Definition

Let MM be a closed with a and complex SS. Its canonical spin spectral triple is

(C(M),L2(M,S),),\bigl(C^\infty(M),\,L^2(M,S),\,\not D\bigr),

where C(M)C^\infty(M) acts faithfully by pointwise multiplication and \not D is the self-adjoint closure of the . The operator \not D has compact resolvent, and for every fC(M)f\in C^\infty(M),

[,f]=c(df)[\not D,f]=c(df)

extends to a bounded operator. These facts make the displayed data a .

Why the spectral-triple axioms hold

Ellipticity on the compact manifold implies that the resolvent of \not D is compact. The first-order Leibniz rule identifies its commutator with multiplication by the Clifford field c(df)c(df), which is bounded because MM is compact. Smooth functions are used rather than all of C(M)C(M): a merely continuous function need not preserve the domain of \not D or have a .

Parity and additional structures

If dimM\dim M is even, chirality grades L2(M,S)L^2(M,S), commutes with functions, and anticommutes with \not D, producing an . In odd dimension the canonical triple is ungraded. can supply a real structure, but that operator and its dimension-dependent signs are additional data, not part of the three-component spectral triple defined in the core.

Geometry recovered from the triple

The growth of the eigenvalues of |\not D| records the dimension through Weyl asymptotics. Connes's distance formula recovers the Riemannian geodesic distance from

d(p,q)=sup{f(p)f(q):[,f]1}.d(p,q)=\sup\{|f(p)-f(q)|:\lVert[\not D,f]\rVert\leq1\}.

The reconstruction theorem requires further regularity, finiteness, orientability, and reality hypotheses; compact resolvent and bounded commutators alone do not characterize canonical manifold triples.

References
  1. Alain Connes, Noncommutative Geometry, Academic Press, 1994. Author-hosted book. Relevant: Part VI.1 on the spectral geometry of compact spin manifolds.
  2. José M. Gracia-Bondía, Joseph C. Várilly, and Héctor Figueroa, Elements of Noncommutative Geometry, Birkhäuser, 2001. DOI record. Relevant: §10.2 on the canonical commutative spectral triple.