The Observer as Gyroscope: Why No Potential Holds the World, and Why Spin Does (INDYNA Research Note 27)

In the flag universe (Notes 24–25) the homogeneous sector has exactly eight unstable directions at the curved home r*: the ℓ=1 drift triplet and the ℓ=2 shear quintet (edge dynamics with Nambu threads), each odd-dimensional. We prove: (i) no potential-type holder — rigid pins or springs of any strength on antipodal pairs or flags — stabilises both; holders conflict (what holds shape disturbs motion). (ii) By the Kelvin–Tait–Chetaev theorem gyroscopic forces stabilise a saddle only for an even number of unstable directions: drift (3) and shear (5) are unholdable alone and holdable together (8). The minimal spin strengths are closed: g=√(μd|ks|)+√(μs|kd|)=0.8272 for mixed pairs and g=2√(μs|ks|)=0.5504 for the shear pair, about one tick-momentum. (iii) Any icosahedrally symmetric spin (including the Galois inversion σ) acts as exactly zero on the unstable sector; antipodally symmetric flag rotations cannot couple drift (odd parity) to shear (even); local flag rotations stabilise only at ∼ 60× the ideal cost. (iv) Breaking I→ T (two triangle spins lock a tetrahedron, two tetrahedra a cube — one of the five cubes in the dodecahedron) decomposes the shear H→ T⊕ E, so drift and shape become the same representation and may rotate into each other (x↔ yz, y↔ xz, z↔ xy; E symplectically with itself): this cube-locked observer holds the whole sector at g=0.8275, all frequencies purely imaginary (cycles only), and the spin does no work — locally time-asymmetric, globally balanced. (v) In four dimensions the vector family is even (4) and self-holding; the 24-cell locking leaves the 600-cell's time families irreducible and splits its present into squares. Holding is an act, not a potential. Part of the INDYNA Notes series (Notes 16–34). All numbers reproduce from the INDYNA law register (master data record, doi:10.5281/zenodo.21891088). Licence CC-BY-4.0.