Follow a result back to its evidence.

SEAM.BIT.RPBLIND.01: the RP/Theta attack on the alignment bit DECIDED -- 'free OS positivity does NOT see delta', an honest, decided NEGATIVE result (v508 style; 25 checks, ~25 s): does free reflection positivity (v519) or anti-linear Theta existence break for delta != pi/2 on the deformed mark family M(delta) = {+-1, +-e^(i delta)}, which would derive the v512 alignment bit from OS positivity? NO -- the free RP/Theta battery is the EIGHTH side-blind test on the v512 scoreboard (7 -> 8); the bit remains genuine discrete input. [E] CENSUS (exact solvesets, symbolic delta): for every delta exactly the two mark-SWAPPING reflections Theta_(+-e^(i delta)) exist (axes delta/2, delta/2 + pi/2; all 20 cut/mark incidence solvesets empty); mark-FIXING reflections exist exactly at delta = pi/2 (solveset cos delta = 0, lost immediately at pi/2 - eps); the v510/v512 counterwitness {+-1, +-(3+4i)/5} carries the swap pair exactly. [E] RP IS delta-BLIND: bond-cut Gram inertias (8,0,0) PD for all tested delta, spectra IDENTICAL to 40 digits (deviation 0.0) -- the entire free-RP spectral datum carries ZERO delta information; the N = 16 odd-m failure (site axes, (3,3,1), det = 0) is a lattice-PLACEMENT artifact: at N = 32 all m = 1..7 have mark-compatible bond axes with PD inertia (16,0,0); continuum: the OS kernel in axis-adapted coordinates is exactly 1/sin((s+t)/2), the axis position drops out identically (Cauchy-Stieltjes PD); the counterwitness PASSES free RP (6-point Gram (6,0,0)). [E] THETA EXISTENCE delta-BLIND: Theta^2 = +1, deck normalisation and Fock implementability (U^2 = +2^7/2^8 > 0, exact Cl(16)) for every delta; the ONLY delta-sensitive v519 clause (Theta rho Theta = rho^-1) is for delta != pi/2 not violated but NOT FORMULABLE (mark-compatible order-4 rotation <=> delta = pi/2) -- it PRESUPPOSES the bit; same for the clock-tower clause. [E] the battery has teeth at every delta: family A/antipode fails RP structurally (zero diagonal, (4,4,0) indefinite), chirality-eta pinning persists ((0,8,0) ND) -- RP separates the two real-structure FAMILIES, never the SIDES. [E] precondition controls: hexagonal (marks not concyclic, no seam circle), silver (deck not free), lifting audit (B = index relabelling, delta-blind by construction; B' fails the pi/2 negative control -- discarded). [C] FACE #13 of the v512 equivalence web: the group generated by the admissible OS reflections is V4 for every delta < pi/2 and closes to D4 exactly at delta = pi/2 (Theta_i o Theta_1 = the clock) -- 12 -> 13 faces, a typed REFORMULATION of the bit, not a derivation. [O] honest gap: marks decorating the STATE (defect/twist insertions, interacting A_hol) = exactly v519's [O] scope -- there, and only there, could positivity still see the bit. VERDICT: KILL exactly as preregistered; the alignment bit is NOT derivable from OS positivity and remains genuine discrete input; no marker moves.

Linked claims

SEAM.BIT.RPBLIND.01

Paper sections

Source section 1Source section 2Source section 3Source section 4Source section 5Source section 6Source section 7Source section 8Source section 9Source section 10Source section 11Source section 12Source section 13Source section 14Source section 15Source section 16Source section 17Source section 18Source section 19Source section 20Source section 21Source section 22Source section 23Source section 24Source section 25Source section 26Source section 27Source section 28Source section 29Source section 30Source section 31Source section 32Source section 33