The claim
A164, answering the owner’s U47: “apparent disagreement between local descriptions does not establish a gluing obstruction until the appropriate transition maps have been specified”. And: “We should not test whether two interpretations are identical before specifying how their observational frames correspond.”
A173 stated the same proposition twice, once as a ready-to-file claim and once as a principle: “Different local descriptions do not constitute a global compatibility obstruction merely because they assign different values to apparently corresponding observables. … A genuine obstruction is the failure of compatible global realization after those correspondences and the empirical admissibility constraints have been imposed.” A178 §1 made observational correspondence “the hinge of the manuscript”. A203 §13 repeats it: “A disagreement between local descriptions cannot be interpreted as a structural incompatibility until the appropriate observational correspondences have been established.” A214 §10 and A218 §6 and Ch9.2 keep it.
A164 sorts disagreements into three classes, and “These categories should not be conflated”:
| Class | A164’s test | Example |
|---|---|---|
| Coordinate difference | ”compatible transition maps reconcile them” | inertial frames (CASE-039) |
| Structural transformation | ”descriptions may remain comparable but no longer equivalent” | literary translation, cross-modal generation |
| Global obstruction | ”cannot be jointly extended under the specified compatibility conditions” | sheaf contextuality (CASE-035) |
A178 §1 lists the same three as “Representational differences that disappear under appropriate correspondence maps. Transformations that conserve some relations while altering others. Empirical constraint systems that do not admit a global realization satisfying all local requirements.”
The order of tests is the claim. First fix the correspondence between the two descriptions’ contexts and outcomes, the context part of a transport (TERM-008), and check that it is coherent with restriction, as equivariance (TERM-004) requires. Then ask whether what is left of the disagreement is a change of the act or a failure of joint extension. Without the first step a transport theory cannot tell the three classes apart. The first two classes are A18’s passive and active transformations (CLAIM-004), which return here with a third class and an order.
The proviso
A correspondence that is fitted to make the disagreement disappear decides nothing. A173: “Transition maps must satisfy appropriate coherence requirements and be independently justified or validated. They must not be chosen solely to eliminate an observed incompatibility.” A164: the correspondences “must be independently anchored and empirically validated”. This is CLAIM-011’s remedy, carried from fidelity to compatibility: anchor the correspondence to independently measured distinctions, and evaluate it on tasks other than those it was fitted on.
What it does not say
- It does not say translation is a change of coordinates. Lorentz maps are invertible and form a group; translation correspondences are directed and lossy. A164: “unlike Lorentz transformations, arbitrary translation maps can collapse distinctions”. A178: “unlike changes of inertial coordinates, translation can erase distinctions and reorganize communicative structure”. A203 §13: “Lorentz coordinate transformations are invertible symmetries, whereas translation can be directed and lossy.” This is CLAIM-035, and the reason the analogy of ARG-004 stays Rejected.
- It does not say relativity is evidence for the claim. The relativity case shows a disagreement with no obstruction once the maps are known. It does not show that translation has such maps.
- It does not say that every disagreement some map removes is a mere change of description. The map has to be anchored first.
- It does not say pragmatic judgements are contextual (CLAIM-009).
The claim carries a circularity risk, which its second defeater names. The correspondence τ is the same object whose fit decides fidelity (CLAIM-005). If τ can only be validated by the fidelity verdict it is used to reach, the order of tests is empty. That is QUESTION-005 for correspondences. A164’s coordinate difference and structural transformation are also the two horns of QUESTION-021, which the claim does not settle.
Note of 2026-10-13: the general content is filed separately
Under ADR-040, the order of tests for disagreeing local descriptions is filed on its own as CLAIM-265, in the formal-apparatus line, resting on no claim. This claim now rests on it for the general content, and keeps CLAIM-011, CLAIM-004 and CLAIM-035, for applying the order of tests to translation. Claims in other lines that need only the general content cite CLAIM-265, so setting pragmatic-transport aside does not take it from them.
Note of 2026-10-27 (evening): the readings of the U15–U23 works
Three readings are engineering instances of this claim’s order of tests, and none validates its correspondences independently.
- Hansen and Ghrist (NOTE-1233, §8.7, p. 40) propose to “denoise the measured relationships themselves using the condition of cycle-consistency”, adjusting the correspondences until a section exists. They also show that the Laplacian does not determine the correspondences (Fig. 3.1).
- Singer (NOTE-1224): “if some oracle whispers to our ears which equations are good … In reality, we have to be able to tell the good from the bad on our own” (p. 2). He validates jointly, by consistency around cycles, under a noise model in which outliers are uniform and independent.
- Robinson (NOTE-1221) trusts his maps. In his Case 2 a sensor’s 2° bias cannot be told apart from an error in that station’s coordinates inside the maps. He calls it a sensor fault because the map is trusted.
Note of 2026-10-29: what the experiments test
CLAIM-238 granted, and CLAIM-239 narrowed, that the second disjunct says something cannot be done. No finite record of failed validations shows that no correspondence can be validated independently. An argument can, for instance a proof that every validation procedure takes a fidelity judgement as an input. So the programme’s studies test only the first disjunct, and only argument can defeat this claim on the circularity horn named above.
Note of 2026-10-29: what holds it back
Its premise CLAIM-004 is Proposed. The first disjunct needs pragmatic data with a correspondence validated independently, which no study has collected, and QUESTION-005, which the second disjunct turns on, is Open.