Sard's theorem

Sard's theorem
Name	Sard's theorem
Field	Analysis, Differential Topology
First proved	1942
Proved by	Arthur Sard
Related	Morse theory, Transversality, Implicit function theorem

Sard's theorem is a result in mathematical analysis and differential topology asserting that the set of critical values of a smooth map between manifolds has measure zero. The theorem connects ideas from calculus on Euclidean space, global analysis, and topology, and plays a central role in the study of smooth maps, Morse theory, and transversality. It provides a bridge between local differential properties and global measure-theoretic behavior for maps studied in the traditions of classical analysis and modern topology.

Statement

Sard's theorem concerns a smooth map f: M -> N between smooth manifolds M and N of dimensions m and n respectively; it states that the set of critical values f(Crit(f)) has Lebesgue measure zero in N under appropriate differentiability hypotheses. The usual precise formulation requires f to be C^k with k > max(0, m - n), which ensures that points where the differential Df fails to be surjective (critical points) map to a negligible set of values. This statement refines classical theorems such as the inverse function theorem and the implicit function theorem, and it underpins results in the frameworks of Morse theory, transversality theory, and singularity theory.

History and context

Arthur Sard proved the theorem in 1942 while working on problems related to real analysis and differential mappings, contributing to a lineage that includes Élie Cartan, Hassler Whitney, and Marston Morse. The theorem arose in the milieu of early 20th-century developments in topology and analysis that involved figures such as Henri Lebesgue, Norbert Wiener, and John von Neumann. Subsequent advances and clarifications involved contributions from Solomon Lefschetz, René Thom, Stephen Smale, and William Thurston, connecting Sard's result to ideas developed by Jean Leray, Henri Poincaré, and Henri Cartan. The theorem influenced the formalization of transversality in the work of René Thom and Stephen Smale and plays a role in global analysis alongside figures like Raoul Bott and Michael Atiyah.

Proofs and techniques

Proofs of Sard's theorem employ techniques from measure theory, differential calculus, and geometric analysis. Classical proofs use covering lemmas inspired by Henri Lebesgue and Vitali, partition of unity arguments familiar from Hassler Whitney and Laurent Schwartz, and estimates based on Taylor expansions originating in Augustin-Louis Cauchy and Bernhard Riemann. Alternative approaches draw on microlocal analysis and Sobolev space methods developed by Lars Hörmander, Jean-Michel Bony, and Elias Stein, while modern expositions exploit transversality methods associated with René Thom and Stephen Smale and variational techniques influenced by Marston Morse. Harmonic analysis perspectives invoke maximal function estimates and Littlewood–Paley theory related to Alberto Calderón and Antoni Zygmund. Each proof balances local differential approximations with global measure estimates.

Generalizations and variants

There are numerous extensions of Sard's theorem in the literature. Whitney's extension theorems and Whitney stratification, due to Hassler Whitney, provide stratified versions compatible with singular spaces studied by René Thom and John Mather. Thom's transversality theorem generalizes Sard's measure-zero conclusion to parameterized families of maps, with crucial contributions by Mather in stability and singularity theory. Sard–Smale theorems extend the result to infinite-dimensional settings such as Banach and Fréchet manifolds, developed by Stephen Smale and applied in global analysis tied to Michael Atiyah, Isadore Singer, and Raoul Bott. There are also rectifiable and fractal variants connected to geometric measure theory advanced by Herbert Federer and Kenneth Falconer, and quantitative versions linking to entropy methods used by Yakov Sinai and David Ruelle.

Applications

Sard's theorem underlies many fundamental applications. In Morse theory, developed by Marston Morse and refined by John Milnor, the theorem ensures regular values exist abundantly for Morse functions, enabling computation of homology groups and attachment of handles in the spirit of René Thom and Stephen Smale. In differential topology, it guarantees generic transversality used in results of Hassler Whitney and René Thom and in proofs of existence theorems by Mikhail Gromov and Michael Freedman. The theorem is instrumental in singularity theory studied by Vladimir Arnold and John Mather, in bifurcation theory influenced by Henri Poincaré and Andronov, and in global analysis and index theory as in the work of Atiyah and Singer. It also informs geometric measure theory and PDE regularity results pursued by Elias Stein and Luis Caffarelli.

Examples and counterexamples

Classical examples illustrating Sard's conclusion include smooth submersions such as projections from Euclidean space studied by Carl Friedrich Gauss and Adrien-Marie Legendre, and Morse functions on compact manifolds featured in Milnor's examples. Counterexamples show the necessity of differentiability hypotheses: functions of class C^k with insufficient k can have sets of critical values with positive measure, as constructed in the style of Henry Lebesgue and later sharpened by examples related to Whitney and John Nash. Infinite-dimensional counterexamples arise in naive attempts to extend Sard's theorem to Fréchet manifolds without Smale-type hypotheses; these prompted the Sard–Smale formulation by Stephen Smale and contemporary work by Andreas Floer and Kris Witten in symplectic topology and gauge theory.

Category:Theorems in differential topology