Logic is indispensable and local. The institute opposes logical imperialism, not logic — and treats the cheapening of deduction as a redistribution of scarcity toward judgement.
The institute's foundational claim is deliberately unglamorous. Logic works. It works so well that people forget what it is working inside. A valid deduction is a machine for preserving truth relative to a framing: a vocabulary, a set of admissible operations, a decision about what counts as a premise and what counts as noise. Change the framing and the same machine can preserve a different truth with equal fidelity. This is not a scandal. It is the operating manual. The scandal begins when someone treats the local maximum as the global one and calls every other summit a valley.
We are not anti-logic. Anti-logic is a mood, usually worn by people who have been bruised by a pedant and decided that the bruise is a philosophy. The institute keeps the pedant and fires the imperialism. Logical imperialism is the habit of exporting one framing's validities into every neighbouring problem as if validity travelled without a passport. It produces clean slides and dirty epistemology. Reviewer 2 notices the slides first and the dirt later, which is why this note exists.
1 — Locality without apology
To say that logic is a local maximum is to say two things at once, and the institute insists on both.
First: within a well-specified frame, logical structure is not optional decoration. It is the difference between an argument and a mood board. If your premises are admitted and your rules are sound, the conclusion is not a vibe you earned by typing confidently. It is constrained. That constraint is precious. Civilisation spends a lot of energy pretending it can be replaced by charisma.
Second: the frame is not free. Vocabularies smuggle ontologies. Admissible operations smuggle values. The choice of what may be assumed is often where the real work — and the real disagreement — already happened. A proof that begins after that choice can be impeccable and still be a local triumph: the highest point available once the mountain range has been selected.
Source ·Paraconsistent Logic (opens in new tab)· Stanford Encyclopedia of Philosophyatlas
Source ·Non-monotonic Logic (opens in new tab)· Stanford Encyclopedia of Philosophyatlas
The comic move, if one is required, writes itself: people who treat classical logic as the only adult in the room often discover, mid-argument, that they have been using defeasible reasoning all afternoon and calling it common sense. Common sense is not a refutation of logic. It is frequently non-monotonic logic wearing weekend clothes.
2 — Against logical imperialism
Opposition to logical imperialism is easy to mishear as opposition to logic. The mishearing is useful to imperialists, so we refuse it in advance.
Logical imperialism has a tell. It answers a framing dispute with a validity check, as if the dispute were already about inference rather than about what may be inferred from. It treats abduction — the generation of explanatory hypotheses — as a failed deduction rather than a different act. It treats relevance as a soft skill. It treats responsibility for the frame as someone else's problem, preferably someone junior.
Source ·Abduction (opens in new tab)· Stanford Encyclopedia of Philosophyatlas
Abduction is not deduction with lower standards. It is the move that proposes what would have to be the case for the data to make sense. Peirce's line of work, and the long philosophical literature that follows it, does not ask us to confuse hypothesis with proof. It asks us to stop pretending that all epistemic labour is proof-shaped. When deduction becomes abundant — when machines can check, expand, and remix formal chains at industrial scale — the scarce goods shift. Judgement about which frame is worth inhabiting does not become cheaper because entailment did. Relevance does not become cheaper. The willingness to answer for a choice of premises does not become cheaper. If anything, abundance of deduction makes the scarcity more embarrassing, because the spreadsheet looks finished while the question is still badly posed.
The institute's stance is therefore conservative in one register and radical in another. Conservative: keep the machinery of validity; do not romanticise contradiction as depth; do not treat hallucination as negative capability with better marketing. Radical: refuse the export of one local maximum as the map of all terrain. A proof can be perfect and still be answering the wrong mountain.
3 — What abundance redistributes
It is tempting to narrate the present as a crisis of logic: models that sound certain, publics that cannot tell entailment from eloquence, institutions that reward the appearance of closure. The institute's diagnosis is slightly different and less flattering to everyone involved. Logic is not dying. Deduction is becoming cheap. Cheap things get overused. Overused tools get mistaken for complete toolkits.
When deduction is expensive, people ration it. They save formal care for the places where a mistake is costly. When deduction is cheap, the temptation is to run it everywhere and call the residue of judgement a soft leftover. That leftover is where framing, abduction, and responsibility live. Soft is the wrong word. Soft is what you call the part of the work that does not fit in your demo.
Another fair objection is that "local maximum" sounds like optimisation talk wearing a toga. We keep the phrase anyway. It names something the toga alone cannot: that improvement relative to a basin is not the same as escape from the basin. Climbing harder does not always change the mountain. Sometimes it only improves your view of the same peak while the useful door is on a ridge you have declared off-map because it was not entailed by yesterday's axioms.
4 — What we are not claiming
We are not claiming that consistency is optional as a moral posture. We are not claiming that contradictions are deep. We are not claiming that vibes are a research method. We are claiming that the prestige of validity has been allowed to stand in for the harder labour of choosing, defending, and revising the conditions under which validity is informative.
One more refusal, because this note will be misquoted: locality is not relativism with tenure. Relativism says any frame is as good as any other. Locality says a frame has a scope, and outside that scope the same moves may stop being informative even when they remain formally pretty. You can still argue that one frame is better for a purpose. You are then arguing about purposes, costs, and responsibilities — which is where adult disagreement usually lived before someone tried to win it with a proof that began too late.
The institute's benches and notes repeatedly return to this redistribution of scarcity. Premature coherence resolves a tension before the tension has finished teaching. Logical imperialism resolves a framing dispute by pretending it was never a dispute. Both look like competence from a distance. Both are expensive up close.
“The authors oppose imperialism with admirable clarity and then ask us to trust their local maximum. Noted.”
What survives the quip is the obligation it half-sees. Trust is not the point. Scope is the point. State the frame. State what the frame excludes. State where abduction is doing work that deduction cannot. Then, and only then, let logic do the thing it is good at: refusing to let you smuggle a conclusion past a premise you already accepted.
Logic is a local maximum. Treat it as one and it remains indispensable. Treat it as the only altitude worth measuring and you will publish beautiful proofs of questions nobody needed answered — which, as institutional failures go, is among the most publishable.