Kant’s Transcendental Arguments (SEP)

The Stanford Encyclopedia entry on the argument form Kant invented, and on the century of attempts to break it. sep-kant gave this spoke the shape of Kant’s system; this gives the machinery of its central proof, and the objection that has governed the debate since 1968.

The form

A transcendental argument starts from “a compelling premise about our thought, experience, or knowledge” and reasons to “a substantive and unobvious presupposition and necessary condition” of that premise. The necessity is usually best explanation rather than strict entailment, the argument is meant to be a priori, and the target is normally a sceptic.

That shape is what makes it attractive. It does not argue with the sceptic about evidence; it tries to show that the sceptic’s own starting point already presupposes what they doubt.

What the Transcendental Deduction is trying to do

The Deduction aims to show that certain a priori concepts, the Categories, “correctly apply to objects featured in our experience” — aimed squarely at Hume’s denial that a concept like causality can be justified from experience. The Categories are the twelve from the Table: unity, plurality, totality (quantity); reality, negation, limitation (quality); inherence and subsistence, causality and dependence, community (relation); possibility, existence, necessity with their opposites (modality).

The argument runs through two claims this spoke has been naming without working through.

Transcendental unity of apperception. “It must be possible for the ‘I think’ to accompany all my representations.” Whatever else is true of my mental life, I must be able to represent myself as one subject across its states.

Synthesis. That unity has to be produced, and the producing is synthesis: “the act of putting different representations together, and grasping what is manifold in them in one cognition.” The Categories are the modes of ordering that synthesis uses, so if synthesis is necessary the Categories apply.

The B-Deduction runs this twice over: an “argument from above” (§§15–16) deriving synthesis from self-consciousness, and an “argument from below” (§§17–20) deriving it from what representing an object requires.

The step that does the work

§18 is where the anti-Humean force sits. Representations of objects are objectively valid and carry features that are “universal and necessary” — the parts of a house are represented as simultaneous, not as merely happening to occur together for me. Association cannot deliver that, and the entry’s example is exact: word association differs by person, since “one person connects the representation of a certain word with one thing, the other with another thing.” Association gives regularity in a mind; it cannot give necessity in an object.

So: we represent objects with necessary universality; association cannot explain it; synthesis under a priori concepts can; therefore the Categories legitimately apply. §§19–20 then tie synthesis to judgment, “the manner in which given cognitions are brought to the objective unity of apperception,” which is how the twelve logical forms of judgment are supposed to yield the twelve Categories.

The Refutation of Idealism, and why it is a separate argument

Against Descartes rather than Hume, and worth keeping distinct. On George Dicker’s reconstruction: I am conscious of my experiences in a determinate temporal order; I can only fix that order against something permanent outside my consciousness; no mental state and not time itself can serve; so I must perceive persisting objects in space. It leans on the fact that “time by itself is not perceived” and that we have no internal clock of the right kind.

Stroud’s objection

Barry Stroud’s 1968 challenge is the reason the field looks the way it does. The argument may show that we must believe in external objects; it does not show that they exist. “The existence of the external feature will not be a necessary condition of the aspect of experience or knowledge in question, for a belief about the external feature would always suffice.”

A sceptic can accept the whole transcendental apparatus and add that we are compelled to believe falsely. Nothing in the argument rules that out.

Modest and ambitious

The response was to retreat, deliberately. Ambitious transcendental arguments claim the existence of external objects. Modest ones (Stroud 1994, 1999; Robert Stern 1999) claim only necessary connections within thought: think this way and you must think that way. The resulting beliefs are defended as indispensable (not abandonable “consistently with our conception of the world”) and invulnerable (they “could not be found to be false consistently with its being found to be held by people”).

Anthony Brueckner (1996) then pressed the retreat. Hold both that we must believe in mind-independent objects and that no such belief is true, and you cannot describe a possible world where both obtain. So the modest version either collapses back into the ambitious one or quietly makes its objects “metaphysically ideal” rather than mind-independent — which is a substantive metaphysical result, not the neutral one modesty promised.

Cited across the entry: Kant, Hume, Strawson, Stroud, Brueckner, Guyer, Allison, Patricia Kitcher, Henrich, Longuenesse, Stern.

Why it matters here

immanuel-kant holds the critical turn and names the synthetic a priori without working it. This supplies the proof that was supposed to secure it, and the reason philosophers stopped expecting such proofs to reach all the way to the world. That bears directly on epistemology‘s anti-sceptical thread and on rationalism: Kant’s answer to Hume is neither rationalist derivation nor empiricist evidence, but an argument from what experience must presuppose — and its best-known critic says that form can reach our beliefs and stop there.

Connections

the philosopher: immanuel-kant · the system-level entry: sep-kant
the sceptic it answers: epistemology · the tradition it breaks with: rationalism

immanuel-kant · sep-kant · epistemology · rationalism · rene-descartes · western-philosophy