Canonical: https://knowlpedia.com/analysis/separated-support-products/

# Vanishing products from separated auxiliary supports

Closed support separation makes all differentiated cross products vanish, including after evaluation along a smooth auxiliary map.

Let \(F_\gamma(x,Y)\) be globally [smooth functions](https://knowlpedia.com/real-analysis/class-ck-map/) with
\[
\operatorname{supp}F_\gamma\subset K_\gamma\times R_\gamma.
\]
Assume that, for distinct labels, \(K_\gamma\cap K_{\gamma'}\ne\varnothing\) implies \(R_\gamma\cap R_{\gamma'}=\varnothing\). Then every pair of mixed derivatives satisfies
\[
(\partial^\alpha F_\gamma)(\partial^\beta F_{\gamma'})=0
\qquad(\gamma\ne\gamma').
\]

## Why derivatives and evaluation preserve the conclusion

A derivative of a smooth function has [support](https://knowlpedia.com/shared-foundations/support-of-a-function/) contained in the support of the function. The assumed product supports are disjoint, proving the identity. After [evaluation](https://knowlpedia.com/real-analysis/auxiliary-variable-evaluation/) at \(Y=\Psi(x)\), the chain rule expresses each derivative using these same mixed derivatives, so cross products still vanish.

For a finite or locally finite sum this removes all cross-label terms from a quadratic differential expression. It says nothing about different harmonics with the same label. Smooth extension across the support boundaries is essential: differentiating a discontinuous cutoff can create boundary distributions, outside the hypotheses here.
