Finite order growth of the pole-subtracted L-function
OpenLL_Decomp530a.leaf_regularPart_finite_order_growthLet , let and let be a Dirichlet character modulo . Then there exist real constants and such that for every complex with one has , where is with the pole of the principal character subtracted.
import Mathlib import Definitions.Def_LL_Decomp530a_regularPart open Classical Complex Set Filter ZMod open scoped Topology
namespace LL_Decomp530a
theorem leaf_regularPart_finite_order_growth {δ : ℝ} (hδ : 0 < δ) (hδhalf : δ < 1 / 2)
{q : ℕ} [NeZero q] (χ : DirichletCharacter ℂ q) :
∃ C c : ℝ, 0 < C ∧ 0 < c ∧ ∀ s : ℂ,
-δ ≤ s.re → s.re ≤ 1 + δ →
‖regularPart χ s‖ ≤ C * Real.exp (|s.im| ^ c) := by sorry
end LL_Decomp530aRead-back
What the Lean code literally says, in plain math · claude-haiku
Read-back: LL_Decomp530a.leaf_regularPart_finite_order_growth
Statement. Let be a real number with . Let be a natural number with (the Lean instance NeZero q), and let be a Dirichlet character modulo with values in . Then there exist real numbers and such that , , and the following holds for every complex number : whenever , the modulus is at most .
Binders and quantifier order. The real parameter and the modulus (with its nonzero instance) and the character are universally quantified first, with constrained by and . The constants and are then existentially chosen, so they may depend on , and , but not on . The bound is then universal over all in the vertical strip . In symbols, the shape is (strip condition implies bound). Because the strip condition is an implication, points outside the strip are unconstrained.
Conclusion. The conclusion is a uniform growth bound in the imaginary direction: on the strip, grows at most like as . Here is a real power of a non-negative real number, and since the value at is , so the bound there reads . The exponent is not fixed in the statement; the claim is that some positive exponent works.
Definitions not visible in this file. The item imports Definitions.Def_LL_Decomp530a_regularPart, which defines regularPart. That file was not read, so I cannot state what regularPart χ s equals. The reading above treats it as a complex-valued function of depending on . Whether it is the intended "regular part" of a Dirichlet -function (for example, with a pole or Euler-type factor removed) cannot be checked from this item alone. The statement's content depends entirely on that definition.
Edge cases and vacuity.
- No hypothesis is contradictory: for example , (the trivial character, which is allowed by
NeZero) satisfies every hypothesis, so the statement is not vacuous on its hypotheses. - Because the existential is over as well as , a larger gives a weaker bound, so the claim is equivalent to the existence of some exponent with the exponential-of-power bound. The chosen form is weaker than polynomial growth and does not give a polynomial bound directly.
- All norms are the complex modulus and all exponentials are real, so there are no junk-value issues from total functions in the statement itself.
- The strip bounds are inclusive on both ends.
- The proof is
sorry, so the statement is currently unproven in this file. Nothing in the statement itself makes it trivially true, but whether it is trivial depends onregularPart, which is not visible here.
Summary of concerns. The statement is not vacuous as written and its quantifiers and inequalities are precise. The main open point is the unread definition of regularPart, without which the mathematical meaning of the conclusion cannot be confirmed. The item is also unproven (sorry).
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.