Joint likelihood for correlated observations O-34, O-2 (shared basis: S-66, S-80). Step 8 writes the single ## Likelihood block here; the member edges point back via group.
Likelihood
# CG-1 (HC-4) — ONE joint estimate over E-40 + E-41: both observations trace to the same pulsar-timing
# basis S-66/S-80 (rule 1). Members in HC-4.hypotheses order: [H-8 catastrophic, H-22 negligible over the
# Sun's remaining lifetime, H-36 residual intermediate]. The whole pattern judged jointly: a POPULATION of
# neutron stars survives intact for >1e9 yr at supranuclear density (O-34), and ONE concretely dated
# instance — PSR J1614-2230, ~2 Msun, characteristic age 5.2 Gyr, held at ~2-5x nuclear saturation for that
# span (O-2). Anchored on H-22 = 1 (rule 7); H-8 and H-36 priced as ratios to it.
lik_ns_H22 = 1.0 # anchor: old gigayear neutron stars are exactly what H-22 predicts — a captured hole
# accretes too slowly to matter, and their survival is the observation H-22 expects.
lik_ns_H8 = 0.12 # 0.12x the anchor: under catastrophic fast-accretion physics a hole captured in the
# densest matter accretes FASTEST of all — A-52 (approved) puts consumption at <~10 Myr
# (D=11 worst case; <~50 yr for D<=7) << the Gyr ages, so a whole Gyr-old population plus
# a 5.2-Gyr instance is strongly improbable. Kept well off zero, NOT via the prior's
# A-7/A-8 gaps (already counted), but by A-52's OWN internalized escape hatch: the bound
# fails only if cosmic rays are almost purely heavy AND the cosmogenic-neutrino flux is
# absent/non-reactive — two independent unlikely failures at once, so residual-small but
# nonzero. A-53 (approved) removes the one out that would raise it: no Eddington throttle
# lengthens the neutron-star destruction time, so H-8 cannot cheaply explain survival.
lik_ns_H36 = 0.5 # 0.5x: unlisted intermediate outcome, unconstrained (rule 3) — an accretion regime slow
# enough to spare Gyr neutron stars yet non-negligible for Earth can accommodate the
# pattern, but sub-gigayear intermediate regimes (E-40 body: "likewise disfavored") would
# consume such dense stars, so middling, below the anchor. Higher would assert the middle
# predicts old-NS survival as well as H-22; lower would deny any middle could produce it.
t_ns = 0.82 # cap = min trust_score over the named observations' sources: O-34 via S-37 = 0.82,
# O-2 via S-66 = 0.90 -> min 0.82. At the cap: the raw datum (neutron stars survive Gyr
# at ~2 Msun) is a clean, independently re-timed Shapiro-delay / spectroscopic measurement;
# S-37's 0.82 already books the model-dependence of the survival chain, so no further dock.
evidence("HC-4", ["O-34", "O-2"], [lik_ns_H8, lik_ns_H22, lik_ns_H36], t=t_ns)