Equal areas and odd Monsky corner parity give a contradiction
ProvedProofsInTheBook.Chapter20.monsky_false_of_odd_corner_parityauxiliary-lemmageometrylean4monsky-theoremproofs-from-the-book
Let n be an odd natural number and let be any family indexed by . Suppose each triangle has area , with the quotient formed in the rationals and embedded in the reals. Color all corners by the fixed Monsky coloring from the chosen extension of the rational 2-adic valuation. Let r_i count, with multiplicity, the red–green pairs among . If is odd, these hypotheses imply a contradiction. There is no covering, disjointness, or unit-square hypothesis; the odd corner-parity condition is supplied explicitly.
Preamble
import Init import Mathlib import Definitions.Def_P2MAssembly_Chapter20 set_option autoImplicit true open ProofsInTheBook.Chapter20 open MonskyColor
Formal statement
theorem ProofsInTheBook.Chapter20.monsky_false_of_odd_corner_parity
{n : ℕ} (hn : Odd n) (tri : Fin n → (ℝ × ℝ) × (ℝ × ℝ) × (ℝ × ℝ))
(harea : ∀ i, realTriangleArea (tri i).1 (tri i).2.1 (tri i).2.2 =
(((1 : ℚ) / n : ℚ) : ℝ))
(hodd : Odd (∑ i : Fin n, triangleLocalRGCount
(realTwoAdicColor (tri i).1, realTwoAdicColor (tri i).2.1,
realTwoAdicColor (tri i).2.2))) :
False := by sorrySource
Original declaration: https://github.com/xiangyazi24/proof_in_the_book/blob/88d88d141768cded75e782c525ef1bf04b8fe220/ProofsInTheBook/Chapter20DissectionSperner.lean#L24. Repository topic: Monsky’s theorem, “One square and an odd number of triangles.”