Repository navigation
Companion note, version 1.39
tex/diaz_prove2me.tex, 44 pages. The PDF is attached as diaz_prove2me-v1.39.pdf.
In this version: the multiplicity estimate (ME) is machine-checked, so Roy's strong six exponentials theorem is
machine-checked with no hypothesis.
Since v1.36 Roy's theorem (Roy 1992, Cor. 2) is machine-checked from one explicit hypothesis, (ME): the case G⁻ = {e},
G⁺ = G of the multiplicity estimate of DALAG Th. 8.1, for G = 𝔾ₐ^{A₀} × 𝔾ₘ^{A₁}, written in coordinates. Since v1.38
its Hilbert side holds over ℂ. Thirty-two results of 30 October 2026 prove its multiplicity side and assemble it
(DiazModulus.multiplicityEstimateElem), and with it Roy's theorem holds with no hypothesis
(DiazModulus.strong_six_exponentials_unconditional).
- What is proved: (ME), by the proof of DALAG §8.4 (chapter 8 is by D. Roy) at finite level, through elementary
arguments: zero sets and heights on the torus, characters of its subgroups, its invariant derivations, the tangent
duals of DALAG Lemmas 8.14–8.16 with Wüstholz's argument, and the descent of §8.4.1 to an obstruction subgroup. For
G⁻ = {e} the estimate is a case of Philippon 1986, Th. 2.1. - What follows: every result of the note that was machine-checked with Roy's theorem as a hypothesis now holds
formally without it, among them all of §2 of Diaz 2007 and the case of Waldschmidt's question in Section 2. The
library proves Roy's theorem asDiaz.strongSixExponentialsand restates each of the 25 results that took it as a
hypothesis without it, inDiaz/RoyUnconditional.lean. The passages of Section 1 and elsewhere that called
(ME) or Roy's theorem unproved now say what is proved. - What is not proved: Th. 8.1 of DALAG in general; only the case that Roy's theorem uses.
- These are classical results; this is a routine formalisation of their known proofs.
Appendix A. Two more rows with one identifier each; 226 identifiers in all. The mirror holds 596 results, all
building in CI on Lean's three standard axioms.