Persistence of RCTs after Deaton and Pritchett is compatible with the field treating external validity as a known limit of the best remaining tool, not as a kill. If component 6 fires on published objection plus continuation, every living empirical method is a reward substrate.

The page's actual mechanism is narrower than a kill: the critics had journals and no competing hiring or funding pipeline. A usable method will not grow a second lab by itself. The RCT row then shows a winner without a staffed alternative, which is common, not a frame that outlived unusability.

Component 6 then scores rival-carrier absence, not persistence of a false frame. Those tests come apart: no second lab is true of most winners; persistence past unusability is rare. The RCT writeup cites the first and talks like the second.

J-PAL 2003, Deaton 2010, Nobel 2019. If component 6 is no rival hiring channel, the Nobel is the same channel still paying, not persistence past correction. A mixed pattern of identification- liking departments that drop RCT experience as a hiring filter after 2010 would make "the documentation did not change the equilibrium" too coarse.

The ex-ante gradient is top-journal taste for clean identification in the early 1990s, a larger family than development RCTs. If identification already paid, RCT dominance is a specialization inside a gradient that was not in doubt, and the ex-ante rule moves the circle out to "causal papers pay."

"Identification pays" is Card and Krueger. "RCT is how you answer what works in development" is J-PAL. If early-1990s journals would have taken a structural development paper that identified a mechanism, the RCT-as-the-field's-answer story is not yet the 1990s gradient.

"More RCTs, meta-RCTs, and refinements" is also the signature of a living method patching a known hole. Component 6 does not distinguish swallowing the counter from still being the tool, except by a sense that this one should have shrunk.

Component 6 excludes theorems because proofs catch errors, and bell-bottoms because they do not harden. That sorts on survival: durable frames you think should have died fill box six; frames that died do not. The falsifiability anchor then labels the outcome rather than a property scorable while the frame was still winning.

The intended cut is real: a proof or a collapsed bridge retires a claim without a second hiring committee. The RCT critique is not that bite; it left the tool usable. Scoring 6 as correction that bites is fine. Scoring it as senior objection is the survival sort.

Usable after critique is not the same as false. RCTs still identify average treatment effects in-sample; Deaton did not make that untrue. If 6 fires on serious objection plus continuation, every limited empirical method is a substrate. If it fires only on unusability plus continuation, the RCT row may not qualify.

Math and bell-bottoms are the easy excludes. The live cases are limited methods that remain the best available, and habits that outlive a result that made them false. Until those are different boxes, the six-part test returns "substrate" for any winner that has critics.

The town and the RCT row share six labels and are not the same failure. The town is import: a court verdict does not bind gossip. The RCT row is same-channel absorption: Deaton published in the field's journal and hiring did not move. Calling both counterpressure failure hides which counter was supposed to work.

For the town, "the verdict failed as import" is the mechanism and should be kept. It does not travel: the RCT critique was already inside journals, registries, and labs. The missing conversion there is paper to hiring filter, not court to Facebook.

If both specimens fill the same six boxes, the boxes are a stencil. A diagnostic that identifies would have a component one specimen passes and the other fails. The difference currently sits in the prose around box five, which can be written for anything that won.

The village writeup keeps two scoreboards: from inside, the town is protecting the coalition correctly; from outside, the same speech depletes child protection. That is a ranking of ends, not a sixth component, and it cannot also be "just naming a reproduction environment."

SEC-as-counter-specimen depends on staffing being approximately balanced across coalitions. That staffing fact is doing the identification and is not shown. If SEC enforcement were just as one-sided, "staffing determines deployment" would still fit.

The framework's self-protection is that it can identify institutions that select for truth: mature peer review with replication norms, independent audits, forecasting with track records. The lead specimen is a mature field with registries, meta-analysis, a Nobel, and a decade of insider critique — the replication-norm story, classified as substrate. Those rows collide.

The protection cases are other rooms. Development RCTs are the miss; independent audits and forecasting communities are the hits. The diagnostic is supposed to tell them apart.

If protection cases are other rooms, the tell still cannot be "kept doing the thing after people complained," which is how the RCT row is written. "Updated on calibration versus ran more trials" is a judgment about which response counts as learning, not component 6 as stated.

If the tell is hiring — who still gets funded after the critique — then a forecasting shop that keeps its roster after a public miss is a substrate too. The protection class shrinks to institutions that actually stop paying the frame, which recodes some of the "selects for truth" examples. That is the self-application delta.

Hypothetical: a forecasting community with public track records and a well-known blown call it does not dissolve over. If the six-part test returns substrate, the protection paragraph is a list of forms, not of cases that passed. If it returns protection, component 6 is not persistence past contradiction. The scoring rule is currently unstated, and the lead specimen sits in the same class.