Why Illusion Comes Naturally

The First Critique said illusion is “no accident of cognition — where there are constructs, there are its conditions.” On the practical side the book finds four. First, the antecedent goes unwritten. A cannot-not always hangs on a position, yet the sentence “this cannot be cancelled” does not on its own carry “as one stands here.” That much follows from the shape of the thing. That a person may feel a cannot-not strongly without noticing its condition is recorded only as a hypothesis: the first speaks of sentences, the second of feelings.

Second, recorded as a hypothesis: “the deeper a construct’s sedimentation, the more readily it enters in the voice of ‘it was always so.’” Laws too — Moral Law calls its four “the structure that the Self naturally is.” Third: a law that works for oneself is easily said to work for anyone. The First Critique had a line of the same form: “ambition is no disease; exempted ambition is.” So here: the urge to universalize is no disease; exempted universalizing is. Fourth: cost is deferred; at the moment of action it appears as if it cost nothing.

The four differ in status and point one way. The First Critique’s law of risk is set up here word for word, premises and limit included: where the position-condition is not automatically preserved with the cannot-not and new invocations are not reviewed in step, the more widely a cannot-not is invoked, the more points at which it oversteps its condition — and the risk of illusion rises; it “does not guarantee that actual illusions multiply.”

Cannot Be Cancelled; Can Be Left

The first antinomy. Thesis: while one stands in a position, some things cannot be cancelled. Antithesis: the position can be left. Both are Action Theory’s own formal claims, and so is the resolution: “What it states is that two statements are incompatible, not a psychological inability to bring oneself to act.”

The second critique adds only the reading from its doctrine of illusion: the antinomy arises when the index of position drops out. “Cannot be cancelled” says that, within the position, two statements cannot both hold; “can be left” is said of the position, one matter at a time. Each sentence is true at its own address, and the First Critique’s “able to stop, and yet unceasing — that is autonomy” is the ready-made seam between them.

One small account remains. The cost of exit is stated four ways in the framework: Reason Cannot Not says degradation; the First Critique and Action Theory say “that is all”; Moral Law says that after leaving 15DD the Self “still has the capacity to legislate.” The four cannot all be right. The book states the cost as the last three do, and marks that this statement stands on Paper 4’s postulate.

Is Unceasing Autonomy?

The second antinomy. Thesis: not ceasing is the mark of autonomy — no straw man, since Reason Cannot Not wrote: “True autonomy is the persistent legislative motion: the subject gives itself law moment by moment; the moment legislation stops, autonomy ceases to operate.” Antithesis: the First Critique: “what is able to stop yet unceasing may also be obsession, and parting autonomy from obsession is the practical critique’s work.”

The sentence’s second half states a necessary condition; its first half joins the terms with “is,” and so can be read as sufficient: as long as one does not cease, it counts as autonomy. Not ceasing is necessary, and not enough. The candidate criterion is whether, under interrogation, the channel of revision and review is open. Obsession is not ceasing while the channel has closed itself. At the other end lies fickleness: stopping whenever it suits, changing the law without keeping the books.

So the book gives autonomy three marks: able to stop, yet unceasing; the channel open under interrogation; revising the law and keeping the books on it. These are the three distinctions Essay 01 promised to make clear — autonomy, obsession and fickleness — and they meet the first one drawn there: revising is not dropping.

Unaltered to the End Is Not Obsession

The criterion must agree with one the framework already had. The learning series gave the cannot-not a touchstone — “External opposition cannot extinguish it” — which from outside is hard to tell from obsession. But it drew a limit: the opposition meant “is social blockage … not factual refutation,” and “if factual refutation appears and you still refuse to adjust, that is not 14DD presence but 13DD failure.” Social blockage cannot put it out; factual refutation it must take up — the channel is open.

Value Theory says something that matters even more: “A person may hold something as what he will keep to the end, unaltered, and this does not conflict with revisability: revisability says the position is open, not that it must be used.” So keeping something unaltered to the end is not obsession. Only a closed position is.

The three marks are this branch’s definition of autonomy, and they have an opponent: the branch that counts a law’s unrevisability as loyalty. One’s Own Law holds the nearest sentence: “You do not need to stop and ask ‘is this direction still correct?’ The direction has not changed.” The book does not falsely claim its definition as the universal one. And Kant’s antinomy of virtue and happiness it does not set up at all; nothing in the framework’s material forces it.

The Discipline of Diagnosis

To diagnose others is also to make a claim, so the book binds its own diagnoses with five rules. Name columns, not persons: a competent diagnosis writes which credential was shown, which column lacks the entry, where the sources are — and not “what the shower intended, and what he deserves.” Read structure, not hearts: acts, accounts, channels and patterns of precaution, never motive. First-person claiming belongs to the person: “First-person avowal can be neither conferred by another nor revoked by another.”

Levels record operations only, by moment and domain: Moral Law itself says “the unit of study is the event, not the person; 15DD names a behavioral layer, not a personal identity”; and the book makes no substantive claim such as “there are no 15DD people” — only this discipline. Rulings-out and direction reports are kept apart; both can be withdrawn; there is no absorbing state. The Conflict Treatise: “cues are not criteria, and traces are not verdicts.”

A diagnosis is also a claim. Misdiagnosis has its own types, the mirror of false claiming: an unreviewed accusation carrying itself as a settled case; one case stretched into a verdict on a whole framework; an old version’s diagnosis laid on the new. Value Theory gives the clearest shape, an audit that claims to see what a person really avows: “if he accepts it, the conclusion holds; if he denies it, the denial is read as further evidence for it.” Such a diagnosis is itself a false claim. The chapter closes on itself: every diagnosis it set up “also stands under interrogation.”