Assumption Testing Protocol¶
Protocol — instantiates Paradox Reframing
Isolates the single hidden premise that makes a contradiction binding and runs a discriminating test on whether that premise actually has to hold.
Most apparent contradictions are held up by one load-bearing premise that nobody has said out loud. Assumption Testing Protocol is the disciplined procedure that goes looking for it: it enumerates the unspoken premises the current frame depends on, ranks them by how much of the contradiction each one carries, and then subjects the top candidate to a discriminating test — a cheap observation or experiment whose result would differ depending on whether the premise is true. Its defining move is not to propose a nicer reframe or to convene a room; it is to falsify. A paradox that survives the removal of a premise was not resting on that premise; a paradox that dissolves the moment a premise is dropped has just revealed exactly where the frame was too small. The protocol treats "which assumption must hold for this to be a contradiction?" as an empirical question with a right answer.
Example¶
An airline's safety board is deadlocked. The reliability engineers hold that adding more automated cockpit warnings makes flights safer; the line-check captains hold that pilots on the newest, most automated fleet are missing more genuine emergencies. Both claims come with real incident data, and the board keeps splitting along those lines.
The Assumption Testing Protocol refuses to pick a side. It first inventories the premises the "more automation is safer" frame silently assumes, and one stands out as load-bearing: pilots monitor an automated system as attentively as they would hand-fly it. The board designs a discriminating test — a set of simulator sessions plus a review of the specific incidents, scoped in advance to the newest fleet only (the evidence boundary that keeps unrelated older-fleet events out of the comparison). The prediction: if the premise holds, missed-emergency rates should be flat across automation levels; if it fails, they should climb with automation. They climb. The premise is false — automation induces monitoring complacency.[n1] That single result reframes the fight: the problem was never "more or less automation" but sustaining pilot engagement inside an automated loop, and the board now has a validated basis to act on rather than two entrenched camps.
How it works¶
- Inventory the premises. List what the current frame must be assuming for the two claims to be mutually exclusive — about measurement, scope, time horizon, causality, or human behavior. Vague dread ("something's off") is turned into named, individually inspectable statements.
- Rank by load-bearing weight. For each premise ask: if this were false, would the contradiction still stand? The premises whose removal collapses the paradox are the ones worth testing; the rest are noise.
- Design a discriminating test. Specify, in advance, an observation whose outcome differs under "premise true" versus "premise false." A test that both sides would predict the same way proves nothing.
- Fix the evidence boundary. State which data bears on the premise and which is out of scope, so the result cannot be rescued or attacked by importing unrelated evidence.
- Keep or kill. Run the test, read it against the pre-stated prediction, and record whether the premise survives — which decides whether the contradiction was real or an artifact of the assumption.
Tuning parameters¶
- Premise granularity — how finely the frame's assumptions are split. Fine granularity isolates the exact load-bearing premise but multiplies candidates to test; coarse premises are fast to list but often bundle a true and a false assumption together.
- Test rigor vs. cost — a five-minute records check versus a funded trial. More rigor narrows what the result can mean but spends time; match it to the stakes of the reframe.
- Kill criterion — how decisively the test must go against a premise before it is discarded. A strict criterion resists premature dissolution; a loose one clears the deadlock faster but risks discarding a premise the evidence only dented.
- Boundary width — how much surrounding evidence counts as bearing on the premise. Too narrow and the test is unfalsifiable-by-exclusion; too wide and unrelated noise drowns the signal.
When it helps, and when it misleads¶
Its strength is that it kills semantic handwaving: instead of admiring a paradox or arguing frames, it names the one premise doing the work and puts it at risk. Because the discriminating prediction is fixed before the test, the result is hard to rationalize away, and a dissolved contradiction comes with evidence rather than rhetoric.
Its failure mode is testing a strawman premise — quietly picking the assumption easiest to knock down while the real load-bearing one goes untested, so the paradox "dissolves" on paper and reappears in practice. It is also vulnerable to confirmation pressure: a board that wants a particular side to win can design a test whose outcome it already knows. The guarding discipline is to pre-register the discriminating prediction and the evidence boundary before looking, and to rank premises by load-bearing weight rather than by how convenient they are to refute — the point is to find the assumption that would hurt to lose, and test that one.
How it implements the components¶
Assumption Testing Protocol realizes the diagnose-and-falsify slice of the archetype — the part that decides whether a contradiction is real:
hidden_assumption_inventory— its opening move: enumerating and ranking the unspoken premises the frame depends on.reframe_validation_check— the discriminating test is the validation, applied to the premise: does dropping it dissolve the contradiction without erasing evidence?evidence_boundary— fixing in advance which data bears on the premise and which is out of scope, so the test result cannot be smuggled either way.
It does not generate or stress-test a full reframe across cases — that reframe candidate (reframe_hypothesis) and its action reading (action_implication) belong to Scenario Reframe Probe, which validates the reframe rather than the premise; this protocol tests only whether the underlying assumption must hold.
Related¶
- Instantiates: Paradox Reframing — it supplies the falsification step that tells the archetype whether a contradiction is genuine or an artifact of a hidden premise.
- Consumes: Contradiction Map supplies the laid-out claims and context this protocol mines for candidate premises.
- Sibling mechanisms: Contradiction Map · Both/And Design Session · Double-Bind Analysis · Level-of-Analysis Shift · Paradox Workshop · Polarity Mapping · Scenario Reframe Probe · Sense Distinction Matrix · Decision Record with Residue
Editorial Notes¶
Form Classification¶
Form family: Experiment, Test & Rehearsal
Rationale: Isolates the single hidden premise that makes a contradiction binding and runs a discriminating test on whether that premise actually has to hold, making its operative form a deliberate probe, variation, simulation, or practiced execution used to generate evidence or readiness.
Independent corroboration: The frozen evidence defines Assumption Testing Protocol as 'Isolates the single hidden premise that makes a contradiction binding and runs a discriminating test on whether that premise actually has to hold', so its operative form is Experiment, Test & Rehearsal.
Nearest alternative: Assessment, Review & Assurance — It runs a discriminating test to generate new evidence about a premise, rather than only reviewing what is already known.
Review outcome: Independent reviewer agreement; medium confidence.
Origin Attribution¶
Primary origin: Philosophy
Origin pattern: Cross-disciplinary synthesis
Present-day reach: Multi-domain
Rationale: Philosophy of science supplies premise exposure and falsification by discriminating evidence.
Related originating lineages:
- Engineering & Design — Safety and automation cases provide the operational setting for testing hidden assumptions.
- Statistics & Experimental Design — Experimental design contributes tests whose outcomes distinguish competing premise states.
Review resolution: Philosophy is the agreed primary source of explicit premise testing. Engineering validation and experimental design materially supply operational test and falsification steps, while the exact single-premise protocol is synthesized rather than a canonical named method.
Attribution caveat: The protocol is an Encyclopedia operationalization of philosophical premise-testing using engineering and experimental controls.
Encyclopedia synthesis: The exact catalogued form synthesizes established practice rather than reproducing a single standard historical label.
Review outcome: Reconciled after independent review; medium confidence.
Notes¶
[n1] The ironies of automation — Lisanne Bainbridge's observation that automating a task can degrade the human operator's ability to monitor and recover it, because vigilance decays when there is nothing active to do. It is the standard name for exactly the premise this example puts to the test. ↩