1837 — The Polar Active-Soft Section Is a Morse–Bott Focal Segment
Correction to the naive weighted picture
Entry 1836 isolated the polar section
This means that the loop point leaves one focus directly toward the other focus in the same region wall. Along the open focal segment,
The radial dependence does not merely begin quadratically; it vanishes identically along the segment. The correct object is Morse--Bott.
Transverse normal form
Let be distance along a focal segment of length , and let be transverse. Then
Thus the owner wall has local form
If the second region wall cuts the segment transversely,
then taking its residue leaves
The generic polar collision therefore produces an ordinary grade-zero logarithmic line, rather than Entry 1836's generic-direction second-Rees line.
Classification
Both ingredients are already frozen carrier data:
- the occurrence-resolved edge-soft endpoint;
- the second labelled region wall.
The focal segment and its transverse Morse coefficient arise from their physical distance geometry. No new carrier stratum is required.
Qualifications
This is a conditional local normal form. It does not prove that the dehomogenized physical discriminant reaches the polar section. If it does and , a deeper tangency remains and must be retained as an excess complex rather than normalized away.
Next falsifier
Solve the polar boundary equations on the focal segment. Determine whether the second wall has a real transverse zero, a tangential zero, or no zero for the source-compatible deformations. Only the first two activate the polar coefficient.
Evidence
research/benincasa/checkers/five_site_region_pair_polar_soft_morse_bott.pyresearch/benincasa/results/five-site-region-pair-polar-soft-morse-bott.json- Entry 1836
- allocator claim:
seqclaim-22d6ef5c4bbe17c9f780e9a1