O-22 - Earth, Sun, white dwarfs, neutron stars and the stars at large have survived billions of years of cosmic-ray bombardment
This is the survival half of the cosmic-ray safety argument: the exposure count is a measurement (see the flux observation), while this node records the uncontested background fact that the exposed bodies persist. The white-dwarf and neutron-star part carries the load for the hypothetical stable-neutral-black-hole case: their large observed populations at Gyr ages are what the Giddings-Mangano accretion argument turns on, since dense stars stop even neutral black holes and would be destroyed on short timescales, ending in highly visible explosive events that are not observed.
Link to original
Why this is evidence
Prompt-decay members (H-10 canonical Hawking, H-14 near-threshold states decaying on microscopic timescales) predict this survival with certainty. H-24 (stable neutral holes) does not: cosmic-ray collisions on dense stars produce the same objects, white dwarfs and neutron stars stop even neutral holes, and stopped stable holes would consume their hosts in highly visible events on timescales short against the observed Gyr ages — so the persistence of large white-dwarf and neutron-star populations is unexpected under H-24 conditional on production. The long-lifetime tail of H-7 (metastability up to years or beyond) is strained the same way, though its shorter-lived range is untouched. Stable variants inside the residual H-35 are disfavored likewise.
Likelihood
# E-33 (HC-3) — lone edge, one observation O-22 "broad astronomical survival": Earth (4.5 Gyr), the Sun, ~1e22
# stars incl. white dwarfs and neutron stars persist through Gyr of cosmic-ray exposure, and conventional
# astrophysics accounts for all detected holes. This is the COARSE, qualitative survival witness — LSAG's
# restatement of Giddings-Mangano (A-30 approved). Its real quantitative teeth against H-24 are the WD/NS
# destruction chain, which E-34/E-35 price precisely off G-M; to avoid over-counting the same physics three
# times this edge is kept milder than those. Prompt-decay members predict survival with certainty; anchor
# H-10 = 1 (rule 7). Members in HC-3.hypotheses order [H-7, H-10, H-14, H-24, H-35].
lik_surv_H7 = 0.85 # metastable: its ms-to-years bulk decays before doing astrophysical damage, so survival
# is nearly expected — only the long-lifetime tail (holes brought to rest, then accreting)
# is strained (A-30). Mild ratio just below the anchor
lik_surv_H10 = 1.0 # anchor: canonical prompt Hawking decay — objects gone in ~1e-27 s, survival certain
lik_surv_H14 = 1.0 # near-threshold stringy state also decays promptly and is astrophysically safe — survival
# certain, indistinguishable from H-10 on this observation
lik_surv_H24 = 0.30 # stable neutral: cosmic-ray-produced neutral holes on dense stars would destroy them in
# highly visible events on <<Gyr timescales (A-29, A-30), so broad survival is unexpected.
# Not near zero — this qualitative statement's force IS the WD/NS quantitative chain priced
# by E-34/E-35, and D-range escape hatches (A-49 D<=7 vs A-52 D>=8) remain open; kept coarse
# here to price only this edge's own increment above the sharp compact-star edges
lik_surv_H35 = 0.60 # residual: unconstrained — its stable-accreting variants are cut like H-24, but most
# unmodelled fates (Planck remnant, stalled evaporation, prompt-like decay) survive fine.
# Middling, tilted toward survival-compatible (rule 3)
t_surv = 0.82 # cap = trust_score(S-1 LSAG) = 0.82; held at cap — the persistence of Earth/Sun/stars is
# an uncontested, generally-known background fact with no data-pipeline weakness. The
# stopping/production inferential chain enters as reasoning, not as a trust dock (rule 5)
evidence("HC-3", ["O-22"], [lik_surv_H7, lik_surv_H10, lik_surv_H14, lik_surv_H24, lik_surv_H35], t=t_surv)