Picard–Vessiot theory

Overview
Picard–Vessiot theory is a branch of differential algebra that provides a Galois-type correspondence for linear ordinary differential equations (ODEs) with coefficients in a differential field. It studies algebraic relations among the solutions of such equations by introducing Picard–Vessiot extensions—minimal differential field extensions generated by a fundamental set of solutions—and their associated linear algebraic groups, called differential Galois groups. The theory parallels classical (finite) Galois theory for polynomial equations, but operates in the context of differential fields rather than purely algebraic fields.

Historical background
The foundations of the theory were laid independently by Émile Picard (1883) and Ernest Vessiot (1904) in the late 19th and early 20th centuries. Their work was later formalized and extended by mathematicians such as Joseph J. Kolchin, who established the modern framework of differential algebra in the 1940s, and by R. M. C. Kovacic, M. F. Singer, and others who contributed to algorithmic aspects in the late 20th century.

Key definitions

Differential field.
A field $K$ equipped with a derivation $\partial: K \to K$ (a map satisfying $\partial(a+b)=\partial a+\partial b$ and $\partial(ab)=a\partial b + b\partial a$). The set of elements $c\in K$ with $\partial c = 0$ is called the field of constants, denoted $C_K$.

Linear differential equation.
An $n$th‑order linear ODE over $K$ can be written in operator form $$ L(y)=y^{(n)} + a_{n-1}y^{(n-1)} + \cdots + a_0 y = 0, $$ with coefficients $a_i\in K$.

Picard–Vessiot extension.
For a given linear differential equation $L(y)=0$, a Picard–Vessiot extension $L$ of the base differential field $K$ is a differential field satisfying:

  1. $L$ is generated over $K$ by a fundamental matrix of solutions $Y$ (i.e., an $n\times n$ matrix whose columns form a basis of the solution space).
  2. The field of constants of $L$ coincides with that of $K$ ($C_L = C_K$).
  3. $L$ has no new constants and is minimal with these properties.

Differential Galois group.
The differential Galois group $\operatorname{Gal}(L/K)$ consists of all differential field automorphisms of $L$ that fix $K$ pointwise. Via its action on a fundamental matrix $Y$, each automorphism corresponds to an invertible constant matrix $g\in \mathrm{GL}_n(C_K)$ such that $\sigma(Y)=Yg$. Consequently, $\operatorname{Gal}(L/K)$ is identified with a linear algebraic subgroup of $\mathrm{GL}_n(C_K)$.

Fundamental results

  1. Galois correspondence.
    There is a one‑to‑one inclusion‑reversing correspondence between intermediate differential fields $K \subseteq F \subseteq L$ and Zariski‑closed subgroups $H$ of $\operatorname{Gal}(L/K)$: $$ F \longleftrightarrow \operatorname{Gal}(L/F)={ \sigma\in\operatorname{Gal}(L/K) \mid \sigma|_{F}= \mathrm{id}}. $$

  2. Solvability by quadratures.
    A linear differential equation is solvable in terms of exponentials, integrals, and algebraic functions (i.e., by quadratures) precisely when its differential Galois group is a solvable linear algebraic group.

  3. Ritt–Kolchin theorem.
    Every Picard–Vessiot extension is finitely generated as a differential field, and its differential Galois group is an algebraic group of finite type over the constant field.

  4. Liouvillian solutions.
    The existence of Liouvillian (i.e., built from exponentials, integrals, and algebraic functions) solutions can be decided algorithmically via the structure of the differential Galois group (Kovacic’s algorithm for second‑order equations).

Applications

  • Transcendence theory: Results on the algebraic independence of special functions (e.g., exponential, Bessel, and hypergeometric functions) often rely on Picard–Vessiot theory.
  • Integrability of dynamical systems: The differential Galois group provides obstructions to the existence of first integrals or analytic solutions for Hamiltonian systems (the Morales‑Ramis theory).
  • Symbolic computation: Algorithms for solving linear ODEs, factorizing differential operators, and determining Liouvillian solutions are grounded in Picard–Vessiot theory.
  • Number theory: Connections with the theory of motives and periods arise through differential equations satisfied by periods of algebraic varieties.

References (selected)

  • E. Picard, Sur les équations différentielles linéaires du second ordre, J. Math. Pures Appl. 1883.
  • E. Vessiot, Sur la théorie générique des intégrales de certaines équations différentielles, C. R. Acad. Sci. Paris 1904.
  • J. F. Ritt, Differential Algebra, American Mathematical Society, 1950.
  • J. Kolchin, Differential Algebra and Algebraic Groups, Academic Press, 1973.
  • M. F. Singer & M. van der Put, Galois Theory of Linear Differential Equations, Springer, 2003.
  • J. J. Kovacic, An Algorithm for Solving Second Order Linear Homogeneous Differential Equations, J. Symbolic Computation 1986.
Browse

More topics to explore