LeanProofs.Rosetta.CliffordInteractionObstruction
DefinitionGrade4_LeanProofs_Rosetta_CliffordInteractionObstructionclifford-algebraleanproofs-grade4rosetta
Foundational definitions and structural lemmas from LeanProofs.Rosetta.CliffordInteractionObstruction, required by the selected LeanProofs theorem graph.
Definition code
import Definitions.Def_Grade4_LeanProofs_Rosetta_ObstructionToDescent
import Definitions.Def_Grade4_LeanProofs_Foundations_InnerProduct
import Definitions.Def_Grade4_LeanProofs_Foundations_Pseudoscalar
import Definitions.Def_Grade4_LeanProofs_Foundations_BivectorIndependence
import Definitions.Def_Grade4_LeanProofs_Foundations_VacuumState
import Mathlib.Tactic
namespace LeanProofsGrade4P2M
/-!
# G3X: interaction-factorization obstruction
This module asks which information a norm-only observation of two typed
Clifford inputs forgets about their interaction. It proves two independent
obstructions:
* equal self-norm data can hide a different bivector commutator;
* even when the commutator vanishes, equal self-norm data can hide a different
pseudoscalar product.
The globally sufficient repair is the Jordan--Lie encoding
`(AB + BA, AB - BA)`, which is linearly equivalent to retaining both ordered
products `(AB, BA)`. Transporting those two coordinates through the existing
Clifford/exterior linear equivalence gives an equally sufficient alternative
encoding. Thus the result concerns sufficient interaction structure, not a
unique Clifford representation.
The norm observation here is the algebra-valued reverse product. A scalar
projection corollary is also proved. Neither is definitionally identified
with the Killing form; that requires a separate normalization theorem.
These are algebraic outcome-factorization results. They do not establish a
dynamical descent, stochastic lumpability, all-horizon prediction, causal
feedback, a physical field equation, a cascade, or a preferred orientation.
-/
namespace Rosetta.CliffordInteractionObstruction
open BivectorProofs
open Rosetta.ObstructionToDescent
noncomputable section
/-! ## Representation-neutral interaction coordinates -/
section AlgebraicEncoding
variable {A : Type*} [Ring A] [Algebra ℝ A]
/-- A typed pair of interaction inputs. -/
abbrev AlgebraInteractionState := A × A
/-- Both orders of the product, retained as the complete registered outcome. -/
def algebraOrderedProducts (state : AlgebraInteractionState (A := A)) : A × A :=
(state.1 * state.2, state.2 * state.1)
end AlgebraicEncoding
/-! ## Registered STA observation and outcomes -/
abbrev InteractionState := STA × STA
/-- The complete registered outcome retaining both product orders. -/
def orderedProducts : InteractionState → STA × STA :=
algebraOrderedProducts
/-! ## Lie/grade-2 obstruction -/
/-! ## Commuting pseudoscalar/product obstruction -/
def hodgeLeft : InteractionState := (γ01, γ23)
def hodgeRight : InteractionState := (γ01, -γ23)
theorem pseudoscalar_ne_zero : pseudoscalar ≠ 0 := by
intro hzero
have hsquare := pseudoscalar_sq
rw [hzero, mul_zero] at hsquare
exact one_ne_zero (neg_eq_zero.mp hsquare.symm)
theorem pseudoscalar_ne_neg : pseudoscalar ≠ -pseudoscalar := by
intro hsame
have hzero : (2 : ℝ) • pseudoscalar = 0 := by
calc
(2 : ℝ) • pseudoscalar = pseudoscalar - (-pseudoscalar) := by module
_ = 0 := sub_eq_zero.mpr hsame
rcases smul_eq_zero.mp hzero with htwo | hpseudo
· norm_num at htwo
· exact pseudoscalar_ne_zero hpseudo
theorem hodgeWitness_different_orderedProducts :
orderedProducts hodgeLeft ≠ orderedProducts hodgeRight := by
intro hsame
apply pseudoscalar_ne_neg
have hfirst := congrArg Prod.fst hsame
simpa [orderedProducts, algebraOrderedProducts, hodgeLeft, hodgeRight,
pseudoscalar] using hfirst
/-! ## Global repair and universal fiber statement -/
/-! ## Exterior-algebra alternative encoding -/
/-! ## Compression and frozen controls -/
end
end Rosetta.CliffordInteractionObstruction
end LeanProofsGrade4P2M
Source