The interchange tensorμ intertwines the braidings #
In a symmetric monoidal category the interchange morphism
tensorμ a b c d : (a ⊗ b) ⊗ (c ⊗ d) ⟶ (a ⊗ c) ⊗ (b ⊗ d) makes the
tensor product a braided functor: braiding the two tensor pairs and
then interchanging agrees with interchanging and then braiding
slotwise. Mathlib records only the diagonal special case
SymmetricCategory.tensorμ_braid_swap; this file proves the general
four-object statement.
The proof decomposes the block braiding β_ (a ⊗ b) (c ⊗ d) into the
four elementary crossings β_ b c, β_ b d, β_ a c, β_ a d via
the hexagon identities; the crossing β_ d a supplied by
tensorμ c d a b then cancels β_ a d by the symmetry axiom, and one
exchange of the disjoint crossings β_ a c and β_ b d produces the
right-hand side.
In a symmetric category the interchange tensorμ makes the tensor
product a braided functor: braiding the tensor pairs and then
interchanging equals interchanging and then braiding slotwise.
In a symmetric category the interchange tensorμ makes the tensor
product a braided functor: braiding the tensor pairs and then
interchanging equals interchanging and then braiding slotwise.
The tensor evaluation of ExactPairing.tensor is compatible with
the braidings: crossing the two dual pairs converts the arc nesting of
the double evaluation, inner (Q, Y)/outer (P, X) against inner
(P, X)/outer (Q, Y).
The tensor evaluation of ExactPairing.tensor is compatible with
the braidings: crossing the two dual pairs converts the arc nesting of
the double evaluation, inner (Q, Y)/outer (P, X) against inner
(P, X)/outer (Q, Y).