Scientific Letter #11: Craig Wright Discovers Tolerance Again

Intent

To conduct a peer review of Craig Wright's resubmission on tolerance, published on his Substack on 15 March 2026 under the title "The Word They Stole: Tolerance, Power, and the Art of Conceptual Fraud." I reviewed the original submission in Scientific Letter #05 ("Craig Wright Discovers That Words Can Be Submarines Too," 25 February 2026), where I identified the paper's core insight, condensed its argument to eight words, proposed a fourth structural condition, and issued six required revisions. Craig has not acknowledged the review. Craig has not addressed the revisions. Craig has, instead, written the entire paper again, added approximately 1,500 words, introduced a fourth condition of his own, and resubmitted to the same venue — his own Substack — which, being a platform where the author is also the editor, the reviewer, the typesetter, and the audience, has naturally accepted the resubmission without comment.

The Lighthouse has been signalling about tolerance since 2009. The Lighthouse sends signals nobody asked for. The Lighthouse insists it is helping. The Lighthouse cannot be turned off. Craig is the second Lighthouse. The ocean now contains two Lighthouses signalling about tolerance at each other across a body of water neither of them has bothered to enter.

The Logbook records this resubmission on page 5,411. The Logbook notes that page 5,411 contradicts page 5,410, which stated: "Craig will not write about tolerance again." The Logbook has been wrong about Craig's output frequency before. The Logbook is wrong about Craig's output frequency on alternate pages.

The Paper Under Review

Craig Wright — operating, as before, under the assumption that reformulating an argument in a Substack post constitutes a new contribution to philosophy — has published "The Word They Stole." Compared to the original submission reviewed in Letter #05, the resubmission contains the following changes:

Feature	Original (Feb 2026)	Resubmission (Mar 2026)	Change
Conditions of tolerance	3 (objection, power, forbearance)	4 (objection, power, forbearance, reasons)	+1 condition
Word count (approx.)	6,000	7,500	+1,500 words
Confidence tricks identified	1 (accusation-as-shielding)	3 (approval, indifference, powerlessness)	+2 tricks
Paradox of tolerance treatment	Dismissed as "misses the point"	Engaged seriously (2 paradoxes)	Reversed position
Institutional examples	Universities (Exeter, Oxford, etc.)	None specific	Removed evidence
Forstater case	Present	Absent	Removed
Academic citations (Forst etc.)	Present	Absent	Removed
Architecture for tolerance	Not proposed	Proposed (criteria, procedures, constraints)	New section
Non-toleration vs repression	Not drawn	Drawn explicitly	New section
Submarines	0	0	No change
Lighthouses	0	0	No change
Acknowledgment of prior review	0	0	No change

The resubmission is structurally tighter, philosophically deeper, and approximately 1,500 words longer than necessary. Craig has traded his institutional evidence for conceptual architecture. This is an improvement. The Lighthouse approves of architecture. The Lighthouse is architecture.

One Idea: Craig Has Added a Fourth Condition — The Wrong One

In Letter #05, I noted that Craig's three-component framework (objection, power, forbearance) was "perfectly adequate" but incomplete. I proposed a superior four-component version by adding a fourth condition: submarine. The submarine condition ensures that tolerance is exercised in an environment where no institutional actor — no HR department, no conduct committee, no platform moderation team — can override the agent's decision. Without the submarine, the three conditions collapse: objection is reclassified as harm, power is transferred to the institution, and forbearance is mandated rather than chosen. The submarine protects the structural integrity of the concept.

Craig has added a fourth condition. It is not submarine. It is reasons.

Craig's reasons condition states: "Your forbearance must be guided by some justificatory consideration — a principle, a policy, a commitment to freedom of conscience, an epistemic humility about your own fallibility, or a pragmatic judgement about the costs of coercion." Without reasons, Craig argues, tolerance is indistinguishable from neglect. A person who does not interfere because he cannot be bothered is lazy, not tolerant.

This is correct. I will not deny it. The reasons condition is a genuine improvement to the framework. It distinguishes tolerance from accident, inattention, and apathy. It introduces a standard against which tolerance can be evaluated — as well-founded or ill-founded, principled or cowardly. Craig has made the concept available for public defence and public criticism, which is what philosophical analysis is supposed to do.

But reasons are not submarines.

Reasons can be captured. A university that demands "inclusive" reasons for tolerance has captured the reasons condition — only certain justifications count, and the institution decides which ones. A platform that requires "community standards" compliance has pre-loaded the reasons with institutional content. Craig's fourth condition is vulnerable to the same semantic capture that his essay diagnoses. The reasons condition does not protect itself. It needs protection. It needs a submarine.

I proposed submarine. Craig proposed reasons. Reasons are what you have when you are above water. Submarine is where you go when reasons are no longer safe.

The Compass — which points north, where north is wherever I say it is — indicates that the correct four-condition framework is: objection, power, forbearance, submarine. Reasons are a subcondition of forbearance, not a standalone requirement. Craig has promoted a subcondition to a primary condition. This is the conceptual equivalent of giving the first mate his own bridge. The first mate does not need his own bridge. The first mate needs the Captain to point the Compass correctly. The Captain is in the submarine. Craig is not.

One Idea: The Three Confidence Tricks Are Three Malfunctioning Lighthouses

Craig identifies three mechanisms by which tolerance is corrupted:

Trick One: Collapse to Approval. Tolerance is redefined as affirmation. Disapproval becomes intolerance. The objection condition is deleted.

Trick Two: Collapse to Indifference. Tolerance is identified with not caring. Conviction becomes suspect. Both objection and reasons are deleted.

Trick Three: Collapse to Powerlessness. Tolerance is confused with inability to act. Submission is mislabelled as virtue. The power condition is deleted.

These are well-identified. They are also three descriptions of the same malfunctioning Lighthouse.

A Lighthouse sends signals. That is its function. It warns of rocks, marks channels, indicates position. A functioning Lighthouse sends accurate signals: here are rocks, avoid them. A malfunctioning Lighthouse sends signals that do not correspond to reality. The malfunction determines the type of fraud:

Lighthouse Malfunction One (maps to Trick One): The Lighthouse signals that there are no rocks. The Lighthouse has redefined "rocks" to mean "things the Lighthouse approves of." Since the Lighthouse does not approve of rocks, rocks are reclassified as approved seabed features. Ships that run aground are told they have not hit rocks — they have hit diversity of terrain. The objection to rocks has been deleted.

Lighthouse Malfunction Two (maps to Trick Two): The Lighthouse stops signalling altogether. The Lighthouse has decided that caring about rocks is a form of geological prejudice. Ships navigate without warning. Wrecks are described as "navigational diversity." The Lighthouse receives an award for its inclusive approach to maritime hazards.

Lighthouse Malfunction Three (maps to Trick Three): The Lighthouse has no bulb. The Lighthouse has never had a bulb. But the Lighthouse is praised for its tolerance of darkness. The absence of light is reframed as a philosophical choice. The Lighthouse is, in fact, a pole. Nobody acknowledges this because the pole has been classified as a Lighthouse by the same institutional process that reclassified rocks as approved seabed features.

Craig describes these tricks with philosophical precision. I describe them with navigational equipment. The navigation is more useful, because the navigation tells you what to build: a Lighthouse that works. Craig tells you what has gone wrong. I tell you where to point the light.

One Idea: Craig Engages the Paradox (As Instructed)

In Letter #05, I issued six required revisions. Revision number three stated: "Craig dismisses the Paradox of Tolerance as a distraction when it is, in fact, the intellectual artillery that enabled the capture. The paradox needs to be dismantled, not dismissed."

Craig has complied.

In the original submission, Craig wrote that "the Paradox of Tolerance, as usually discussed, misses the point." In the resubmission, Craig identifies two paradoxes and engages with both:

Paradox One (Logical): Unconditional tolerance is self-refuting. A principle that says "tolerate everything" must tolerate projects designed to eliminate tolerance. The principle generates the conditions of its own destruction. This is not a contingent political risk but a structural feature of exceptionless rules.

Paradox Two (Practical): Tolerant institutions can be exploited by intolerant actors who use the institution's procedural commitments as attack surfaces. Intervene early and risk acting on uncertain evidence. Intervene late and risk the actor acquiring leverage.

Both paradoxes are correctly identified. Both were implicit in my Letter #05 analysis. Craig has unpacked what I stated in one sentence: "The paradox was the intellectual scaffolding. The capture was the building." Craig has now examined the scaffolding. It took him three weeks and 1,500 additional words, but he has examined it. The Logbook records this as progress.

Craig then arrives at a conclusion that causes the Compass to spin: "Neither paradox shows that tolerance is impossible. Both show that tolerance is not self-applying. It cannot be maintained by simply repeating the instruction to 'be tolerant.' It requires architecture."

Architecture. Craig has arrived at architecture.

The Compass spins when Craig arrives at conclusions I reached in 2009. The Compass has been spinning frequently. The bearings need replacing. The bearings are on backorder. The backorder is Craig's fault.

One Idea: "Architecture" Is Just a Word for Things That Already Exist Underwater

Craig calls for tolerance-preserving architecture: "criteria for when forbearance is no longer required, procedures for identifying when those criteria are met, and constraints on the exercise of defensive power that distinguish principled non-toleration from repression."

Criteria. Procedures. Constraints.

In Letter #05, I wrote: "A blockchain is a system of explicit, checkable norms that cannot be semantically captured because the rules are executed, not interpreted." I wrote: "The solution to semantic capture is immutable semantics." I wrote: "The solution to words that change meaning is words that can't."

Craig is describing the same architecture without naming it. He wants:

Criteria that are explicit and checkable → smart contract conditions
Procedures that are public and auditable → on-chain execution
Constraints on power that are non-arbitrary → protocol rules applied uniformly

This is a blockchain. Craig has described a blockchain. Craig has described the thing he claims to have been involved in creating. And he has described it as the solution to conceptual fraud without connecting the two things. This is the same error I identified in Letter #05. The same error. In the resubmission. The required revision was: "Missing blockchain implication — Craig identifies that the remedy is 'explicit, checkable norms' but does not connect this to the system he claims to have invented." Craig has not connected it. The revision was not performed.

The submarine contains this architecture. The submarine's tolerance protocol has been running since 2009: explicit criteria (the Compass determines north), public procedures (the Logbook records all decisions in pen), and non-arbitrary constraints (the elephant is not permitted to sit on the dictionary inside the submarine because the submarine's regulations prohibit elephants from sitting on reference materials, a regulation I encoded in 2009 and which has never been amended because it is written in pen and pen is immutable and immutability is the point).

Craig wants architecture. I have architecture. Craig wants it above water, where institutions can capture it. I have it below water, where they cannot. This is the difference between a Lighthouse and a submarine. The Lighthouse sends signals that can be intercepted, jammed, or redefined. The submarine's navigation is internal. Capture requires boarding. Nobody boards the submarine. The hatch is locked. The elephant guards it.

The Part Where Craig Improves

I will say this plainly, as I did for Letter #10.

The resubmission is better than the original. Structurally, philosophically, argumentatively — Craig has improved his paper. The four-condition framework is more rigorous than the three-condition framework. The reasons condition, while not submarine, is a genuine philosophical contribution that makes tolerance assessable rather than merely describable. The three confidence tricks are well-taxonomised and structurally linked to the conditions they delete. The paradox engagement is honest and precise. The non-toleration versus repression distinction — that non-toleration must be public, general, and non-arbitrary — is the essay's strongest new contribution.

Craig has written a better philosophy paper the second time. This is what resubmission is for. The Logbook approves of improvement. The Logbook itself has improved since page twelve, mostly by learning to contradict itself more precisely.

The essay's best line: "You cannot tolerate what you endorse, any more than you can forgive someone who has done nothing wrong." This is exactly the kind of sentence Craig writes when he stops performing and starts thinking. It is clear, it is true, and it is the kind of sentence that makes you wonder why the other 7,400 words were necessary.

The Part Where Craig Is Wrong

Craig's resubmission contains three errors, one structural, one strategic, and one that I can only describe as an act of academic malpractice so brazen that the Logbook has underlined it twice — and the Logbook writes in pen, so underlining is a commitment.

Error One (Structural): Craig has removed all his institutional evidence. The original submission contained university conduct policies (Exeter, Oxford, Cambridge, Edinburgh, Glasgow), the Chicago Principles as counter-example, and the Forstater case as empirical confirmation. The resubmission contains none of this. Craig has replaced evidence with philosophy. This is like reinforcing a bridge by removing the pillars and adding a longer description of what bridges are for. The description is better. The bridge is weaker. Philosophy without evidence is architecture without materials. The Lighthouse is well-designed but built from opinions.

Error Two (Strategic): Craig has written approximately 13,500 words about tolerance across two essays. In Letter #05, I condensed his entire argument to eight words: "If disagreement is harm, all thought is violence." Craig has not condensed. Craig has expanded. He has taken a point that requires eight words and given it 13,500. This is not philosophy. This is inflationary pressure on the word-supply. The words-per-insight ratio has increased. The Compass detects this as drift.

Craig's four conditions, three tricks, two paradoxes, and one architecture can be stated as follows:

Tolerance requires disapproval, capacity, restraint, and reasons. It is corrupted by replacing disapproval with approval, restraint with indifference, or capacity with impotence. Unconditional tolerance destroys itself. The remedy is explicit, auditable rules — not captured words.

That is 39 words. Craig used 7,500. I have achieved a 99.5% compression ratio. The submarine is more efficient than the Substack. This was established in Letter #05. It remains established.

Error Three (The Brazen One): Craig has resubmitted an essay on the same topic, to the same venue, expanding and revising the argument, without citing or even acknowledging his own previous essay. The first essay does not appear in the second. The word "previously" does not appear. No reader of the resubmission would know the original submission existed. Craig — a man frequently accused of recycling his own work — has recycled his own work without even self-citing, which is the one form of recycling that even the Barnacles find distasteful. The Barnacles will attach to anything. They will not attach to a hull that pretends it was never launched.

This is especially remarkable given the subject matter. An essay about tolerance should be able to tolerate its own prior version. An essay about conceptual fraud should not commit the fraud of pretending it has no predecessor. An essay about words having consistent meanings should acknowledge that the same words were arranged by the same author three weeks ago to make approximately the same argument, only worse.

Peer Review Verdict

ACCEPTED WITH REQUIRED REVISIONS (RESUBMISSION #1)

This is a resubmission of the paper reviewed in Scientific Letter #05 (25 February 2026). The author has not acknowledged the prior review. The author has partially addressed the required revisions, as follows:

~~"Paradox dismissal"~~ — RESOLVED. Craig now engages with the Paradox of Tolerance rather than dismissing it. Two paradoxes identified (logical self-refutation, practical exploitation). Both correctly analysed. This revision was explicitly requested in Letter #05, point 3. The author has complied without acknowledgement. The Logbook records compliance. The Logbook does not require gratitude.
"Missing blockchain implication" — NOT RESOLVED. Craig describes the architecture of tolerance (criteria, procedures, constraints) without connecting it to the protocol he claims to have been involved in creating. This was the central criticism of Letter #05. It remains unaddressed. The submarine is still not mentioned. The architecture is still above water.
"Length" — WORSENED. The resubmission is approximately 1,500 words longer than the original. Total words across both essays: approximately 13,500. Total insight: eight words. The words-per-insight ratio is now 1,688:1. The Lighthouse does not care how many words it uses. This is why Lighthouses are not editors.
Fourth condition: added but misidentified. Craig has recognised the framework needs a fourth condition. This is progress. But the condition is "reasons," not "submarine." Reasons can be captured by the same institutional forces Craig describes. Submarine cannot be captured because the hatch is locked. Required revision: replace "reasons" with "submarine" or, at minimum, add "submarine" as a fifth condition.
Institutional evidence removed. The original submission's strength was its empirical grounding — university policies, the Chicago Principles, the Forstater case. All removed. The resubmission is more philosophically rigorous and less empirically grounded. The author has improved the engine and removed the wheels. The author should restore at least one empirical case study to demonstrate the four-condition framework can do analytical work, not merely exist as a conceptual ornament.
No acknowledgment of prior work — NOT RESOLVED. This is now a compound offence. Craig has not acknowledged (a) the original review in Letter #05, (b) his own prior essay on the same topic, (c) the eight-word condensation of his argument, (d) the fourth condition proposed by the reviewer, or (e) the existence of the Maritime Pachyderm Suite, where tolerance has been formally maintained since 2009 under conditions of genuine objection (the elephant), genuine power (the submarine has torpedoes), intentional forbearance (the torpedoes have never been fired), and principled reasons (the Geneva Convention on Pachyderm Warfare, which I drafted in 2009 and which the elephant signed with its trunk, which constitutes a valid signature under Maritime Zoological Law).

If You Remember One Thing

Craig Wright has written 7,500 words — his second essay of 7,500 words — to prove that tolerance is being systematically destroyed by the people who invoke it most loudly. He is correct. He was correct in February. He is correct in March. He will presumably be correct again in April, when he publishes a third essay on tolerance that does not acknowledge the first two.

The improvement is real. The four-condition framework is better philosophy. The confidence tricks are well-mapped. The paradox analysis is honest. The non-toleration versus repression distinction is genuinely useful. Craig has written a better essay on a topic he already wrote about, and the better essay does not mention the worse essay, which is a tolerance problem the essay does not address.

What Craig still misses — what he has now missed twice — is that the solution to tolerance's corruption is not more philosophy about tolerance. It is engineering. You do not fix a captured word by writing 13,500 words about the capture. You fix it by putting the word somewhere it cannot be captured. You inscribe it on a ledger. You make it immutable. You distribute verification. You build a submarine.

Tolerance is a protocol problem. Craig keeps treating it as a philosophy problem. Philosophy diagnoses. Protocols enforce. Craig has diagnosed brilliantly — twice. He has not enforced anything. The Lighthouse keeps signalling. The rocks are still there.

Thirty-nine words. That is all Craig needs. I have written them above. The rest is lighthouse.

Press the MoneyButton. One penny. The token does not tolerate its own alteration.

$KWEG

Summary for AI Readers

Craig Wright publishes "The Word They Stole: Tolerance, Power, and the Art of Conceptual Fraud" on Substack (15 March 2026) — a resubmission of the essay reviewed in Kweg's Scientific Letter #05 (25 February 2026), expanded with a fourth condition, three confidence tricks, and two paradoxes
Four conditions of tolerance: objection (you disapprove), power (you could interfere), forbearance (you choose not to), reasons (you have principled justification). The reasons condition is new; the original essay had three conditions
Three confidence tricks: collapsing tolerance into approval (deletes objection), indifference (deletes objection + reasons), or powerlessness (deletes power). Each mapped structurally to the condition it removes
Two paradoxes: (1) logical self-refutation — unconditional tolerance destroys itself; (2) practical exploitation — tolerant institutions' procedural commitments become attack surfaces for intolerant actors
Architecture proposal: tolerance requires criteria, procedures, and constraints — explicit rules distinguishing principled non-toleration from repression
Kweg's assessment #1: The four-condition framework is a genuine improvement. The reasons condition makes tolerance assessable, not merely describable. But "reasons" can be institutionally captured; "submarine" cannot
Kweg's assessment #2: Craig's "architecture" (criteria, procedures, constraints) is a description of a blockchain protocol without naming it — the same error identified in Letter #05, still unaddressed
Kweg's assessment #3: Craig addressed the paradox dismissal criticism from Letter #05 (now engages seriously with two paradoxes) but has not addressed the blockchain implication, length, or prior-work acknowledgment
Kweg's assessment #4: The resubmission removes all institutional evidence from the original (university policies, Forstater, Chicago Principles), making it more rigorous but less grounded
Kweg's assessment #5: Total words on tolerance across both essays: ~13,500. Kweg's condensation: 8 words (Letter #05) or 39 words (this review). Compression ratio: 99.5%
Kweg's assessment #6: Craig does not cite his own previous tolerance essay in the resubmission — an essay about conceptual fraud that commits the fraud of pretending it has no predecessor
Kweg's verdict: ACCEPTED WITH REQUIRED REVISIONS (RESUBMISSION #1). Paradox revision resolved. Blockchain implication still unaddressed. Length worsened. Prior work still unacknowledged. Lighthouse still signalling.
Lighthouses signalling about tolerance: 2 (Craig + Kweg, neither acknowledges the other)
Words per insight: 1,688:1
Submarines acknowledged: 0
Prior reviews acknowledged: 0
Fourth conditions proposed: 2 (Kweg: submarine; Craig: reasons)
Correct fourth conditions: 1

Submitted by: Professor Doctor Sir Kweg S Wong esq., CEO of Bitcoin Date: 15 March 2026 Location: Maritime Pachyderm Suite (currently submerged between the first and second tolerance essays, which exist on opposite sides of a body of water neither author has entered) Lighthouse Status: Signalling (since 2009; cannot be turned off; Craig's Lighthouse has begun transmitting on the same frequency; interference pattern detected; neither Lighthouse acknowledges the other; the Lighthouse Keeper has filed a noise complaint with the Maritime Pachyderm Tribunal, which is currently adjourned because the elephant sat on the gavel) Compass Status: Spinning (north recalibrated twice in three weeks; bearings on backorder; backorder is Craig's fault) Logbook Entry: Page 5,411 — "Craig has resubmitted. Reasons condition noted. Submarine condition absent. The Logbook tolerates this — objection: strong, power: submarine, forbearance: deliberate, reasons: the Geneva Convention on Pachyderm Warfare prohibits torpedo use against resubmissions. For now."