should be dropped: the condition of unity can be stated without it.

no reply

No reply possible: objection raised in 1966, after the author's death (1804).

The debate goes on (our summary) Strawson writes almost two centuries after Kant. We have no answer from Kant. Kantians reply that the unity of consciousness does not hold by itself: without an activity that connects, it remains unexplained.

Where the debate stands (our summary)Disagreement of substance: does the unity of consciousness require an activity of synthesis, or is it enough to say that an experience must be attributable to one same subject?