IndisputableMonolith.Numerics.Interval.AlphaBounds
Module Numerics.Interval.AlphaBounds supplies machine-verified interval bounds on the fine-structure constant via the seed quantity 4 pi * 11. Scorecard and prediction modules for electron g-2, masses, CKM angles, and Hartree-Rydberg quantities import these bounds to certify agreement with measured intervals. The module consists of Taylor-expansion lemmas with explicit remainder controls together with direct comparisons that establish the seed inequality and related exponential bounds at points near 0.48.
claim$\alpha_{\rm seed} = 4\pi\cdot 11 > 138.230048$, with companion upper and lower bounds obtained from Taylor remainders of $\exp$ at arguments 0.48, 0.481, 0.483.
background
The module imports W8Bounds, whose central object is the closed-form eight-tick weight $w_8 = (348 + 210\sqrt{2} - (204 + 130\sqrt{2})\phi)/7 \approx 2.490569$, and Constants.Alpha. These supply the numerical constants needed to convert the Recognition Science phi-ladder and J-cost into concrete interval statements for $\alpha^{-1}$. The local setting is the numerics layer that turns the abstract T5 J-uniqueness and T7 eight-tick octave into machine-checked inequalities lying inside the target band (137.030, 137.039).
proof idea
The module is a collection of definition and lemma blocks. Each block introduces a Taylor polynomial of degree 10 for exp together with an explicit error term, then applies interval arithmetic to obtain strict inequalities at the three evaluation points 0.48, 0.481, 0.483. The seed comparison alpha_seed_gt is obtained by direct arithmetic on the imported constants.
why it matters in Recognition Science
The bounds are imported by HartreeRydbergScoreCard (P1-C02, P1-C03, P1-C04), Masses.NumericalPredictions, CKMGeometry, ElectronGMinus2ScoreCard (P1-C05), and ElectronMass.Necessity. They close the numerical verification step that links the eight-tick octave and phi-ladder to the experimental alpha band required by the downstream physical derivations.
scope and limits
- Does not derive the value of alpha from the functional equation; only certifies interval membership.
- Does not contain floating-point computations; all statements are formal inequalities.
- Does not address higher-order QED corrections beyond the leading Schwinger term.
- Does not replace the abstract T5 J-uniqueness; supplies only its numerical consequences.
used by (5)
depends on (2)
declarations in this module (31)
-
theorem
alpha_seed_gt -
theorem
alpha_seed_lt -
def
exp_taylor_10_at_048 -
def
exp_error_10_at_048 -
lemma
exp_048_taylor_ceiling -
lemma
exp_048_lt -
def
exp_taylor_10_at_0481 -
def
exp_error_10_at_0481 -
lemma
exp_0481_taylor_ceiling -
lemma
exp_0481_lt -
def
exp_taylor_10_at_0483 -
def
exp_error_10_at_0483 -
lemma
exp_0483_taylor_floor -
lemma
exp_0483_gt -
lemma
log_phi_gt_048 -
lemma
log_phi_gt_0481 -
lemma
log_phi_lt_0483 -
theorem
f_gap_gt -
theorem
f_gap_gt_strong -
theorem
f_gap_lt -
def
exp_taylor_10_at_neg_00871 -
def
exp_error_10_at_neg_00871 -
lemma
exp_neg_00871_taylor_floor -
lemma
exp_neg_00871_gt -
def
exp_taylor_10_at_neg_00866 -
def
exp_error_10_at_neg_00866 -
lemma
exp_neg_00866_taylor_ceiling -
lemma
exp_neg_00866_lt -
theorem
alphaInv_gt -
theorem
alphaInv_lt -
theorem
alphaInv_lt_strong