IndisputableMonolith.Numerics.Interval.PhiBounds
Certified rational endpoint bounds on √5 and on the golden ratio φ = (1+√5)/2. Mass, CKM, neutrino, and interval log/power modules import these inequalities whenever φ-ladder arithmetic must stay inside Lean’s exact rational kernel. Proofs are elementary square comparisons of rationals against 5, then linear transport to φ.
claimStrict rational bounds $a^2 < 5 < b^2$ for explicit positive rationals $a,b$ (e.g. $2.236^2 < 5 < 2.237^2$ and multi-digit refinements), hence $a < \sqrt{5} < b$, and the induced bounds on the golden ratio $\varphi = (1+\sqrt{5})/2$ such as $1.618 < \varphi < 1.6185$ and tighter digit strings.
background
Recognition Science forces φ as the unique self-similar fixed point (forcing step T6). Downstream formulas (mass ladder, electroweak rungs, CKM geometry, observability cutoffs) therefore need concrete numerical control of φ and of √5 without floating-point gaps.
The parent numerics layer (Interval.Basic) supplies verified interval arithmetic whose endpoints are rationals Lean can decide exactly. This module specializes that pattern to the golden-ratio constants imported from Mathlib: compare squares of explicit rationals to the integer 5, obtain strict bounds on √5, then push them through the affine map $x \mapsto (1+x)/2$ to bound φ.
Sibling lemmas come in coarse and fine packs (four-digit and eight-digit style endpoints), so callers can pick the precision they need.
proof idea
Each bound is a short chain. First prove rational square inequalities such as $(2236/1000)^2 < 5$ and $5 < (2237/1000)^2$ by exact integer arithmetic on cleared denominators. Monotonicity of square root on positives yields $2236/1000 < \sqrt{5} < 2237/1000$. Adding 1 and dividing by 2 transports the same endpoints to φ. Tighter packs repeat the pattern with longer decimal rationals. No transcendental analysis is required beyond Mathlib’s √ and goldenRatio definitions.
why it matters in Recognition Science
φ appears in the RS-native constants ($\hbar = \varphi^{-5}$, $G = \varphi^5/\pi$) and in the mass formula yardstick · φ^(rung−8+gap(Z)). This module is the shared numerical substrate for those claims.
Importers include Masses.ElectroweakMasses (Z at rung 1, m_Z ~ φ^51), Masses.Verification (PDG comparison), Physics.CKMGeometry (T11 mixing angles), Physics.ElectronMass.Necessity, Physics.NeutrinoSector, and Astrophysics.ObservabilityLimits (recognition-length cutoffs). Interval modules Log and Pow also import it so exp/log/power enclosures can start from certified φ endpoints rather than ad-hoc floats.
Without these lemmas, every φ-ladder inequality would re-prove the same rational squares.
scope and limits
- Does not derive φ from the forcing chain; only bounds Mathlib’s goldenRatio.
- Does not prove uniqueness or self-similarity of φ (that is T5–T6).
- Does not supply interval enclosures for arbitrary reals, only √5 and φ.
- Does not claim decimal strings are optimal; only that the stated rationals work.
- Does not connect bounds to experimental PDG values (Verification stays quarantined).
used by (12)
-
IndisputableMonolith.Astrophysics.ObservabilityLimits -
IndisputableMonolith.Masses.ElectroweakMasses -
IndisputableMonolith.Masses.Verification -
IndisputableMonolith.Numerics.Interval.Log -
IndisputableMonolith.Numerics.Interval.Pow -
IndisputableMonolith.Physics.CKMGeometry -
IndisputableMonolith.Physics.ElectronMass.Necessity -
IndisputableMonolith.Physics.NeutrinoSector -
IndisputableMonolith.Physics.QuarkMasses -
IndisputableMonolith.Physics.WBosonAbsoluteScoreCard -
IndisputableMonolith.StandardModel.NeutrinoMassHierarchy -
IndisputableMonolith.Verification.MassComparison
depends on (1)
declarations in this module (98)
-
theorem
sq_2236_lt_5 -
theorem
five_lt_sq_2237 -
theorem
sqrt5_gt_2236 -
theorem
sqrt5_lt_2237 -
theorem
phi_gt_1618 -
theorem
phi_lt_16185 -
theorem
sq_22360679_lt_5 -
theorem
five_lt_sq_22360680 -
theorem
sqrt5_gt_22360679 -
theorem
sqrt5_lt_22360680 -
theorem
phi_gt_161803395 -
theorem
phi_lt_16180340 -
theorem
phi_tight_bounds -
def
phiIntervalTight -
theorem
phi_in_phiIntervalTight -
def
phi_quarter_lo -
def
phi_quarter_hi -
lemma
phi_quarter_lo_pow4_lt_phi_lo -
lemma
phi_hi_lt_phi_quarter_hi_pow4 -
theorem
phi_quarter_gt -
theorem
phi_quarter_lt -
theorem
phi_quarter_bounds -
theorem
phi_neg_quarter_bounds -
theorem
phi_sq_gt -
theorem
phi_sq_lt -
theorem
phi_neg2_gt -
theorem
phi_neg2_lt -
theorem
phi_inv_eq -
theorem
phi_inv_gt -
theorem
phi_inv_lt -
def
phi_inv_interval_proven -
theorem
phi_inv_in_interval_proven -
theorem
phi_cubed_eq -
theorem
phi_cubed_gt -
theorem
phi_cubed_lt -
theorem
phi_pow4_eq -
theorem
phi_pow4_gt -
theorem
phi_pow4_lt -
theorem
phi_pow5_eq -
theorem
phi_pow5_gt -
theorem
phi_pow5_lt -
theorem
phi_pow6_eq -
theorem
phi_pow7_eq -
theorem
phi_pow8_eq -
theorem
phi_pow8_gt -
theorem
phi_pow8_lt -
def
phi_pow8_interval_proven -
theorem
phi_pow8_in_interval_proven -
theorem
phi_inv2_gt -
theorem
phi_inv2_lt -
theorem
phi_inv3_gt -
theorem
phi_inv3_lt -
def
phi_inv3_interval_proven -
theorem
phi_inv3_in_interval_proven -
theorem
phi_inv3_zpow_bounds -
theorem
phi_inv5_gt -
theorem
phi_inv5_lt -
def
phi_inv5_interval_proven -
theorem
phi_inv5_in_interval_proven -
theorem
phi_pow16_eq -
theorem
phi_pow16_gt -
theorem
phi_pow16_lt -
theorem
phi_pow51_eq -
theorem
phi_pow51_gt -
theorem
phi_pow51_lt -
def
phi_pow51_interval_proven -
theorem
phi_pow51_in_interval_proven -
theorem
phi_pow54_eq -
theorem
phi_pow54_gt -
theorem
phi_pow54_lt -
theorem
phi_neg54_gt -
theorem
phi_neg54_lt -
theorem
phi_pow58_eq -
theorem
phi_pow58_gt -
theorem
phi_pow58_lt -
theorem
phi_neg58_gt -
theorem
phi_neg58_lt -
lemma
qhi_pos -
lemma
qlo_pos -
theorem
phi_neg2174_gt